针对电力日最高负荷受多种因素影响,变化趋势复杂,难以通过建立准确的数学模型进行预测的问题,提出灰色动态模型对电力日最高负荷进行预测,在此基础上构造了灰色神经网络组合预测模型。该模型避免了变权组合预测模型的主观与繁琐,能有效地将灰色预测弱化数据序列波动性的优点和神经网络较强的非线性适应能力相融合。算例结果表明该方法的可行性和有效性,预测精度也得到了改善。 In view of the fact that the maximum daily load of electric power is affected by many factors and the trend of change is complex, it is difficult to predict by establishing an accurate mathematical model. The gray dynamic model is proposed to predict the maximum load of electric power day. On this basis, the gray neural network model. This model avoids the subjectivity and cumbersomeness of variable weight combination forecasting model, and can effectively combine the advantages of gray forecasting weakening data sequence volatility with the strong nonlinear adaptability of neural network. The results show that the method is feasible and effective and the prediction accuracy is also improved.