Scholarlink

<p>This paper presents the Cerebellum-ANN Intelligent Controller (CAIC), a novel controller designed to enhance the dynamic performance of a Space Vector Modulation based Direct Torque Control (SVM-DTC) of an induction motor (IM) drive. The CAIC uniquely integrates the adaptive learning capabilities of the human cerebellum with the sophisticated sensory data interpretation strengths of artificial neural networks (ANNs). Traditional SVM-DTC controllers struggle to adapt to fluctuating conditions, often resulting in suboptimal performance; the CAIC addresses these limitations and a key deficiency in conventional cerebellum controllers – the imprecise estimation of sensory signals – through an embedded ANN to enhance sensory estimation and signal generation. This integration enhances system robustness and refines the control signals, thus optimizing performance. Through simulations using the IM under conditions of constant speed and torque, variable speed with constant torque, variable torque with constant speed and motor reversing, the superior performance of CAIC over a cerebellum and PI controller was observed across key metrics. Specifically, compared to conventional PI controllers the settling time for the speed response was improved to 0.5824&#xa0;s from 1.225&#xa0;s, rise time became 0.4313&#xa0;s from 0.9441&#xa0;s, overshoots reduced from before 0.2553% to became 0.0011%, initial transient torque peaks mitigated to 366.4 N-m from before 443.582 N-m. CAIC significantly reduced the current Total Harmonic Distortion (THD) from before 140.34% to now 22.05% and compared with state of art cerebellum controller the settling time improved to 0.5824&#xa0;s from 0.9802&#xa0;s and a significant drop was observed for current THD to 22.05% from 39.56%. Moreover, the CAIC showcased amplified resilience to system disturbances and uncertainties, thereby confirming its reliability as an ideal control solution for induction motors.</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

A Novel Intelligent Controller for Enhancing the Dynamic Performance of an SVM-DTC Based Induction Motor Drive

  • 期刊论文

  • 发表日期：

    2025年8月16日

查看全文
  • Manikanta Raju Velpula,
  • Venkateswara Rao Veluvolu

摘要

This paper presents the Cerebellum-ANN Intelligent Controller (CAIC), a novel controller designed to enhance the dynamic performance of a Space Vector Modulation based Direct Torque Control (SVM-DTC) of an induction motor (IM) drive. The CAIC uniquely integrates the adaptive learning capabilities of the human cerebellum with the sophisticated sensory data interpretation strengths of artificial neural networks (ANNs). Traditional SVM-DTC controllers struggle to adapt to fluctuating conditions, often resulting in suboptimal performance; the CAIC addresses these limitations and a key deficiency in conventional cerebellum controllers – the imprecise estimation of sensory signals – through an embedded ANN to enhance sensory estimation and signal generation. This integration enhances system robustness and refines the control signals, thus optimizing performance. Through simulations using the IM under conditions of constant speed and torque, variable speed with constant torque, variable torque with constant speed and motor reversing, the superior performance of CAIC over a cerebellum and PI controller was observed across key metrics. Specifically, compared to conventional PI controllers the settling time for the speed response was improved to 0.5824 s from 1.225 s, rise time became 0.4313 s from 0.9441 s, overshoots reduced from before 0.2553% to became 0.0011%, initial transient torque peaks mitigated to 366.4 N-m from before 443.582 N-m. CAIC significantly reduced the current Total Harmonic Distortion (THD) from before 140.34% to now 22.05% and compared with state of art cerebellum controller the settling time improved to 0.5824 s from 0.9802 s and a significant drop was observed for current THD to 22.05% from 39.56%. Moreover, the CAIC showcased amplified resilience to system disturbances and uncertainties, thereby confirming its reliability as an ideal control solution for induction motors.