Abstract: A novel multi-object Particle Swarm Optimization algorithm（PSO） based on supervising and controlling principles of endocrine system is proposed.According to the modulation principle of controlling method between stimulation hormones and releasing hormones，considering the supervision and controlling of individual in the set of non-dominated for the nearest class of swarm，the global optimization position of class is introduced in the position updating process of particles.The new position of particles is not only determined by the best position which it achieves so far and the global best position in current generation，but also influenced by the best position of class，so the global information and local information is combined completely.The new method of updating for particle swarm is designed.In order to validate the effectiveness of given method three benchmark multi-objective problems are simulated by the given method，MOPSO and NSGA-II，the results indicate that the given method has better performance in convergence and distribution than MOPSO and NSGA-II.

Key words: endocrine system, non-dominated solution, global optimization of the class, Particle Swarm Optimization algorithm（PSO）

摘要： 借鉴内分泌系统的监督控制机制，提出了一种基于内分泌思想的多目标粒子群优化算法。根据生物体激素调节机制中促激素和释放激素间的相互作用原理，考虑当前非劣解集中的个体对其最临近一类群体的监督控制，在粒子位置的更新过程中，引入当前粒子的类全局最优位置，使粒子位置的更新不仅受粒子运动到当前的最好位置和当前代粒子的最好位置影响，而且受其所属类中最好位置粒子的影响，实现粒子全局信息和局部信息的结合。设计了新的粒子群更新方案，为验证方法的有效性，对几个典型的多目标优化问题进行了仿真实验，通过与MOPSO、NSGA-II两种方法的结果对比，表明算法有较好的收敛性和分布性。

关键词: 内分泌系统, 非劣解, 类全局最优, 粒子群优化算法