Neural Network Control Of Chemical Reactor System With Constraints

Continuously Stirred Tank Reactor(CSTR)is a widely used reaction instrument in the chemical process.It has the advantages of low cost and strong heat exchange capacity,and so on.However,since many different chemical reactions occur simultaneously in the reactor during the production process,how to control these reactions effectively and practicably becomes the primary problem to be solved.Based on constrained control and adaptive control strategies,this paper uses the modeling characteristics of neural networks or fuzzy logic systems to design a stable intelligent adaptive controller for constrained continuous a constrained continuous stirred reactor to ensure that the temperature,flow and pressure in the kettle remain within a certain range,the parameters are maintained stable,so that the reactants in the kettle are evenly distributed.This article mainly researches from the following two aspects:The continuous stirred reactor system is a typical non-linear system.During the reaction,various parameters such as reactant concentration are prone to fluctuate and are time-varying,the pressure and temperature in the chemical reaction system must be controlled within certain safety limits,and the data is binding.For this reason,an adaptive fuzzy logic control method for the continuous stirred reactor system with time-varying output-constrained nonlinearity is proposed.By constructing a novel time-varying Barrier Lyapunov function and designing an adaptive controller to ensure the stability of the system and all state variables of the system must not violate their constraint bounds.Finally,a simulation study of a non-linear continuous stirred reactor system is performed to verify the effectiveness and feasibility of the proposed control method.For the case of cascade reaction of two reactors,in which the materials in the kettle will inevitably have a time lag during transmission,which will cause the reaction time of the next stage to be delayed accordingly.It has a strong time lag characteristic.At the same time,parameters such as concentration will still fluctuate with it,with time-varying.Therefore,an adaptive neural network control method is proposed for multiple continuous stirred reactor systems with time-varying delays.By constructing a novel time-varying obstacle Lyapunov function and Lyapunov-Krasovsky functional,an adaptive controller is provided to ensure the stability of the system,and the effects of state delays and constraints on the stability of the system are overcome.A simulation study of the continuous stirred reactor system is conducted to verify the effectiveness of the proposed control method.