发电技术 ›› 2020, Vol. 41 ›› Issue (1): 9-18.DOI: 10.12096/j.2096-4528.pgt.19173

• 泛在电力物联网与综合能源系统关键技术 • 上一篇    下一篇

风光水多能源电力系统互补智能优化运行策略

胡伟1(),戚宇辰1,张鸿轩1,董凌2,李延和2   

  1. 1 清华大学电机工程与应用电子技术系, 北京市 海淀区 100084
    2 国网青海省电力公司, 青海省 西宁市 810003
  • 收稿日期:2019-11-26 出版日期:2020-02-29 发布日期:2020-03-03
  • 作者简介:胡伟(1976),男,副教授,博士生导师,研究方向为电力系统稳定与控制、电力大数据及人工智能、综合能源系统, huwei@mail.tsinghua.edu.cn|戚宇辰(1996),男,硕士研究生,研究方向为综合能源系统优化运行
  • 基金资助:
    国家重点研发计划项目(2017YFB0902200);国家电网公司科技项目(5228001700CW)

Complementary Intelligent Optimization Operation Strategy of Wind-Solar-Hydro Multi-energy Power System

Wei HU1(),Yuchen QI1,Hongxuan ZHANG1,Ling DONG2,Yanhe LI2   

  1. 1 Department of Electrical Engineering, Tsinghua University, Haidian District, Beijing 100084, China
    2 State Grid Qinghai Electric Power Company, Xining 810003, Qinghai Province, China
  • Received:2019-11-26 Published:2020-02-29 Online:2020-03-03
  • Supported by:
    National Key Research and Development Program of China(2017YFB0902200);Science and Technology Project of State Grid Corporation of China(5228001700CW)

摘要:

高比例可再生能源接入电网,采用风光水互补发电可以提供平滑和稳定的电力供应。针对风光水多能源电力系统,采用大数据和人工智能技术,提出基于随机规划的短期优化运行方法。首先,基于变分自编码器(variational autoencoder,VAE)提出了可再生能源场景生成方法,能够生成符合可再生能源出力特性的多样化场景,并精确刻画可再生能源出力的相关性。其次,基于场景法建立了风光水互补短期优化运行模型,并采用分段线性化的方法将多种非线性约束转化为线性约束,可以保证模型精度的同时实现快速求解。最后,通过对雅砻江下游风光水发电基地的算例仿真,验证了所提的智能优化运行策略的有效性。

关键词: 电力系统, 风光水互补, 场景生成, 变分自编码器, 优化运行

Abstract:

For high-ratio renewable energy access to the grid, the utilization of wind-solar-hydro complementary power generation can provide a smooth and stable power supply. This paper proposed a short-term optimal operation method based on stochastic programming for the windsolar-hydro multi-energy power system, using big data and artificial intelligence technology. Firstly, based on the variational autoencoder (VAE), a renewable energy scenarios generation method was proposed, which can generate diverse scenarios that meet the characteristics of renewable energy output and accurately describe the correlation of renewable energy output. Secondly, based on the scenarios method, a short-term optimization operation model of wind-solar-hydro was established and a piecewise linearization method was used to transform various nonlinear constraints into linear constraints, which can achieve fast solution while ensuring the accuracy of the model. Finally, through the simulation of the wind-solarhydro power system in downstream of the Yalong River, it demonstrates the effectiveness of the intelligent optimization operation strategy proposed in this paper.

Key words: power system, complementation of windsolar-hydro, scenarios generation, variational autoencoder, optimal operation

中图分类号: