ACOA-RBF网络模型在短期负荷预测中的应用  

Application of Ant Colony Optimization Algorithm and Radial Basis Function Network Model to Short-Term Load Forecasting

作　　者：王涛[1] 王洋洋[1] 郭长娜[1] 

机构地区：[1]辽宁工程技术大学电气与控制工程学院,葫芦岛125000

出　　处：《计算机系统应用》2012年第1期127-131,共5页Computer Systems & Applications

摘　　要：为了进一步提高RBF神经网络的性能,实现准确、快速预测短期电力负荷的目的,将蚁群优化算法(ACOA)作为RBF神经网络的学习算法,建立了一种新的蚁群优化算法的RBF(ACOA-RBF)网络预测模型,利用山西某地区电网的历史数据进行短期负荷预测。仿真表明,这一算法与传统的RBF神经网络预测方法相比,能达到更好的预测效果。该优化算法改善了径向基神经网络的泛化能力,提高了山西电网短期负荷预测的精度,可有效用于电力系统的短期负荷预测。To improve the capacity of RBF neural network and make short-term load forecasting more accurate and faster,a neural network ant colony optimization algorithm and Radial Basis Function neural network forecasting model is established by using the ant colony optimization algorithm to train the RBF neural network.Using the method and history load data of shanxi power system,the short-term load forecasting was carried out.The simulation results show that the forecasting results by the proposed method are better than those by RBF neural network method.The optimization algorithm improves the RBF neural network generation capacity,and the short-term load forecasting accuracy is improved in Shanxi power system.So it can be effectively used in short-term load forecasting of power system.

关 键 词：蚁群优化算法 径向基神经网络 短期负荷预测 预测精度 隐含层 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TM715[自动化与计算机技术—控制科学与工程]