基于一维卷积神经网络的某网区月度负荷预测研究

Research on the Monthly Load Forecast of a Power Grid Based on One-dimensional Convolutional Neural Network

对网区月度最大负荷预测是调控方式专业的常规工作,对涉及转供负荷的线路和设备进行有效的负荷预测也是实施停电计划工作的关键,本文通过搭建卷积神经网络,预测下一个月的每日最大负荷及月度最大负荷,通过人工智能算法优化停电计划编排,验证了该方法提高月度最大负荷预测准确度的可行性。

The maximum monthly load forecast of power grid is a routine work of operation mode dispatching & control.And effective load forecast of lines and equipment involved in load transfers is also the key to the implementation of power outage planning.A convolutional neural network to forecast the maximum daily load and the maximum monthly load of the next month is built in this paper.An artificial intelligence algorithm is used to optimize the schedule of power outage plans.The feasibility of the method to improve the forecast accuracy of maximum monthly load is verified.