Research On PH Control System Based On Penicillin Extraction Process

Penicillin has always occupied an important position in the pharmaceutical industry since its emergence,and has made a significant contribution to human health.However,in recent years,the profit of penicillin production industry has been reduced again and again,and the competition among enterprises is fierce.In this paper,the extraction process of penicillin extraction in a pharmaceutical factory is used as the research background.Because the extraction process is still manually adjusted sulfuric acid liquid valve,the production process has a large lag and instability,which leads to the penicillin product quality is not high,but also consumes more production materials.Based on this,this paper conducts research on the p H value of fermentation filtrate during the extraction process of penicillin as the control object.First of all,based on the analysis of the penicillin extraction principle and the hysteresis,nonlinearity and time-variation of the p H control process,this paper determined the structure of the mathematical model of the p H value of the fermentation filtrate in the penicillin extraction process,namely the first order plus delay time model,and identified the parameters in the model by using the two-point step response method and part of the historical production data.Thus,the mathematical model of the p H control system of penicillin fermentation filtrate was established.Then,aiming at the control system of p H value of fermentation filtrate during penicillin extraction,the control strategy was designed in this paper,and the improved Smith-PID algorithm,improved PSO-PIDNN algorithm and improved JGPC algorithm were adopted and proposed successively to control the mathematical model of p H value of penicillin fermentation filtrate.Among them:1)The improved Smith-PID algorithm is a control strategy by adding a low-pass filter to the Smith predictive controller and combining the PID controller,which weakens the adverse effects on the system control caused by the traditional Smith predictive compensator failing to adapt to the mismatch of the model;2)The improved PSO-PIDNN algorithm is a control strategy to update PIDNN weight based on the improved PSO algorithm instead of the traditional BP algorithm which is dependent on initial weight and easy to fall into local extreme value.Among them,the improved PSO algorithm realizes the nonlinear updating of inertia weight by Logistic chaotic mapping,the updating formula of particle position and velocity is determined by state measurement parameters,and the Pade approximation and recursive least square method with forgetting factor are introduced to calculate the particle fitness;3)The improved JGPC algorithm is a control strategy that uses JGPC algorithm to simplify the traditional GPC algorithm and make the GPC algorithm no longer restricted by the stability of model parameters.At the same time,it combines with the approximate calculation method for other elements in the optimal control sequence except the first element,which is used to reduce the computational amount of the algorithm.Finally,the Matlab tool was used to implement and simulate each algorithm,and overshoot,adjustment time,and steady-state error were selected as evaluation indicators for the performance of each algorithm.The simulation results show that the improved JGPC algorithm has faster response speed,no overshoot,no oscillation,no static error,and the adjustment time is reduced by 40.66% and 37.12% respectively compared with the improved Smith-PID algorithm and the improved PSO-PIDNN algorithm.Therefore,it can be concluded that the improved JGPC algorithm can effectively control the p H value of the fermentation filtrate in the penicillin extraction process,and the control effect is the best among the above algorithms.