Periodic Reporting for period 1 - SPOCEI (Smart Planning, Operation and Control for Energy Internet)
Berichtszeitraum: 2022-01-19 bis 2024-01-18
In the project “SPOCEI: Smart Planning, Operation and Control for Energy Internet”, we investigate the problems of the structure planning, distributed optimal control for energy internet. To this end, the researcher focuses on pioneering work for energy internet to achieve intelligent, distributed, reliable, flexible and cooperative system operation via reasonably planning the overall system structure, designing distributed model-free control methods and developing a trading mechanism with a novel distributed optimization algorithm.
Such topics are important, since the society and European Union’s (EU) requires increasing the renewable energy resources in the global energy mix. It is encouraged to develop distributed and smart technologies, e.g. flexibility markets, next generation of low-carbon energy technologies and the application of AI. The emerging energy internet provides potential solutions to developing sustainable energy resources, hybrid energy utilization model, flexible energy management and secure system control, which fits well with the society requirement and the EU energy development strategy. Our research topics are the key issues in energy internet, which are eager to be solved.
Objectives of SPOCEI have been to : 1) controllable structure planning for energy internet under different environments (before and after fault) from the concept of controllability view; 2) distributed and adaptive cooperated control strategies for energy internet in the scenario that no expert knowledge and no requirement for system model are available; 3) multiple timescale power trading model and distributed optimization algorithm to achieve fully peer-to-peer energy trading in the processes of energy-generation, energy-consumption and energy-delivery.
Such topics are important, since the society and European Union’s (EU) requires increasing the renewable energy resources in the global energy mix. It is encouraged to develop distributed and smart technologies, e.g. flexibility markets, next generation of low-carbon energy technologies and the application of AI. The emerging energy internet provides potential solutions to developing sustainable energy resources, hybrid energy utilization model, flexible energy management and secure system control, which fits well with the society requirement and the EU energy development strategy. Our research topics are the key issues in energy internet, which are eager to be solved.
Objectives of SPOCEI have been to : 1) controllable structure planning for energy internet under different environments (before and after fault) from the concept of controllability view; 2) distributed and adaptive cooperated control strategies for energy internet in the scenario that no expert knowledge and no requirement for system model are available; 3) multiple timescale power trading model and distributed optimization algorithm to achieve fully peer-to-peer energy trading in the processes of energy-generation, energy-consumption and energy-delivery.
Work was conducted via 6 work packages (WPs)
WP1 refers to the project management. The researcher managed the whole project and reported the process to the supervisor through two-weekly meeting, determining on all aspects of the project.
WP2 refers to the training and career development. The researcher performed diverse actions: 1) wrote a mandatory professional development plan following the requirement of University of Oslo; 2) talked with Vice Head of Department at Department of Informatics about the future plan; 3) lectured “Deep Reinforcement Learning for Demand Response and Energy Management” in University of Oslo to increase the lecturing experience; 4) helped to supervise 2 PhD students to improve the supervised ability. With those effort, the researcher promotes to the Assistant Professor and will join in Aalborg University after the fellowship is ended.
WP3 refers to the research on controllable structure planning. We have established typical structures and proposed structure planning method by designing a mixed maximum matching and deep deterministic policy gradient method. Research studies yielded one conference paper that receives the Best Paper Award from 2023 ICCSIE.
WP4 refers to the research on distributed and adaptive control strategies. We have proposed the data-driven load forecasting method, and several distributed and adaptive control strategies, which enables energy internet to obtain the optimal control operation without requiring systems’ dynamics and models. Research studies yielded 2 journal publications, with additional one magazine and one journal papers under review.
WP5 refers to research on distributed power trading mechanism. We have established several energy trading models and designed diverse distributed algorithms to achieve multiple timescale energy trading. Research studies yielded 2 journal publications, with additional two journal papers under review. With the achievements in research, the researcher received the 2023 Excellent Young Expert Award from journal of Modern Power Systems and Clean Energy (MPCE).
WP6 refers to dissemination and communication. The researcher performed diverse actions: 1) gave tutorials at the 11th ISGT-Asia, and invited talk at CAA Youth e-Summit and three universities; 2) co-organized one special section at the 4th SPIES, one track at the 33th ISIE 2024, one workshop at the 15th SmartGridComm, and one on-line seminar; 3) served as the guest editors of IET Electronics Letter and Frontiers in Energy Research, served as reviewers for several journals, and received the 2023 Excellent Reviewer Award from MPCE. ; 4) served as the Student Video / PHD Thesis Competition Chair of the 15th SmartGridComm; 5) attended a lot of seminars organized by University of Oslo.
WP1 refers to the project management. The researcher managed the whole project and reported the process to the supervisor through two-weekly meeting, determining on all aspects of the project.
WP2 refers to the training and career development. The researcher performed diverse actions: 1) wrote a mandatory professional development plan following the requirement of University of Oslo; 2) talked with Vice Head of Department at Department of Informatics about the future plan; 3) lectured “Deep Reinforcement Learning for Demand Response and Energy Management” in University of Oslo to increase the lecturing experience; 4) helped to supervise 2 PhD students to improve the supervised ability. With those effort, the researcher promotes to the Assistant Professor and will join in Aalborg University after the fellowship is ended.
WP3 refers to the research on controllable structure planning. We have established typical structures and proposed structure planning method by designing a mixed maximum matching and deep deterministic policy gradient method. Research studies yielded one conference paper that receives the Best Paper Award from 2023 ICCSIE.
WP4 refers to the research on distributed and adaptive control strategies. We have proposed the data-driven load forecasting method, and several distributed and adaptive control strategies, which enables energy internet to obtain the optimal control operation without requiring systems’ dynamics and models. Research studies yielded 2 journal publications, with additional one magazine and one journal papers under review.
WP5 refers to research on distributed power trading mechanism. We have established several energy trading models and designed diverse distributed algorithms to achieve multiple timescale energy trading. Research studies yielded 2 journal publications, with additional two journal papers under review. With the achievements in research, the researcher received the 2023 Excellent Young Expert Award from journal of Modern Power Systems and Clean Energy (MPCE).
WP6 refers to dissemination and communication. The researcher performed diverse actions: 1) gave tutorials at the 11th ISGT-Asia, and invited talk at CAA Youth e-Summit and three universities; 2) co-organized one special section at the 4th SPIES, one track at the 33th ISIE 2024, one workshop at the 15th SmartGridComm, and one on-line seminar; 3) served as the guest editors of IET Electronics Letter and Frontiers in Energy Research, served as reviewers for several journals, and received the 2023 Excellent Reviewer Award from MPCE. ; 4) served as the Student Video / PHD Thesis Competition Chair of the 15th SmartGridComm; 5) attended a lot of seminars organized by University of Oslo.
This MSCA project has advanced energy internet planning, operation, and control in three main areas:
1. Structure Planning: In WP3, we discovered that loss of controllability is a key factor in energy internet cascading failures. Our new structure planning model, integrating network controllability and economic operation, is pioneering in using a controllability index for this purpose. This model has the potential to significantly reduce the risk of large-scale failures.
2. Control Strategy: WP 4 introduced novel distributed and adaptive control strategies for the energy internet. Compared with the existing studies, the proposed methods, fully distributed and not requiring expert knowledge or system models, are particularly effective for energy networks with high renewable energy penetration. This achievement has the potential to accelerate the digital decision-making.
3. Trading Mechanism. In WP5, we presented unique distribute optimization methods to achieve the optimal energy generation/consumption and optimal energy flow allocation. Compared with the existing studies, our proposed methods can simultaneously hold the features of initialization-free, distributed implementation, asynchronous communication, and strong robustness to cyberattacks. In addition, we also established the theory foundation. With this effort, this project generated new distributed power trading method and theory to solve a class of new distributed optimization problem.
The outputs of the project possess good potential impacts, since our research are capable of contributing to two UN Sustainable Development Goals (SDGs): 1) Affordable and Clean Energy (SDG 7); 2) Climate Action (SDG 11). Specifically, we developed distributed and mode-free control method to deal with the uncertainty of the renewable energy and enhance the system adaptivity. Meanwhile, we designed distributed trading mechanisms that are beneficial for the local utilization and integration of renewable energy sources. Thus, the outputs of the project provide potential solutions to increase substantially the share of renewable energy in the global energy mix. It fits well with the 2th targets of SDG 7. Moreover, we opened up new method to reconfigure the energy internet from the concept of controllability view against system cascading failure. Thus, it can effectively strengthen resilience and adaptive capacity to climate-related hazards and natural disasters, which fits well with the 1th target of SDG 11.
1. Structure Planning: In WP3, we discovered that loss of controllability is a key factor in energy internet cascading failures. Our new structure planning model, integrating network controllability and economic operation, is pioneering in using a controllability index for this purpose. This model has the potential to significantly reduce the risk of large-scale failures.
2. Control Strategy: WP 4 introduced novel distributed and adaptive control strategies for the energy internet. Compared with the existing studies, the proposed methods, fully distributed and not requiring expert knowledge or system models, are particularly effective for energy networks with high renewable energy penetration. This achievement has the potential to accelerate the digital decision-making.
3. Trading Mechanism. In WP5, we presented unique distribute optimization methods to achieve the optimal energy generation/consumption and optimal energy flow allocation. Compared with the existing studies, our proposed methods can simultaneously hold the features of initialization-free, distributed implementation, asynchronous communication, and strong robustness to cyberattacks. In addition, we also established the theory foundation. With this effort, this project generated new distributed power trading method and theory to solve a class of new distributed optimization problem.
The outputs of the project possess good potential impacts, since our research are capable of contributing to two UN Sustainable Development Goals (SDGs): 1) Affordable and Clean Energy (SDG 7); 2) Climate Action (SDG 11). Specifically, we developed distributed and mode-free control method to deal with the uncertainty of the renewable energy and enhance the system adaptivity. Meanwhile, we designed distributed trading mechanisms that are beneficial for the local utilization and integration of renewable energy sources. Thus, the outputs of the project provide potential solutions to increase substantially the share of renewable energy in the global energy mix. It fits well with the 2th targets of SDG 7. Moreover, we opened up new method to reconfigure the energy internet from the concept of controllability view against system cascading failure. Thus, it can effectively strengthen resilience and adaptive capacity to climate-related hazards and natural disasters, which fits well with the 1th target of SDG 11.