A novel Kangaroo Escape Algorithm for efficient combined heat and power economic dispatch: Feasibility analysis and validations

This paper introduces a novel Kangaroo Escape Algorithm (KEA), that parallels the survival-driven escape behaviors of Kangaroos integrating adaptive energy control and chaotic perturbation to achieve an effective balance between exploration and exploitation. Unlike conventional algorithms that rely on multiple tuning parameters, KEA operates with only two user-defined settings, offering a simplified yet highly efficient optimization framework. It uses zigzag escape and long jump escape models, with a unique decoy drop mechanism to improve solution diversity and avoid premature convergence. Its innovative mechanisms enable superior performance in solving large-scale Combined Heat and Power Economic Dispatch (CHPED) problems, ensuring both optimality and feasibility of solutions. CHPED complexities arise from the interdependent generation of both thermal and electrical energy, as well as the physical limitations of cogeneration units. The issue structure is further complicated by modelling the transmission losses. The given conditions guarantee a comprehensive evaluation of KEA's optimisation skills in addressing the multimodal CHPED challenge. The proposed KEA's effectiveness and scalability are evaluated using various benchmark scenarios, including small-scale (5–7 generation units), medium-scale 24-unit, and very large 192-unit configurations. The study is also extended to evaluate the proposed KEA under three different heat and electricity demand profiles, thereby simulating realistic operational environments. Detailed statistical indices, including mean, minimum, standard deviation, and success rate, were analyzed to demonstrate the robustness of the optimizer. Moreover, Wilcoxon signed-rank tests confirmed the statistical significance of the improvements over competing methods with p-values well below 0.05, validating the consistency and reliability of the obtained results.