Scholarlink

<p>In the area of energy management and optimization, economic dispatch (ED) has received considerable interest. Its central goal is to minimize electricity dispatch costs without violating physical operating constraints. Due to their robust optimization capability and straightforward applicability, metaheuristic algorithms (MAs) have become a popular choice for tackling ED problems in recent decades. Among them, differential evolution (DE) consistently demonstrated significant effectiveness and produced the best results on various optimization problems. It has been utilized across a wide kind of areas, owing to its ease of execution, robustness, and rapid convergence. However, DE is not more successful at escaping the local-optimum trap (i.e., insufficient exploration and excessive exploitation) when solving hard optimization problems. To address these limitations and effectively solve ED problems, this paper introduces a modified DE (mDE). It adopted (i) an innovative particle swarm optimization-based mutation strategy with adaptive parameter settings to improve exploration efficiency, (ii) a modified crossover mechanism for better exploitation, and (iii) a refined selection process to elude local optima and sustain diversity balance. Presentation of the offered mDE process is validated on 13 standard test problems. In addition, to assess the mDE successes, six ED generating units (3, 6, 13, 15, 40, and 140) test systems were optimized. The experimental findings reveal that mDE provides highly reliable and competitive results over existing optimization techniques.</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 modified strategy of the differential evolution algorithm for economic dispatch

  • 期刊论文

  • 发表日期：

    2026年7月20日

查看全文
  • Pooja Tiwari,
  • Vishnu Narayan Mishra,
  • Raghav Prasad Parouha

摘要

In the area of energy management and optimization, economic dispatch (ED) has received considerable interest. Its central goal is to minimize electricity dispatch costs without violating physical operating constraints. Due to their robust optimization capability and straightforward applicability, metaheuristic algorithms (MAs) have become a popular choice for tackling ED problems in recent decades. Among them, differential evolution (DE) consistently demonstrated significant effectiveness and produced the best results on various optimization problems. It has been utilized across a wide kind of areas, owing to its ease of execution, robustness, and rapid convergence. However, DE is not more successful at escaping the local-optimum trap (i.e., insufficient exploration and excessive exploitation) when solving hard optimization problems. To address these limitations and effectively solve ED problems, this paper introduces a modified DE (mDE). It adopted (i) an innovative particle swarm optimization-based mutation strategy with adaptive parameter settings to improve exploration efficiency, (ii) a modified crossover mechanism for better exploitation, and (iii) a refined selection process to elude local optima and sustain diversity balance. Presentation of the offered mDE process is validated on 13 standard test problems. In addition, to assess the mDE successes, six ED generating units (3, 6, 13, 15, 40, and 140) test systems were optimized. The experimental findings reveal that mDE provides highly reliable and competitive results over existing optimization techniques.