基于MPSO-RBF混合优化的过热汽温神经网络自适应控制

Neural Network Adaptive Control for Superheated Steam Temperature Based on MPSO-RBF Hybrid Optimization

提出了基于改进PSO算法的RBF神经网络混合优化(MPSO-RBF)方法,该方法将改进PSO算法的全局搜索能力和RBF神经网络局部优化的高效性相融合,克服了普通PSO算法收敛的不稳定性和RBF网络易陷入局部极小值的缺点.针对具有较大惯性和滞后的非线性系统构造出一个基于MPSO-RBF混合优化方法的带输入迟延链的复合神经网络自适应控制系统(MPSO-NNC),针对某超临界600MW直流锅炉高温过热器的过热汽温控制进行了仿真试验,并与GA-RBF和Smith预估控制效果进行了对比,结果表明该方法具有更好的性能指标.

A hybrid optimization algorithm（MPSO-RBF） for radial basis function（RBF） neural network based on modified particle swarm optimization（MPSO） was presented,this method may take full advantage of the global searching performance of MPSO and the local optimized effectiveness of RBF neural network,and it will overcome general PSO algorithm convergent instability and the disadvantage of RBF network with falling into local minimum.In allusion to the property of big inertia,time delay and non-linear system,a compound neural networks adaptive control system with input delay chain was constructed based-on modified particle swarm optimization（MPSO-NNC）,at the same time the typical nonlinear system simulation experiments were done,also this method will be a contrast to RBF neural network based-on genetic algorithm（GA-RBF） and Smith pre-estimated control effects,finally the results have proved that MPSO-RBF method has better performance index.