基于径向基函数和模糊逻辑的短期电力负荷预测

综合考虑温度、日期类型和天气等因素对短期电力负荷的影响,提出了将径向基(RBF)网络和模糊逻辑相结合的预测方法。利用具有非线性逼进能力的RBF神经网络预测出预测日的最大负荷值和最小负荷值,并用模糊逻辑预测出预测日的负荷系数,进而得到预测日的负荷值。实际算例表明:该方法同BP网络相比,具有较高的预测精度,证明了该方法的有效性。

Short-term electric load forecasting based on radial basis function networks and fuzzy logic

A load forecasting strategy based on the radial basis function(RBF) neural network and fuzzy logic is proposed considering the comprehensive effect of temperature, date type and weather on the short-term electric load. The peak load and the least load in one day are inferred by using the RBF network and taking advantage of its nonlinear convergence, and the load coefficient of the day are approximated by means of fuzzy logic, then the load values of the day are got. An application example is given and the presented method is compared with the method which is based on the back-propagation(BP) network and fuzzy logic,and the results show that the precision of short-term load forecasting by the presented method is much high,so the presented forecasting strategy is very effective.