Solving the Economic Load Dispatch Problem by Attaining and Refining Knowledge-Based Optimization

The Static Economic Load Dispatch (SELD) problem is a paramount optimization challenge in power engineering that seeks to optimize the allocation of power between generating units to meet imposed constraints while minimizing energy requirements. Recently, researchers have employed numerous meta-heuristic approaches to tackle this challenging, non-convex problem. This work introduces an innovative meta-heuristic algorithm, named “Attaining and Refining Knowledge-based Optimization (ARKO)”, which uses the ability of humans to learn from their surroundings by leveraging the collective knowledge of a population. The ARKO algorithm consists of two distinct phases: attaining and refining. In the attaining phase, the algorithm gathers knowledge from the population’s top candidates, while the refining phase enhances performance by leveraging the knowledge of other selected candidates. This innovative way of learning and improving with the help of top candidates provides a robust exploration and exploitation capability for this algorithm. To validate the efficacy of ARKO, we conduct a comprehensive evaluation against eleven other established meta-heuristic algorithms using a diverse set of 41 test functions of the CEC-2017 and CEC-2022 test suites, and then, three real-life applications also verify its practical ability. Subsequently, we implement ARKO to optimize the SELD problem considering several instances. The examination of the numerical and statistical results confirms the remarkable efficiency and potential practical ability of ARKO in complex optimization tasks.