基于改进粒子群优化技术的拜耳法物料平衡计算

Bayer material balance computation based on improved particle swarm optimization algorithm

分析了氧化铝生产工艺中物料平衡计算的特点,研究了拜耳法生产工艺流程和拜耳法物料平衡计算的数学模型.针对传统粒子群算法(PSO)存在的不足,给出了具有变异因子并能很好跳出局部最优解的新型粒子群算法(IPSO),并应用于氧化铝生产中的物料平衡计算.计算结果表明:优化后的粒子群算法具有较强的全局搜索能力和较高的收敛精度,是进行拜耳法物料平衡计算的有效方法.

This paper firstly analyzed the characteristic of the material balance computation （MBC） in alumina production, and Bayer process was briefly introduced. Then, math model of MBC was presented. In order to solve the problems of easily falling into local optimum solution and slow convergence speed of the traditional PSO, an improved particle swarm optimization （IPSO） with stochastic mutation was proposed based on the gathering degree and the steady degree. During the iterating process, the mutation probability of the current particle was determined by the means of all the particlefs fitness, the gathering degree and the steady degree. The exploration ability was efficiently improved by the mutation, and the probability of falling into local optimumwas greatly decreased. The practical results of the MBC show that the new algorithm was better than the traditional PSO with both a better stability and a steady convergence. Most importantly, results demonstrateed that IPSO was more feasible and efficient in practical application, and also shed new light on the further improvement of PSO.