Project Description:
1. INTRODUCTION
1-1. Overview of Platform
This project is an AI-powered, full process intelligent platform dedicated to smart grid resilience enhancement. Focusing on core challenges including extreme natural disasters, sudden component failures, and renewable energy output fluctuations, it builds a closed-loop management system covering pre-fault proactive prevention, mid-fault precise control, and post-fault efficient restoration. The platform comprehensively improves the anti-disturbance capability, safe operation level, and emergency response efficiency of general power systems (including transmission networks, distribution networks, and integrated energy systems), providing a complete, industry-proven solution for full-process grid resilience management.
Figure 1. Framework of Platform
1-2. Background
With the increasing frequency of extreme natural disasters worldwide, the high-proportion integration of renewable energy, and the rising complexity of modern power grid operation, power systems are facing unprecedented challenges to safe and stable operation. Traditional grid operation and management modes have obvious shortcomings in three core dimensions: insufficient proactive risk prevention capability for extreme scenarios, limited real-time fault suppression and cascading failure blocking ability during emergencies, and low intelligence and efficiency of post-blackout restoration decision-making. Large-scale blackouts caused by extreme events have brought huge economic losses and social impacts globally. Against this background, this project develops the AI-Driven Full Process Smart Grid Resilience Enhancement Platform, to build a full process intelligent management system for grid resilience, and fundamentally solve the core pain points of grid resilience enhancement.
1-3. Core Objectives
This project aims to build a full-process, modular, highly compatible smart grid resilience enhancement platform for general power systems, with the following core objectives:
Build a full process closed-loop management system for grid resilience, covering the three core stages of pre-fault proactive prevention, mid-fault emergency control, and post-fault efficient restoration, to realize full-chain intelligent management of grid risks.
Significantly improve the proactive defense capability of the power grid against extreme events, accurately identify potential risks in advance, reduce the probability of large-scale power outages from the source, and enhance the inherent anti-disturbance capability of the system.
Realize ultra-fast and reliable emergency disposal during grid disturbances, effectively block the propagation of cascading failures, ensure the continuous power supply of critical loads, and minimize the scope and impact of power outages during emergencies.
Improve the intelligence and efficiency of post-blackout grid restoration, shorten outage duration, reduce economic and social losses caused by power outages, and realize rapid and reliable recovery of the power system.
Achieve full compatibility with mainstream power grid operation systems and industrial standards, balance academic research value and industrial application prospects, and provide an easy-to-deploy, customizable solution for grid resilience enhancement for both academic institutions and power industry practitioners.
1.4. Holistic System Architecture
The platform adopts a three-tier core architecture that fully covers the whole process of grid resilience enhancement, with a unified underlying system, standardized data interaction mechanism, and closed-loop collaborative logic among all tiers. The overall architecture forms a complete "prevention-suppression-restoration" full-cycle resilience management system, with the following three core tiers:
(1) Pre-Fault Proactive Defense Tier
As the foundational "first line of defense" of the platform, this tier focuses on proactive risk management before disturbances occur. Through panoramic situational awareness of the power grid, extreme scenario risk assessment, and weak link identification, it formulates scientific and cost-effective grid infrastructure optimization and reinforcement plans, fundamentally reduces the probability of grid failures, and provides basic risk data support for the emergency control and restoration stages.
(2) Mid-Fault Active Control & Suppression Tier
As the core hub of the platform, this tier undertakes real-time emergency disposal during disturbances. It connects the pre-fault prevention and post-fault restoration stages, realizes ultrafast fault isolation, dynamic grid topology optimization, and safe and reliable dispatch decisionmaking during extreme events and sudden equipment failures, effectively blocks the propagation of cascading failures, ensures the power supply of critical loads, and provides a clear surviving grid topology and fault information for the post-fault restoration stage.
(3) Post-Fault Intelligent Restoration Tier
As the final closure of the platform's full-cycle resilience management, this tier focuses on rapid and reliable recovery of the power system after a blackout. Based on pre-fault risk data and midfault grid state information, it automatically generates the optimal grid black-start and restoration sequence, realizes step-by-step safe recovery of generators, loads and the grid network, minimizes system restoration duration, and feeds back restoration bottleneck information to the pre-fault defense tier to realize iterative optimization of the whole system.
1.5. Core Advantages
Full-Cycle Closed-Loop Resilience Management System
Different from traditional solutions that focus on a single stage of grid emergency disposal, this platform fully covers the whole process of grid resilience enhancement from pre-fault prevention to mid-fault control and post-fault restoration. The three core tiers are deeply coordinated, with unified data standards and closed-loop iterative optimization, realizing fullchain intelligent management of grid risks, and filling the industry gap of full-process grid resilience enhancement solutions.
AI-Driven Intelligent Decision-Making Capability
The platform deeply integrates artificial intelligence technology with the physical laws and operation rules of the power grid, breaks free from the dependence on traditional manual experience in grid operation and emergency disposal, and realizes intelligent, efficient and accurate decision-making throughout the whole process of risk assessment, emergency control and restoration planning, significantly improving the response speed and disposal efficiency of the power grid to various disturbances.
High Adaptability and Wide Compatibility
Designed for general power grid scenarios, the platform is fully applicable to transmission networks, distribution networks, integrated energy systems and other types of power systems. It is compatible with mainstream power grid operation data formats and industrial standards, supports flexible deployment and customized expansion according to the actual needs of different scenarios, and can meet the needs of both academic research and industrial field application.
Safe, Reliable and Interpretable Decision-Making Mechanism
Aiming at the high safety requirements of the power industry, the platform builds a safe and credible decision-making mechanism. All decisions strictly comply with the physical laws of the power grid and industrial operation specifications, with full traceability and clear logical basis, completely solving the trust problem of intelligent decision-making in high-risk industrial scenarios, and ensuring the safe and reliable operation of the power grid.
Significant Cost-Effectiveness and Practical Value
The platform balances investment cost and resilience enhancement effect throughout the whole process of grid operation. On the one hand, it optimizes pre-fault grid reinforcement investment through scientific risk assessment, and improves the cost-effectiveness of infrastructure investment; on the other hand, it significantly reduces economic and social losses caused by power outages through efficient emergency disposal and rapid restoration, bringing long-term and stable comprehensive benefits to grid operators and the whole society.
Figure 2. AI-Driven Intelligent Decision-Making Capability
1.6. Application Effects
After verification and application, the platform has achieved significant effects in the whole process of grid resilience enhancement, specifically as follows:
Proactive Risk Prevention Effect: The platform can accurately identify more than 95% of potential weak links in the power grid under extreme scenarios, and the optimized grid reinforcement plan can reduce the failure probability of key lines by more than 90% under extreme disasters while also cutting the projected investment by an expected 60%, significantly improving the inherent anti-disturbance capability of the power grid from the source.
Mid-Fault Emergency Control Effect: The platform can realize rapid fault isolation and emergency disposal during disturbances, effectively block the propagation of cascading failures, reduce the risk of large-scale blackouts by more than 90%, and ensure the continuous power supply of critical loads during extreme events, minimizing the scope and impact of power outages.
Post-Fault Restoration Effect: The platform greatly shortens the decision-making time of grid restoration from hours to minutes, and the optimized restoration sequence can shorten the full system restoration duration compared with the traditional manual scheduling mode, significantly reducing the economic loss caused by power outages.
1.7. Comprehensive Benefits
Safety and Reliability Benefits
The platform comprehensively improves the resilience of the power grid against extreme natural disasters, sudden equipment failures and renewable energy fluctuations, effectively reduces the probability and scale of large-scale blackouts, and strongly guarantees the safe and stable operation of the power system, as well as the reliable power supply for economic development and people's livelihood.
Economic Benefits
On the one hand, the platform realizes precise investment in grid infrastructure through scientific risk assessment and targeted reinforcement planning, avoids inefficient investment, and improves the cost-effectiveness of grid construction and operation investment; on the other hand, it significantly reduces the direct and indirect economic losses caused by power outages through efficient emergency disposal and rapid restoration, bringing considerable economic benefits to power grid operators and the whole society.
Academic and Industrial Application Benefits
The platform adopts a modular and open design, which can be widely used in academic research of power system resilience and actual operation and management of industrial power grids. At the same time, it provides an excellent training and education tool for power grid operators and related professional students, effectively promoting technological progress and talent training in the power industry.
1.8. Application Prospects
With the global energy transition and increasing attention to power grid safety, the demand for grid resilience enhancement is growing rapidly. With its full-process closed-loop design, advanced intelligent decision-making capability and wide applicability, this platform has broad application prospects in the following fields:
• Large-scale regional power grids and urban power grids in areas frequently hit by extreme natural disasters;
• New power systems with high proportion of renewable energy integration;
• Integrated energy systems and microgrids for key industrial parks, important institutions and urban core areas;
• Academic research and experimental teaching in the field of power system resilience;
• Emergency disposal capacity building for power grid enterprises in various countries and regions.
Figure 3. Typical Application Scenarios
2. STRATEGY
2-1. Innovativeness
2.1.1 Is the innovative concept come from the project itself or other existing programs?
The innovative concept of this project is entirely original and independently developed by the project team, filling the global industry gap of a full process closed-loop solution for grid resilience enhancement, with all core innovations rooted in the team's decades of accumulated research in power system restoration and cascading failure prevention.
Traditional grid resilience solutions in the industry mostly focus on a single segment (either prefault planning, in-fault control, or post-fault restoration), with fragmented data and no crosssegment synergy. They also rely heavily on manual experience and static physical topology analysis, unable to adapt to the dynamic operating state changes brought by extreme disasters and high-proportion renewable energy integration. In contrast, this project has pioneered the world's first AI-driven smart grid resilience enhancement platform covering the full process of prefault proactive prevention, in-fault precise control, and post-fault efficient restoration. The three core subsystems, including the Intelligent Network Planning for Resilience (INPR), the Cloud-Edge Collaborative Defense and Credible Decision Engine (CECD), and the Generic Restoration Intelligent Milestone (GRIM), adopt a modular and specialized design, sharing unified data formats, mathematical foundations, and interactive interfaces to achieve deep collaboration across the entire chain.
Figure 4. Full-Cycle Data Closed-Loop Flow
The originality of each core subsystem is further reflected in:
1) Pre-fault INPR Subsystem: It innovatively integrates multi-source data-driven situational awareness, probabilistic extreme scenario simulation, and targeted line reinforcement strategies, breaking through the limitations of traditional grid planning that only focuses on static N-1 security criteria. It realizes accurate identification of weak links in the grid under extreme scenarios and scientific reinforcement planning, while achieving seamless data connection with the in-fault and post-fault subsystems.
2) In-fault CECD Subsystem: It innovatively proposes a cloud-edge collaborative trusted decision engine architecture, integrating physics-constrained digital twins and RAG-enhanced whitebox large model simulation. It solves the industry pain points of high communication latency, difficult state estimation with missing data, and untrusted black-box AI decision-making in traditional emergency control, achieving millisecond-level fault isolation and second-level cascading failure blocking.
3) Post-fault GRIM Subsystem: As the released tool, it originally builds an intelligent decision system for grid black-start restoration based on milestones. Through a backtracking solver and step-by-step OPF feasibility verification, it realizes automatic optimization of restoration sequences and full-process safety verification, breaking through the pain points of traditional black-start relying on manual experience, low decision efficiency, and no feasibility guarantee.
The core technical achievements of the project have formed relevant tools, multiple academic papers, and industrial application cases. All innovative concepts come from the original research accumulation of the project team, rather than borrowing from other existing programs.
2.1.2. How the innovative policy design encourages financial support and public-private partnership?
The project has built a multi-dimensional and full-chain innovation mechanism to fully leverage financial support and public-private partnership from the government, industry, academia, and research institutions, forming a sustainable industry-university-research-user collaborative innovation ecosystem:
Industry-university-research collaborative innovation mechanism to connect government R&D funds
The project team has established in-depth joint R&D mechanisms with universities, as well as power grid operation enterprises and power research institutions, to jointly apply for government science and technology innovation funds. Referring to the successful experience of projects in the power system field, this model has successfully promoted the implementation of 3+ special projects, providing sufficient government financial support for the technological iteration of the project.
Smart procurement mechanism to empower the participation of SMEs
The project adopts a "smart procurement" model. In the process of engineering implementation, it gives priority to opening technical interfaces and encourages innovative SMEs to participate in the localized development and engineering implementation of solutions. It not only reduces the cost of project implementation, but also provides technology landing scenarios and business opportunities for SMEs, further expanding the coverage of public-private cooperation, and forming an industrial ecosystem of integrated development of large, medium and small enterprises.
ESG-oriented funding channel expansion
The project focuses on core ESG topics such as grid security, energy saving and carbon reduction, and urban resilience enhancement, actively connects with green finance and ESG-related special funds, aligns the project's carbon reduction benefits and livelihood security value with the green investment needs of financial institutions, broadens the longterm funding sources of the project, and promotes the large-scale implementation of the project's achievements.
2.1.3. How does the innovative concept catch the trend of future development?
The innovative concept and technical system of the project are highly consistent with the four core development trends of the global power industry in the future, and forward-looking solve the core pain points of future grid development, leading the technical development direction of smart grid resilience enhancement:
Aligns with the development trend of new power systems with high-proportion renewable energy integration
With the acceleration of global energy transition, the large-scale grid connection of wind power, photovoltaic and other renewable energy has become an inevitable trend. The volatility and uncertainty of grid operation state have been greatly intensified, and the traditional static resilience management model can no longer adapt. The full process closedloop system of the project: the pre-fault INPR can quantify the system risk brought by renewable energy fluctuations; the in-fault CECD can realize fast fault suppression under renewable energy fluctuations; the post-fault GRIM can optimize the restoration strategy adapting to high-proportion renewable energy scenarios. It comprehensively solves the core pain points of resilience management in new power systems, and provides key technical support for the large-scale safe consumption of renewable energy.
Aligns with the global trend of digital and intelligent transformation of power grids
The global power industry is in a critical stage of comprehensive digital transformation, and the deep integration of big data, artificial intelligence, digital twins and other technologies with grid business is the core direction of future development. The project deeply integrates cutting-edge digital technologies such as spatio-temporal graph neural networks, physicsinformed neural networks, RAG large models, and cloud-edge collaboration, builds a datadriven full-process intelligent decision-making system, breaks through the limitations of traditional grid operation relying on manual experience, and is fully compatible with the development direction of grid intelligent transformation. At the same time, through standardized data interfaces and modular design, it can be seamlessly integrated into existing grid digital systems, with strong implementability and scalability.
Figure 5. Cloud-Edge Collaboration
Aligns with the trend of grid resilience enhancement under frequent extreme disasters
Global climate change has led to frequent extreme natural disasters such as typhoons, ice disasters and rainstorms. The impact of large-scale blackouts on the social economy and people's livelihood has become increasingly serious. Grid resilience enhancement has become the core priority of energy policies around the world. The full process resilience management system built by the project fundamentally solves the pain point that traditional solutions only focus on a single link and cannot cope with the full-process impact of extreme disasters. It comprehensively improves the grid's anti-disturbance capability, emergency disposal capability and rapid recovery capability in response to extreme events, which is fully in line with the future development needs of urban and grid infrastructure resilience enhancement.
Aligns with the trend of APEC regional energy security and low-carbon transition cooperation
APEC economies are accelerating the promotion of energy intensity reduction targets and carbon neutrality vision, and regional energy interconnection and smart grid technology cooperation have become core topics. The project adopts an open and compatible design concept, is compatible with IEEE standard grid data formats and mainstream power system simulation tools, can adapt to grid scenarios of different economies and different scales. At the same time, through open source sharing, international competitions, academic exchanges and other models, it promotes the promotion and application of technology in the APEC region, helps economies improve grid security and energy efficiency, and is fully in line with the development trend of APEC regional energy cooperation and low-carbon transition.
2.2. Inspiration
2.2.1. Whether the idea can inspire later/subsequent cases?
The innovative concept, technical system and implementation model of the project have strong demonstration and leading significance, and can fully inspire and guide subsequent cases related to grid resilience enhancement and smart grid construction, becoming a replicable benchmark paradigm in the industry:
Replicable technical paradigm provides a new framework for subsequent R&D
The project's pioneering "full process closed-loop" grid resilience enhancement technical paradigm breaks the limitation of fragmented "pre-, in-, post-fault" links in traditional research and engineering, and provides a new R&D framework for subsequent various grid resilience enhancement projects. Whether it is regional transmission grids, urban distribution networks, park integrated energy systems, or microgrid scenarios, subsequent projects can directly learn from this closed-loop framework, carry out targeted optimization combined with their own scenario characteristics, and avoid repeated underlying technical exploration. At present, research teams from many universities have carried out extended research on renewable energy microgrid resilience enhancement based on the full process framework of the project, which fully verifies the inspiration of the concept.
Referenceable implementation model provides a mature path for industrial landing
The project's implementation model of "industry-university-research-user collaboration + ecosystem cultivation + licensed commercial promotion" provides a mature and referenceable path for the industrial landing of subsequent smart grid technology projects. Subsequent power industry digital and intelligent innovation projects can refer to this model, reduce the technical threshold through openness, connect R&D resources through industry-university-research cooperation, and balance inclusiveness and commercial value through the licensed model, to achieve rapid landing of technology from the laboratory to the engineering site.
Scalable engineering application provides reference for cross-scenario applications
The three core subsystems of the project are modularly designed, which can be deployed independently or combined, and can inspire subsequent resilience enhancement cases in different scenarios. For example, the weak link identification method of the pre-fault INPR can inspire the planning and design of subsequent grid infrastructure reinforcement and extreme disaster prevention projects; the cloud-edge collaborative control method of the in-fault CECD can inspire the development of subsequent distribution network automation and microgrid emergency control projects; the intelligent restoration decision-making method of the post-fault GRIM can inspire the construction of subsequent power outage emergency disposal and black-start drill training projects.
Promotable international cooperation provides a template for regional technical cooperation
The project's model of promoting cross-border technology promotion through international academic exchanges, open-source communities, and global competitions can inspire subsequent smart grid technology cooperation projects between APEC economies, and provide a replicable cooperation template for regional energy technology exchanges and capacity building, helping to improve the overall level of energy security in the APEC region.
2.2.2. What domain has been enlightened by this policy?
The innovative concept, technical system and implementation model of the project have not only had a far-reaching impact within the power industry, but also radiated and inspired a number of related fields, forming a wide driving effect across industries and fields:
Whole industrial chain of the power system
The project has comprehensively inspired the intelligent transformation of the whole industrial chain of the power industry from planning, design, construction, operation, emergency to restoration. In the field of grid planning, it has promoted the transformation of the industry from static safety planning to resilience-oriented dynamic planning; in the field of grid dispatching and operation, it has promoted the large-scale application of cloudedge collaboration and trusted AI decision-making in emergency dispatching; in the field of grid emergency management, it has promoted the standardized construction of intelligent black-start restoration systems based on intelligent decision-making; in the field of power equipment manufacturing, it has inspired equipment manufacturers to transform to intelligent equipment integrating intelligent sensing, edge control and digital twins, driving the intelligent upgrading of power equipment.
Critical infrastructure resilience enhancement
The project's full process closed-loop resilience enhancement concept, as well as datadriven weak link identification, dynamic emergency control, and intelligent restoration decision-making methods, can be widely extended to other critical infrastructure fields such as urban water supply, gas, transportation, and communication. These infrastructure systems also face common pain points such as extreme disaster impacts, cascading failure propagation, and low emergency restoration efficiency. The technical paradigm and implementation model of the project provide a new idea and method for resilience enhancement in these fields. At present, we have carried out preliminary technical exchanges with the Water Supplies Department and gas companies of Hong Kong, China, which has inspired the R&D of urban water supply network resilience assessment and emergency restoration systems.
Energy digital technology and industrial AI application
The project's pioneering industrial-grade AI decision-making method of "physical constraints + data-driven", as well as the RAG-enhanced white-box large model industrial application paradigm, solves the core pain points of "untrusted black box, non-compliance with physical laws, and difficult landing" of AI technology in high-risk industrial scenarios. It provides an important technical reference and practical template for the large-scale application of artificial intelligence in process industries such as energy, chemical industry, and manufacturing, and inspires the technological transformation of the industrial AI field from "black-box prediction" to "white-box trusted decision-making".
Urban safety and emergency management
The project's full process disaster response closed-loop system has inspired the concept upgrade of the urban comprehensive emergency management field, and promoted the transformation of urban emergency management from "post-disaster disposal" to full process closed-loop management of "pre-disaster prevention, in-disaster control, and postdisaster restoration". The project's extreme scenario simulation, cascading failure blocking, and rapid restoration optimization technologies can be integrated into the urban comprehensive emergency command system, providing technical support for the city to respond to extreme disasters and large-scale emergencies, and improving the overall safety and resilience of the city.
Green finance and ESG investment
The project has established a quantitative evaluation system between grid resilience enhancement and energy saving and carbon reduction, livelihood security, and economic loss avoidance, providing a quantifiable evaluation standard and landing scenario for green credit and ESG investment in the field of grid infrastructure resilience enhancement in the green finance field. It has inspired financial institutions to incorporate infrastructure resilience into the ESG investment evaluation system, and broadened the coverage and application scenarios of green finance.
Figure 6. Domains Enlightened by the Project
2.3. Clearness
2.3.1. Is there any open and transparent channel of public communication?
The project has established a multi-dimensional, full-coverage, open and transparent public communication and information release channel, ensuring the full disclosure of project information, inclusive sharing of technical achievements, and unimpeded industry exchanges, while providing convenient consultation and interaction channels for the public, industry practitioners, and academic institutions:
International academic conferences and industry forum exchanges
The project team continuously releases the project's research results and application practices through global top academic conferences and industry forums, including IEEE ISGT Europe, IEEE PES General Meeting, and other international conferences. It has published more than 10 academic papers and carried out open technical exchanges and discussions with global industry experts and scholars, ensuring the openness and transparency of the project's technical achievements and industry consensus.
Standardized achievement release and reporting system
The project has established a standardized achievement release mechanism, regularly releasing public documents, application case reports, and engineering implementation guidelines, which disclose the technical principles, implementation effects, quantitative indicators and best practices of the project in detail.
Figure 9. Relevant Publications
Unimpeded public consultation and interaction channels
The project has set up a special consultation email and online message board on the official website, which is responsible for responding to technical consultations, cooperation needs and suggestions from users around the world by a professional technical team. At the same time, the project has established a normalized interactive exchange mechanism with global users through academic email groups, industry communities, etc., to timely answer user questions, collect improvement suggestions, and realize continuous iterative optimization of the project, ensuring that the development of the project always meets the needs of the industry and the public.
2.3.2. Is there any difference between this policy and other similar policies?
Compared with similar solutions at home and abroad, the project has essential differences in core concept, technical system, and collaboration capability, forming a unique core competitive advantage and filling the industry gap. The specific differences are reflected in four core dimensions:
Core concept difference: from single link optimization to full process closed-loop management
Most similar solutions around the world mostly focus on a single link of grid resilience management, either only focusing on pre-fault grid planning and weak link reinforcement, or only focusing on post-fault black-start restoration decision-making, or only focusing on in-fault fault emergency control. The data between each link is fragmented and lacks collaboration, which cannot form a full-process resilience management closed loop. In contrast, this project pioneered the concept of full process closed-loop grid resilience enhancement of "pre-fault proactive prevention - in-fault precise control - post-fault efficient restoration". The three core subsystems are deeply coordinated, data interoperable, and iteratively optimized: the weak link data of the pre-fault INPR provides fault prevention and control priority for the in-fault CECD, and provides restoration bottleneck prediction for the post-fault GRIM; the surviving grid topology of the in-fault CECD provides the initial restoration conditions for the post-fault GRIM; the restoration bottleneck data of the post-fault GRIM is fed back to the pre-fault INPR to realize iterative optimization of the grid reinforcement scheme. This full closed-loop concept fundamentally solves the industry pain point of fragmented links and limited resilience enhancement effect of traditional solutions, and is the first integrated solution in the industry to achieve fullprocess coverage.
Technical system difference: from experience-driven static solutions to dynamic intelligent decision-making
Traditional similar solutions mostly rely on manual experience and static physical topology analysis, which cannot adapt to the dynamic changes of grid operation state, and have the problems of low decision efficiency, poor adaptability, untrusted results, and no feasibility verification. In contrast, this project has built a full-link AI-driven dynamic intelligent decision-making technical system, and each subsystem has achieved breakthrough innovations in technology: the pre-fault INPR realizes dynamic and accurate identification of grid weak links through multi-source data fusion and probabilistic extreme scenario simulation; the in-fault CECD realizes millisecond-level fault isolation and trusted emergency decision-making through cloud-edge collaborative architecture, digital twins and RAG white-box large models; the post-fault GRIM realizes automatic optimization of restoration sequences and full-process safety guarantee through intelligent backtracking solver and step-by-step AC-OPF feasibility verification. At the same time, the entire platform shares a unified mathematical foundation and data standards, is compatible with mainstream industry tools, and its technical maturity, reliability and adaptability are far superior to traditional static solutions.
Collaboration capability difference: from independent tool deployment to modular collaborative ecological architecture
Most similar solutions mostly adopt the design mode of independent tools and closed architecture, with incompatible data between different functional modules, unable to link, and difficult to adapt to grid scenarios of different scales and types. In contrast, this project adopts the design philosophy of "modular specialization and ecological collaboration". The three core subsystems can be deployed independently to meet the user's needs for a single link, or seamlessly combined to form an end-to-end full-process resilience enhancement solution. All modules share unified data formats, interface specifications and interaction paradigms, are compatible with various grid scenarios such as transmission grids, distribution networks, integrated energy systems, and microgrids, and support user-defined expansion and secondary development. This open, modular ecological architecture is essentially different from traditional closed independent tools, and greatly reduces the application threshold and deployment cost for users.
3. MEASURE
3.1. Practicability
3.1.1 Has any effective measure for moving ahead been made?
Around the four core objectives of technology R&D, engineering landing, ecosystem cultivation, and international promotion, the project has formulated and implemented a series of effective promotion measures to ensure the smooth progress of the project from technology R&D to engineering landing, and achieved remarkable phased results:
Completed multi-scenario simulation verification and engineering pilot landing, verified technical practicability
First, based on IEEE standard test systems, the project completed full-scenario simulation verification, and carried out thousands of groups of extreme scenario simulation, fault prevention and control, and restoration decision simulation tests, fully verifying the effectiveness and reliability of the technology. On this basis, the project has carried out engineering pilot landing in cooperation with power grid operation enterprises in Hong Kong, China and Chinese Mainland, such as deploying the CECD cloud-edge collaborative emergency control system in an urban distribution network. All pilot projects have achieved remarkable application effects, verifying the practicability and stability of the technology in actual engineering scenarios.
Built an open-source and open platform, and formed an industry-university-research-user collaborative ecosystem
The project released the partial code and documents and established a normalized technical training and seminar mechanism. Up to now, the project has established joint R&D cooperation with more than 5 universities, and established industrial cooperation with more than 3 power grid enterprises and power engineering service providers, forming an in-depth collaborative innovation ecosystem of "industry-university-research-user", laying a solid ecological foundation for the continuous iteration and large-scale landing of the technology.
Formulated a standardized promotion path, and promoted the standardized application of technology
The project has started the standardization work of technology, and is compiling industry technical standards. At the same time, it has formulated a phased and large-scale promotion plan: first, realize large-scale application in the Guangdong-Hong Kong-Macao Greater Bay Area power grid, then gradually promote it nationwide, and then promote it to Asia-Pacific economies through the APEC energy cooperation mechanism. At the same time, for economies of different scales and development levels, differentiated technology landing plans have been formulated to ensure that the technology can adapt to the actual situation of power grids in different regions and achieve implementable and promotable results.
3.1.2 Is there any numerical goal for reference?
At the beginning of the project, clear, hierarchical and measurable numerical goals were formulated, covering four dimensions: simulation, engineering application, ecological promotion, and economic benefit. All of the goals have a clear implementation schedule:
(1) Simulation Numerical Goals
(2) Engineering Application Numerical Goals
(3) Ecological Promotion Numerical Goals
(4) Economic Benefit Numerical Goals
3.2 Replicability
3.2.1 Could the ideas, methods or techniques be applied internationally?
The core concept, technical methods and implementation model of the project have strong universality, adaptability and replicability, and can be fully promoted and applied in the power grids of APEC economies and other countries and regions around the world. At present, preliminary international application verification has been carried out with good results:
The core concept and technical methods have global universality
The core concept of the project, "full process closed-loop grid resilience enhancement", addresses the common pain points faced by the global power industry, including extreme disaster prevention and control, cascading failure blocking, and low black-start restoration efficiency. Whether developed or developing economies, large interconnected grids or small isolated grids, all face the same industry challenges, so this core concept has global universality. At the same time, the core technical methods of the project are based on the general physical laws of power systems, IEEE international standards, and industry mainstream simulation tools, do not rely on specific grid architectures, equipment models and management systems, and can be directly adapted to various grid scenarios in different countries and regions around the world, including large regional transmission grids, urban distribution networks, and island microgrids, with strong scenario adaptability.
Modular open architecture ensures flexibility for international implementation
The project adopts a modular and open architecture design. The three core subsystems can be deployed independently or in combination, and can flexibly select adapted solutions according to the grid development level, technical foundation and actual needs of different economies:
• For developed economies with a good grid digital foundation, the full-process closed-loop platform can be deployed to realize full process intelligent management of grid resilience;
• For developing economies, single modules such as the GRIM black-start restoration tool and INPR weak link identification system can be deployed independently to quickly improve grid emergency disposal capacity, and then gradually expand to the full-process system;
• For small island grids and microgrids in remote areas, the algorithm model can be optimized for the characteristics of small-capacity and weak-connection grids, to achieve low-cost and high-reliability resilience enhancement.
At the same time, the platform is compatible with mainstream global grid data formats, communication protocols and simulation tools, supports multiple languages, and can quickly interface with existing grid dispatching systems and automation systems in various countries, greatly reducing the technical threshold and implementation cost of international implementation.
A complete international technical support and capacity building system ensures replicability
The project has established a complete international technical support and capacity building system, providing full-process technical support and talent training services for various economies through international seminars, online and offline training courses, localized translation of technical manuals, joint R&D cooperation, etc., to help local technical personnel master the project's technical methods and tool applications, and realize localized technology implementation and continuous optimization. At the same time, the project actively promotes the regional promotion and standard coordination of technology through multilateral cooperation platforms, providing institutional guarantee for the large-scale replication and application of technology in the APEC region.
3.3. Cost-effectiveness
3.3.1. Will it be cost-effective to implement?
The implementation of the project has extremely high cost-effectiveness, which is reflected in the extremely low upfront investment, significant investment optimization, huge economic benefits, as well as a long-term sustainable cost reduction space. Whether it is a single project deployment or large-scale promotion, it shows a cost-effectiveness advantage far exceeding that of traditional solutions:
Extremely low upfront investment, almost zero marginal deployment cost
The core R&D investment of the project has been basically completed, and a mature software platform and technical system have been formed. It can be deployed in the existing grid system without large-scale hardware equipment replacement and infrastructure reconstruction. For users, the upfront investment only requires a small amount of software deployment, personnel training and data docking costs, which is far lower than the infrastructure investment of traditional grid transformation projects. At the same time, the project adopts a modular design, and users can choose to deploy a single module according to their own needs, further reducing the upfront investment threshold. For academic research and non-commercial use, the core tools of the project are completely open source and free, with almost zero upfront investment.
Greatly optimize grid infrastructure investment and avoid ineffective capital waste
Traditional grid resilience enhancement mainly relies on large-scale line reinforcement, equipment replacement and other infrastructure investments, with the problem of "blind expansion, ineffective investment" and extremely low capital utilization efficiency. The prefault INPR subsystem of the project can accurately identify the key weak links affecting grid resilience, formulate targeted line reinforcement schemes, and avoid ineffective investment in full-line non-differential transformation. In numerical tests, the optimized reinforcement scheme saves 60% of the grid transformation investment compared with the traditional scheme.
Bring huge direct and indirect economic benefits
Large-scale blackout accidents will bring huge economic losses to the social economy, and the economic loss of a single large-scale blackout can reach hundreds of millions or even billions of HKD. Through full-process resilience enhancement, the project can reduce the probability of large-scale grid blackout accidents by more than 90%, greatly avoiding the huge economic losses caused by power outages. In pilot applications, the project can avoid more than 8 million HKD of economic losses caused by power outages every year for the local area. At the same time, through predictive maintenance and fault early warning, it can reduce the operation and maintenance cost of grid equipment, extend the service life of equipment, and bring long-term cost savings.
Long-term cost continues to decline, with sustainable economic benefits
With the large-scale promotion and continuous iteration of the technology, the implementation cost of the project will further decrease, while the benefits will continue to increase. On the one hand, large-scale application will dilute the R&D and maintenance costs, and the technical threshold and implementation cost of localized adaptation will continue to decrease; on the other hand, the open-source ecosystem of the project will attract global developers to continuously optimize the algorithm model, and continuously improve the benefit output of the technology. At the same time, the technical system of the project can be continuously expanded with the development of the power grid, adapting to the future development needs such as high-proportion renewable energy grid connection and grid digital upgrading, without frequent system reconstruction and large-scale re-investment, with long-term cost-effectiveness advantages.
3.3.2. Is there any measurable reduction of emission or energy use? Please describe the measurement method.
The implementation of the project can achieve measurable, verifiable and repeatable energy saving and carbon emission reduction. We have established a measurement method that strictly follows international standards to accurately quantify the energy saving and carbon reduction benefits of the project.
Measurable Energy Saving and Carbon Reduction Results
The energy saving and carbon reduction benefits of the project mainly come from two core dimensions, both of which have achieved quantifiable and significant results in engineering pilots:
Direct energy saving and carbon reduction: from grid line loss reduction
Through the grid topology optimization of the pre-fault INPR, the real-time operation mode optimization of the in-fault CECD, and the power flow control during the restoration process of the post-fault GRIM, the project realizes the reduction of the comprehensive line loss rate of the grid in the whole life process. In the engineering pilot application, the project reduced the comprehensive line loss rate of the pilot area grid with an annual electricity saving of 8 million kWh. After large-scale promotion, it is expected to achieve an annual electricity saving of more than 50 million kWh.
Indirect energy saving and carbon reduction: from reducing energy waste caused by power outages
Large-scale blackout accidents will lead to a large amount of energy waste and invalid carbon emissions from industrial production interruption, equipment start-stop loss, and high-emission operation of emergency power generation equipment. By improving grid resilience, the project reduces the probability of large-scale blackout accidents by more than 90%, greatly reducing the energy waste caused by power outages. At the same time, the project shortens the recovery time after grid failures, reduces the operation time of diesel emergency generators during power outages, and further reduces fossil energy consumption and carbon emissions.
In addition, by optimizing the grid operation mode, the project improves the renewable energy consumption capacity, can promote the large-scale grid connection of clean energy such as wind power and photovoltaic, replace fossil energy power generation, and achieve larger-scale indirect carbon emission reduction, with long-term carbon reduction potential.
Measurement Method for Energy Saving and Carbon Reduction
We strictly follow the International Performance Measurement and Verification Protocol (IPMVP), the global authoritative standard for energy saving project measurement, combined with the relevant requirements of the Guidelines for Greenhouse Gas Emission Accounting and Reporting, and established a scientific, rigorous and repeatable measurement method. The core steps are as follows:
1) Determine the measurement boundary and baseline period
First, clarify the geographical boundary, grid equipment scope and time cycle of the measurement, taking 12 consecutive months before the project implementation as the baseline period, and 12 consecutive months after the project implementation as the measurement period, to ensure that the meteorological conditions, grid load level, power supply structure and other boundary conditions of the two periods are basically consistent, and exclude the interference of external factors on the measurement results.
For the energy saving measurement of line loss reduction, the boundary covers all transmission and distribution lines, transformers and other grid equipment in the pilot area; for the indirect energy saving measurement of power outage avoidance, the boundary covers all industrial, commercial, residential electricity loads and emergency power generation equipment in the pilot area.
2) Establish the baseline energy consumption model
The Artificial Neural Network is used to build the grid energy consumption baseline model in the baseline period. The input variables of the model include: grid load level, meteorological parameters, power supply structure, renewable energy output, grid topology and other key factors affecting grid line loss and energy consumption.
• The model is trained with the actual grid operation data in the baseline period to ensure the prediction accuracy of the model, and the error between the predicted value of the model and the actual value is controlled within 2%, meeting the accuracy requirements of the IPMVP standard.
• This model is used to calculate the baseline line loss electricity and baseline power outage duration of the grid without the project implementation in the measurement period, and the corresponding baseline energy consumption.
3) Collect actual operation data in the measurement period
Through the grid dispatching automation system, smart meters, sensors and other equipment, collect the actual operation data of the grid in the measurement period, including: actual line loss data of each line, total power supply, total electricity sales, actual power outage times and duration, equipment operation parameters, meteorological data, etc. The data sampling frequency is consistent with the baseline period, to ensure the integrity, continuity and accuracy of the data.
4) Uncertainty analysis and third-party verification
• Conduct a complete uncertainty analysis on the measurement results, including data acquisition error, model prediction error, error caused by boundary condition changes, etc., to ensure that the overall uncertainty of the measurement results is controlled within 5%, meeting the international standard requirements.
• Entrust a qualified third-party authoritative institution to independently verify the measurement process and results, and issue an official verification report to ensure the authenticity, accuracy and traceability of the energy saving and carbon reduction data.
Figure 12. Smart Grid Energy Saving and Carbon Reduction Measurement Methodology
3.4. Consistency
3.4.1. Are adopted measures consistent with energy policy and strategy?
All technical measures, implementation paths and promotion models adopted by the project are highly consistent with the energy policies and development strategies of the Hong Kong Special Administrative Region, China, the national level of China, and the APEC regional energy cooperation vision, and are the specific implementation and important support for energy strategic objectives at all levels in engineering practice:
Highly consistent with the energy policy and development strategy of the Hong Kong Special Administrative Region, China
The project is fully in line with the core objectives of Hong Kong's Climate Action Blueprint 2050, and Hong Kong Smart City Blueprint 2.0:
• Hong Kong's Climate Action Blueprint 2050 clearly puts forward the goals of improving grid security and resilience, improving energy use efficiency, promoting the digital transformation of the power industry, and achieving carbon neutrality by 2050. Through fullprocess grid resilience enhancement, the project greatly reduces the risk of large-scale blackout accidents and ensures the security of energy supply; it achieves significant energy saving and carbon reduction by optimizing grid operation and reducing line loss; it promotes the digital and intelligent transformation of the power industry through the deep integration of AI technology and grid business, which is fully in line with the core requirements of Hong Kong's carbon neutrality action.
• Hong Kong's Smart City Blueprint 2.0 takes smart energy and resilient infrastructure as the core construction content. The AI-driven full-process smart grid resilience enhancement platform built by the project is a core component of the resilience improvement of smart city infrastructure, and provides key technical support for the construction of Hong Kong's smart city.
Figure 13. Hong Kong's Climate Action Blueprint 2050 and Smart City Blueprint 2.0
Website:
https://cnsd.gov.hk/wp-content/uploads/pdf/CAP2050_booklet_en.pdf
https://www.smartcity.gov.hk/modules/custom/custom_global_js_css/assets/files/HKSmartCityBlueprint(ENG)v2.pdf
Highly consistent with Chinese Mainland’s energy strategy and power development plan
The implementation of the project is fully in line with the core requirements of China’s 14th Five-Year Plan for Modern Energy System, 15th Five-Year Plan’s outline for carbon peaking and carbon neutrality, and the Blue Book for the Development of New Power Systems:
• The China’s 14th Five-Year Plan and the 15th Five-Year Plan’s outline clearly put forward the need to build a new power system, improve the security and resilience of the power grid, enhance the ability to resist disturbances, prevent and resolve the risk of large-scale blackouts, and ensure the security of energy supply. The full process grid resilience enhancement system built by the project fundamentally improves the ability of the power grid to cope with extreme disasters and renewable energy fluctuations, and is a key technical measure to implement the grid security strategy.
• The Chinese Mainland’s "dual carbon" policy system requires promoting the digital and intelligent transformation of the energy sector, improving energy use efficiency, and significantly reducing carbon emissions in the energy industry. The project achieves significant energy saving and carbon reduction by optimizing grid operation, reducing line loss, and improving renewable energy consumption capacity, while promoting the large-scale application of AI technology in the energy field, which is fully in line with the development direction of the dual carbon strategy.
• The Blue Book for the Development of New Power Systems proposes to promote the indepth application of artificial intelligence, digital twins, cloud-edge collaboration and other technologies in the power system, and improve the intelligent operation level of the power grid. The core technical system of the project fully integrates these cuttingedge digital technologies, and provides an important technical guarantee for the safe and stable operation of the new power system, which is completely consistent with the strategic direction of the Chinese Mainland’s new power system construction.
Figure 14. China’s 14th Five-Year Plan for Modern Energy System
Website:
https://www.gov.cn/zhengce/zhengceku/2022 - 03/23/content_5680759.htm
https://www.nea.gov.cn/download/xxdlxtfzlpsgk.pdf
Highly consistent with the APEC regional energy cooperation vision and development goals
The core concept, technical achievements and promotion model of the project are fully in line with the core requirements of the APEC Energy Smart Communities Initiative (ESCI), the APEC Energy Intensity Reduction Goal, and the APEC Smart Grid Development Roadmap:
• The core goal of the APEC ESCI Initiative is to promote the innovation and application of advanced energy technologies, improve energy efficiency, promote low-carbon transformation, and enhance the resilience of communities and infrastructure. As an innovative technical achievement in the field of smart grids, the project is a typical best practice of the ESCI Initiative, which can help APEC economies improve the resilience of grid infrastructure, improve energy use efficiency, and promote low-carbon transformation, fully in line with the core purpose of ESCI.
• APEC has set a goal of reducing energy intensity by 45% by 2035. The project achieves significant energy saving benefits by reducing grid line loss and energy waste, and after largescale promotion, it will make an important contribution to the achievement of the APEC regional energy intensity reduction goal.
• The APEC Smart Grid Development Roadmap proposes to promote the cooperation and exchange of smart grid technologies among economies, and improve the security and reliability of power grids, renewable energy consumption capacity and digital level. The project promotes the inclusive promotion of smart grid resilience technology in the APEC region through open source, international training and multilateral cooperation, which is fully in line with the development vision of APEC regional energy technology cooperation.
Figure 15. APEC’s Urban Development Smart Grid Roadmap and Energy Intensity Reduction Target
Website:
https://www.apec.org/docs/default - source/publications/2014/3/urban - development - smart - grid - roadmap - christchurch - re covery - project/s - ewg - 08 - 12_urban - development - smart - grid - roadmap - christchurch - recovery - project.pdf?sfvrsn=a5966e40_1 https://www.apec.org/publications/2025/07/apec - energy - overview - 2025
3.4.2. Is there any long-term measure or implementing organization for this project?
The project has established a complete long-term implementation guarantee system, with a stable implementation organization, a clear long-term development plan and a continuous resource guarantee mechanism, to ensure the long-term sustainable development and large-scale promotion of the project:
Stable Long-Term Implementation Organization System
The project has built a four-in-one long-term implementation organization system of "core R&D institution + industrial landing institution + international cooperation institution + expert advisory committee", which provides a solid organizational guarantee for the long-term development of the project:
• Core R&D Institution: Hong Kong Productivity Council and The University of Hong Kong are the core R&D subjects, and a special R&D center has been established in conjunction with Tsinghua University, with a fixed core R&D team responsible for the continuous iteration, algorithm optimization and new product R&D of the project's core technologies. The R&D center has fixed R&D sites, experimental equipment and computing power resources, and has established a normalized R&D management mechanism, providing stable support for the long-term technological innovation of the project.
• Industrial Landing Institution: The industrial promotion center has established an engineering implementation team covering the Guangdong-Hong Kong-Macao Greater Bay Area and radiating across the country, and has formulated a standardized engineering landing process to ensure the stable and efficient application of the technology in actual grid scenarios.
• International Cooperation Institution: Relying on international platforms such as the APEC energy cooperation mechanism and the IEEE Power & Energy Society, an international promotion office will be established, responsible for the global promotion, international technical cooperation and capacity building of the project's technologies.
• Expert Advisory Committee: Top power system experts, energy policy experts, and artificial intelligence experts have been invited to form an expert advisory committee to provide professional guidance and decision-making suggestions for the long-term development strategy, technology R&D direction, and international promotion path of the project, ensuring that the development of the project always fits the global industry development trend and the actual needs of various economies.
Clear Long-Term Development Plan and Implementation Measures
The project has formulated a long-term development plan from 2026, with clear phased development goals and implementation measures, to ensure the long-term sustainable development of the project:
1) First Phase (2026-2030): Technology Deepening and Large-Scale Landing in the Guangdong-Hong Kong-Macao Greater Bay Area
• Technical level: Complete the in-depth optimization of the three core subsystems, adapt to the new power system scenario with high-proportion renewable energy access, complete the development of special versions for microgrids and integrated energy systems; complete the compilation and release of 2 industry technical standards, and realize the standardized application of technology.
• Landing level: Achieve large-scale promotion in the Guangdong-Hong Kong-Macao Greater Bay Area, with cumulative application to no less than 5 urban-level power grids, forming a replicable GBA promotion model; complete pilot application verification in no less than 2 APEC economies.
• Ecological level: Improve the construction of the open-source community, train no less than 100 professional talents cumulatively, and form a complete industry ecosystem.
2) Second Phase (2031-2036): National Promotion and In-Depth Cooperation in the APEC Region
• Technical level: Complete the technical adaptation of cross-border interconnected power grids and AC/DC hybrid power grids, develop multi-language versions of the platform software, and adapt to the grid standards and management systems of different economies; promote the technology to become the recommended technical specification of the APEC region.
• Landing level: Promote the application nationwide, covering no less than 15 provincial/urban-level power grids; realize the landing application in no less than 6 APEC economies.
• Cooperation level: Establish long-term cooperative relations with energy departments and power grid enterprises of various APEC economies, carry out cross-border technical cooperation projects, and promote the overall improvement of regional grid resilience level.
3) Third Phase (2037-): Global Promotion and Industry Leadership
• Technical level: Complete the comprehensive upgrade of the technical system, form an intelligent resilience enhancement solution covering the full scenario of the power system, and maintain the global leading position of the technology; promote the technology to become an international standard.
• Landing level: Realize the promotion and application of the technology in no less than 20 countries and regions around the world, and become a benchmark solution in the field of global grid resilience enhancement.
• Industry level: Build a world-leading technology innovation center for grid intelligent resilience, form a complete ecological system of "technology R&D - standard setting - industrial landing - international promotion", and lead the technological development of the global grid resilience enhancement field.
Figure 16. Long-Term Development Plan with Different Phases
Continuous Long-Term Resource Guarantee Mechanism
• Policy and Funding Guarantee: The project has obtained long-term policy support and financial guarantee from the Hong Kong government; at the same time, through industryuniversity-research joint application, it continuously obtains government scientific research funds; through commercial authorization and engineering services, it will achieve stable commercial income, forming a diversified long-term funding guarantee system of "government funds + commercial income", ensuring continuous capital investment in project R&D and promotion.
• Talent Guarantee: It has established a long-term joint talent training mechanism with universities, set up joint laboratories, and trains interdisciplinary professionals in power systems and artificial intelligence, providing continuous talent supply for the long-term development of the project.
• Intellectual Property Guarantee: A complete intellectual property management system has been established, and comprehensive invention patent and software copyright protection is carried out for the core technologies, algorithms and software platforms of the project, ensuring the intellectual property security of the project, and providing legal guarantee for long-term technological innovation and commercial promotion.
4. PERFORMANCE
4.1. Completeness
4.1.1. Is the achievement scale measurable?
All achievements of the project have established a multi-dimensional and quantifiable evaluation system, which can be accurately measured and fully traced. Core measurable achievements cover four dimensions:
%5) Technology R&D: Three core subsystems have been fully developed, with key indicators meeting or exceeding targets: INPR's weak link identification accuracy reaches above 95% under extreme scenarios, CECD achieves cascading failure blocking effective rate above 90%, GRIM shortens the restoration time by more than 40%.
%5) Engineering Application: 3 grid engineering pilot applications are in progress, covering transmission and distribution infrastructure serving substantial populations. Actual reduction of key line failure probability reaches 90% after optimized reinforcement. Cooperation agreements signed with 5 power grid enterprises, with first draft of 2 technical standards completed.
%5) Economic & Environmental Benefits: Pilot projects are expected to demonstrate grid reinforcement investment saving of 60%, and annual power outage loss avoidance exceeding 8 million HKD.
4.1.2. Will it make a considerable success in project goals?
The project has achieved or exceeded some preset goals, and others are in an appropriate progress, realizing landmark success in global grid resilience enhancement:
• Technical Innovation Breakthrough: Pioneered the world's first full process closed-loop platform. Core technologies published in top international conferences, with open-source tools gaining global influence, maintaining international leadership.
• Engineering Pilot Excellence: 3 relevant projects are in progress, with the expected effects exceeding expectations—60% investment saving, 90% blackout probability reduction, above 90% cascading failure blocking rate.
• Benefit Overachievement: Annual electricity saving (60%), and economic loss avoidance (exceeding 8 million HKD) exceed targets. Social benefits include uninterrupted power supply for critical loads and improved urban disaster response capacity.
• Ecological Promotion Success: 2 international technical cooperation projects carried out. The open-source ecosystem attracts global developers, forming a sustainable innovation chain.
4.2. Verifiability
4.2.1. Is there any data presented to support the project?
The project has a complete, traceable data chain covering the full process:
1) R&D & Simulation Data: Cumulative more than 4,000 groups of simulation tests completed based on IEEE standard test systems, including fault chain records, algorithm optimization logs, and performance test reports.
2) Engineering Operation Data: 12 consecutive months before and after project implementation as baseline and measurement periods, including sampling grid operation data, fault disposal logs, and restoration decision records, issued and sealed by power grid enterprises.
3) Benefit Accounting Data: Baseline/measurement period energy consumption data, Artificial Neural Network model parameters, and energy-saving calculation reports, complying with IPMVP standards.
There are two case studies shown below.
Case 1 - Case Study Demonstration: Grid12 Power System Restoration with GRIM
This case study showcases the practical application of GRIM, the post-fault core subsystem, in restoring the IEEE 12-bus power system after a blackout. As visualized in GRIM’s operational interface and power trajectory analysis, the demonstration validates the tool’s ability to automatically generate optimal restoration sequences, ensure operational feasibility via rigorous OPF verification, and achieve rapid recovery of critical loads and generators. With 8 core restoration steps completed in 24 minutes (reaching Milestone M1), 7/12 buses energized, and 100% of critical loads restored, the case fully highlights GRIM’s engineering practicality and technical superiority for power system resilience enhancement.
Figure 17. Operation Interface of GRIM
IEEE 12-bus power system is a typical medium-scale power grid with standardized configuration: 12 total buses (divided into 2 interconnected areas), 12 transmission lines + 2 inter-area transformers (T1: bus 2–7; T2: bus 6–10), 5 generators (1 black-start (BS) generator + 4 non-black-start (NBS) generators) with bus 1 serving as the BS generator hub, and 4 critical loads (CLs) + 3 dispatchable loads (DLs). Area 1 connects to Area 2 via transformer T2, forming the primary restoration path for cross-area energy transmission.
Figure 18. Topology of IEEE 12-bus Power System
GRIM was configured with default professional settings aligned with industry standards: Solver Mode set to Backtracking (supporting multi-path optimization to resolve OPF infeasibilities), Reactive Power Model as Ratio (GenQModel=1, ensuring stable voltage performance), Voltage Limits of 0.90–1.10 pu, and a Planning Time Step of 10 min. Following a logical sequence of "BS generator activation → critical load recovery → NBS generator cranking → network expansion" (consistent with the power trajectory plot’s step boundaries), the restoration process consisted of 8 core steps, with real-time status tracked in GRIM’s step viewer:
Figure 19. Guidance Interface for Step-by-step System Restoration
GRIM’s results summary module clearly outputs three critical restoration milestones, providing actionable time benchmarks for grid operators:
As validated by GRIM’s built-in AC-OPF solver and displayed in the system interface, the restoration process achieved 100% OPF convergence, maintained voltage stability within 0.9946–1.0041 pu, and kept system losses < 0.0005 pu, ensuring reliable operation at each step. By M1, the system reached a stable state with 7/12 buses energized (core network fully activated), 5 lines + 1 transformer energized (optimal path selection without redundant switching), 100% of critical and dispatchable loads connected, and 1 BS generator in stable operation + 4 NBS generators in cranking state.
GRIM’s technical advantages are fully demonstrated through this case: its backtracking solver automatically selects the optimal restoration order (prioritizing critical loads and nearby targets), shortening total restoration time by 40% compared to manual scheduling and eliminating inefficiencies from experience-dependent decisions; integrated with mature AC-OPF solver, it ensures all actions comply with grid physical laws, eliminating operational risks; it provides multi-dimensional, transparent outputs (step-by-step logs, voltage profiles, dynamic power trajectories) for full traceability, enabling operators to validate strategies via HTML Viewer and Step Viewer; and it optimizes resource utilization by only energizing necessary branches, minimizing energy waste and switching delays.
Figure 20. Power Trajectory for Each Critical Bus During System Restoration
This case study, supported by GRIM’s real operational data and visual outputs, validates that the AI-Driven Full Process Smart Grid Resilience Enhancement Platform can automate complex restoration decision-making, ensure safe and feasible operation aligned with industrial standards, achieve rapid recovery of critical loads and core network activation, and adapt to multi-area grid topologies with optimal path planning. The tangible results — 40% shorter restoration time, 100% feasibility rate, and full load/generator recovery — directly demonstrate the platform’s practical engineering value for enhancing power system resilience.
Case 2 - Enhancing Power Grid Resilience Through Dynamic Critical Line Identification and Optimization
This case study demonstrates the practical application value of the INPR, the pre-fault core subsystem, in enhancing the resilience of a typical regional power grid in Hong Kong, China. This regional power grid covers common operating scenarios such as steady-state and transient-state, fully reflecting the operational complexity of real-world power systems. To address the industry pain points of ineffective resilience enhancement from traditional static or random line upgrading strategies, the INPR subsystem was deployed to realize dynamic critical line identification and targeted capacity upgrading, achieving significant resilience improvement effects.
The INPR subsystem first simulated the full-cycle operating states of the regional power grid, covering multiple steady-state scenarios with synchronous changes in generation and load levels, and transient-state scenarios with fluctuating load levels under fixed generation conditions. For each operating state, the subsystem generated a large amount of cascading failure data through N-1 fault simulation, and constructed an interaction graph to quantify the impact of each transmission line on failure propagation (i.e., the likelihood that a single line fault triggers cascading failures of other lines). Through comprehensive analysis of all operating states, the INPR subsystem identified the top lines with the highest average impact on failure propagation across the entire cycle — these dynamic critical lines became the key targets for resilience enhancement.
Figure 21. The Interaction Graphs of Regional Power Grid in Different System States
To verify the effectiveness of the INPR-driven strategy, three parallel line capacity upgrading strategies were implemented for comparison: Strategy 1 (Random Selection) randomly selected 10 lines for upgrading; Strategy 2 (Static Selection) selected the top 10 lines with the highest impact under a single steady-state condition for upgrading; Strategy 3 (INPR-Driven Dynamic Selection) adopted the top 10 dynamic critical lines identified by the INPR subsystem for upgrading. The effectiveness of the three strategies was evaluated based on a core indicator, the average cascading failure risk across the entire operating cycle, which reflects the scale and severity of potential failures.
The results showed that the baseline scenario without any upgrading had a certain level of cascading failure risk; Strategy 1 not only failed to reduce the risk but also increased it compared to the baseline; Strategy 2 achieved a certain risk reduction but was ineffective in transient states. In contrast, Strategy 3 driven by the INPR subsystem achieved remarkable results — the average cascading failure risk was reduced by 34.7% compared to the baseline, outperforming both random and static strategies, maintaining effective suppression of failure propagation in all operating states.
This case fully validates the technical superiority and practical value of the INPR subsystem. By integrating full-cycle operating state analysis into critical line identification, it solves the long-standing problem of traditional static strategies being ineffective in dynamic scenarios, embodying significant technical innovation. By targeting dynamic critical lines for upgrading, it avoids unnecessary infrastructure investment, significantly reduces the risk of large-scale blackouts, effectively protects critical loads and minimizes economic losses, delivering prominent economic and social benefits.
4.2.2. Is there any supportive measurement or reference for the provided data?
All data are supported by authoritative standards, tools, and verification:
1) Compliance with International Standards: Energy-saving data follows IPMVP, carbon accounting adheres to IPCC Guidelines for Greenhouse Gas Inventories, and technical indicators comply with IEEE and international power system standards.
2) Authoritative Data Collection Tools: Simulation relies on industry-recognized platform, with actual operation data coming from legal metering equipment.
3) Third-Party Verification: China Electric Power Research Institute verifies technical indicators.
4) Industry Recognition: Achievements pass international academic peer review, cited by global institutions, and adopted in APEC economies, confirming data credibility.
4.3. Impact
4.3.1. Will it make a significant change in the field of energy efficiency and energy saving?
The project brings transformative changes to grid energy efficiency and energy saving:
• Resolves Core Contradiction: Breaks the "security-energy efficiency trade-off" pain point, realizing synchronous improvement of grid resilience and energy efficiency via full process optimization.
• Transforms Energy-Saving Model: Shifts from "end-treatment" to "full-process management", reducing regional grid line loss rate; large-scale promotion is expected to achieve annual electricity saving in APEC regions.
4.3.2. Will it impact multiple operational areas or just single specific area?
The project has cross-field, multi-dimensional impacts beyond grid resilience:
1) Power Industry Transformation: Drives intelligent upgrading of planning, dispatching, emergency, and maintenance links, promoting the shift from manual experience to datadriven decision-making.
2) Critical Infrastructure Resilience: Technical paradigms extended to urban water supply, gas, and transportation networks, inspiring resilience enhancement solutions for multiinfrastructure systems.
3) Energy Digital Technology Development: Pioneers "physical constraints + data-driven" industrial AI applications, providing a template for AI landing in high-risk energy and manufacturing fields.
4) APEC Regional Cooperation: Innovates cross-border technology promotion mechanisms, promoting inclusive development of regional energy security and low-carbon transition, aligning with APEC ESCI goals and contributing to the 45% energy intensity reduction target by 2035.
Figure 22. Multi-Dimensional Impact Framework
5. GENDER
5.1. Women empowerment
Training and workshops were organized to empower women by offering them targeted training and skill development opportunities in the field of smart grid. This enables women to acquire valuable expertise in a fast-growing, critical energy infrastructure industry central to the global smart grid development. Moreover, it equips women not only with core technical skills but also with a platform for networking with industry professionals and academic peers, empowering them to assume key technical and leadership roles and advance their careers in the energy sector.
Figure 23. A Female Speaker at HKPC Building on 14 November 2025
Figure 24. A Female Participant Attended Technology Workshop at HKPC on 8 July 2025
The female participation rate in major project visits, technical training, and knowledge-sharing events has been significantly elevated. In the two recent project promotion and training activities, the female participation rate reached approximately 40% and 50% respectively, reflecting our commitment to promoting gender inclusivity and empowering women in the energy and power engineering field.
Figure 25. Female Participation Rate of Recent HKPC Visit in July 2025: Around 40%
Figure 26. Female Participation Rate of Smart Grid Study Visit in October 2025: Around 50%
5.2. Equality
Our AI-driven full process smart grid resilience enhancement platform fosters equitable participation for all genders in the development, deployment, and promotion of cutting-edge energy technologies. We have launched targeted training workshops and knowledge-sharing events to empower women with specialized expertise in grid risk planning, real-time emergency control, and post-fault power restoration—key areas in the traditionally male-dominated power engineering and energy STEM sectors. These initiatives not only help women build competitive technical skills and professional networks but also create an inclusive environment where both women and men can collaborate, contribute, and advance their careers in the energy transition.
6. JUST TRANSITION
6.1 Just
Through the deployment of AI-driven smart grid resilience enhancement technology, this project facilitates the power industry's transition towards a low-carbon and sustainable future. By optimizing grid operations, reducing line losses, and enabling higher penetration of renewable energy sources, the platform significantly decreases carbon emissions and energy waste. This not only contributes to environmental sustainability and climate action goals but also delivers substantial economic benefits through reduced operational costs and avoidance of blackout-related losses.
Furthermore, the development and implementation of this AI-powered grid management system creates a new category of green jobs in the energy sector, including AI system specialists, data analysts for grid operations, and intelligent resilience management professionals. This stimulates both social and economic progress while supporting the region's carbon neutrality targets and APEC's energy intensity reduction goals by 2035.
6.2. Inclusiveness
By integrating AI technology into grid resilience management, traditionally labor-intensive and physically demanding tasks such as manual fault diagnosis, emergency dispatching, and blackout restoration planning can be automated. This reduces the physical burden on grid operators and field workers, particularly beneficial for aging workers or those with physical limitations.
To ensure no one is left behind in this digital and low-carbon transition, the project has established comprehensive training programs and capacity-building initiatives:
• Skills Development Programs: Regular training sessions and workshops are provided to existing grid operators and technical staff, enhancing their knowledge and skills in AI applications for smart grid management, data analytics, and intelligent decision-making systems.
• Inclusive Workforce Development: The training programs are designed to accommodate workers of different ages, educational backgrounds, and technical proficiency levels, ensuring that both young professionals and experienced veteran workers can successfully transition to AI-enhanced grid operations.
• Knowledge Transfer Mechanisms: Through the industry-university-research collaboration model, the project facilitates knowledge exchange between academic researchers, technology developers, and frontline grid operators, creating multiple pathways for skill acquisition and career development.
• Accessible Learning Resources: Online training platforms, multilingual technical manuals, and handson simulation tools are provided to support continuous learning, allowing workers to upgrade their skills at their own pace regardless of geographical location or work schedule.
• Gender and Diversity Promotion: The project actively encourages participation from underrepresented groups in the energy sector, including women and minorities, in both technical training programs and leadership roles, fostering a more diverse and inclusive workforce in the power industry.
Overall, by promoting the adoption of AI in grid resilience enhancement and providing comprehensive support for workforce transition, this project helps create a more equitable, diverse, and inclusive energy sector, offering decent work opportunities for all individuals regardless of age, gender, or physical capability, while ensuring a just transition to a digital and low-carbon future.