Optimization Of Wastewater Treatment Process Based On Improved Artificial Immune Algorithm

With the development of economy and the promotion of urbanization,the discharge of industrial wastewater and domestic sewage has become increasingly intensified.Wastewater treatment has become a hotspot and difficult issue in current research.Wastewater treatment has characteristics of complexity,high coupling,strong nonlinearity,time-varying,large delay and many interference factors,which make it difficult to control,so the traditional treatment methods are not very effective.Due to the complex growth mechanism of microorganisms,fluctuating influent flow,large amount of uncertainties in inflow wastewater and so on,and relatively late research on wastewater treatment in our country and the relatively low control level,so wastewater treatment control research is necessary.The main purpose of wastewater treatment is to improve the treatment capacity,and meet the effluent quality standards,as much as possible to reduce energy consumption.Therefore,about how to make the wastewater treatment process run smoothly,meet effluent quality standards,optimize the key variable set values,save energy and other issues,an artificial immune algorithm with global optimization ability is selected in this paper.Based on the international benchmark simulation model BSM1,carry out analysis and research.The main research contents are as follows:1.In view of the non-linearity and complexity of the wastewater treatment process and the difficulty and large errors in predicting some important effluent water quality parameters,a water quality prediction model based on artificial immune algorithm training extreme learning machine(AIA-ELM)is proposed.Firstly,artificial immune algorithm(AIA)is used to optimize the hidden node parameters of ELM,and ELM input weights and hidden layer node thresholds are mapped into AIA antibodies to improve the generalization ability of ELM.Then generalized inverse MP(Moore-Penrose)is used to obtain the output connection weights of the network rapidly.The effluent biochemical oxygen demand(BOD)and the effluent chemical oxygen demand(COD)are important indexes for the water quality detection.Because of the coupling relationship,the AIA-ELM is finally applied to the international benchmark simulation platform BSM1 to predict two important parameters of BOD and COD.The simulation results show that the proposed method has a good prediction effect.2.Dissolved oxygen(DO)concentration is a very important parameter in wastewater treatment process.Aiming at the problems of time-varying,non-linear and difficult to track for the control of dissolved oxygen concentration in wastewater treatment,a TS fuzzy neural network control method based on artificial immune algorithm(AIA-TSFNN)is proposed.Firstly,a controller is designed by using the self-learning ability of TS fuzzy neural network.Then the artificial immune algorithm is used as the learning method of TS fuzzy neural network to optimize the center value,width value and connection weight of the network,which ensures the convergence of controller and improves the control accuracy.Finally,this control method is applied to international Benchmark Simulation platform BSM1.The experimental results show that the proposed method has good adaptability and robustness,and improves the tracking control performance of dissolved oxygen.3.Wastewater treatment is a complex dynamic response process.Aiming at the problems that how to determine the dynamic optimal set value of key variables in optimization control and the running energy consumption is reduced while the effluent discharge standards are satisfied,a sewage treatment control strategy based on artificial immune algorithm Copt-ai Net(Artificial Immune Network for Combinatorial Optimization)is proposed.Firstly,the optimum set values of the second partition dissolved oxygen concentration and the fifth zone nitrate concentration are determined by Copt-ai Net.Secondly,the controller follows the optimal set values to control the oxygen transfer coefficient and the internal reflux flow,and then the pumping energy and aeration energy are simultaneously optimized.Finally,the simulation experiments are carried out on the simulation platform BSM1.The results verify the effectiveness of the proposed control strategy,which can reduce the energy consumption under the premise of ensuring the effluent quality.