Scholarlink

<p>This paper proposes a secure, bidirectional control framework for a separately excited DC motor over a wireless link, targeting modern IoT applications such as automated guided vehicles (AGVs), drones, and industrial IoT systems. The core of the approach is a novel genetic algorithm-enhanced self-organizing double function-link brain emotional learning controller (GA-SO-DFL-BELC), which extends the standard SO-DFL-BELC by (i) integrating a genetic algorithm to optimize learning rates for faster convergence and improved stability, and (ii) embedding a disturbance observer (DOB) to actively estimate and compensate for model uncertainties and external perturbations. For communication security, both PWM control signals and speed feedback are encrypted through two synchronized 4D chaotic systems—operating in a master–slave configuration—to counteract cyberattacks. The SO-DFL-BELC structure combines the brain emotional learning model with a double function-link network for high weight accuracy, and a self-organizing mechanism that adjusts neuron count dynamically to handle abrupt input changes with reduced computational cost. Simulation and hardware tests on three ESP32 microcontrollers confirm robust chaotic synchronization, high encryption security, and feasible closed-loop operation, making the proposed method highly suitable for secure Internet of Things (IoT) control systems.</p>

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

Secure Bidirectional Wireless Control of a DC Motor Using 4D Chaotic Systems Based on a Self-Organizing Double Function-Link Brain Emotional Learning Controller Optimized by Genetic Algorithm and Disturbance Observer

  • 期刊论文

  • 发表日期：

    2026年2月17日

查看全文
  • Ngoc Ha An,
  • Duc Hung Pham,
  • Van Nam Giap,
  • Xuan Thang Trinh,
  • Chih-Min Lin

摘要

This paper proposes a secure, bidirectional control framework for a separately excited DC motor over a wireless link, targeting modern IoT applications such as automated guided vehicles (AGVs), drones, and industrial IoT systems. The core of the approach is a novel genetic algorithm-enhanced self-organizing double function-link brain emotional learning controller (GA-SO-DFL-BELC), which extends the standard SO-DFL-BELC by (i) integrating a genetic algorithm to optimize learning rates for faster convergence and improved stability, and (ii) embedding a disturbance observer (DOB) to actively estimate and compensate for model uncertainties and external perturbations. For communication security, both PWM control signals and speed feedback are encrypted through two synchronized 4D chaotic systems—operating in a master–slave configuration—to counteract cyberattacks. The SO-DFL-BELC structure combines the brain emotional learning model with a double function-link network for high weight accuracy, and a self-organizing mechanism that adjusts neuron count dynamically to handle abrupt input changes with reduced computational cost. Simulation and hardware tests on three ESP32 microcontrollers confirm robust chaotic synchronization, high encryption security, and feasible closed-loop operation, making the proposed method highly suitable for secure Internet of Things (IoT) control systems.