微型电脑应用

为了提高输电线路设备异常状态检测的准确性，设计一种基于粒子群算法的输电线路设备集中供电异常状态检测方法。设计输电线路设备边缘智能终端，采集集中供电异常状态相关数据。基于长短时记忆神经网络、排列熵以及集合经验模态分解设计数据缺失修复模型，修复缺失数据。基于BP神经网络与粒子群算法设计异常状态检测算法，实现输电线路设备集中供电异常状态检测。测试结果表明：设计方法的供电异常状态检测均方误差较低，整体低于0.015;在异常状态个数达到40时，设计方法的迭代次数仍低于400次。

To improve the accuracy of abnormal state detection for transmission line equipment, this article designs a centralized power supply abnormal state detection method for transmission line equipment based on particle swarm optimization algorithm. The article designs an edge intelligent terminal for transmission line equipment to collect data related to abnormal state of centralized power supply. A data missing repair model is constructed based on long short-term memory neural networks, permutation entropy and ensemble empirical mode decomposition is used to repair missing data. An abnormal state detection algorithm is designed based on BP neural network and particle swarm optimization algorithm to achieve centralized power supply abnormal state detection for transmission line equipment. The test results show that the mean squared error of power supply abnormal state detection of the designed method is lower than 0.015 as a whole. When the number of abnormal states reaches 40, the number of iterations of the designed method is still less than 400.