Scholarlink

In this paper, SPM chaotic mapping, weighted randomization, and reverse learning are integrated to improve the sparrow search algorithm (SSA), which alleviates the local optimal problem of basic SSA and improves its global search ability and stability. Then, this paper introduced mixer structure, Uout, and other mechanisms to improve the multi-layer perceptron model (MLP), applied the improved sparrow search algorithm (ISSA) to optimize the initial parameters of the mixed multi-layer perceptron model (MMLP), and obtained a better anomaly detection model of power control system. The experimental results show that improved SSA optimization improved MLP (ISSA-MMLP) has better performance.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Anomaly Detection Method of Power Control System Based on Improved SSA Optimization and MLP

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xiaoyu Wang,
  • Wenhai Liu,
  • Chongyao Wang,
  • Xinhao Mao

摘要

In this paper, SPM chaotic mapping, weighted randomization, and reverse learning are integrated to improve the sparrow search algorithm (SSA), which alleviates the local optimal problem of basic SSA and improves its global search ability and stability. Then, this paper introduced mixer structure, Uout, and other mechanisms to improve the multi-layer perceptron model (MLP), applied the improved sparrow search algorithm (ISSA) to optimize the initial parameters of the mixed multi-layer perceptron model (MMLP), and obtained a better anomaly detection model of power control system. The experimental results show that improved SSA optimization improved MLP (ISSA-MMLP) has better performance.