Combination optimization method of grid sections based on deep reinforcement learning with accelerated convergence speed

被引:1
作者
Zhao, Huashi [1 ]
Wu, Zhichao [2 ]
He, Yubin [1 ]
Fu, Qiujia [1 ]
Liang, Shouyu [1 ]
Ma, Guang [1 ]
Li, Wenchao [1 ]
Yang, Qun [2 ]
机构
[1] China Southern Power Grid Dispatching & Control Ct, Guangzhou, Peoples R China
[2] Nanjing Univ Aeronaut & Astronaut, Coll Comp Sci & Technol, Coll Artificial Intelligence, Coll Software, Nanjing, Peoples R China
关键词
grid section; deep reinforcement learning; convergence speed; discretize; loss function; annealing optimization algorithm; OPTIMAL POWER-FLOW; ECONOMIC-DISPATCH;
D O I
10.3389/fenrg.2023.1269854
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
A modern power system integrates more and more new energy and uses a large number of power electronic equipment, which makes it face more challenges in online optimization and real-time control. Deep reinforcement learning (DRL) has the ability of processing big data and high-dimensional features, as well as the ability of independently learning and optimizing decision-making in complex environments. This paper explores a DRL-based online combination optimization method of grid sections for a large complex power system. In order to improve the convergence speed of the model, it proposes to discretize the output action of the unit and simplify the action space. It also designs a reinforcement learning loss function with strong constraints to further improve the convergence speed of the model and facilitate the algorithm to obtain a stable solution. Moreover, to avoid the local optimal solution problem caused by the discretization of the output action, this paper proposes to use the annealing optimization algorithm to make the granularity of the unit output finer. The proposed method in this paper has been verified on an IEEE 118-bus system. The experimental results show that it has fast convergence speed and better performance and can obtain stable solutions.
引用
收藏
页数:9
相关论文
共 25 条
  • [1] Hybrid Chaotic Maps-Based Artificial Bee Colony for Solving Wind Energy-Integrated Power Dispatch Problem
    Alshammari, Motaeb Eid
    Ramli, Makbul A. M.
    Mehedi, Ibrahim M.
    [J]. ENERGIES, 2022, 15 (13)
  • [2] Ardakani AJ, 2018, 2018 POWER SYSTEMS COMPUTATION CONFERENCE (PSCC)
  • [3] Solution to non-convex economic dispatch problem with valve point effects by incremental artificial bee colony with local search
    Aydin, Dogan
    Ozyon, Serdar
    [J]. APPLIED SOFT COMPUTING, 2013, 13 (05) : 2456 - 2466
  • [4] Optimal power flow by enhanced genetic algorithm
    Bakirtzis, AG
    Biskas, PN
    Zoumas, CE
    Petridis, V
    [J]. IEEE TRANSACTIONS ON POWER SYSTEMS, 2002, 17 (02) : 229 - 236
  • [5] Diehl F., 2019, P 33 C NEUR INF PROC, P1
  • [6] Gherbi F., 2011, P 2011 INT C COMM CO, P1
  • [7] Real-time Decision Making for Power System via Imitation Learning and Reinforcement Learning
    Guo, Lei
    Guo, Jun
    Zhang, Yong
    Guo, Wanshu
    Xue, Yanjun
    Wang, Lipeng
    [J]. 2022 IEEE/IAS INDUSTRIAL AND COMMERCIAL POWER SYSTEM ASIA (I&CPS ASIA 2022), 2022, : 744 - 748
  • [8] Economic dispatch with network and ramping constraints via interior point methods
    Irisarri, G
    Kimball, LM
    Clements, KA
    Bagchi, A
    Davis, PW
    [J]. IEEE TRANSACTIONS ON POWER SYSTEMS, 1998, 13 (01) : 236 - 242
  • [9] Investigation on New Metaheuristic Algorithms for Solving Dynamic Combined Economic Environmental Dispatch Problems
    Larouci, Benyekhlef
    Ayad, Ahmed Nour El Islam
    Alharbi, Hisham
    Alharbi, Turki E. A.
    Boudjella, Houari
    Tayeb, Abdelkader Si
    Ghoneim, Sherif S. M.
    Abdelwahab, Saad A. Mohamed
    [J]. SUSTAINABILITY, 2022, 14 (09)
  • [10] Lei Jiang, 2021, 2021 IEEE 5th Conference on Energy Internet and Energy System Integration (EI2), P2883, DOI 10.1109/EI252483.2021.9713491