基于神经网络-模糊推理综合模型的短期负荷预测

Short-term load forecasting based on the hybrid model of neural network & fuzzy inference

针对由于神经元网络泛化能力不足等原因造成的预测精度不高甚至出现坏数据从而难以适用于负荷波动厉害的电网的情况,提出一种基于神经网络-模糊推理综合模型的短期负荷预测方法。该方法结合了神经网络和模糊推理的优点,通过模糊推理来修正神经网络输出的预测结果,能有效地提高预测精度。特别是对于受天气影响比较明显而天气变化又比较剧烈的电网,能有效防止不合理预测结果的出现。在武汉电网的实际运行情况说明了本算法的有效性。

The ability of generalization of artificial neural networks (ANN) is limited, so it is inevitable that the precision of forecasting result based on ANN is low, sometimes bad. Therefore method based on ANN only can hardly be used to the power grid which load fluctuates hard. A method of short-term load forecasting based on the hybrid model of neural network and fuzzy inference was presented in the paper. The method which combines the advantages of neural network and fuzzy inference can prevent unreasonable predict results and thus greatly improve the precision of forecasting, especially when the load of the power system is remarkably affected by the weather. A factual example in Wuhan power grid shows the validity of the method presented in this paper.