Aeration control method, system, equipment and medium based on support vector regression machine

Abstract

The present application provides an aeration control method, system, equipment and medium based on a support vector regression machine, which relates to the field of sewage treatment. In the method, the operation data of a preset time period is obtained, and the operation data includes historical data and current data, and the historical data and the current data are merged to obtain input data; the input data is screened according to a preset screening method to obtain screened data; the screened data is subjected to a first data preprocessing to obtain first preprocessing feature data; the first preprocessing feature data is input into a pre-trained support vector regression machine for multiple times to obtain a prediction result, and the prediction result includes dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening; according to the dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening, an objective function is constructed, and the objective function is optimized to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value, which can accurately control the aeration level and improve energy efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of sewage treatment, and in particular to an aeration control method, system, equipment and medium based on a support vector regression machine. Background Art

[0002] In the activated sludge process, aeration is a necessary component of the biological treatment stage. Too low aeration concentration will inhibit the metabolic rate of microorganisms, affect the removal effect of microorganisms on pollutants in water, and affect the effluent quality; too high aeration concentration will not only lead to excessive aeration volume, which is easy to cause energy waste, but also the excess aeration will increase the consumption of carbon sources through internal and external reflow, and affect the denitrification of anoxic tanks and the phosphorus release effect of polyphosphate bacteria in anaerobic tanks. Aeration control is a key component in sewage treatment plants, and its energy consumption accounts for 30% to 60% of the total energy consumption. Reasonable aeration control can not only save energy, but also improve the sewage treatment effect. At present, neural networks are used for aeration control to avoid too high or too low aeration concentration. However, in aeration control that requires rapid response, the computational complexity and processing time of neural networks may be limited, and accurate control cannot be achieved due to the inability to respond quickly. Summary of the invention

[0003] The present application provides an aeration control method, system, equipment and medium based on a support vector regression machine, which can respond quickly, accurately regulate the aeration level and improve energy utilization efficiency.

[0004] The technical solution of the embodiment of the present application is as follows:

[0005] In a first aspect, an embodiment of the present application provides an aeration control method based on a support vector regression machine, the method comprising:

[0006] Acquire operation data of a preset time period, the operation data including historical data and current data, and merge the historical data and the current data to obtain input data, wherein the current data and the historical data are continuously updated;

[0007] The input data is screened according to a preset screening method to obtain screened data, wherein the preset screening method is to perform a second data processing on the input sample data in the training phase, and obtain the screened data according to the correlation relationship between the data obtained after the second data processing;

[0008] Performing first data preprocessing on the screened data to obtain first preprocessing feature data;

[0009] Inputting the first preprocessed feature data multiple times into a pre-trained support vector regression machine to obtain a prediction result, wherein the prediction result includes a dissolved oxygen value, an ammonia nitrogen value, a predicted air volume, and a valve opening;

[0010] According to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening, an objective function is constructed, and the objective function is optimized to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value.

[0011] In the above technical solution, the operation data of a preset time period is obtained, and the operation data includes historical data and current data, and the historical data and the current data are merged to obtain input data, wherein the current data and the historical data are constantly updated. By obtaining the historical data and the current data and combining the data, the collected data can be fully utilized, which is beneficial to the subsequent data processing and improves the resource utilization rate; the input data is filtered according to the preset filtering method to obtain the filtered data, wherein the preset filtering method is to perform a second data processing on the input sample data in the training stage, and obtain the data according to the correlation relationship between the data obtained after the second data processing, first filter out the data that meets the conditions The first preprocessing feature data is used to obtain the first preprocessing feature data. Since the data in the historical data and the current data may not be unified, the first preprocessing feature data can be unified. The first preprocessing feature data is input into the pre-trained support vector regression machine for multiple times to obtain the prediction results. The support vector regression machine can better fit the correlation between the data for subsequent precise control. The prediction results include dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening. According to the dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening, the objective function is constructed and optimized to obtain the valve opening, optimized air volume and optimized dissolved oxygen value. By constructing and optimizing the objective function, it is possible to better adapt to changes, learn mutual relationships, and respond quickly to achieve precise aeration control and improve resource utilization.

[0012] In some embodiments of the present application, the training process of the pre-trained support vector regression machine includes:

[0013] Acquire a sample data set for a preset time period, the sample data set including input sample data and real values ​​corresponding to the input sample data, the input sample data being obtained by merging historical sample data and current sample data;

[0014] Performing second data preprocessing on the input sample data to obtain second preprocessing feature data;

[0015] Segmenting the second preprocessed feature data to obtain segmented preprocessed data, and inputting the segmented preprocessed data into a preset initial support vector regression machine to obtain a predicted value;

[0016] Calculate a loss function based on the predicted value and the true value to obtain a value of the loss function, and adjust parameters of the initial support vector regression machine according to the value of the loss function to obtain a target support vector regression machine;

[0017] The target support vector regression machine is cross-validated using the second preprocessed feature data to obtain a test evaluation value, and when the test evaluation value meets a preset training condition, the pre-trained support vector regression machine is obtained.

[0018] In the above technical solution, the support vector regression machine is used to capture nonlinear relationships, performs well in processing data sets with complex structures and nonlinear relationships, maintains a certain degree of flexibility during the training process, and avoids overfitting of training data.

[0019] In some embodiments of the present application, performing second data preprocessing on the input data to obtain second preprocessed feature data includes:

[0020] Performing standardization processing on the input sample data, and then performing normalization processing on the standardized data to obtain normalized feature data;

[0021] Performing correlation calculation on the normalized feature data to obtain a correlation coefficient;

[0022] Calculate the time series relationship between the data in the operation data according to the correlation coefficient to obtain the degree of correlation;

[0023] A causal check is performed on the time series between the data in the operation data according to the correlation degree to obtain the second preprocessing characteristic data.

[0024] In the above technical solution, the second data preprocessing includes standardization, normalization, correlation coefficient calculation, correlation analysis, Granger causality verification and other processing. By performing the above processing on the input sample data, the second preprocessing feature data is obtained, which can not only unify the data, but also analyze the correlation between the data, which is conducive to the subsequent model to better learn the data, so as to achieve accurate control of aeration.

[0025] In some embodiments of the present application, the performing correlation calculation on the normalized feature data to obtain a correlation coefficient includes:

[0026] Using a preset Pearson correlation coefficient to perform correlation calculation on the normalized feature data to obtain a correlation coefficient;

[0027] The formula of the Pearson correlation coefficient is expressed as: ,

[0028] Wherein, r represents the correlation coefficient, n represents the number of the normalized feature data, represents the normalized feature data, represents the average value, Represents standard deviation.

[0029] In the above technical solution, through the correlation operation, each normalized data is calculated with the mean value and the standard deviation, so that the correlation degree between every two variables can be calculated and the direction of the relationship can be pointed out.

[0030] In some embodiments of the present application, constructing an objective function according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening includes:

[0031] Constructing an objective function using a preset function construction formula according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening;

[0032] The preset function construction formula is expressed as:

[0033] G(x)= min(αmax(|f(x1)−a|−0.5,0)+βmax(f(x2)−b,0)+γf(x3))

[0034] Among them, G(x) represents the objective function, f(x1) represents the dissolved oxygen value, f(x2) represents the ammonia nitrogen value, f(x3) represents the air volume, a represents the preset dissolved oxygen setting value, b represents the preset ammonia nitrogen setting value, α, β, and γ represent the adjustment parameters respectively.

[0035] In the above technical solution, by constructing an objective function to find a set of optimal solutions, the possibility of overexposure is avoided, and there is the possibility of selecting an optimal dissolved oxygen-ammonia nitrogen-air volume combination solution from a wider variety of combinations, thereby improving the accurate control of aeration.

[0036] In some embodiments of the present application, the objective function is optimized to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value, including:

[0037] The objective function is optimized for minimum value so that the objective function obtains the minimum value; when the objective function is the minimum value, a corresponding set of the valve opening, the optimized air volume and the optimized dissolved oxygen value is obtained.

[0038] In the above technical scheme, by performing minimum value optimization, a set of solutions corresponding to the minimum of the objective function is selected to obtain the valve opening, optimized air volume and optimized dissolved oxygen value. The valve opening, optimized air volume and optimized dissolved oxygen value can make the energy utilization efficiency higher.

[0039] In some embodiments of the present application, cross-validating the target support vector regression machine using the second preprocessed feature data to obtain a test evaluation value includes:

[0040] Divide the second preprocessed feature data into K parts, and in each iteration process, select 1 / K parts as a validation test set, and use the input sample data except 1 / K parts as a validation training set;

[0041] The target support vector regression machine is trained using the verification training set, and the trained target support vector regression machine is verified using the verification test set to obtain a verification value, which is repeated K times, and the test evaluation value is obtained based on the K verification values.

[0042] In the above technical solution, K-fold cross validation is performed to increase the model's ability to learn data and improve the model's robustness and generalization.

[0043] In a second aspect, an embodiment of the present application provides an aeration control system based on a support vector regression machine, the system comprising:

[0044] A data acquisition module is used to acquire operation data of a preset time period, wherein the operation data includes historical data and current data, and merge the historical data and the current data to obtain input data, wherein the current data and the historical data are continuously updated;

[0045] A data screening module, used to screen the input data according to a preset screening method to obtain screened data, wherein the preset screening method is to perform a second data processing on the input sample data in the training phase, and obtain the data according to the correlation relationship between the data obtained after the second data processing;

[0046] A data processing module, used for performing a first data preprocessing on the screened data to obtain first preprocessing feature data;

[0047] A model processing module, used for inputting the first preprocessed feature data into a pre-trained support vector regression machine for multiple times to obtain a prediction result, wherein the prediction result includes a dissolved oxygen value, an ammonia nitrogen value, a predicted air volume and a valve opening;

[0048] The function construction optimization module is used to construct an objective function according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening, and optimize the objective function to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value.

[0049] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, a user interface, a communication bus and a network interface, wherein the processor, the memory, the user interface and the network interface are respectively connected to the communication bus, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods provided in the first aspect.

[0050] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any one of the methods provided in the first aspect is executed.

[0051] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0052] 1. The method adopts the technical means of obtaining historical data and current data, merging the data, filtering the input data and performing the first data preprocessing, inputting the processed data into the support vector regression machine to obtain the prediction results, constructing the objective function according to the prediction results and optimizing, so as to effectively solve the problem that the related technologies cannot achieve accurate control. By obtaining the operation data of a preset time period, the operation data includes historical data and current data, and merging the historical data and the current data to obtain the input data, wherein the current data and the historical data are constantly updated, by obtaining the historical data and the current data and combining the data, the collected data can be fully utilized, which is beneficial to the subsequent data processing and improving the resource utilization rate; the input data is filtered according to the preset filtering method to obtain the filtered data, wherein the preset filtering method is to perform the second data processing on the input sample data in the training stage, and obtain the data according to the correlation between the data obtained after the second data processing, first filter out the data that meets the conditions, and then filter out the data that meets the conditions. It is beneficial to reduce the amount of calculation in the future; the first data preprocessing is performed on the filtered data to obtain the first preprocessing feature data. Since the data in the historical data and the current data may not be unified, the first data preprocessing can unify the data; the first preprocessing feature data is input into the pre-trained support vector regression machine for multiple times to obtain the prediction results. The support vector regression machine can better fit the correlation between the data for subsequent precise control. The prediction results include dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening; according to the dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening, the objective function is constructed, and the objective function is optimized to obtain the valve opening, optimized air volume and optimized dissolved oxygen value. By constructing and optimizing the objective function, it can better adapt to changes, learn mutual relationships, and respond quickly to achieve precise aeration control and improve resource utilization.

[0053] 2. The dual-model structure is combined with the search for the minimum air volume to avoid the possibility of overexposure, and there are more permutations and combinations to select an optimal aeration-ammonia nitrogen-air volume combination solution.

[0054] 3. The support vector regression machine can capture nonlinear relationships and perform well in processing data sets with complex structures and nonlinear relationships. It maintains a certain flexibility during the training process, avoids overfitting of training data, and improves the generalization and robustness of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of an aeration control method based on a support vector regression machine provided by an embodiment of the present application;

[0056] Figure 2 is a flow chart of an aeration control method based on a support vector regression machine provided by another embodiment of the present application;

[0057] Figure 3 yes Figure 2 A schematic flow chart of a sub-step of step S220;

[0058] Figure 4 Schematic diagram of a support vector regression machine of an aeration control method based on a support vector regression machine provided in one embodiment of the present application;

[0059] Figure 5 It is a structural schematic diagram of an aeration control system based on a support vector regression machine provided by an embodiment of the present application;

[0060] Figure 6 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0062] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0063] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0064] The embodiment of the present application provides an aeration control method, system, electronic device and readable storage medium based on a support vector regression machine. The aeration control method based on the support vector regression machine first obtains the operation data of a preset time period, the operation data includes historical data and current data, and merges the historical data and the current data to obtain input data, wherein the current data and the historical data are constantly updated, and by obtaining the historical data and the current data and combining the data, the collected data can be fully utilized, which is beneficial to the subsequent data processing and improves the resource utilization rate; the input data is filtered according to the preset filtering method to obtain the filtered data, wherein the preset filtering method is to perform a second data processing on the input sample data in the training stage, and according to the second data processing The correlation between the data obtained later is obtained, and the data that meets the conditions is first screened out, which is conducive to reducing the amount of calculation later; the first data preprocessing is performed on the screened data to obtain the first preprocessing feature data. Since the data in the historical data and the current data may not be unified, the data can be unified through the first data preprocessing; the first preprocessing feature data is input into the pre-trained support vector regression machine for multiple times to obtain the prediction results. The support vector regression machine can better fit the correlation between the data for subsequent precise control. The prediction results include dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening; according to the dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening, the objective function is constructed, and the objective function is optimized to obtain the valve opening, optimized air volume and optimized dissolved oxygen value. By constructing and optimizing the objective function, it can better adapt to changes, learn mutual relationships, and respond quickly to achieve precise aeration control and improve resource utilization.

[0065] It should be noted that the aeration control method based on support vector regression machine is applied to the sewage treatment process in sewage treatment and sewage treatment. By controlling the aeration, the concentration of dissolved oxygen in the sewage can be increased, which helps microorganisms to oxidize and decompose organic matter in the sewage under aerobic conditions, improve the efficiency of biochemical reactions, and enhance the overall operating performance and energy-saving effect of the sewage treatment system.

[0066] The technical solution provided in the embodiments of the present application is further described below in conjunction with the accompanying drawings.

[0067] Reference Figure 1 , Figure 1 1 is a flow chart of an aeration control method based on a support vector regression machine provided in an embodiment of the present application. The aeration control method based on a support vector regression machine is applied to an aeration control system based on a support vector regression machine, and the aeration control method based on a support vector regression machine is executed by a processor in an electronic device or a readable storage medium. The aeration control method based on a support vector regression machine includes step S110, step S120, step S130, step S140 and step S150.

[0068] Step S110, obtaining operation data of a preset time period, the operation data including historical data and current data, and merging the historical data and the current data to obtain input data, wherein the current data and the historical data are continuously updated.

[0069] In one embodiment, the preset time period can be the 30 minutes before the current time, or the hour before the current time period. The data of the current time period is the current data, and the data of the previous preset time is the historical data. The historical data and the current data can be the chemical oxygen demand COD of the influent, ammonia nitrogen NH4-N, total phosphorus TP, total nitrogen TN, instantaneous influent flow, internal reflux flow, external reflux flow, anaerobic ORP, anoxic ORP, anoxic effluent nitrate nitrogen, aerobic effluent nitrate nitrogen, activated sludge concentration MLSS in the bioreactor tank, fan flow, air volume, temperature, outlet total phosphorus TP, outlet total nitrogen TN, etc., which are stored in the form of a matrix. The historical data and the current data are added in the form of a matrix, the data is merged, and the input data is obtained to provide data support for subsequent processing. It should be noted that the current data and the historical data are constantly updated as time goes by, but the historical data and the current data are both data within the set time. As the input data is constantly updated, subsequent model training is also continuously conducted within a preset time period to adapt to the constant changes in data, so that it can adapt to changes in the environment, respond in a timely manner, and achieve accurate control of aeration to achieve energy-saving effects.

[0070] Step S120, filtering the input data according to a preset filtering method to obtain filtered data, wherein the preset filtering method is to perform a second data processing on the input sample data during the training phase, and obtain the filtered data based on the correlation relationship between the data obtained after the second data processing.

[0071] In one embodiment, the preset screening method is to obtain the screening method through the correlation relationship between the data obtained in the second data processing stage after performing the second data processing on the input sample data in the training stage. The second data processing is specifically introduced in detail and will not be repeated here. The input data is screened according to the preset screening method to obtain the screened data, which can screen out data with strong correlation, reduce the amount of calculation, and improve resource utilization.

[0072] Step S130, performing first data preprocessing on the filtered data to obtain first preprocessing feature data.

[0073] In one embodiment, the first data preprocessing includes standardization and normalization. Since the units or data representations of the acquired historical data and current data are not uniform, and the dimensions of different indicators are different so that they are not comparable, the data are unified within a range, and all data are de-dimensionalized and consistent. First, the screened data is standardized, and the z-score is used for standardization, and then the standardized data is normalized to obtain the first preprocessed feature data. By standardizing and normalizing the data, the data standard can be unified so that the data can be calculated later.

[0074] Step S140, inputting the first preprocessed feature data into the pre-trained support vector regression machine for multiple times to obtain prediction results, which include dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening.

[0075] In one embodiment, the pre-trained support vector regression machine is a trained model, and the first pre-processed feature data is input into the trained support vector regression machine for multiple times, and the prediction results are output, so that the output prediction results may include dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening. Exemplarily, the relevant data, ammonia nitrogen value, air volume and valve opening are input into the pre-trained support vector regression machine, and the predicted dissolved oxygen value is output. The relevant data, dissolved oxygen value, air volume and valve opening are input into the pre-trained support vector regression machine, and the predicted ammonia nitrogen value is output. The prediction of air volume and valve opening is similar and will not be repeated here. The obtained dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening are conducive to the subsequent construction of the objective function, achieving optimal aeration control, and thus realizing accurate control of aeration. It should be noted that the pre-trained support vector regression machine is continuously trained according to the set interval time over time, and the trained support vector regression machine is used for data prediction, which can better adapt to environmental changes, respond in time, and realize accurate control of aeration to achieve energy-saving effects.

[0076] like Figure 2 As shown, the training process of the pre-trained support vector regression machine includes but is not limited to the following steps:

[0077] Step S210, obtaining a sample data set of a preset time period, the sample data set including input sample data and real values ​​corresponding to the input sample data, the input sample data being obtained by merging historical sample data and current sample data.

[0078] In one embodiment, the preset time period can be the first 30 minutes of the current time, or the first hour of the current time period. The data of the current time period is the current sample data, and the data of the previous preset time period is the historical sample data. The historical sample data and the current sample data can be the chemical oxygen demand COD of the influent, ammonia nitrogen NH4-N, total phosphorus TP, total nitrogen TN, instantaneous influent flow, internal reflow flow, external reflow flow, anaerobic ORP, anoxic ORP, anoxic effluent nitrate nitrogen, aerobic effluent nitrate nitrogen, biological reaction tank activated sludge concentration MLSS, fan flow, air volume, temperature, outlet total phosphorus TP, outlet total nitrogen TN, etc., which are stored in the form of a matrix. The input sample data is obtained by adding the historical sample data and the current sample data in the form of a matrix to complete the data merging. The sample data set also includes the true value corresponding to the input sample data, which provides data support for subsequent model training. It should be noted that the sample data set is also the data collected at the preset time interval, and the data in the sample data set may change over time.

[0079] In another embodiment, the process of collecting input sample data is as follows: collecting chemical oxygen demand COD, ammonia nitrogen NH3-N, total phosphorus TP, total nitrogen TN, pH value, and water volume at the water inlet; collecting anaerobic ORP in the anaerobic tank; collecting aeration valve, air pressure, gas temperature, ammonia nitrogen, aeration DO, gas flow meter, and activated sludge concentration MLSS in the aerobic tank; collecting orthophosphorus, total phosphorus TP, and total nitrogen TN in the high-efficiency tank; and collecting chemical oxygen demand COD, ammonia nitrogen NH3-N, total phosphorus TP, total nitrogen TN, pH value, and water volume at the outlet.

[0080] Step S220, performing second data preprocessing on the input sample data to obtain second preprocessing feature data.

[0081] In one embodiment, the second data preprocessing includes standardization, normalization, correlation coefficient calculation, correlation analysis, Granger causality verification and other processing. By performing the above processing on the input sample data, the second preprocessing feature data is obtained, which can not only unify the data, but also analyze the correlation between the data, which is conducive to the subsequent model to better learn the data to achieve accurate control of aeration.

[0082] like Figure 3 As shown, performing second data preprocessing on the input data to obtain second preprocessed feature data includes but is not limited to the following steps:

[0083] Step S221, standardize the input sample data, and then normalize the standardized data to obtain normalized feature data.

[0084] In some possible embodiments of the present application, due to the problem of inconsistent units or data representation between the collected historical sample data and the current sample data, the dimensions of different indicators are different so that they are not comparable, and the data are unified within a range, and all data are de-dimensionalized and consistent. First, the input sample data is standardized, and the z-score is used for standardization, and then the standardized data is normalized to obtain the first normalized feature data. By standardizing and normalizing the data, the data standard can be unified so that the data can be calculated later.

[0085] Step S222, performing correlation calculation on the normalized feature data to obtain a correlation coefficient.

[0086] In one embodiment, performing correlation calculation on the normalized feature data to obtain a correlation coefficient includes but is not limited to: performing correlation calculation on the normalized feature data using a preset Pearson correlation coefficient to obtain a correlation coefficient;

[0087] The formula for the Pearson correlation coefficient is: ,

[0088] Among them, r represents the correlation coefficient, n represents the number of normalized feature data, represents the normalized feature data, represents the average value, Represents standard deviation.

[0089] In some possible embodiments of the present application, through correlation operations, each normalized data is calculated with the mean and standard deviation, so that the degree of association between every two variables can be calculated, and the direction of the relationship can be pointed out. A positive correlation means that when one variable increases, the other variable also tends to increase; while a negative correlation indicates that when one variable increases, the other variable tends to decrease, so that the potential relationship in the data can be quickly understood.

[0090] Step S223, calculating the time series relationship between the data in the operating data according to the correlation coefficient to obtain the correlation degree.

[0091] In some possible embodiments of the present application, since historical data is involved and in order to further explore the relationship between data in the time series, autocorrelation analysis is performed on the data calculated in step S222 to find a certain degree of correlation between the hidden sequence and the sequence formed by itself after a certain order of lag, and obtain the degree of correlation, that is, the autocorrelation coefficient. According to the correlation coefficient, the time series relationship between the data in the running data is calculated by the autocorrelation analysis algorithm to obtain the degree of correlation. The autocorrelation analysis algorithm can be an ACF algorithm; it can also be a PACF algorithm; it can also be performed first with the ACF algorithm and then with the PACF algorithm (this case is adopted when the influence of other variables is mixed in the correlation coefficient). Taking the ACF algorithm as an example, the formula of the autocorrelation analysis algorithm is expressed as:

[0092] Among them, ACF(k) represents the degree of correlation at lag k, k is the number of lags, and n is the total number of samples. is the value of the time series at time point t, is the mean of the time series.

[0093] Step S224, performing causal verification on the time series between the data in the operation data according to the degree of correlation to obtain second preprocessing feature data.

[0094] In some possible embodiments of the present application, the Granger causality test is used to test whether one set of time series is the cause of another set of time series, that is, to determine whether the change of variable x is the cause of the change of variable y. In the running data, the causal check of the time series between the data in the preset time period is calculated, and the Granger causality check algorithm is used to perform causal check on the time series between the data in the running data according to the degree of correlation, and the second preprocessed feature data is obtained. By performing the Granger causality check algorithm, data with strong correlation is selected from the running data for subsequent model training, which improves the quality of the training data, makes the robustness and generalization of the support vector regression machine better, and thus better adapts to changes.

[0095] In some possible embodiments of the present application, the process of selecting data with strong correlation has certain rules. For example, assuming that there are 10 columns of data, through the above-mentioned correlation calculation, 7 columns of data with strong correlation are selected to obtain the second preprocessing feature data, and the selected 7 columns of data correspond to corresponding category data, such as dissolved oxygen and air volume, etc. The corresponding category data is used as a screening method, and the corresponding category is selected in actual application to input the trained model.

[0096] Step S230, segmenting the second preprocessed feature data to obtain segmented preprocessed data, and inputting the segmented preprocessed data into a preset initial support vector regression machine to obtain a predicted value.

[0097] In one embodiment, the second preprocessed feature data is first divided into a training set and a test set according to a preset ratio. The preset ratio can be 7:3, 6:4, or 8:2, which will not be described in detail here. Taking the preset ratio of 7:3 as an example, the training set accounts for 70% of the total data, and the test set accounts for 30% of the total data. The training set is input into the preset initial support vector regression machine, and the initial support vector regression machine outputs a predicted value so that the model training of the initial support vector regression machine can be performed according to the predicted value later. The initial support vector regression machine is then tested with the test set. Through training and testing, it is beneficial to obtain the target support vector regression machine later.

[0098] Step S240, calculating the loss function based on the predicted value and the true value to obtain the value of the loss function, and adjusting the parameters of the initial support vector regression machine according to the value of the loss function to obtain the target support vector regression machine.

[0099] In one embodiment, according to the predicted value obtained in step S230, a preset loss function is used to calculate the loss function of the predicted value and the true value to obtain the value of the loss function. The preset loss function is a mean-square error loss function (MSE), and the value of the loss function is calculated in each round of iteration. Then, according to the value of the loss function, the parameters of the initial support vector regression machine are adjusted, that is, back propagation is performed to obtain the target support vector regression machine.

[0100] It should be noted that if Figure 4 As shown in the figure, the loss of the initial support vector regression machine is zero only when f(x) is exactly the same as y. For the support vector regression machine, it is usually assumed that there is a tolerable deviation of at most ε between f(x) and y. The loss is calculated only when the absolute value of the difference between f(x) and y is greater than ε. At this time, it is equivalent to building an interval band with a width of 2ε with f(x) as the center. If the training sample falls into this interval band, it is considered to be predicted correctly. The purpose is not only to maximize the interval band ε, but also to minimize the loss, so as to determine the parameters w and b. The initial support vector regression machine is trained according to the above process to obtain a pre-trained support vector regression machine.

[0101] Step S250, cross-validating the target support vector regression machine using the second preprocessed feature data to obtain a test evaluation value, and obtaining a pre-trained support vector regression machine when the test evaluation value meets the preset training conditions.

[0102] In one embodiment, the target support vector regression machine obtained in step S240 is verified to ensure the accuracy of model prediction. The target support vector regression machine is cross-validated using the second preprocessed feature data to obtain a test evaluation value. When the test evaluation value meets the preset training conditions, the training is terminated to obtain a pre-trained support vector regression machine for subsequent aeration control. The preset training condition is that the test evaluation value no longer fluctuates greatly, that is, it tends to a stable state, indicating that the obtained model is relatively stable and has good generalization.

[0103] In another embodiment, when the test evaluation value does not meet the preset training conditions, the obtained target support vector regression machine is still unstable and needs to be trained again. Through continuous iterative training, the test evaluation value meets the preset training conditions, and a pre-trained support vector regression machine is obtained. In order to subsequently use the pre-trained support vector regression machine to predict the dissolved oxygen value and ammonia nitrogen value, it is beneficial to construct and optimize the objective function.

[0104] It should be noted that each time the data input is different, the category of the predicted value is also different. When the input data is related data, ammonia nitrogen value, air volume, etc., the output prediction result is the dissolved oxygen value. When the input data is related data, dissolved oxygen value, and air volume, the output prediction result is the ammonia nitrogen value. A dual model is constructed. By inputting different data, different prediction results can be obtained, so that the pre-trained support vector regression machine can be used to output the dissolved oxygen value, ammonia nitrogen value, air volume and valve opening to obtain a set of optimal aeration control solutions.

[0105] In one embodiment, the target support vector regression machine is cross-validated using the second preprocessed feature data to obtain a test evaluation value, including but not limited to: dividing the second preprocessed feature data into K parts, and in each iteration process, selecting 1 / K parts as a verification test set, and using the input sample data except 1 / K parts as a verification training set; using the verification training set to train the target support vector regression machine, and using the verification test set to verify the trained target support vector regression machine to obtain a verification value, repeating K times, and obtaining a test evaluation value based on the K verification values.

[0106] In some possible embodiments of the present application, K-fold cross validation is used to perform verification and evaluation of the target support vector regression machine. First, the second preprocessed feature data is divided into K parts. In each iterative training process, 1 / K parts are selected as the test set, and the input sample data except 1 / K is used as the training set. Among them, K can be 10. Exemplarily, the second preprocessed feature data is divided into 10 parts, and one of them is taken as the test set each time without repetition, and the other nine parts are used as the training set to train the model. Through K-fold cross validation, the distribution of data can be better learned to avoid overfitting. The target support vector regression machine is trained using the verification training set, and the trained target support vector regression machine is verified by the verification test set. Through K cycles of iteration, K verification values ​​are obtained, and the K verification values ​​are added and averaged to obtain the test evaluation value, which is conducive to the subsequent use of the test evaluation value to determine whether the model is available.

[0107] Step S150, constructing an objective function according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening, optimizing the objective function, and obtaining the valve opening, the optimized air volume and the optimized dissolved oxygen value.

[0108] In one embodiment, a dual model is constructed for the above steps. Exemplarily, an air volume-dissolved oxygen model is constructed using air volume and dissolved oxygen value, and the air volume is adjusted so that when the dissolved oxygen reaches a suitable level, the minimum air volume is output to obtain the optimized air volume. The optimized air volume can make the dissolved oxygen value at a suitable level, so that it is fully dissolved, and resources are saved by controlling the air volume. The air volume-ammonia nitrogen model can also be constructed using air volume and ammonia nitrogen value, and the air volume is adjusted so that when the ammonia nitrogen value reaches a suitable level, the appropriate minimum air volume is output to obtain the optimized air volume. The optimized air volume can make the ammonia nitrogen value at a suitable level, so that it is fully dissolved, and resources are saved by controlling the air volume. It should be noted that the above-mentioned dual model construction combined with the search for the minimum value avoids the possibility of overexposure, and has more permutations and combinations to select an optimal dissolved oxygen-ammonia nitrogen-air volume combination solution. Then, the objective function is constructed by predicting the dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening, and then the objective function is optimized to find a set of optimal solutions that minimize the objective function, so as to accurately control the aeration.

[0109] In yet another embodiment, constructing an objective function according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening includes but is not limited to: constructing an objective function using a preset function construction formula according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening;

[0110] The preset function construction formula is expressed as:

[0111] G(x)= min(αmax(|f(x1)−a|−0.5,0)+βmax(f(x2)−b,0)+γf(x3))

[0112] Among them, G(x) represents the objective function, f(x1) represents the dissolved oxygen value, f(x2) represents the ammonia nitrogen value, f(x3) represents the air volume, a represents the preset dissolved oxygen setting value, b represents the preset ammonia nitrogen setting value, and α, β, and γ represent the adjustment parameters respectively.

[0113] In some possible embodiments of the present application, the above objective function is optimized, and the dissolved oxygen value is limited by a preset dissolved oxygen setting value a, and the dissolved oxygen value is limited to a normal range of 0.5 to the dissolved oxygen setting value a to avoid exceeding the standard. The ammonia nitrogen value is limited to a normal range, which is less than the preset ammonia nitrogen setting value b to avoid exceeding the standard. The proportion of each value is adjusted by α, β, and γ, and G(x) is subsequently optimized to obtain the valve opening, the optimized air volume, and the optimized dissolved oxygen value.

[0114] In one embodiment, the objective function is optimized to obtain the valve opening, optimized air volume and optimized dissolved oxygen value, including but not limited to: performing minimum value optimization on the objective function so that the objective function obtains the minimum value; when the objective function is at the minimum value, a corresponding set of valve opening, optimized air volume and optimized dissolved oxygen value is obtained.

[0115] In some possible embodiments of the present application, according to the objective function G(x) constructed above, the objective function G(x) is optimized for minimum value, and when the dissolved oxygen value reaches the preset dissolved oxygen set value and the ammonia nitrogen value is less than the preset ammonia nitrogen set value, the ammonia nitrogen value is avoided from exceeding the standard so that the value of the objective function is minimized, and the dissolved oxygen value, ammonia nitrogen value, air volume and valve opening corresponding to the minimum of the objective function are selected to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value, and the valve opening, the optimized air volume and the optimized dissolved oxygen value can make the energy utilization efficiency higher. Among them, the valve opening remains unchanged during the optimization process of the objective function, and only the air volume, dissolved oxygen value and ammonia nitrogen value are optimized so that the objective function reaches the minimum value.

[0116] like Figure 5As shown, the embodiment of the present application provides an aeration control system 100 based on a support vector regression machine. The aeration control system 100 based on the support vector regression machine obtains the operation data of a preset time period through a data acquisition module 110. The operation data includes historical data and current data, and the historical data and the current data are merged to obtain input data, wherein the current data and the historical data are constantly updated. By acquiring the historical data and the current data and combining the data, the collected data can be fully utilized, which is beneficial to the subsequent data processing and improves the resource utilization rate; and then the data screening module 120 is used to screen the input data according to the preset screening method to obtain the screened data, wherein the preset screening method is to perform a second data processing on the input sample data in the training stage, and according to the correlation between the data obtained after the second data processing The relationship is obtained, and the data that meets the conditions is first screened out, which is conducive to reducing the amount of calculation in the future; the data processing module 130 is used to perform the first data preprocessing on the screened data to obtain the first preprocessing feature data. Since the data in the historical data and the current data may not be unified, the data can be unified through the first data preprocessing; the first preprocessing feature data is input into the pre-trained support vector regression machine for multiple times through the model processing module 140 to obtain the prediction result. The support vector regression machine can better fit the correlation between the data for subsequent precise control. The prediction results include dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening; finally, the function construction optimization module 150 is used to construct the objective function according to the dissolved oxygen value, ammonia nitrogen value, predicted air volume and valve opening, and the objective function is optimized to obtain the valve opening, optimized air volume and optimized dissolved oxygen value. By constructing and optimizing the objective function, it is possible to better adapt to changes, learn mutual relationships, achieve precise aeration control, and improve resource utilization.

[0117] It should be noted that the data acquisition module 110 is connected to the data screening module 120, the data screening module 120 is connected to the data processing module 130, the data processing module 130 is connected to the model processing module 140, and the model processing module 140 is connected to the function construction optimization module 150. The aeration control method based on the support vector regression machine is applied to the aeration control system 100 based on the support vector regression machine. The aeration control system 100 based on the support vector regression machine obtains historical data and current data, merges the data, and then performs data screening and first data preprocessing on the input data, and inputs the processed data into the support vector regression machine to obtain a prediction result, constructs an objective function according to the prediction result and optimizes it, so as to accurately perform aeration control and improve resource utilization efficiency.

[0118] It should also be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0119] The present application also discloses an electronic device. Figure 6 , Figure 6 The electronic device 500 may include: at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 , and at least one communication bus 502 .

[0120] The communication bus 502 is used to realize the connection and communication between these components.

[0121] The user interface 503 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 503 may also include a standard wired interface and a wireless interface.

[0122] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0123] Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Optionally, the processor 501 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 501 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501, and it can be implemented separately through a chip.

[0124] Among them, the memory 505 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 may also be optionally at least one storage device located away from the aforementioned processor 501. Refer to Figure 6 The memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of an aeration control method based on a support vector regression machine.

[0125] exist Figure 6In the electronic device 500 shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call the application program of an aeration control method based on a support vector regression machine stored in the memory 505. When executed by one or more processors 501, the electronic device 500 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0126] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0128] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0129] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0130] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0131] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0132] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. An aeration control method based on support vector regression machine, characterized in that: The method comprises: Acquire operation data of a preset time period, the operation data including historical data and current data, and merge the historical data and the current data to obtain input data, wherein the current data and the historical data are continuously updated; The input data is screened according to a preset screening method to obtain screened data, wherein the preset screening method is to perform a second data preprocessing on the input sample data in the training phase, and obtain the screened data according to the correlation relationship between the data obtained after the second data preprocessing; Performing first data preprocessing on the screened data to obtain first preprocessing feature data; Inputting the first preprocessed feature data multiple times into a pre-trained support vector regression machine to obtain a prediction result, wherein the prediction result includes a dissolved oxygen value, an ammonia nitrogen value, a predicted air volume, and a valve opening; According to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening, construct an objective function, optimize the objective function, and obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value; The constructing of an objective function according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening comprises: Constructing an objective function using a preset function construction formula according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening; The preset function construction formula is expressed as: G(x)=min(αmax(|f(x1)−a|−0.5,0)+βmax(f(x2)−b,0)+γf(x3)) Among them, G(x) represents the objective function, f(x1) represents the dissolved oxygen value, f(x2) represents the ammonia nitrogen value, f(x3) represents the air volume, a represents the preset dissolved oxygen setting value, b represents the preset ammonia nitrogen setting value, α, β, and γ represent the adjustment parameters respectively.

2. The method according to claim 1, characterized in that The training process of the pre-trained support vector regression machine includes: Acquire a sample data set for a preset time period, the sample data set including input sample data and real values ​​corresponding to the input sample data, the input sample data being obtained by merging historical sample data and current sample data; Performing second data preprocessing on the input sample data to obtain second preprocessing feature data; Segmenting the second preprocessed feature data to obtain segmented preprocessed data, and inputting the segmented preprocessed data into a preset initial support vector regression machine to obtain a predicted value; Calculate a loss function based on the predicted value and the true value to obtain a value of the loss function, and adjust parameters of the initial support vector regression machine according to the value of the loss function to obtain a target support vector regression machine; The target support vector regression machine is cross-validated using the second preprocessed feature data to obtain a test evaluation value, and when the test evaluation value meets a preset training condition, the pre-trained support vector regression machine is obtained.

3. The method according to claim 2, characterized in that The performing second data preprocessing on the input data to obtain second preprocessing feature data includes: Performing standardization processing on the input sample data, and then performing normalization processing on the standardized data to obtain normalized feature data; Performing correlation calculation on the normalized feature data to obtain a correlation coefficient; Calculate the time series relationship between the data in the operation data according to the correlation coefficient to obtain the correlation degree; A causal check is performed on the time series between the data in the operation data according to the correlation degree to obtain the second preprocessing characteristic data.

4. The method according to claim 3, characterized in that The performing correlation calculation on the normalized feature data to obtain a correlation coefficient includes: Using a preset Pearson correlation coefficient to perform correlation calculation on the normalized feature data to obtain a correlation coefficient; The formula of the Pearson correlation coefficient is expressed as: ， Wherein, r represents the correlation coefficient, n represents the number of normalized feature data, represents the normalized feature data, represents the average value, Represents standard deviation.

5. The method according to claim 1 or 4, characterized in that: The optimizing the objective function to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value includes: The objective function is optimized for minimum value so that the objective function obtains the minimum value; when the objective function is the minimum value, a corresponding set of the valve opening, the optimized air volume and the optimized dissolved oxygen value is obtained.

6. The method according to claim 2, characterized in that The cross-validation of the target support vector regression machine using the second pre-processed feature data to obtain a test evaluation value includes: Divide the second preprocessed feature data into K parts, and in each iteration process, select 1 / K parts as a validation test set, and use the input sample data except 1 / K parts as a validation training set; The target support vector regression machine is trained using the verification training set, and the trained target support vector regression machine is verified using the verification test set to obtain a verification value, which is repeated K times, and the test evaluation value is obtained based on the K verification values.

7. An aeration control system based on support vector regression machine, characterized in that: The system comprises: A data acquisition module (110) is used to acquire operation data of a preset time period, wherein the operation data includes historical data and current data, and merge the historical data and the current data to obtain input data, wherein the current data and the historical data are continuously updated; A data screening module (120) is used to screen the input data according to a preset screening method to obtain screened data, wherein the preset screening method is to perform a second data preprocessing on the input sample data during the training phase, and obtain the data according to the correlation relationship between the data obtained after the second data preprocessing; A data processing module (130) is used to perform first data preprocessing on the screened data to obtain first preprocessing feature data; A model processing module (140) is used to input the first preprocessed feature data into a pre-trained support vector regression machine for multiple times to obtain a prediction result, wherein the prediction result includes a dissolved oxygen value, an ammonia nitrogen value, a predicted air volume and a valve opening; A function construction optimization module (150) is used to construct an objective function according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening, and optimize the objective function to obtain the valve opening, the optimized air volume and the optimized dissolved oxygen value; A function construction optimization module (150), specifically configured to construct an objective function using a preset function construction formula according to the dissolved oxygen value, the ammonia nitrogen value, the predicted air volume and the valve opening; The preset function construction formula is expressed as: G(x)=min(αmax(|f(x1)−a|−0.5,0)+βmax(f(x2)−b,0)+γf(x3)) Among them, G(x) represents the objective function, f(x1) represents the dissolved oxygen value, f(x2) represents the ammonia nitrogen value, f(x3) represents the air volume, a represents the preset dissolved oxygen setting value, b represents the preset ammonia nitrogen setting value, α, β, and γ represent the adjustment parameters respectively.

8. An electronic device, characterized in that: The electronic device (500) comprises a processor (501), a memory (505), a user interface (503), a communication bus (502) and a network interface (504), wherein the processor (501), the memory (505), the user interface (503) and the network interface (504) are respectively connected to the communication bus (502), the memory (505) is used to store instructions, the user interface (503) and the network interface (504) are used to communicate with other devices, and the processor (501) is used to execute the instructions stored in the memory (505) so that the electronic device (500) executes the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 6 is performed.

Images