基于神经网络逆系统的发酵过程多变量解耦控制

发酵过程是时变、非线性、不确定的多变量耦合系统，高性能的解耦控制一直是追求的目标。将逆系统方法与神经网络相结合，提出了一种基于神经网络逆系统的发酵过程解耦控制方法。根据发酵过程的特点，给出了相应的数学模型，并证明了系统的可逆性，进一步构造神经网络逆系统并与发酵系统串联复合成伪线性系统，再设计线性闭环调节器实现高性能解耦控制。仿真结果表明，提出的解耦控制方法能够适应过程模型的不确定性和参数的时变性，具有较强的鲁棒性，克服了解析逆系统解耦控制方案依赖于过程模型和对模型参数的变化很敏感的缺点。

Multivariable Decoupling Control Based on Neural Network Inverse System in a Fermentation Process

Fermentation process is a time-variable,nonlinear, uncertain and multivariable coupling system,and high performance decoupling control is a target to seek.A decoupling control strategy based on neural network inverse system for a multivariable fermentation process is proposed,in which the inverse system combines with the neural networks.Based on the characteristics of fermentation process,the model of fermentation system is obtained,and the reversibility of system is testified.Constructing a neural network inverse system and combining it with fermentation process,a pseudo-linear system is completed.Then a linear close-loop adjustor is designed to obtain the good control performance.The simulation experiments demonstrate that good control performance(high accuracy and good robust)can be obtained in multivariable fermentation process based on neural network inverse system,and the disadvantages of inverse system method relied on the exact process model and parameters are overcome.