Optimization algorithms are essential for solving complex problems in engineering and data science. This study introduces an improved algorithm called the Levy Energy Valley Optimizer (LEVO), which combines the principles of the original Energy Valley Optimizer (EVO) with Levy flight. EVO, a metaheuristic algorithm, draws inspiration from particle stability and decay processes, aiming to optimize solutions by guiding particles toward a state of stability.
LEVO enhances EVO's capabilities by incorporating Levy flight, a random walk technique that enables particles to make large jumps in the search space. This jump allows for more extensive exploration, helping particles escape local optima and ensuring a broader search across the solution space. Tested on fifteen benchmark functions, LEVO showed superior performance in both unimodal and multimodal functions compared to other optimization algorithms, including Genetic Algorithms, Particle Swarm Optimization, and Cuckoo Search.
The application of LEVO is particularly relevant for high-dimensional and complex optimization tasks, where traditional algorithms may struggle to find global optima efficiently. With LEVO, researchers can achieve high accuracy in problem-solving while reducing computational requirements.
full Text: https://www.igminresearch.com/articles/html/igmin172
DOI link: 10.61927/igmin172
Comments
Post a Comment