电器与能效管理技术 ›› 2024, Vol. 0 ›› Issue (3): 30-35.doi: 10.16628/j.cnki.2095-8188.2024.03.005

• 研究与分析 • 上一篇    下一篇

基于蛇算法优化的改进RBF神经网络的航天电磁继电器贮存寿命预测方法

李久鑫, 王召斌, 朱佳淼   

  1. 江苏科技大学 自动化学院, 江苏 镇江 212003
  • 收稿日期:2023-10-02 出版日期:2024-03-30 发布日期:2024-04-22
  • 作者简介:李久鑫(1998—),男,硕士研究生,研究方向为继电器可靠性及寿命预测。|王召斌(1982—),男,副教授,研究方向为电器贮存可靠性、加速试验及寿命预测技术。|朱佳淼(1999—),男,硕士研究生,研究方向为航天电器小子样情况下可靠性评估问题。
  • 基金资助:
    国家自然科学基金项目(51507074);江苏省研究生科研与实践创新计划资助项目(KYCX23_3875)

Storage Life Prediction Method of Aerospace Electromagnetic Relay with Improved RBF Neural Network Based on Snake Algorithm Optimization

LI Jiuxin, WANG Zhaobin, ZHU Jiamiao   

  1. College of Automation,Jiangsu University of Science and Technology, Zhenjiang 212003, China
  • Received:2023-10-02 Online:2024-03-30 Published:2024-04-22

摘要:

针对航天电磁继电器的接触电阻预测和预测精度问题,提出了一种基于蛇优化(SO)算法改进BRF神经网络的模型。在传统径向基函数(RBF)模型基础上,通过SO算法对其权值参数进行优化,从而更好地预测继电器接触电阻值。基于SO-RBF模型与RBF模型、GA-RBF模型分别预测接触电阻,对比分析预测结果,表明所提模型具有较高的预测精度。

关键词: RBF神经网络, 退化试验, 贮存, 继电器

Abstract:

Aiming at the prediction and prediction accuracy of contact resistance of aerospace electromagnetic relays,a radial basis function (BRF) neural network model based on snake optimization (SO) algorithm is proposed.On the basis of the traditional RBF model,the SO algorithm is used to optimize the weight parameters so as to better predict the contact resistance value of the relay.The constructed SO-RBF prediction model is compared with RBF model.The models are used to predict the change trend of contact resistance.The comparison and analysis of the prediction results show that the proposed model has high prediction accuracy.

Key words: radial basis function (RBF) neural network, degradation test, storage, relay

中图分类号: