Study Of Anaerobic Fermentation Reactor Based On Fuzzy Neural Network PID Control

China's livestock and poultry industry is developing rapidly,the annual output of livestock and poultry manure has reached 5 billion tons,is the world's largest production of the country.Animal and poultry feces is rich in nutrition,but also contains arsenic,cobalt,phosphorus and other harmful substances,which makes the harmless treatment of feces and resource utilization has become an important subject of research.Anaerobic fermentation is an energy-based treatment and one of the most effective treatments for the degradation of organic matter.There are many factors affecting anaerobic fermentation,such as temperature,humidity,p H,carbon-nitrogen ratio,load rate,etc.,which effectively control these factors in the optimal range of anaerobic fermentation,improve the fermentation process and effect is meaningful.In the above background,this topic began to design intelligent anaerobic fermentation tank,aimed at improving the effect and rate of anaerobic fermentation,to achieve anaerobic fermentation automation.This topic starts with algorithmic innovation and control platform,and carries on the research on the internal environmental control scheme of anaerobic fermentation tank with anaerobic fermentation tank as the object.Because anaerobic fermentation is accompanied by biological reaction and chemical reaction,the internal environmental control of fermentation tank has a great lag,time variability and nonlinearity.Therefore,a PID control algorithm based on fuzzy neural network is proposed,which combines the advantages of PID algorithm,fuzzy algorithm and neural network algorithm,and constitutes an advanced intelligent control algorithm.In order to ensure the stability and reliability of the anaerobic tank control platform,the Siemens S7-200 SMART of programmable logic controller PLC(Programmable Logic Controller)is used as the core,and the upper machine MCGS(Monitor and Control Generated System's producer interface validates the feasibility of the design.The main research content is as follows:(1)Anaerobic fermentation reactor tank design.The process of anaerobic fermentation is studied,and the overall structure design is carried out according to the characteristics,influencing factors,control structure and equipment selection,including the main design of the tank,the insulation device,the stirring device,the selection and layout arrangement of multi-point sensor,and the design of the servo mixing system.(2)Control the selection of the algorithm.Because the process of anaerobic fermentation is a typical multivariable,large lag,nonlinear process,its accurate mathematical model is difficult to obtain.In this paper,the fuzzy control theory is introduced,the parameters are blurred and controlled,and finally the fuzzy output control parameters are unsolvable and the adaptive adjustment of fuzzy PID control is realized.Because of the problems of fuzzy PID control itself,the neural network algorithm is optimized,first of all,the fuzzy reasoning system FIS is established in Matlab software,the three parameters of PID Kp,Ki and Kd are trained in neural network,and the Simulink module is used for simulation and debugging.The fuzzy neural network program is written by Matlab software to achieve temperature control.Considering that the process of anaerobic fermentation is accompanied by gas production,BP neural network is used to predict the gas production of anaerobic fermentation.(3)Control platform construction.Using PLC to design anaerobic fermentation tank control system for the control core,using STEP 7 software to program and debug PLC,using MCGS configuration software for the human-machine interface design,and finally the servo system,execution equipment and sensors after the wiring work for test and verification,Finally,the temperature control accuracy of intelligent anaerobic fermentation tank is 99.17%,and the system is robust and meets the requirements of anaerobic fermentation control equipment.