摘要
BP神经网络是故障检测的常用方法,但是在装备故障诊断的运用中,其很难快速有效地获得正确数量的网络层和神经元,而且学习速度较慢,导致检测效率低、稳定性差。为提高某型通信设备的故障诊断效能,论文在以BP神经网络作为诊断算法的基础上,引入麻雀搜索算法(SSA)这种新型优化算法改进BP神经网络参数的选取,优化参数设置,提高检测速度。相关改进算法运用于某型通信设备故障检测,与传统的BP神经网络算法相对比,采用麻雀搜索优化的BP神经网络故障检测算法在诊断效能方面优势明显,为开展装备维护,提供了很好的指导作用,也为后续相关诊断软件平台的开展提供了算法基础。
In the communication equipment fault diagnosis,BP neural network is usually used.But it struggles to swiftly and efficiently obtain the right number of network layers and neurons,as well as having a slow learning rate,which leads to low detec⁃tion efficiency and poor stability.In order to improve the fault diagnosis efficiency of a certain type of communication equipment,this paper introduces a new optimization algorithm-Sparrow Search Algorithm(SSA)to improve the selection of BP neural network pa⁃rameters,optimize the parameter settings,and improve the detection speed on the basis of BP neural network as the diagnostic algo⁃rithm.On contrast with the traditional BP neural network algorithm,the BP neural network fault detection based on sparrow search algorithm has obvious advantages in diagnosis efficiency,which provides a good guidance for further equipment maintenance and provides an algorithm basis for the development of subsequent relevant diagnosis software platform.
作者
姜峰
鞠建波
JIANG Feng;JU Jianbo(Naval Aviation University,Yantai 264000)
出处
《舰船电子工程》
2024年第2期157-160,共4页
Ship Electronic Engineering
基金
军队科研专项(编号:41512322)
国家自然科学基金项目“基于尺度的综合导航系统信息融合方法研究”(编号:60874112)资助。
关键词
故障诊断
麻雀搜索算法
通信设备
设备安全
参数优化
神经元
数据分类
BP神经网络
fault diagnosis
sparrow search algorithm
communication equipment
equipment safety
parameter optimiza⁃tion
neuron data classification
BP neural network
