发电技术

• •    下一篇

电力系统智能计算的关键技术及应用展望




张俊1,蒲天骄2,高文忠1,刘友波3,裴玮4,许沛东1,高天露1,白昱阳1   

  1. 1.武汉大学电气与自动化学院,湖北省 武汉市 430072;2.中国电力科学研究院有限公司,北京市 海淀区 100192;3.四川大学电气工程学院,四川省 成都市 610065;4.中国科学院电工研究所,北京市 海淀区 100080
  • 基金资助:
    新一代人工智能国家科技重大专项(No.2021ZD0112700)

Key Technologies and Prospects for Application of Intelligent Computing in Power Systems

ZHANG Jun1, PU Tianjiao2, GAO Wenzhong1, LIU Youbo3, PEI Wei4, XU Peidong1, GAO Tianlu1, BAI Yuyang1   

  1. 1. School of Electrical Engineering and Automation, Wuhan University, No.299 Bayi Road, Wuhan 430072, China; 2. China Electric Power Research Institute, Haidian District, Beijing 100192, China; 3. School of Electrical Engineering, Sichuan University, Chengdu 610065, Sichuan Province, China; 4. Institute of Electrical Engineering, Chinese Academy of Science, Haidian District, Beijing 100080, China

摘要: 【目的】随着我国“双碳”目标所驱动的新型电力系统的建设和应用,以高比例可再生能源和高电力电子化等为特征的电力系统展现出前所未有的复杂动态特性,给电力系统安全稳定运行提出新的挑战。电力系统智能计算利用新一代人工智能技术,尤其是大模型技术、预训练技术等,实现了物理模型、数据驱动模型、知识模型的融合,形成了一种新型的电力系统计算分析方法。因此综述和展望电力系统智能计算方法和技术在新型电力系统分析、优化、运行、调度、控制等方面的应用。【方法】首先,阐述了电力系统智能计算的概念和关键技术;然后,以电力系统时序预测、安全域优化、多能微网协同优化等关键领域的智能化处理为例阐述了其技术路线;最后,展望了其应用前景。【结论】所提出的智能化算法在多个评估指标上优于传统方法,显著提高了电力系统的预测精度和控制效率,验证了智能计算在电力系统计算分析中的应用潜力,为新型电力系统提供了一种有效的智能化技术路线,对于电力系统的智能化发展具有重要意义。[1]



基金项目:新一代人工智能国家科技重大专项(No.2021ZD0112700)。

Project Supported by the National Key Research and Development Program of China (No. 2021ZD0112700).

关键词: 智能计算, 新型电力系统, 大模型, 人工智能, 电网运行控制, 优化调控, 智能评估, 关键技术

Abstract:

[Objectives] With the construction and application of new power systems driven by my country's "double carbon" goals, power systems characterized by high proportions of renewable energy and high power electronics have shown unprecedented complex dynamic characteristics, providing the safe and stable operation of the power system. New challenges. Power system intelligent computing utilizes a new generation of artificial intelligence technology, especially large model technology, pre-training technology, etc., to realize the fusion of physical models, data-driven models, and knowledge models, forming a new type of power system calculation and analysis method. Therefore, the application of power system intelligent computing methods and technologies in new power system analysis, optimization, operation, scheduling, control, etc. is reviewed and prospected. [Methods] First, the concept and key technologies of intelligent computing in power systems are elaborated; then, the technical route is illustrated by taking intelligent processing in key areas such as power system timing prediction, security region optimization, and multi-energy microgrid collaborative optimization as examples; Finally, its application prospects are prospected. [Conclusions] The proposed intelligent algorithm is superior to traditional methods in multiple evaluation indicators, significantly improving the prediction accuracy and control efficiency of the power system, verifying the application potential of intelligent computing in power system calculation and analysis, and providing a new power system. Effective intelligent technical route is of great significance to the intelligent development of the power system.

Key words:

text-align:justify, ">  , intelligent computing, new power system, large model, artificial intelligence, operation and control of power systems, optimization and dispatching, intelligence evaluation, key technology