电力系统负荷数据预测的设计与实现

Design and Implementation of Load Data Prediction of Power System

为了确保送电质量,提升供电可靠性,通过负荷预测为电力企业的规划策略提供数据依据,从电力负荷数据特性着手,详细分析了影响因素以及预测难点。采用PSO算法对LSTM长短期记忆神经网络进行参数寻优,并综合考虑温度、时段、电价等因素构建电力负荷数据预测模型;采用C#语言设计了包含数据仓库、模型构建与择优、负荷预测模块的C/S模式负荷数据预测平台。通过实际数据验证,预测精度较高,误差在允许范围内,为电力系统负荷数据预测提供了可靠的信息化手段。

In order to ensure the power quality,improve power supply reliability,and provide data through the load forecasting for the electric power enterprise planning strategy,this paper makes a feature analysis of power load data,detailed analyzes of influencing factors and predicts the difficulty,uses PSO algorithm to both short-term and long-term memory neural network tooptimize parameters by considering factors such as temperature,time,electricity prices,and to build power load data forecast model.The C#language is used to design the load data prediction platform of C/S mode,which includes data warehouse,model building and optimization,and load prediction modules.Through the example data verification,the prediction accuracy is high and the error is in the allowable range,which provides a reliable information method for the power system load data prediction.