海洋微生物酶反应器智能控制系统的研制

    Development of intelligent control system for bioreactor of marine microbial enzymes

    • 摘要: 生物发酵过程具有严重非线性、高度时变性、高阶多变量和大型不确定的特点,为了实现生物发酵过程的自动化控制,提高生物技术产品的生产水平,使得发酵过程的参数检测、操作监视、自动控制等智能化,分析了影响发酵过程的主要因素以及各变量之间的耦合关系,综合应用传统的PID控制方法、模糊神经网络技术,构建了一种多变量模糊神经控制系统的前馈解耦算法并将其应用在发酵过程的思想,同时采用了冷凝、回收、利用发酵尾气技术,解决了尾气排放、罐体泄漏染菌的问题。模糊神经控制器和解耦部分独立设计,在模糊控制器中引入神经网络,解耦网络采用一层隐层,利用简化的学习算法,根据系统输出误差,在线调整网络权值,从而实现动态解耦而无需辨识被控对象的模型。该方法结构简单且计算量小,经实际应用结果表明这种控制算法具有很好的控制效果。

       

      Abstract: The bio-fermentation process has highly nonlinear, time-varying, higher and more variable and large uncertain characteristics. In order to achieve automatic control of the fermentation process, improve the level of production of biotechnology products, and make the detection of fermentation process parameters, operation monitoring and automatic control intelliget, we constructed a feedforward decoupling method for the multi-variable fuzzy neural control system, with analysising of the main factors affecting the fermentation process as well as the variable coupling and applicating the integrated application of the traditional PID control and fuzzy neural network technology. And the method is applicated in the fermentation process. While the problem of emissions and bacterial contamination of the tank leakage are solved with using the condensation, recovery, and the tail gas of fermentation technology. Fuzzy neural controller and decoupling are independent designed. Neural Network is introduced in the fuzzy controller, and the decoupling network applys a layer of hidden layer. Network weights are on-line adjusted based on a simplified learning algorithm according to the system output error.with the purpose of dynamic decoupling without the need to identify the controlled object model. The method is simple and computation, and the actual application results show that this algorithm has good control effect.

       

    /

    返回文章
    返回