Distributed power supply output sensing method combined with CNN neural network

The invention provides a distributed power supply output sensing method combined with a CNN (Convolutional Neural Network) neural network, which is characterized in that on the basis of a photovoltaic output decomposition method of minimum mutual information, distributed power supply output is decom...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: HUANGFU WUJUN, KOU QILONG, JI ZHONGHAO, ZHANG JIANGTAO, PENG GANG, DUAN XIAOCHEN, ZHOU LEIYUE, HUANG LEIFENG, GENG XIN, SHE YANJIE, LIU YI, ZHANG MINGKE, KONG XIANGWEN, WANG QING, LIANG SHICHENG
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention provides a distributed power supply output sensing method combined with a CNN (Convolutional Neural Network) neural network, which is characterized in that on the basis of a photovoltaic output decomposition method of minimum mutual information, distributed power supply output is decomposed by further combining the CNN for a contradictory relationship between dynamic response and observation noise when photovoltaic capacity changes, and the distributed power supply output sensing method based on the CNN neural network is obtained. The method has high reliability and effectiveness under the conditions of different photovoltaic permeability and measurement noise, and can complete normal photovoltaic sensing and monitoring functions of the transformer area. 本发明提供一种结合CNN神经网络的分布式电源出力感知方法,其在最小互信息的光伏出力分解方法基础之上,针对光伏容量变化时的动态响应与观测噪声之间的矛盾关系,进一步结合CNN(Convolutional Neural Network)神经网络对分布式电源出力进行分解,本方法在不同光伏渗透率且有测量噪声情况下具有较高的可靠性和有效性,能够完成正常的台区光伏感知、监控功能。