Application Research Of PID Algorithm Based On Neural Network In Biomass Fermentation Control

With the continuous advancement of agricultural technology,China's grain output has also increased year by year.By 2018,China's grain output ranked first in the world.While effectively guaranteeing the relationship between supply and demand in the grain market,it also brought about problems such as high inventory.Therefore,relevant experts suggested that the commercialization of biomass fuel ethanol should be realized to promote structural reform of agriculture.In this context,according to the requirements of the Party Central Committee,it is believed that there are already opportunities and conditions for expanding the production of biofuel ethanol and promoting ethanol gasoline for vehicles.However,in the process control of biomass fermentation fuel ethanol,the influence of temperature on enzymatic fermentation is very important.It has the characteristics of large lag,nonlinearity and time-varying.Therefore,in order to avoid the unnecessary influence caused by temperature change,The main research work of this thesis is as follows:Firstly,this thesis introduces the working principle and process flow of biomass fermentation fuel ethanol,summarizes the difficulties and problems in process control,and then designs the hardware monitoring system of biomass fermentation to realize the recording and storage of temperature,PH and dissolved oxygen data.Secondly,in the process of biomass fermentation,the traditional PID controller parameters are often difficult to be tuned and optimized for the characteristics of large temperature lag and nonlinearity.In order to meet the high index control requirements,a neural network based PID control algorithm is proposed to realize PID.Self-tuning of parameters.Finally,the biomass fermentation experiment environment was set up 1000 sets of temperature data sets were obtained in the biomass fermentation experiment.The mathematical model parameters of temperature were identified by MATLAB toolbox.Then the simulation of BP neural network PID control method and RBF neural network PID control method is carried out on MATLAB.The simulation results show that the PID self-tuning method using RBF neural network has better control effect in fermentation temperature control.