BP神经网络在解决非线性复杂系统中存在很大的优势.针对家庭用电设备自身的负荷特点,以广州供电局用户用电设备能耗数据作为训练样本,利用BP神经网络构建用电设备能耗分析模型,选定能够反映对象特性的能效指标,确定神经元数,构建用户分类指标,依据训练的BP神经网络进行用户划分,实现用户间的能效对比分析.结果表明,模型收敛性较好,所得分析结果绝对误差较小.因此,利用BP神经网络进行用户能效分析的结果具有实用性和有效性.%The BP neural network has great advantage to solve the nonlinear complex system. According to the characteris-tics of the household electricity load itself,the electrical equipment energy consumption data of the users attaching to Guang-zhou Power Supply Bureau is taken as the training sample. The BP neural network is used to construct the energy consumption analysis model of the electrical equipment. The energy efficiency index which can reflect the target features is selected to deter-mine the quantity of the neurons,and construct the users classification index. The users are classified according to the trained BP neural network to realize the energy efficiency contrastive analysis among users. The results show that the model has good convergence,and the analysis result has small absolute error. Therefore,the BP neural network used to analyze the users' ener-gy efficiency has practicability and availability.
展开▼
