Scholarlink

<p>This research presents a novel hybrid optimization technique combining the Crayfish Optimization Algorithm (COA) and Hamiltonian Deep Neural Networks (HDNN) for improving power quality in electrical systems. The technique optimizes both series and shunt active power filters in Unified Power Quality Conditioners (UPQC), significantly reducing Total Harmonic Distortion (THD) to 1.2%, compared to 3.2% and 2.2% achieve by Particle Swarm Optimization (PSO) and Ant Lion Optimization (ALO), respectively. The method also mitigates voltage sags and swells, ensuring enhanced voltage stability under dynamic load conditions. This work fills a gap in previous studies by offering a scalable, robust solution that can be applied across industrial, commercial, and residential systems, thereby improving power quality and supporting sustainable energy practices. The results highlight the COA-HDNN technique as an effective alternative to traditional methods in mitigating power quality issues.</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

Enhancing power quality in UPQC systems using crayfish optimization and deep neural power networks

  • 期刊论文

  • 发表日期：

    2025年11月4日

查看全文
  • T. Hajmunnisa,
  • N. P. Subramaniam

摘要

This research presents a novel hybrid optimization technique combining the Crayfish Optimization Algorithm (COA) and Hamiltonian Deep Neural Networks (HDNN) for improving power quality in electrical systems. The technique optimizes both series and shunt active power filters in Unified Power Quality Conditioners (UPQC), significantly reducing Total Harmonic Distortion (THD) to 1.2%, compared to 3.2% and 2.2% achieve by Particle Swarm Optimization (PSO) and Ant Lion Optimization (ALO), respectively. The method also mitigates voltage sags and swells, ensuring enhanced voltage stability under dynamic load conditions. This work fills a gap in previous studies by offering a scalable, robust solution that can be applied across industrial, commercial, and residential systems, thereby improving power quality and supporting sustainable energy practices. The results highlight the COA-HDNN technique as an effective alternative to traditional methods in mitigating power quality issues.