Low efficiency and the tendency to fall into local optimum are the major obstacles to web service composition optimization. In this paper, we propose a multi-objective optimization algorithm based on the subgroup stratified coarse-grained model to improve the performance of web service composition optimization. Compared to the general particle swarm optimization algorithms, the proposed algorithm improves population structure, dynamically adjusts evolution strategy and increases the local extremum's perturbation. We demonstrate a solution to the constrained multi-objective web service composition optimization problem based on the proposed algorithm. Theoretical analysis and experimental results show that it improves the performance of web service composition optimization.

低效率和陷入局部最优的趋势是Web服务组成优化的主要障碍。在本文中，我们提出了一种基于子组分层粗粒度模型的多目标优化算法，以提高Web服务组合优化的性能。与一般的粒子群优化算法相比，该算法改善了种群结构，动态调整了进化策略，增加了局部极值的扰动。我们提出了一种基于所提出算法的约束多目标Web服务组合优化问题的解决方案。理论分析和实验结果表明，该方法提高了Web服务组合优化的性能。