模型参考自适应-Smith预估控制在发酵温度控制中的应用研究

Application of model reference adaptive control-improved Smith-predictor in temperature control of fermentation

针对存在严重积分纯滞后的生物发酵系统,利用Smith预估器对纯时间延迟具有补偿作用的原理,在对Smith预估器进行改进的基础上,建立了模型参考自适应串级Smith预估器温度控制模型,推导了自适应Smith预估器调节律和设计算法.仿真结果表明,改进Smith预估控制不但降低了纯时间延迟特性对控制性能产生的不利影响,也提高了系统的鲁棒性.模型参考自适应串级Smith预估控制,通过模型输出与被控对象的输出之差来在线调整控制器参数,加快了响应速度,降低了超调,使被控对象更能够适应对象特性或扰动变化多变的特点.

With an aim to address serious integral plus time-delay in biochemical fermentation process, the paper established a model-reference-adaptive cascade Smith predictor temperature control model based on the compensating principle of Smith predictor for time delay process and the improvement of Smith predictor, and deduced the adjust law and design algorithm for adaptive smith predictor. Simulation results indicate that the improved Smith predictor can not only reduce the adverse influence on control performance caused by pure time-delay, but also enhance system robustness. By on-line adjusting controller parameters on the basis of the difference between the model output and the controlled object output, the model-reference-adaptive cascade Smith predictor control can improve response speed and reduce overshoot, thus the controlled object can adapt to the various change from object properties and disturbance more easily.