摘要
实现了BP神经网络电力负荷预测模型和小波神经网络电力负荷预测模型。通过对两种神经网络的算法进行理论分析以及两种模型的预测结果比较发现,小波神经网络在神经网络节点数目相同的情况下,小波神经网络比BP神经网络具有更高的预测精度。小波神经网络是一种建立在小波理论基础上的一种新型前馈神经网络,具有许多优良特性。文中所指的小波神经网络的优点,例如所需网络节点少和预测精度高,已经在电力负荷预测中得到验证。表明小波神经网络模型预测精度高,自适应性好,收敛速度也明显快。
Achieve a BP neural network load forecasting model and wavelet neural network load forecasting model. By analyzing two nanral network algorithms and comparing two model forecasting result,it shows that when they have same number of network node, wavelet neural network is better than BP network in forecast accuracy. Wavelet neural network is a new feedforward neural network based on wavelet theory with many advantage. It points out that the advantages of WNN, such as requiring less network nodes and achieving accu- rate forecasting,are validated in power load forecasting research. Wavelet neural network model shows that the prediction accuracy is high, the adaptability is good, the convergence speed is significantly fast.
出处
《计算机技术与发展》
2012年第10期237-241,共5页
Computer Technology and Development
基金
云南省自然科学基金项目(2009CD028)
昆明理工大学科学研究基金(201001)
