关键词: 微粒群优化算法, 模拟退火, 神经网络, 软测量

Abstract: A novel cooperative evolutionary algorithm（SAPSO） is proposed by taking advantage of both Particle Swarm Optimization（PSO） and Simulated Annealing（SA） algorithm.It can validly overcome the premature problem in PSO through cooperative search between PSO and SA.Then，SAPSO is employed to train artificial neural network and applied to soft-sensing of gasoline endpoint of delayed coking plant and melt-index of High Pressure Low-density Polyethylene yield.Its performance is compared with existing soft sensor modeling methods.The simulation results show that this model has higher measuring precision as well as better generalization ability，and can satisfy the need of spot measurement.

Key words: Particle Swarm Optimization（PSO） algorithm, Simulated Annealing（SA）, Neural Network（NN）, soft-sensor