Multi-Lever EarlyWarning forWind and Photovoltaic Power Ramp Events Based on Neural Network and Fuzzy Logic  

作　　者：Huan Ma Linlin Ma Zengwei Wang Zhendong Li Yuanzhen Zhu Yutian Liu 

机构地区：[1]Power Grid Technology Center,State Grid Shandong Electric Power Research Institute,Jinan,250003,China [2]Shandong Electric Power Dispatching and Control Center,State Grid Shandong Electric Power Company,Jinan,250012,China [3]Key Laboratory of Power System Intelligent Dispatch and Control of Ministry of Education,Shandong University,Jinan,250061,China

出　　处：《Energy Engineering》2024年第11期3133-3160,共28页能源工程（英文）

基　　金：funded by State Grid Shandong Electric Power Company Technology Project(520626220110).

摘　　要：With the increasing penetration of renewable energy in power system,renewable energy power ramp events(REPREs),dominated by wind power and photovoltaic power,pose significant threats to the secure and stable operation of power systems.This paper presents an early warning method for REPREs based on long short-term memory(LSTM)network and fuzzy logic.First,the warning levels of REPREs are defined by assessing the control costs of various power control measures.Then,the next 4-h power support capability of external grid is estimated by a tie line power predictionmodel,which is constructed based on the LSTMnetwork.Finally,considering the risk attitudes of dispatchers,fuzzy rules are employed to address the boundary value attribution of the early warning interval,improving the rationality of power ramp event early warning.Simulation results demonstrate that the proposed method can generate reasonable early warning levels for REPREs,guiding decision-making for control strategy.

关 键 词：Early warning machine learning power system security renewable energy power ramp event smart grid 

分 类 号：TM61[电气工程—电力系统及自动化]