粒子群算法的进化式实时优化方法

针对传统实时优化方法需要等待稳态以及模型更新和优化计算复杂等缺点,提出了一种进化式实时优化方法.该方法将模型更新与控制器设定点优化计算表达为一个多目标的进化过程,把扰动后等待稳态的过程划分为若干个“拟稳态”区间,并根据测量值进行连续的模型更新和优化计算.采用粒子群多目标优化算法对模型参数和操作参数进行统一编码和进化操作,详细给出了算法的实现步骤.经Williams-Otto反应器为对象进行了算法的仿真和分析,证明了该优化方法的有效性.

PSO-based Evolutionary Real-time Optimization Approach

It is acknowledged that traditional real-time optimization approaches may suffer certain deficiencies such as tremendous waiting time for steady states and complexities concerning model updates and optimization computations.In response to these challenges,an evolutionary real-time optimization method is proposed.With this method,model update and controller set-point optimization calculation are expressed as a multi-objective evolutionary procedure;and the process corresponding waiting time for steady state ...