Multi-objective Optimization Model of Forest Spatial Structure Based on Dynamic Multi-Group PSO Algorithm

The multi-objective optimization problem, as one of the most popular hotspots in the current research, is facing both a big opportunity and a great challenge. Multi-group particle swarm optimization is often used to solve multi-objective optimization problems, however, the multi-group particle swarm optimization algorithm is more commonly used in solving static multi-objective optimization problems and less frequently used in solving dynamic multi-objective optimization problems. In solving dynamic multi-objective optimization problems, the algorithm lacks the corresponding environment detection mechanism. In this work, Dynamic Multi-Group Particle Swarm Optimization Algorithm is proposed and verified to solve multi-objective optimization problems of forest spatial structure. The proposed algorithm introduces a new mechanism for environmental detection, which can sense the changes of the situated environment and make the multi-objective optimization results more suitable for the dynamic actual situation. The results show that the average generation distance (\(\stackrel{-}{GD}\)) of the proposed algorithm is less than 0.0753, and the average metric of maximum spread (\(\stackrel{-}{MS}\)) is greater than 0.9852. The spatial structure of the target forest has been optimized three times. The mingling intensity, volume, neighborhood comparison, and open ratio are increased by 40.6%, 2.5%, 18.9%, and 11.8%, respectively; while the competition index and angle index are decreased by 37.9% and 13.1%, respectively. Obviously, the various indicators of the forest improved by our scheme are better than those before optimization.