Kalman Filter Based Improved Beluga Whale Optimization Algorithm

Beluga Whale Optimization (BWO) algorithm is a novel heuristic approach that has shown promising performance in optimization tasks. However, it still exhibits certain limitations in terms of convergence. To address this, we propose an improved version called Kalman Filter-based Beluga Whale Optimization algorithm (KBWO). KBWO algorithm utilizes a logistic growth model to update the balance factor. It facilitates a more balanced distribution of whales engaged in both global and local searches. Moreover, by incorporating the Kalman Filter during the process of state updating, KBWO achieves a more comprehensive exploration of the searching space during the local exploitation stage. Comparative test results demonstrate that the KBWO algorithm outperforms the original BWO algorithm in terms of stability and convergence, enabling it to approach the global optima more effectively. This provides a more reliable solution for applying the Beluga Whale Optimization algorithm to complex optimization problems.