基于BP神经网络的电力系统日负荷预测，模型采用24个输入，24个输出，训练次数为10000，训练目标为0.5，学习速率设置为0.1。(Load forecasting based on BP neural network, the model uses 24 inputs, 24 outputs, the number of training is 10000, the training target is 0.5, the learning rate is set to 0.1.)