Multi-objective forest succession planning algorithm: a new metaheuristic algorithm for multi-objective operation optimization of marine power plants

To address the issues that traditional multi-objective algorithms are prone to falling into local optima and suffer from insufficient population diversity in the maneuverability and noise reduction optimization of marine power plants, this paper proposes a novel metaheuristic algorithm based on the principles of forest ecosystem succession—the multi-objective forest succession planning algorithm (MO-FSPA). The algorithm innovatively introduces optimization strategies including a dynamic weighted crowding distance calculation method and a directed elite migration mechanism driven by nonlinear equilibrium to simulate key ecological stages such as pioneer tree species colonization, forest fire disturbance diffusion, and human intervention, thereby achieving a dynamic balance between global exploration and local exploitation. To comprehensively evaluate the algorithm’s performance, the DTLZ, ZDT, and WFG standard test suites are employed, and comparative experiments are conducted against NSGA-II, NSGA-III, MOEA/D, MOPSO, and SPEA2, with verification using the Wilcoxon rank-sum test, Friedman global ranking, Holm correction, and critical difference diagrams. The results demonstrate that MO-FSPA significantly outperforms the comparison algorithms in terms of IGD, HV, and SP metrics. In the multi-objective engineering optimization problem of maneuverability and noise reduction for marine power plants, MO-FSPA also achieves a superior Pareto front, providing an efficient novel solution approach for strongly coupled, high-dimensional, and heavily constrained marine engineering optimization problems.