基于最小二乘状态估计和模糊神经的中长期负荷预测

Medium and Long-term Load Forecasting Based on Least Squares State Estimation and Fuzzy Neural Network

可靠准确的中长期电力负荷预测可以使电力公司和发电公司更好地分配配电网络,并有助于在可再生能源并网时提高对电力系统的保护稳定性。为此,提出一种基于加权最小二乘状态估计和模糊神经网络的中长期电力负荷预测方案。基于最小二乘状态估计得到的潮流信息和来自神经网络得到的预测负荷作为模糊神经网络的输入,借由模糊神经网络生成高质量的中长期电力负荷预测结果,最后在IEEE 30总线系统上进行验证和评估。试验结果表明,所提方案可以实现平均绝对百分比误差低于2.55%,比单独的最小二乘状态估计法和模糊神经网络法拥有更低的平均绝对百分比误差。

Reliable and accurate medium and long-term power load forecasting can enable power companies and generation companies to better allocate distribution networks and help them improve the protection and stability of the power system when renewable energy is connected to the grid.To this end,a medium and long-term power load forecasting scheme based on weighted least squares state estimation and fuzzy neural network was proposed.Based on the least squares state estimation,the power flow information obtained and the predicted load obtained from the neural network were used as inputs to the fuzzy neural network.Through the fuzzy neural network,high-quality medium and long-term power load prediction results were generated,and finally validated and evaluated on the IEEE 30 bus system.The experimental results show that the proposed scheme can achieve an average absolute percentage error of less than 2.55%,which is lower than the individual least squares state estimation method and fuzzy neural network method.