浙江大学梅竞成等：一种通过学习分支求解机组检修问题的方法

原创英文期刊编辑部全球能源互联网期刊 2022-09-08 08:00

摘要

为了最大限度地提高电力系统可靠性，本文考虑电力系统网络安全约束和机组检修工作合理性约束对机组检修问题进行了建模。鉴于机组检修问题的计算复杂性，本文基于支持向量机的变量选择方法来求解0-1混合整数规划问题。该方法通过搜集强分支策略的决策数据，然后基于支持向量积学习一个函数模仿强分支策略函数。然后，在后续求解过程中采用学习拟合到的排序函数进行分支变量选择。测试算例表明，本文提出的基于机组检修问题分支定界过程提出的特征学习到的分支变量选择方法能够在保证求解精度的基础上提高机组检修问题的求解速度。

Learning to branch in the generation maintenance scheduling problem

一种通过学习分支求解机组检修问题的方法

Jingcheng Mei¹, Jingbo Hu², Zhengdong Wan³, Donglian Qi¹

(1.College of Electrical Engineering, Zhejiang University, Hangzhou 310027, P. R. China

2.Jiaxing Hengchuang Electric Power Group Co., Jiaxing 314000, P. R. China

3.Energy Development Research Institute, China Southern Power Grid, Guangzhou 510623, P. R. China )

收听作者1分钟语音介绍

Learning to branch in the generation maintenance scheduling

Abstract

To maximize the reliability index of a power system, this study modeled a generation maintenance scheduling problem that considers the network security constraints and rationality constraints of the generation maintenance practice in a power system. In view of the computational complexity of the generation maintenance scheduling model, a variable selection method based on a support vector machine (SVM) is proposed to solve the 0–1 mixed integer programming problem (MIP). The algorithm observes and collects data from the decisions made by strong branching (SB) and then learns a surrogate function that mimics the SB strategy using a support vector machine. The learned ranking function is then used for variable branching during the solution process of the model. The test case showed that the proposed variable selection algorithm — based on the features of the proposed generation maintenance scheduling problem during branch-and-bound — can increase the solution efficiency of the generation-scheduling model on the premise of guaranteed accuracy.

Keywords

Generation maintenance scheduling, Support vector machine (SVM), Variable selection, Strong Branching (SB).

Fig.1 Flow of the algorithm

Fig.2 System annual load profile

Fig.3 Generation maintenance schedule for 52 weeks

Fig.4 Comparison of the computation time

Fig.5 Comparison of the system with Gap=0.1%

Fig.6 Comparison of the absolute value of the system reserve difference

本文引文信息

Mei JC, Hu JB, Wan ZD, et al. (2022) Learning to branch in the generation maintenance scheduling problem. Global Energy Interconnection, 5(4): 409-417

梅竞成，胡景博，万正东，等 (2022) 一种通过学习分支求解机组检修问题的方法. 全球能源互联网（英文）, 5(4): 409-417

Biographies

Jingcheng Mei

Jingcheng Mei is currently pursuing his Ph.D. degree at the College of Electrical Engineering, Zhejiang University, Hangzhou, China. His current research interests include maintenance optimization with power systems.

Jingbo Hu

Jingbo Hu was admitted to a Master of Engineering program in Electronics and Communication Engineering from Zhejiang University in September 2018. Since 2014, he has been working in power supply companies in cities and counties, and he currently serves as the deputy general manager of the Huachuang branch. He specializes in the dispatching of control automation; the operation, maintenance, and overhaul of the distribution network; material planning management; science and technology management; energy big data analysis; and new energy operation.

Zhengdong Wan

Zhengdong Wan received his Bachelor’s degree from Huazhong University of Science and Technology, Wuhan in 2007, and Master’s degree from Harbin Institute of Technology, Harbin, in 2009. He is working at the Energy Development Research Institute, China Southern Power Grid, Guangzhou. His research interests include economic engineering and power grid engineering cost.

Donglian Qi

Donglian Qi is currently a professor at the Zhejiang University. Her current research focus is on cyber physical power systems (CPPSs).

编辑：王彦博

审核：王伟

郑重声明

根据国家版权局相关规定，纸媒、网站、微博、微信公众号转载、摘编本网站作品，需包含本网站名称、二维码等关键信息，并在文首注明《全球能源互联网》原创。个人请按本网站原文转发、分享。

Learning to branch in the generation maintenance scheduling problem

浙江大学 梅竞成等：一种通过学习分支求解机组检修问题的方法

郑重声明

浙江大学梅竞成等：一种通过学习分支求解机组检修问题的方法