基于改进深度置信网络的变压器诊断模型

电力变压器作为电网运行中具有重要影响的电器设备，具有输送电能和改变电压的重要作用。为提高变压器诊断方法的准确率，文中提出了基于麻雀搜索算法的深度置信网络的变压器诊断方法。该方法应用深度置信网络对故障进行分类，再应用麻雀搜索算法对深度置信网络的结构参数进行优化。仿真结果表明，与优化前的DBN网络相比，该方法的准确率更高，具有一定的可行性。

Transformer Diagnosis Model Based on Improved Deep Confidence Network

As an important electrical equipment in power grid operation, power transformers play an important role in transmitting electrical energy and changing voltage. In order to improve the accuracy of transformer diagnosis method, a transformer diagnosis method based on deep belief network based on sparrow search algorithm is proposed in this paper. The method uses deep belief network to classify faults, and then applies sparrow search algorithm to optimize the structural parameters of deep belief network. The simulation results show that compared with the DBN network before optimization, the method has higher accuracy and has certain feasibility.