基于深度信念网络的供电可靠率预测模型

供电可靠率是供电服务水平的重要指标之一。提出了基于相关性分析和深度信念网络的供电可靠率预测模型。首先利用Pearson系数选取停电次数、最大负荷和用户电费均价系数作为输入特征集。然后将特征集输入到所建立的深度信念网络中，采用逐层无监督训练方法和反向传播训练方法对模型进行参数优化，通过该模型进行供电可靠率预测。最后将所提出的模型与传统人工神经网络、支持向量回归和差分整合移动平均自回归进行比较，结果验证了文章提出的供电可靠率预测模型的有效性。

Power supply reliability forecasting model based on deep belief network

Power supply reliability is an important indicator of power supply service level. A power supply reliability model based on correlation analysis and deep belief network is proposed.Firstly, Pearson coefficient is used to select the outage times, maximum load and average price coefficient of consumer electricity as the input feature set. Then the feature set is put into the established deep belief network, the parameters of the model are optimized by layer- by-layer unsupervised training method and back propagation training method. The model is used to predict the power supply reliability. Finally, the proposed framework is compared with artificial neural network, support vector regression and autoregressive integrated moving average. Simulation results show the effectiveness of the proposed power supply reliability forecasting model.