Research On Optimization And Control Of Aeration System Based On Soft Measurement Of Inlet BOD₅

At present,the centralized treatment goal of "quality and quantity" for urban sewage in China has been achieved,and various industries are minimizing energy consumption under the guidance of the "dual carbon goal".Therefore,the future strategic goal of sewage treatment is inevitably to "ensure quality and reduce consumption".The topic of this article is based on the background of high proportion of aeration energy consumption,low degree of automation control,and outdated control modes in China’s sewage treatment process.It is planned to design an intelligent aeration control system to respond to changes in influent water quality,in order to ensure that sewage plants achieve the low-carbon goals of "low energy consumption"and "low emissions".This article takes a sewage plant in Xi’an as the research object to analyze the oxygen consumption process of activated sludge in sewage treatment,and obtains the BOD5 index of aerobic tank inlet through soft measurement.The BOD5 soft measurement results are used as the basis for controlling the dissolved oxygen concentration in the aeration system,and the aeration control system is optimized and designed.The specific research results are as follows:(1)The analysis of the historical operating data of the sewage treatment plant shows that it has achieved a 100%sewage treatment compliance rate at this stage,and has verified the variability and uncertainty characteristics of the inflow load.Analysis of operational energy consumption shows that the average energy consumption per ton of sewage in this sewage treatment plant 0.470 kW·h/m3 is about 45%higher than the average energy consumption per ton of sewage in China.Analyze the aerobic process and determine that the optimal dissolved oxygen concentration in the aerobic tank of the sewage plant is 1.50mg/L～4.48mg/L.Using"3σ" The method is used to process the operation data with outlier,and the max min standardization is used to normalize the inflow indicators,so as to speed up the model calculation and improve the model generalization ability.(2)Establish a BOD5 soft sensing model using BP neural network,and optimize the BP neural network using genetic algorithm and particle swarm optimization algorithm.The root mean square errors of each soft measurement model are 33.639,29.117,and 28.776,with an average error of 27.619,23.233,and 22.437,and a maximum error of 88.229,51.268,and 48.028.The results show that the prediction error of the particle swarm optimization algorithm optimized BP neural network is significantly reduced.Therefore,the particle swarm optimization algorithm was used to compensate for errors in the BP neural network,further improving its generalization ability for BOD5.(3)Design a cascade PID controller for aeration dissolved oxygen based on the dissolved oxygen concentration model and blower model,and introduce a fuzzy control module to dynamically adjust the Kp,Ki,and Kd parameters of the PID.In MATLAB,the target values of dissolved oxygen were set to 2.Omg/L and 4.5mg/L respectively for simulation testing.The results showed that the fuzzy PID control strategy was 312s and 106s faster than the cascade PID control strategy,and the anti-interference performance of the fuzzy PID control strategy was also better than the cascade PID control strategy.(4)Comparing the aeration dissolved oxygen control strategy based on inflow BOD5 soft measurement with the actual dissolved oxygen control settings of the sewage plant,the results show that the dissolved oxygen control strategy designed in this paper can reduce the aeration volume of the aerobic tank in the sewage plant by about 15%while ensuring that the effluent quality meets the standard.After the planned operation,the energy consumption per ton of sewage is 0.289 kW·h/m3,which is 39%lower than the original control method of 0.470 kW·h/m3 per ton of sewage,and 11.07%lower than the average energy consumption per ton of sewage of sewage plants in China of 0.325 kW·h/m3.