摘要
电力系统负荷预测通过对历史数据分析,预测未来需求。本文先用小波聚类对数据进行负荷分类,再分别用经典的遗传算法、Elman神经网络算法、小波-神经网络算法和组合智能算法建立预测模型。通过比较以上几种短期电力负荷预测模型的仿真结果,验证了混合智能算法可以大大增强负荷预测结果的准确性和可靠性,具有良好的应用前景。
Power system load forecasting through the historical data analysis, forecast future demand. In this paper, we use wavelet clustering to load data. Then we use the classical genetic algorithm, Elman neural network, wavelet neural network and combined intelligent algorithm to build the forecasting model. By comparing the simulation results of several short-term power load forecasting models, the hybrid intelligent algorithm can greatly enhance the accuracy and reliability of the load forecasting results, and has good application prospect.
出处
《电子测试》
2015年第10期104-105,共2页
Electronic Test
基金
安徽工程大学国家级大学生创新创业训练计划"短期电力负荷预测研究"资助
项目编号:AH201210363220
