Research On Intelligent Control Method Of Precise Aeration In Sewage Treatmen

This paper studies the intelligent control method of precise aeration process of sewage treatment.The aeration method that relies on manual experience has waste of energy.In order to reduce the energy consumption of aeration,it is necessary to accurately control the aeration process.The main difficulty in achieving accurate aeration is the nonlinear change of the influent flow and water quality,large disturbance,and aeration.There is a large lag in the gas process.With the transformation of the information age,a large number of online detection instruments and cloud data storage are used in sewage treatment systems.How to use these data becomes the key to precise control systems.Therefore,based on the data,this paper uses Simulink to build the BSM1(Benchmark Simulation Model NO.1)benchmark simulation platform,and studies the intelligent control method of precise aeration in terms of energy saving and consumption reduction in sewage treatment and maintaining the stability of effluent water quality.In terms of energy saving and consumption reduction,in order to ensure the effluent quality,reduce operating costs,and at the same time accumulate and mine the optimal set values of operating variables under different working conditions,a casebased reasoning(CBR)wastewater treatment exposure method was proposed.Gas process intelligent control method.The method first uses historical data to establish a neural network model of working conditions and water output indicators,and uses the model as the fitness function of particle swarm optimization to generate the operation parameters of the initial case database and the case correction part of optimizing the operation parameters;using the inflow flow rate,ammonia nitrogen,sludge concentration,biochemical oxygen demand,effluent ammonia nitrogen,and biochemical oxygen demand are used as working condition descriptions,and the optimized set values of operating parameters under different working conditions are obtained through particle swarm algorithm,and a dynamic case library is established.For a new operating condition,the operating parameters obtained by the case reasoning are input into the simulation model for evaluation,and the neural network model is updated with the new data and the operating parameters of each case are revised.The proposed method was applied to BSM1 for simulation,the optimized system reduced aeration energy consumption by 18.5% compared with the original system,and the effluent quality of the effluent was improved at the same time.In terms of maintaining the stability of the effluent,an aeration air volume control algorithm based on DDPG(Deep Deterministic Policy Gradient)is proposed.The water homogenization tank is added to the BSM1 model,and the Markov process of the sewage treatment process is established,and the air volume is used as the control object.,learn the optimal air volume setting value in the simulated environment interaction data.In the comparative experiment,compared with the stable control method of effluent water quality index based on case reasoning,the proposed method reduces the fluctuation of effluent water quality index caused by interference,and can limit the water quality index within the expected range.Afterwards,considering that the motor performance degradation may lead to the failure or even damage of the motor,by analyzing the characteristics of the motor vibration curve and using the sphere center and radius of the motor vibration data under SVDD under normal conditions,it is judged whether the motor is in a normal operation state.The method can effectively judge whether the motor is in a normal working state.