Intelligent optimal control method and system for effluent quality of sbr process

Abstract

The application discloses an intelligent optimal control method and system for water quality of SBR process effluent, a water quality prediction model is constructed through a BP neural network, the water quality of effluent can be predicted according to the water inflow condition, and the situation that the water quality of effluent does not reach the standard is avoided. The main factors influencing the removal rates of C, N and P and the influence degrees of the main factors are determined through grey correlation analysis, which is helpful to determine the key factors influencing the water quality and to perform process parameter regulation and control in a targeted manner. The control of the prediction model is performed through a non-dominated sorting genetic algorithm, an intelligent optimization model for the water quality of SBR process effluent is constructed, the removal rate of C is taken as a constraint condition, and the removal rates of N and P are taken as targets to perform planning, and the optimization effect of process parameters and the stability of the water quality of effluent are improved. Therefore, the above scheme can solve the problem in the prior art that when the water inflow condition of the SBR process is known, the water quality of effluent cannot be predicted, and controllable process parameters cannot be planned according to the current condition to achieve the best water quality of effluent.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of sewage treatment, and relates to an intelligent optimal control method and system for SBR process effluent water quality. BACKGROUND

[0002] The activated sludge technology, as a common sewage treatment method, effectively removes organic matter, nitrogen, phosphorus and other pollutants in sewage by using flocculent sludge formed by microorganisms in an aerobic environment, and is often applied in a sequencing batch reactor (SBR).

[0003] However, the operation of the SBR process in controlling sewage treatment indicators to achieve a specific effluent standard is complex and difficult to accurately master due to the involvement of multiple influencing factors. In actual operation, engineers often rely on personal experience to regulate and control the SBR process to ensure that the effluent water quality meets the standard. However, the intelligent optimal control method has low stability, resulting in low accuracy and long regulation time when regulating water quality. SUMMARY

[0004] The application aims to solve the problem of low stability of the intelligent optimal control method in the prior art, which results in low accuracy and long regulation time when regulating water quality, and provides an intelligent optimal control method and system for SBR process effluent water quality.

[0005] To achieve the above-mentioned purpose, the application adopts the following technical solutions:

[0006] The application provides an intelligent optimal control method for SBR process effluent water quality, which comprises the following steps:

[0007] Obtain effluent water quality indicators affecting carbon, nitrogen and phosphorus removal rates, and construct an intelligent prediction model for SBR process effluent water quality based on the effluent water quality indicators affecting carbon, nitrogen and phosphorus removal rates;

[0008] Analyze the intelligent prediction model for SBR process effluent water quality to determine the influencing indicators affecting the removal rates of carbon, nitrogen and phosphorus;

[0009] Based on the influencing indicators and the intelligent prediction model for SBR process effluent water quality, an intelligent optimization model for SBR process effluent water quality is constructed, the optimal solution of the intelligent prediction model for SBR process effluent water quality is solved, and intelligent optimal control of SBR process effluent water quality is realized.

[0010] Preferably, the intelligent prediction model for SBR process effluent water quality is built by using a BP neural network.

[0011] Preferably, the intelligent prediction model for SBR process effluent water quality is constructed as follows:

[0012] The input features in the prediction of the effluent quality of the SBR process are taken as the nodes of the input layer, and the input features include static indicators and dynamic indicator data of the effluent quality;

[0013] Each neuron receives input signals from the previous layer, sets weights and biases according to specific problems, selects an activation function, and performs a nonlinear transformation on the weighted sum, and the neurons in the hidden layer and the output layer calculate the weighted sum and the activation function layer by layer to obtain the output result;

[0014] The historical data and real-time monitoring data related to the prediction of the effluent quality of the SBR process form a training set, and the training set is used for the training of the BP neural network, and the weights and biases of the network are optimized through the back propagation algorithm. After the training of the BP neural network is completed, an intelligent prediction model of the effluent quality of the SBR process is obtained.

[0015] Preferably, the method for determining the influence indicators affecting the removal rates of carbon, nitrogen and phosphorus by using grey correlation analysis is as follows:

[0016] The data set of dynamic indicators and static indicators affecting the removal rates of C, N and P is obtained;

[0017] The data set is normalized to map the data range of each indicator to [0, 1];

[0018] According to the normalized data set, the correlation degrees between each indicator and the prediction target of the effluent quality of the SBR process are calculated;

[0019] According to each column of the correlation matrix, the weights of each indicator and the target are calculated.

[0020] According to the calculated indicator weights, the indicators are sorted, and the indicators with higher weights are selected as the influence indicators.

[0021] Preferably, the indicators with higher weights are normalized to obtain optimized influence indicators.

[0022] Preferably, a non-dominated sorting genetic algorithm is used to solve the optimal solution of the intelligent prediction model of the effluent quality of the SBR process.

[0023] Preferably, an intelligent optimization model of the effluent quality of the SBR process is constructed, and the process of solving the optimal solution of the intelligent prediction model of the effluent quality of the SBR process is as follows:

[0024] The objective function is the removal rate prediction model of N and P, and the constraint function is the removal rate prediction model of C;

[0025] An initial population is randomly generated, including a number of individuals; for each individual in the population, the fitness value is calculated according to the objective function and the constraint condition;

[0026] Non-dominant sorting is performed on individuals in the population, the individuals are divided into different levels, and a non-dominant sorted level structure is formed;

[0027] The crowding degree of individuals in the population in the target space is calculated;

[0028] According to the non-dominant sorting and the crowding degree calculation result, a certain number of individuals are selected as parents;

[0029] The selected parents are subjected to crossover and mutation operations to generate new individuals;

[0030] The steps of evaluating fitness, non-dominant sorting, crowding degree calculation, selection operation, crossover and mutation are repeated until a set number of iterations is reached, and the Pareto optimal solution is obtained, and the optimal solution of the intelligent prediction model of the SBR process effluent water quality is obtained.

[0031] The intelligent optimal control system for SBR process effluent water quality provided by the application comprises:

[0032] The model construction module is used to obtain the effluent water quality indexes affecting the removal rates of carbon, nitrogen and phosphorus, and construct the intelligent prediction model of the SBR process effluent water quality based on the effluent water quality indexes affecting the removal rates of carbon, nitrogen and phosphorus;

[0033] The model analysis module is used to analyze the intelligent prediction model of the SBR process effluent water quality and determine the influence indexes affecting the removal rates of carbon, nitrogen and phosphorus;

[0034] The model solving module is used to construct the SBR process effluent water quality intelligent optimization model based on the influence indexes and the intelligent prediction model of the SBR process effluent water quality, solve the optimal solution of the intelligent prediction model of the SBR process effluent water quality, and realize the intelligent optimal control of the SBR process effluent water quality.

[0035] An electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the intelligent optimal control method for SBR process effluent water quality when executing the computer program.

[0036] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the intelligent optimal control method for SBR process effluent water quality.

[0037] Compared with the prior art, the application has the following beneficial effects:

[0038] The intelligent optimal control method of SBR process effluent water quality provided by the application can help engineers make control decisions in advance and avoid the situation that the effluent water quality does not meet the standard by constructing a prediction model, predicting the effluent water quality according to the influent condition, and determining the main factors affecting the removal rates of C, N and P and the influence degree thereof through grey correlation analysis, which helps engineers to determine the key factors affecting the water quality and save the control time. The intelligent optimization model of SBR process effluent water quality is constructed, the removal rate of C is taken as the constraint condition, and the removal rates of N and P are taken as the target for planning, which improves the optimization effect of the process parameters and the stability of the effluent water quality. Therefore, the above scheme can predict the effluent water quality when the influent condition of the SBR process is known, and plan controllable process parameters according to the current condition to achieve the best effluent condition.

[0039] Further, the non-dominated sorting genetic algorithm is used for solving, which can consider multiple targets at the same time, obtain a group of approximate optimal solutions, provide more choices for engineers, and improve the flexibility and efficiency of process control.

[0040] Further, the application of the BP neural network algorithm can more accurately predict the effluent water quality, help engineers better understand the process operation state, timely adjust the parameters, and improve the stability and consistency of the effluent water quality.

[0041] The intelligent control system of SBR process effluent water quality provided by the application divides the system into a model establishment module, a model analysis module and a model solving module, and realizes the intelligent optimal control of SBR process effluent water quality. The modularization idea makes each module independent of each other, which is convenient for unified management of each module. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical scheme of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the application, and therefore should not be regarded as a limitation on the scope, and other related drawings can also be obtained by the drawings without paying creative labor for those skilled in the art.

[0043] Figure 1 The flow chart of the intelligent optimal control method of SBR process effluent water quality of the application.

[0044] Figure 2 The construction structure diagram of the intelligent optimization model of SBR process effluent water quality of the application.

[0045] Figure 3 The specific flow chart of the non-dominated sorting genetic algorithm of the intelligent optimization model of SBR process effluent water quality of the application.

[0046] Figure 4 Figure of intelligent optimal control system for effluent quality of SBR process of the present application.

[0047] Figure 5 Figure of structure of electronic device of the present application. DETAILED DESCRIPTION

[0048] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will be combined with the accompanying drawings for the embodiments of the present application to make a clear and complete description of the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0049] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0050] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0051] In the description of the embodiments of the present application, it should be noted that if the terms "upper", "lower", "horizontal", "inner" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship when the product of the present application is used, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second" and the like are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0052] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly inclined. For example, "horizontal" only means that its direction is relatively more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined.

[0053] In the description of the embodiments of the present application, it also needs to be explained that, unless explicitly specified and limited, if the terms of "arrange", "install", "connect", "connect" appear, they should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0054] The present application will be further described in detail below in combination with the drawings:

[0055] Referring to the drawings, an intelligent optimal control method for SBR process effluent water quality is proposed, as shown in Figure 1 The method comprises the following steps:

[0056] S1, obtaining effluent water quality indexes affecting carbon, nitrogen and phosphorus removal rates, and constructing an intelligent prediction model for SBR process effluent water quality based on the effluent water quality indexes affecting carbon, nitrogen and phosphorus removal rates;

[0057] The intelligent prediction model for SBR process effluent water quality is built by using BP neural network.

[0058] The intelligent prediction model for SBR process effluent water quality is built as follows:

[0059] The input features in the prediction of SBR process effluent water quality are taken as the nodes of the input layer, and the input features include static index and dynamic index data of effluent water quality;

[0060] Each neuron receives input signals from the previous layer, sets weights and biases according to specific problems, selects an activation function, and performs nonlinear conversion on the weighted sum, and the neurons in the hidden layer and the output layer are calculated layer by layer to obtain the output result;

[0061] The historical data and real-time monitoring data related to the prediction problem of SBR process effluent water quality form a training set, and the training set is used for training of the BP neural network, and the weights and biases of the network are optimized by the back propagation algorithm, and the BP neural network is trained to obtain the intelligent prediction model for SBR process effluent water quality.

[0062] S2, analyzing the intelligent prediction model for SBR process effluent water quality by gray correlation analysis to determine the influence indexes affecting the removal rates of carbon, nitrogen and phosphorus;

[0063] The method for determining the influence indexes affecting the removal rates of carbon, nitrogen and phosphorus by using gray correlation analysis is as follows:

[0064] Obtain the data set of dynamic indexes and static indexes affecting the removal rates of C, N and P;

[0065] The data set is normalized, and the data range of each index is mapped to [0, 1];

[0066] According to the normalized data set, the correlation degree between each index and the SBR process effluent water quality prediction target is calculated;

[0067] According to each column of the correlation matrix, the weight of each index and the target is calculated;

[0068] According to the calculated index weight, each index is sorted, and the index with higher weight is selected as the influence index.

[0069] The index with higher weight is normalized to obtain the optimized influence index.

[0070] S3, based on the influence index and the intelligent prediction model of SBR process effluent water quality, an intelligent optimization model of SBR process effluent water quality is constructed, and the optimal solution of the intelligent prediction model of SBR process effluent water quality is solved, and the intelligent optimization control of SBR process effluent water quality is realized.

[0071] The non-dominated sorting genetic algorithm is used to solve the optimal solution of the intelligent prediction model of SBR process effluent water quality.

[0072] The intelligent optimization model of SBR process effluent water quality is constructed, and the process of solving the optimal solution of the intelligent prediction model of SBR process effluent water quality is as follows:

[0073] The objective function is the removal rate prediction model of N and P, and the constraint function is the removal rate prediction model of C;

[0074] Randomly generate an initial population, including a certain number of individuals; for each individual in the population, calculate the fitness value according to the objective function and the constraint condition;

[0075] The individuals in the population are non-dominantly sorted, and the individuals are divided into different levels to form a non-dominantly sorted level structure;

[0076] The crowding degree of individuals in the population in the target space is calculated;

[0077] According to the non-dominant sorting and crowding degree calculation results, a certain number of individuals are selected as parents;

[0078] The selected parents are subjected to crossover and mutation operations to generate new individuals;

[0079] Repeat the steps of evaluating fitness, non-dominant sorting, crowding degree calculation, selection operation, crossover and mutation until the set iteration number is met, get the Pareto optimal solution, and obtain the optimal solution of the intelligent prediction model of SBR process effluent water quality.

[0080] The model framework of the intelligent optimal control method of the SBR process effluent water quality is as shown in Figure 2 The detailed steps are as follows:

[0081] Step 1, obtain the static index and dynamic index data affecting the SBR process effluent water quality, and obtain the effluent water quality index data of the SBR.

[0082] For data acquisition, two laboratory-scale batch reactors (SBR) R1 and R2 are used for water treatment in this example. Activated sludge (AS) is taken from a sewage treatment plant for water treatment. The effective volume of the two reactors is 2.54 L, the height H is 1000 mm, and the inner diameter D is 60 mm. The water inlet and aeration head are arranged on the side wall and the bottom of the reactor, respectively, and the water outlet is arranged in the middle position.

[0083] The reactor is set as follows: the length of each cycle is set to 6 h, including the static water feeding, anaerobic, aeration, sedimentation, and water draining stages, and the length of each stage is shown in Table 1. In the water feeding stage, the simulated wastewater is pumped into the bottom inlet by a peristaltic pump, and a constant speed stirring device is used for stirring to ensure uniform water feeding; in the aeration stage, air is supplied by an air compressor and aeration sand head, and the aeration amount is controlled by a glass rotor flowmeter; in the water draining stage, a water draining valve is used to control the water draining time, and the wastewater is drained from the water outlet 500 mm away from the bottom of the reactor at a volume exchange rate of 50% and collected in a water outlet bucket. The operation of the peristaltic pump, water draining valve, air compressor, and stirring device is automatically controlled by a time controller.

[0084] Table 1 Reactor operating parameters / min

[0085]

[0086] The reactor is operated using synthetic wastewater. The carbon source, nitrogen source, and phosphorus source are sodium acetate, ammonium sulfate, and potassium dihydrogen phosphate, respectively, and the C, N, and P content ranges during operation are shown in Table 2. To study the influence of static and dynamic variables on SBR process water treatment, the dynamic variables of this patent are selected as: dissolved oxygen (DO), temperature (T), mixed liquor suspended solids (MLSS), pH, organic loading rate (OLR), and oxidation-reduction potential (ORP); the static variables are selected as: chemical oxygen demand (COD), ammonia nitrogen (NH4-N), total nitrogen (TN), total phosphorus (TP), and settling performance index (SVI). The above parameters are measured according to standard methods. The specific data range is shown in Table 2.

[0087] Table 2 Input index range of two reactors

[0088]

[0089] Specifically, in step 1, the BP neural network algorithm is used to establish a SBR effluent quality prediction model with 6 static indicators and dynamic indicators as input indicators, and the removal rates of C, N, and P as output indicators, including:

[0090] S111. Set the input layer: the input features in the SBR effluent quality prediction are used as the nodes of the input layer, including 6 static indicators and 6 dynamic indicators, where the dynamic variables are selected as: dissolved oxygen (DO), temperature (T), mixed liquor suspended solids (MLSS), pH, organic loading rate (OLR), and oxidation-reduction potential (ORP); the static variables are selected as: chemical oxygen demand (COD), ammonia nitrogen (NH4-N), total nitrogen (TN), total phosphorus (TP), and settling performance index (SVI). These input features can be represented as X = [x1, x2,.x 11 ].

[0091] S112. Weighted sum of neurons and activation function: each neuron receives input signals from the previous layer and sets weights and biases according to specific problems. Then, select an appropriate activation function to perform nonlinear transformation on the weighted sum. In this way, the neurons in the hidden layer and the output layer will calculate the weighted sum and activation function layer by layer to get the output result. The activation function is ReLU.

[0092] S113. Data set and training: historical data and real-time monitoring data related to the SBR effluent quality prediction problem can be used to form a training set. These data are used for the training process of the BP neural network, and the weights and biases of the network are optimized through the backpropagation algorithm to minimize the difference between the predicted results and the actual results. In this way, the BP neural network can be trained to learn the complex relationship between the data.

[0093] The training of the BP neural network model in this example uses 70% of the data for training, 15% for testing, and 15% for verification. The design uses MATLAB software version 2023b to train the BP neural network.

[0094] S114. Prediction: once the BP neural network training is completed, new input data can be input into the network to obtain the SBR effluent quality prediction results through forward propagation.

[0095] The specific parameter settings of the SBR effluent quality prediction problem need to adjust the structure, loss function, and optimization algorithm of the BP neural network according to the actual situation.

[0096] The structure of the BP neural network training algorithm of the three prediction models of the removal rates of C, N, and P is as follows: single hidden layer structure, the number of neurons of the input layer is 11, the hidden layer is 10, and the output layer is 1; the loss function is the root mean square error plus L2 regularization; and the optimization algorithm is the BP algorithm. The BP algorithm, i.e., the back propagation algorithm, is a common algorithm for training neural networks. It continuously adjusts the weights and biases in the neural network to minimize the error between the network output and the expected output.

[0097] The performance of the prediction model trained by the data is as follows: the R2 of the C removal rate prediction model is 0.901420374; the R2 of the N removal rate prediction model is 0.941581994; and the R2 of the P removal rate prediction model is 0.972175119.

[0098] Step 2, based on the output and input indicators of the intelligent prediction model of the effluent quality of the SBR process of the constructed activated sludge, the main factors affecting the removal rates of C, N, and P and their influence degrees are determined through grey correlation analysis, including:

[0099] S21. Data preparation: collect the data set related to the effluent quality of the SBR process, including 6 dynamic variables and 5 static variable indicators and attributes affecting the removal rates of C, N, and P in the effluent quality of the SBR process. There are 11 input indicators for C, N, and P removal rates, and the data set can be represented as a matrix X, where the size of X is n x 11, and n represents the number of samples.

[0100] In this example, 42 groups of data are obtained, i.e., n = 42.

[0101] S22. Data normalization: normalize the data set to map the data range of each indicator to [0, 1] to ensure that different indicators have the same dimension.

[0102] S23. Construct the correlation matrix: calculate the correlation degree between the 11 indicators and the C, N, and P removal rate targets in the effluent quality of the SBR process according to the normalized data set.

[0103] The C, N, and P removal rate correlation matrix is three 11 x 11 matrices, where each element represents the correlation degree between the corresponding indicator and the target.

[0104] S24. Determine the weight: calculate the weight of each indicator and the target according to each column of the correlation matrix. The greater the weight value, the higher the correlation degree of the indicator and the target, and the greater the influence on predicting the removal rates of C, N, and P.

[0105] S25. Sort and select indicators: sort the indicators according to the calculated indicator weights.

[0106] Through the index selection by the grey correlation analysis, the correlation between the 11 indexes and the C, N and P removal rate prediction can be accurately determined, thereby providing valuable reference for the priority of the regulation and control of the selection pressure.

[0107] Through the index selection by the maximum mutual information coefficient, the index with the highest correlation with the C, N and P removal rate prediction can be found, and the regulation and control priority of the selection pressure can be sorted according to the result of the grey correlation analysis in step 2. The higher the priority is, the greater the influence on the effluent quality is.

[0108] In order to compare the sensitivity of the indexes, the calculation result is normalized. Specifically, the grey correlation analysis of the 11 indexes of the C, N and P removal rates obtained by the MATLAB in the present example is shown in Table 3.

[0109] Table 3 Sensitivity score of each index affecting the C, N and P removal rates based on the grey correlation analysis

[0110]

[0111] Specifically, according to the result in Table 3, the regulation and control priority (from large to small) for the C removal rate is: pH, DO, TP, NH4-N, T, TN, COD, MLSS, ORP, SVI and OLR. The regulation and control priority (from large to small) for the N removal rate is: ORP, pH, T, MLSS, SVI, TN, COD, OLR, TP, NH4-N and DO. The regulation and control priority (from large to small) for the N removal rate is: pH, SVI, ORP, T, COD, OLR, MLSS, TN, TP, NH4-N and DO.

[0112] Step 3, as shown in Figure 3 the SBR process effluent quality intelligent optimization model of the activated sludge is constructed, which comprises:

[0113] S31. Determine the objective function: first, determine the objective function of the multi-objective programming problem, including the optimization target and constraint condition of each target. The objective function should be able to comprehensively reflect the multiple optimization targets of the problem, while considering the constraint conditions of the problem.

[0114] Specifically, the objective function of the present application is: the removal rate prediction model of N and P, and the constraint function is: the removal rate prediction model of C.

[0115] S32. Initialize the population: randomly generate the initial population, including a certain number of individuals, each individual representing a possible solution. Ensure the diversity of the population to better explore the solution space.

[0116] Specifically, the solution space of the present application is the value range of the 6 dynamic variables, and the size of the population is 100.

[0117] S33. Evaluate fitness: For each individual in the population, calculate the fitness value according to the objective function and the constraint conditions, evaluate the individual's degree of excellence. The fitness value should consider the optimization effect of multiple objectives and the satisfaction degree of constraint conditions.

[0118] Specifically, the present application uses the concept of constraint violation degree to establish a penalty function in the objective function to punish solutions that violate the constraint conditions, to ensure that the generated solution meets the constraint conditions.

[0119] S34. Non-dominated sorting: Non-dominated sorting of individuals in the population, the individual is divided into different levels, forming a non-dominated sorting level structure. Ensure that there is no obvious relationship between individuals in the population.

[0120] S35. Crowding degree calculation: Calculate the crowding degree of individuals in the population in the objective space, maintain the diversity of the population, avoid premature convergence, and improve the global search ability of the algorithm.

[0121] Specifically, the present application selects the top 50 individuals according to the non-dominated sorting and crowding degree calculation, and uses them as the parent population to reproduce 50 individual offspring, and then combines the parents and offspring to form a new generation of individual population with a population size of 100. The sorting first looks at the non-dominated sorting, and then uses the crowding degree to calculate the priority order when the individual attributes in the population are in the same non-dominated level.

[0122] S36. Selection operation: According to the results of non-dominated sorting and crowding degree calculation, select a certain number of individuals as parents for crossover and mutation operations to generate new individuals.

[0123] Specifically, the present application selects 50 individuals as parents.

[0124] S37. Crossover and mutation: Crossover and mutation operations are performed on the selected parents to generate new individuals. Crossover and mutation operations help maintain the diversity of the population, promote the evolution and optimization of the population.

[0125] S38. Iterative optimization: Repeat the steps of evaluating fitness, non-dominated sorting, crowding degree calculation, selection operation, crossover and mutation until the stopping condition is met. Finally, a set of approximate optimal solutions (Pareto optimal solutions) is obtained, providing decision makers with a variety of choices.

[0126] Specifically, continue to repeat the process of S33-S37 until the maximum population iteration number of 200 is reached, and select the top 24 population from the last generation of population as the Pareto optimal solution, and finally obtain the control parameters of the dynamic variables of the SBR process C, N and P removal rate.

[0127] In step 3, the dynamic variable parameters of the optimal N and P removal rate that meet the C removal requirement under the condition of the current static index are obtained, and the static variable index of the influent water is known first. The static variables that can simulate the influent water conditions are as follows: COD: 547.66, NH4-N: 41.31, TN: 39.09, TP: 9.30, SVI: 0.89. The outflow water quality simulation under the Pareto optimal solution of the MATLAB simulation of the influent water conditions in the present example is shown in Table 4.

[0128] Table 4. Simulation results of dynamic variable parameter adjustment under the Pareto optimal solution of the static variable condition of the planning model (the planning target is in the brackets)

[0129]

[0130]

[0131] It should be noted that C / N / P-out represents the removal rates of COD, TN and TP, and the present application represents the removal rates of C, N and P.

[0132] Specifically, 24 results can be obtained by the planning model shown in Table 4, and engineers can select the parameters of the controllable variables according to actual requirements. The present application considers the requirements of denitrification and phosphorus removal, and selects the result of the Pareto optimal solution No. 13, and the C, N and P removal rates of the result are 91.45, 95.34 and 97.13, respectively. The specific controllable variable operation parameters are as follows: DO: 6.56, T: 21.55, MLSS: 4.10, pH: 7.12, F / M: 110.81, and ORP: 254.78.

[0133] If the engineer wants to focus on denitrification (phosphorus removal), other Pareto optimal solutions can be selected. The maximum N and P removal rates are 100%, and the N(P) removal rates of a part of the Pareto optimal solutions in the result are greater than 100%, which can be explained as follows: the BP neural network fitting of the N(P) removal rate prediction model appears a certain overfitting, so that the result is greater than 100%. For the problem, the result greater than 100% can be regarded as 100%.

[0134] Compared with the prior art, the intelligent optimal control method provided by the application can find out the dynamic variable control strategy of the optimal C, N and P removal rate meeting the actual requirements in the case that the static variables of the influent are known. The data source of the example is the laboratory, and some problems not considered from the laboratory to the actual situation are not considered. For example, the influent of the laboratory is the blending water, and the pollutants therein are explicit, but the actual sewage often has some pollutants with great toxicity but difficult to detect, and the influence of these pollutants on the effluent quality of the SBR process is very serious. Therefore, when the model is applied to the actual situation, new influence indexes need to be added for improvement. The planning model provides a new potential method for the SBR process effluent quality planning. It has guiding significance in the future intelligent engineering control.

[0135] Embodiment 2

[0136] The intelligent optimal control system for the SBR process effluent quality provided by the application, as shown in Figure 4 , comprises:

[0137] A model construction module is configured to acquire the effluent quality indexes affecting the carbon, nitrogen and phosphorus removal rates, and construct an intelligent prediction model for the SBR process effluent quality based on the effluent quality indexes affecting the carbon, nitrogen and phosphorus removal rates.

[0138] A model analysis module is configured to analyze the intelligent prediction model for the SBR process effluent quality, and determine the influence indexes affecting the carbon, nitrogen and phosphorus removal rates.

[0139] A model solution module is configured to construct an intelligent optimization model for the SBR process effluent quality based on the influence indexes and the intelligent prediction model for the SBR process effluent quality, solve the optimal solution of the intelligent prediction model for the SBR process effluent quality, and realize the intelligent optimal control of the SBR process effluent quality.

[0140] Embodiment 3

[0141] Please refer to Figure 5 , the application further provides an electronic device 100 for the intelligent optimal control method of the SBR process effluent quality; the electronic device 100 comprises a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and capable of running on the at least one processor 102, and at least one communication bus 104.

[0142] The memory 101 can be used to store the computer program 103, and the processor 102 realizes the steps of the intelligent optimal control method of the effluent quality of the SBR process by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data (such as audio data) created according to the use of the electronic device 100. In addition, the memory 101 can include a non-volatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device.

[0143] The at least one processor 102 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor 102 can be a microprocessor or the processor 102 can also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and connects all parts of the electronic device 100 through various interfaces and lines.

[0144] The memory 101 in the electronic device 100 stores a plurality of instructions to realize an intelligent optimal control method of the effluent quality of the SBR process, and the processor 102 can execute the plurality of instructions to realize:

[0145] obtaining effluent quality indexes affecting carbon, nitrogen and phosphorus removal rates, and constructing an intelligent prediction model of the effluent quality of the SBR process based on the effluent quality indexes affecting the carbon, nitrogen and phosphorus removal rates;

[0146] analyzing the intelligent prediction model of the effluent quality of the SBR process to determine influence indexes affecting the carbon, nitrogen and phosphorus removal rates;

[0147] Based on the influence index and the intelligent prediction model of the SBR process effluent quality, the intelligent optimization model of the SBR process effluent quality is constructed, the optimal solution of the intelligent prediction model of the SBR process effluent quality is solved, and the intelligent optimization and control of the SBR process effluent quality are realized.

[0148] Embodiment 4

[0149] The modules / units integrated in the electronic device 100, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory and read-only memory (ROM).

[0150] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0151] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for performing the functions specified in one or more flows and / or blocks. Figure 1 The device for performing the functions specified in one or more flows and / or blocks.

[0152] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow Figure 1 The flow or flows and / or blocks Figure 1 The flow or flows and / or blocks

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the flow Figure 1 The flow or flows and / or blocks Figure 1 The flow or flows and / or blocks

[0154] The above merely provides the preferred embodiment of the present application and is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent optimal control method for effluent quality of SBR process, characterized in that, The method comprises the following steps: Obtain the effluent quality indexes affecting the removal rates of carbon, nitrogen and phosphorus, and construct an intelligent prediction model of the SBR process effluent quality based on the effluent quality indexes affecting the removal rates of carbon, nitrogen and phosphorus; the intelligent prediction model of the SBR process effluent quality is built by using a BP neural network; Analyze the intelligent prediction model of the SBR process effluent quality to determine the influence indexes affecting the removal rates of carbon, nitrogen and phosphorus; Based on the influence indexes and the intelligent prediction model of the SBR process effluent quality, an intelligent optimization model of the SBR process effluent quality is constructed, and the optimal solution of the intelligent prediction model of the SBR process effluent quality is solved to realize intelligent optimization and control of the SBR process effluent quality; The intelligent prediction model of the SBR process effluent quality is constructed as follows: The input features in the prediction of the SBR process effluent quality are taken as the nodes of the input layer, and the input features include static indexes and dynamic index data of the effluent quality; Each neuron receives input signals from the previous layer, sets weights and biases according to specific problems, selects an activation function, and performs nonlinear conversion on the weighted sum, and the neurons in the hidden layer and the output layer are calculated layer by layer to obtain the output result; The historical data and real-time monitoring data related to the prediction problem of the SBR process effluent quality form a training set, and the training set is used for training of the BP neural network, the weights and biases of the network are optimized by the back propagation algorithm, the BP neural network is trained, and the intelligent prediction model of the SBR process effluent quality is obtained; The method for determining the influence indexes affecting the removal rates of carbon, nitrogen and phosphorus by using gray correlation analysis is as follows: Obtain the data set of the dynamic indexes and static indexes affecting the removal rates of C, N and P; Normalize the data set to map the data range of each index to [0, 1]; According to the normalized data set, the correlation degrees between each index and the prediction target of the SBR process effluent quality are calculated; According to each column of the correlation matrix, the weights of each index and the target are calculated; According to the calculated index weights, the indexes are sorted, and the indexes with higher weights are selected as the influence indexes; The process of constructing the intelligent optimization model of the SBR process effluent quality and solving the optimal solution of the intelligent prediction model of the SBR process effluent quality is as follows: The objective function is the removal rate prediction model of N and P, and the constraint function is the removal rate prediction model of C; Randomly generate an initial population including a plurality of individuals; for each individual in the population, calculate the fitness value according to the objective function and the constraint condition; Non-dominated sorting is performed on the individuals in the population, and the individuals are divided into different levels to form a non-dominated sorting level structure; The crowding degree of the individuals in the population in the target space is calculated; According to the non-dominated sorting and crowding degree calculation results, a certain number of individuals are selected as parents; The selected parents are subjected to crossover and mutation operations to generate new individuals; Repeat the steps of evaluating fitness, non-dominated sorting, crowding degree calculation, selection operation, crossover and mutation until the set number of iterations is met, obtain the Pareto optimal solution, and obtain the optimal solution of the intelligent prediction model of the SBR process effluent quality.

2. The intelligent optimal control method for effluent quality of SBR process according to claim 1, characterized in that, The indicators with high weights are normalized to obtain optimized influence indicators.

3. The intelligent optimal control method for effluent quality of SBR process according to claim 1, characterized in that, The non-dominated sorting genetic algorithm is used to solve the optimal solution of the intelligent prediction model of the effluent quality of the SBR process.

4. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor implements the steps of the intelligent optimal control method of the effluent quality of the SBR process as claimed in any one of claims 1 to 3 when executing the computer program.

5. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 4. The computer program is executed by the processor to implement the steps of the intelligent optimal control method of the effluent quality of the SBR process as claimed in any one of claims 1 to 3.

Images