GANO算法下广域电力系统短期负荷预测仿真  

Simulation of Short Term Load Forecasting in Wide Area Power System Based on GANO Algorithm

作　　者：李文武[1] 张李勇 张鹏宇 LI Wen-wu;ZHANG Li-yong;ZHANG Peng-yu(School of Electrical and New Energy,China Three Gorges University,Yichang Hubei 443002,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002

出　　处：《计算机仿真》2025年第1期92-95,110,共5页Computer Simulation

基　　金：梯级水电站运行与控制湖北省重点实验室(三峡大学)开放基金(2019KJX08)。

摘　　要：用电模式的复杂程度随着电力市场和电网技术的发展逐渐增加,在此背景下提高了对电力系统短期负荷预测稳定性和精度的要求。提出GANO算法下广域电力系统短期负荷预测方法,建立自编码器,将电力系统的历史负荷数据输入自编码器中,通过数据重构实现电力负荷数据的去噪处理;分别建立了用于广域电力系统短期负荷预测的GM(1,1)模型和神经网络模型,为了提高负荷预测精度,结合GM(1,1)模型和神经网络模型的预测结果,建立灰色神经网络预测组合预测模型(GANO),实现电力系统短期负荷预测。仿真结果表明,在预测精度和预测效率方面,所提方法表现出良好的性能。Gradually,the complexity of electricity consumption patterns has increased with the development of power markets and grid technologies,which has raised the requirements for stability and accuracy of short-term load forecasting in power systems.In this context,a method for forecasting short-term load in a wide-area power system was proposed based on the GANO algorithm.Firstly,an autoencoder was designed,and then historical load data of the power system was input into the autoencoder.Secondly,the electric load data was denoised by reconstructing the data.Moreover,the GM(1,1)model and neural network model were separately built for short-term load forecasting in wide-area power systems.To improve the load forecasting accuracy,a Grey Neural Network Integrated Forecasting Model(GANO)was constructed by the prediction results of the GM(1,1)model and neural network model.Finally,shortterm load forecasting in the power system was achieved.Simulation results show that the proposed method has good performance in terms of prediction accuracy and efficiency.

关 键 词：广域电力系统 自编码器 负荷预测 神经网络模型 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]