Study On Control Strategy Of Dissolved Oxygen Concentration In Wastewater Treatment

Sewage treatment is a complex nonlinear process with strong coupling and parameter variability,and its effluent quality is affected by the rate influent flow,temperature and concentration of dissolved oxygen,total nitrogen and total phosphorus in biochemical reaction tank.The complexity of the process makes it difficult to establish the accurate mathematical models in the sewage treatment process,which greatly increasing the difficulty of the control process.The dissolved oxygen concentration in the sewage treatment process is the key factor to determine the quality of the effluent.Therefore,the establishment of the sewage treatment process model and the study of dissolved oxygen concentration control has very important significance.In this paper,the establishment of sewage treatment process model and dissolved oxygen concentration control strategy were studied.First,The wastewater treatment process and the dynamic mechanism of sewage treatment process are described.Combined with the act ual sewage treatment process,the reaction mechanism of activated sludge 2 model(ASM2)is described with emphasis on the actual sewage treatment process.According to the classification and reaction rate of ASM2 model,the ASM2 mathematical model is deduc ed.The model is simulated and steady-state result is analyzed.Secondly,Aiming at the instability and large inertia of the dissolved oxygen control system,this paper puts forward the PID control strategy based on the critical ratio method with the disso lved oxygen concentration as the control target.The PID parameters can be set directly in the closed control system,and the PID parameters can be selected quickly and accurately.The simulation results show that the PID control strategy based on the crit ical ratio method is feasible and has good control effect on the dissolved oxygen concentration control,but there is still a problem that the anti-interference is not ideal.Thirdly,Due to the sewage process of oxygen and dissolved oxygen concentration between the strong non-linear,slow response characteristics,a neural network predictive control(NNPC)strategy is proposed.Based on the LM(Levenberg– Marquardt)algorithm,the BP neural network is used as the prediction model of dissolved oxygen,and the control of dissolved oxygen concentration is achieved by feedback correction and rolling optimization.Finally,the anti-jamming ability and control system stability.In addition,in order to improve the water quality and reduce the operating cost,th is paper also proposes a hierarchical control structure based on neural network predictive control.And finally,Considering the characteristics of the sewage treatment process itself and its rapid requirements for the control of dissolved oxygen concentration,the dynamic regulation and poor robustness,a generalized predictive control(GPC)method is introduced.The least squares method with forgetting factor was used to identify the dissolved oxygen concentration model,and the obtained linear model was used as the prediction model.The generalized predictive control algorithm was used to control the dissolved oxygen concentration.The simulation results show that compared with the traditional PID control method,the designed GPC control method has the adv antages of small overshoot and fast response to the control of dissolved oxygen concentration.