Research On Intelligent Detection And Optimal Control Of Wastewater Treatment Process

Wastewater treatment process(WWTPs)has complex biological,physical and chemical characteristics,and it is difficult to control because of the serious coupling relationship among variables and high nonlinearity.At present,activated sludge is used to remove pollutants in water in our country.The power consumption of this biochemical reaction process is high,resulting in a huge operating cost.And few studies focus on reducing the peak concentration of pollutants until they are completely eliminated.In addition,as stricter standards and regulations are implemented around the world,high fines can lead to increased costs.Therefore,in the premise of how to avoid excessive water quality,improving the quality of wastewater treatment and reduce energy consumption is a problem to be solved urgently.It is mainly reflected in:1)ammonia nitrogen(NH₄-N)as a key factor of water pollution control system in wastewater treatment process,online detection of water quality is difficult to achieve and challenging;2)how to obtain the best Pareto solution with the best convergence and distribution,so as to obtain the best set value of dissolved oxygen and nitrate nitrogen;3)the evaluation strategy of WWTPs involves not only one goal,but also multiple goals,such as effluent quality,operation cost and the stability of the system.How to reduce the cost under the premise of ensuring the effluent quality reaching the standard is particularly important;4)the wastewater treatment process is a complex dynamic system,multiple objectives change with time,how to use the dynamic control strategy to achieve the optimal operation of wastewater treatment;5)how to avoid improving the wastewater treatment quality and reducing energy consumption under the premise of the effluent peak exceeding the standard.In view of the problems above,this paper proposes research strategy of intelligent detection and optimal control of wastewater treatment process.Firstly,the characteristics of wastewater treatment process were analyzed,and to establish a data-driven intelligent detection method for wastewater treatment process to predict ammonia nitrogen concentration.Secondly,the multi-objective optimization algorithm of energy consumption and water quality is designed to obtain the optimal setting values of dissolved oxygen and nitrate nitrogen.The self-organizing tracking controller is used to track the set value.In addition,a dynamic multi-objective optimal control method is proposed to deal with the dynamic changes of the environment and obtain better control performance.Finally,According to the intelligent detection results of key water quality,design the knowledge decision scheme,the optimal control strategy for peak suppression is provided.The simulation platform benchmark simulation model no.1(BSM1)is used to verify it.The main research work and innovations are as follows:(1)Design of NH₄-N predictor based on artificial immune self-organizing RBFWater pollution is an important environmental problem,especially the effluent ammonia nitrogen(NH₄-N)exceeds the standard,which has become one of the focuses.Excessive NH₄-N may lead to eutrophication of water body,increase genetic toxicity of wastewater and endanger human health.In order to make the wastewater treatment plant know the concentration of NH₄-N in real time,this paper proposes a NH₄-N predictor based on distance concentration artificial immune self-organizing RBF(DCIA-SORBF).First of all,preprocess the actual collected data,and screen out the process variables with strong correlation with effluent NH₄-N.Secondly,the soft sensing model of effluent NH₄-N was established by RBF,and the structure and parameters of RBF were self-organized adjusted by distance concentration artificial immune algorithm.Finally,the trained DCIA-SORBF model is used to predict the NH₄-N in real time.The experimental results show that the proposed NH₄-N predictor has advantages in efficiency and accuracy.(2)Design of adaptive hybrid evolutionary artificial immune algorithm based on uniform distributionGenerally,in the iterative process of evolutionary algorithm,whether it is a multi-objective optimization problem or a single objective optimization problem,there is a problem of uneven individual distribution in the target space.This uneven distribution greatly reduces the diversity and convergence rate of the population.Therefore,this paper proposes an adaptive hybrid evolutionary immune algorithm(AUDHEIA)based on uniform distribution selection mechanism.In this algorithm,the individuals in the population are mapped to the hyperplane corresponding to the target space,and cluster to increase the diversity of individuals in the population.In order to improve the distribution of the solution,the mapped hyperplane is divided into uniform regions.With the continuous change of distribution in the iterative process,the threshold value of population distribution standard is adjusted adaptively.When the threshold value is not satisfied in the corresponding interval,the distributed enhancement module is activated.Then,select the same number of individuals in each interval.However,in the iterative process,sometimes there are not enough individuals or empty in some intervals.At this time,the limit optimization mutation strategy of the optimal individual is used to supplement the individual.The experimental results show that the algorithm can jump out of the local optimum and has high convergence speed.In addition,the distribution and convergence of the algorithm are better than the similar test algorithms.(3)Design of immune multi-objective optimal control method on wastewater treatmentAn intelligent control system based on immune optimization is proposed to solve the problems of excessive energy consumption and serious effluent quality in the process of wastewater treatment control.First of all,this method takes the energy consumption of wastewater treatment and the quality of effluent as the optimization objective,and establishes the optimization objective function model.Secondly,AUDHEIA multi-objective optimization algorithm is used to obtain the Pareto solution with good convergence and distribution,so as to obtain the optimal set value of dissolved oxygen and nitrate nitrogen.Finally,the self-organizing recursive fuzzy neural network controller is applied to the bottom tracking control of the set value.In order to verify the effectiveness of the algorithm,experiments are carried out on the benchmark simulation model 1(BSM1)of the international standard.The results show that the proposed immune optimization control method can effectively reduce the energy consumption of wastewater treatment process while meeting the effluent quality standards.(4)Design of immune optimal control method for dynamic process of wastewater treatmentDue to the dynamic change of wastewater treatment process,there is a coupling relationship between energy consumption and water quality.In order to quickly realize the optimal operation of wastewater treatment when the operation situation changes,this paper proposes a dynamic multi-objective immune optimization control(DMOIA-OC)method.In this method,dynamic multi-objective immune algorithm(DMOIA)is designed to obtain the best set value of dissolved oxygen and nitrate nitrogen with the dynamic change of environment.In order to improve the performance of evolutionary algorithm in solving dynamic multi-objective optimization problems,the algorithm adopts multi-directional prediction strategy.In order to predict the moving position of Pareto solution set more accurately,the population is clustered into several representative groups by adaptive uniform distribution strategy,and the evolution direction of individuals is predicted according to the environmental change,and the population is re-initialized around the predicted new position.Finally,the method is verified by BSM1 simulation platform.The experimental results show that the proposed DMOIA-OC method has significantly improved the control performance compared with similar methods..(5)Design of immune optimal control method for wastewater treatment based on soft sensing and knowledgeIn order to reduce the water quality and energy consumption in the wastewater treatment process,this paper proposes an intelligent detection and optimal control system of wastewater treatment process(IDOC).First of all,the soft sensing model(DCIA-SORBF)is used to predict the concentration of ammonia nitrogen and total nitrogen.According to the prediction results and expert knowledge,the whole process control strategy of wastewater treatment is designed.When the predicted water quality is up to the standard,the dynamic immune optimization algorithm is used to get the high-quality set values of dissolved oxygen and nitrate nitrogen,so as to achieve the purpose of energy saving and consumption reduction.When the predicted effluent quality is not up to standard,the peak suppression control is started.Finally,BSM1 model is used to verify the method.The experimental results show that the proposed IDOC strategy can achieve the real-time standard of the whole process wastewater treatment,and effectively reduce energy consumption,so it has better practical value.