A method and system for controlling aeration of sewage

Abstract

The present application relates to the technical field of sewage treatment, and relates to a sewage treatment aeration control method and system. According to collected historical data, accurate sample data sets are obtained through intelligent calculation combined with optimization correction to eliminate abnormal values, and then classification is carried out through a clustering model. Each classification is trained, verified and tested to obtain a trained BP neural network model. A subsequent controller outputs accurate aeration quantity through the BP neural network model to control each aeration equipment, and automatically adjusts the increase / decrease of the air volume of a biochemical tank. According to the water quality and quantity, the air volume of the biochemical tank is controlled, and the air volume is matched with the working condition, so that the control precision of the aeration quantity in the sewage treatment process can be effectively improved, energy consumption waste can be reduced, process control efficiency can be improved, the air volume of the biochemical tank under different working conditions can be ensured to be at an optimal value, the water quality stability and the standard rate of effluent can be ensured, and the power consumption cost can be reduced. The present application is suitable for sewage treatment plants of various types and scales, and has a wide market application prospect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sewage treatment, and more particularly to a sewage treatment aeration control method and system. BACKGROUND

[0002] In the activated sludge process, the oxygen aeration amount and the dissolved oxygen control of the biochemical tank determine the denitrification and phosphorus removal effect of the sewage treatment process, and are one of the most important links in the secondary treatment. At present, in the actual operation of most sewage treatment plants, there are two problems: 1. The aeration amount of the biochemical tank is designed only by considering the maximum aeration amount required for removing the design water quality standard under the most unfavorable working condition. However, in the actual operation, the TN of the inflow in the rainy season is about 15-25 mg / L, which is lower than the design value, and the TN of the inflow in the dry season is about 35-40 mg / L, which is higher than the design value. How to adjust the air volume under different working conditions of the rainy season / dry season is not clearly guided in the design specification. At the same time, the display air volume of the air blower is the actual working condition, while the design calculation is under the standard condition. How to convert the standard condition to the actual working condition lacks specific basis. 2. The production scheduling of the present sewage treatment plant relies on the subjective judgment of technical personnel. Due to different operation experiences, the scheduling judgment lacks scientific decision-making and predictive judgment. Frequent manual adjustment and serious lag occur when the water quality fluctuates, which makes it difficult to operate in a fine and standardized manner, resulting in waste of power consumption and increase of labor cost.

[0003] Therefore, how to convert the "experience control" to the "fine control" of the appropriate air volume, improve the effluent water quality, and reduce the power consumption is a problem to be solved. SUMMARY

[0004] The present application aims to overcome the inaccuracy of the aeration amount control of the prior art sewage treatment, and provides a sewage treatment aeration control method and system, which can effectively improve the control accuracy of the aeration amount in the sewage treatment process.

[0005] To solve the above technical problems, the technical scheme adopted by the present application is:

[0006] A sewage treatment aeration control method is provided, comprising the following steps:

[0007] S1. Data acquisition: collecting historical data in a set time period to form a data sample including different inflow working conditions in the rainy season and the dry season;

[0008] S2. Process calculation: establishing a process mechanism calculation model, calculating the theoretical oxygen demand under different inflow working conditions by using the process mechanism calculation model according to the collected historical data, and obtaining a theoretical oxygen demand data set;

[0009] S3. Data optimization: outlier elimination is performed on the theoretical oxygen demand data set and the biochemical pool dissolved oxygen in the historical data to obtain an optimized theoretical oxygen demand data set and a dissolved oxygen sample set;

[0010] S4. Data clustering: the optimized theoretical oxygen demand data set and the biochemical pool dissolved oxygen sample set are classified into different categories using a clustering algorithm.

[0011] S5. Neural network training: a BP neural network model is established, the data of the same category is divided into a training set, a validation set and a test set; the training set is input into the BP neural network model, the BP neural network model is trained, and the trained BP neural network model is obtained, and the BP neural network model is verified and tested by the validation set and the test set.

[0012] S6. Aeration amount prediction: the data of the wastewater pool to be measured are input into the trained BP neural network model to obtain the predicted aeration amount.

[0013] S7. Device control: the aeration device of the wastewater pool is controlled according to the predicted aeration amount to realize accurate aeration.

[0014] The wastewater treatment aeration control method of the present application can obtain accurate sample data set by intelligent calculation combined with optimized correction and outlier elimination according to the collected historical data, and then classify the data through a clustering model. Each classification is trained, verified and tested to obtain a trained BP neural network model. The subsequent controller outputs accurate aeration amount through the BP neural network model according to the latest data to control each aeration device and automatically adjust the increase / decrease of the air volume of the biochemical pool. The present application can control the air volume of the biochemical pool according to the water quality and quantity, match the air volume with the working condition, effectively improve the control precision of the aeration amount in the wastewater treatment process, reduce energy waste, improve the process control efficiency, ensure that the air volume of the biochemical pool is at the optimal value under different working conditions, guarantee the stability and compliance rate of the effluent water quality, and reduce the power consumption cost. The present application is suitable for wastewater treatment plants of various types and scales and has a wide market application prospect.

[0015] As a preferred, in step S1, the historical data of the influent chemical oxygen demand, the influent biochemical oxygen demand, the influent total nitrogen, the effluent total nitrogen, the influent quantity, the biochemical pool dissolved oxygen and the actual air volume are collected at every even integer point time in a certain time period by programming program code automatically, a large number of data sample sets including different influent working conditions in the rainy season and the dry season are formed, and are sorted and saved in an Excel table for subsequent use.

[0016] As a preferred, the process calculation system adopts an actual influent and effluent water quality (chemical oxygen demand, biochemical oxygen demand, total nitrogen), influent quantity operation basic algorithm to calculate the theoretical oxygen demand, and lays a foundation for the subsequent neural network fixed model.

[0017] In step S2, the theoretical oxygen demand is calculated by the following way:

[0018] O2= 0.001aQ(S o - e )-cΔX v +b[0.001Q(N k - ke )-0.12ΔX v ]

[0019] -0.62[0.001Q(N t - ke - oe )-0.12ΔX v ]

[0020] In the formula, O2 is the theoretical oxygen demand, kg O2 / d; a is the oxygen equivalent of carbon, which is taken as 1.47 when the carbon-containing substance is calculated by biochemical oxygen demand; b is a constant, the oxygen demand of oxidase per kilogram of ammonia nitrogen, kg O2 / kg N, taken as 4.57; c is a constant, the oxygen equivalent of bacterial cells, taken as 1.42; Q is the treatment water quantity, m 3 / d; S0 is the influent biochemical oxygen demand, mg / L; Se is the effluent biochemical oxygen demand, mg / L; Nk is the influent total nitrogen, mg / L; Nke is the effluent total nitrogen, mg / L; Nt is the total nitrogen concentration of the influent of the biological reaction tank, mg / L; Noe is the nitrate nitrogen concentration of the effluent of the biological reaction tank, taken as the empirical value of the plant area, mg / L; ΔXv is the microbial quantity discharged from the biological reaction tank system, kg / d; 0.12ΔXv is the nitrogen content of the microorganisms discharged from the biological reaction tank system, kg / d;

[0021] S0=BOD / CODxCOD 进

[0022] Se=BOD / CODxCOD 出

[0023] Wherein, BOD / COD is the ratio of biochemical oxygen demand to chemical oxygen demand in sewage, taken as the monthly average value; COD 进 is the influent chemical oxygen demand, mg / L; COD 出 is the effluent chemical oxygen demand, mg / L.

[0024] As preferred, in step S3, the quartile range IQR criterion is used to detect the abnormal values of each column of the data sample set, the features are the theoretical oxygen demand and the biochemical tank dissolved oxygen quantity, and the label is the actual air volume, the quartile range is calculated, and the upper and lower limits of the critical value are calculated. The abnormal values greater than the upper limit of the critical value and less than the lower limit of the critical value are removed.

[0025] As preferred, in step S4, the optimized and arranged theoretical oxygen demand and biochemical tank dissolved oxygen sample set is subjected to cluster analysis, divided into K categories, and by applying the elbow rule, the K value with the most obvious change in the slope of the descending curve is selected as the optimal clustering number. The cluster algorithm is used to classify the theoretical oxygen demand + reaction tank dissolved oxygen data sample set (not containing the actual air volume). The data sample set is divided into different categories.

[0026] As preferred, in step S5, the BP neural network model system trains and verifies the samples after K-classification, and when the accuracy rate of each test set is weighted and averaged, the total accuracy rate reaches the standard, and then the BP neural network model is saved as the trained BP neural network model. Subsequently, the controller only needs to collect the latest influent water quality, influent water quantity, and biochemical tank dissolved oxygen data of the wastewater treatment plant, and through the background automatic conversion of the programming program code after the theoretical oxygen demand + biochemical tank DO is subjected to the neural network fixed model, an accurate predicted air volume value is obtained and output to the equipment control.

[0027] As preferred, the step S5 comprises:

[0028] Setting the algebra of the training set and the validation set, the input layer, the hidden layer, the output layer, and the number of neurons;

[0029] Using a nonlinear activation, the predicted air volume of each group is obtained through the BP neural network model;

[0030] The loss between the predicted air volume and the actual air volume is calculated through the built-in loss function, and then the gradient of the loss function with respect to the model parameter weight is calculated by back propagation, and the gradient of each parameter is updated;

[0031] The built-in function optimizer of the neural network is used to optimize the weight and bias according to the obtained gradient, so that the loss function is minimized to reduce the error;

[0032] The average error of each generation of training set and validation set is obtained through iteration according to the above steps, and finally the BP neural network model of the generation with the smallest average error is selected, and the test set accuracy rate is tested;

[0033] When the accuracy rate of each test set is weighted and averaged, the total accuracy rate reaches the standard, and then the BP neural network model is saved as the trained BP neural network model.

[0034] As preferred, the equipment control has switching of multiple operation modes, including manual control mode, automatic control mode, and semi-automatic control mode, etc., to meet the needs of different scenarios. The automatic control mode and the semi-automatic control mode can intelligently regulate and control the operation parameters of the biochemical tank blower and the aeration valve according to the latest predicted accurate air volume, to realize accurate aeration of the biochemical tank.

[0035] The application also provides a sewage treatment aeration control system, comprising:

[0036] A data acquisition module is configured to acquire historical data in a set time period to form data samples including different water inflow conditions in rainy season and dry season;

[0037] A process calculation module is configured to establish a process mechanism calculation model, calculate theoretical oxygen demand under different water inflow conditions by using the process mechanism calculation model according to the acquired historical data, and obtain a theoretical oxygen demand data set;

[0038] A data optimization module is configured to remove outliers in the theoretical oxygen demand data set and biochemical pool dissolved oxygen in the historical data to obtain an optimized theoretical oxygen demand data set and a dissolved oxygen sample set;

[0039] Data clustering is performed on the optimized theoretical oxygen demand data set and the biochemical pool dissolved oxygen sample set by using a clustering algorithm to divide them into different categories;

[0040] A BP neural network model module is configured to establish a BP neural network model, divide data in the same category into a training set, a validation set and a test set respectively, input the training set into the BP neural network model, train the BP neural network model to obtain a trained BP neural network model, and verify and test the BP neural network model by using the validation set and the test set;

[0041] An aeration amount prediction module is configured to input data of a sewage pool to be measured into the trained BP neural network model to obtain predicted aeration amount;

[0042] An equipment control module is configured to control aeration equipment of the sewage pool according to the predicted aeration amount to realize accurate aeration.

[0043] The sewage treatment aeration control system comprises a data acquisition module, a process calculation module, an optimization correction module, a clustering algorithm module, a neural network training module, an aeration amount prediction module and an equipment control module, calculates theoretical oxygen demand by acquiring water inflow data, using a classical process algorithm, introducing correction parameters for correction and other technical means, classifies theoretical oxygen demand and its corresponding reaction pool DO historical data sample set by using a clustering algorithm, and finally obtains a fixed model by mining hidden relationships between data through BP neural network model module training and verification. The subsequent controller only needs to acquire the latest water quality, water quantity, biochemical pool DO and air volume, and the precise predicted air volume can be obtained by programming program conversion. The system can be directly connected with the original air blowing aeration system of the sewage plant to participate in the control of the air blower.

[0044] The application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the sewage treatment aeration control method when executing the computer program.

[0045] The application further provides a computer readable storage medium, which stores a computer program, and the computer program realizes the steps of the sewage treatment aeration control method when executed by a processor.

[0046] Compared with the prior art, the application has the following beneficial effects:

[0047] The sewage treatment aeration control method and system can control the air blowing amount of the biochemical tank according to the water quality and water quantity, control the air volume to match the working condition, effectively improve the control precision of the aeration amount in the sewage treatment process, reduce energy waste, improve the process control efficiency, ensure that the air volume of the biochemical tank is at an optimal value under different working conditions, guarantee the water quality stability and compliance rate, reduce the power consumption cost, and is suitable for various types and scales of sewage treatment plants and has a wide market application prospect. BRIEF DESCRIPTION OF DRAWINGS

[0048] Fig. 1 It is a method flowchart of the sewage treatment aeration control method;

[0049] Fig. 2 It is a data processing step schematic diagram of the sewage treatment aeration control method;

[0050] Fig. 3 It is an operation flowchart of the sewage treatment aeration control method. DETAILED DESCRIPTION

[0051] The application will be further described in combination with the specific embodiments. The accompanying drawings are only used for exemplary description, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation on the patent; in order to better illustrate the embodiments of the application, some components in the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some known structures and their descriptions in the drawings can be omitted.

[0052] The same or similar reference numerals in the drawings of the embodiments of the present application correspond to the same or similar components; in the description of the present application, it is understood that if the orientations or positional relationships indicated by the terms "upper", "lower", "left", "right", etc. are based on the orientations or positional relationships shown in the drawings, they are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only used for exemplary illustration, and cannot be understood as a limitation on the present patent, for those skilled in the art, the specific meanings of the above terms can be understood according to the specific circumstances.

[0053] Embodiment one

[0054] The present embodiment is a first embodiment of a sewage treatment aeration control method, as shown in Figs. 1-3 The present embodiment is a first embodiment of a sewage treatment aeration control method, as shown in

[0055] Step S1. Data acquisition: collect historical data in a set time period to form data samples including different water inflow conditions in the rainy season and the dry season.

[0056] In step S1, the historical data of the influent chemical oxygen demand, the influent biochemical oxygen demand, the influent total nitrogen, the effluent total nitrogen, the influent quantity, the dissolved oxygen quantity of the biochemical pool, and the actual air quantity are automatically collected at every even integer point in time in a set time period through a programmed program code to form a large number of data sample sets including different water inflow conditions in the rainy season and the dry season, which are sorted and saved in an Excel table for subsequent use.

[0057] Step S2. Process calculation: establish a process mechanism calculation model, calculate the theoretical oxygen demand under different water inflow conditions according to the collected historical data by using the process mechanism calculation model, and obtain a theoretical oxygen demand data set.

[0058] The process calculation system adopts an actual influent and effluent water quality (chemical oxygen demand, biochemical oxygen demand, total nitrogen), influent quantity operation basic algorithm to calculate the theoretical oxygen demand, which lays a foundation for the subsequent neural network fixed model;

[0059] In step S2, the theoretical oxygen demand is calculated by the following method:

[0060] O2=0.001aQ(S o -S e )-cΔX v +b[0.001Q(N k -N ke )-0.12ΔX v ]

[0061] -0.62[0.001Q(N t -Nke -N oe )-0.12ΔX v ]

[0062] In the formula: O2 is the theoretical oxygen demand, kg O2 / d; a is the oxygen equivalent of carbon, which is 1.47 when the carbon-containing substance is calculated in terms of biochemical oxygen demand; b is a constant, the oxygen demand of oxidase per kilogram of ammonia nitrogen, kg O2 / kg N, which is 4.57; c is a constant, the oxygen equivalent of bacterial cells, which is 1.42; Q is the treatment water quantity, m 3 / d; SO is the influent biochemical oxygen demand, mg / L; Se is the effluent biochemical oxygen demand, mg / L; Nk is the influent total nitrogen, mg / L; Nke is the effluent total nitrogen, mg / L; Nt is the total nitrogen concentration of the influent of the biological reaction tank, mg / L; Noe is the nitrate nitrogen concentration of the effluent of the biological reaction tank, which is an empirical value of the plant, mg / L; ΔXv is the microbial quantity discharged from the biological reaction tank system, kg / d; 0.12ΔXv is the nitrogen content of the microorganisms discharged from the biological reaction tank system, kg / d;

[0063] SO = BOD / COD × COD 进

[0064] Se = BOD / COD × COD 出

[0065] BOD / COD is the ratio of biochemical oxygen demand to chemical oxygen demand in the sewage, and the monthly average value is taken;

[0066] COD 进 is the influent chemical oxygen demand, mg / L; COD 出 is the effluent chemical oxygen demand, mg / L.

[0067] Step S3. Data optimization: the abnormal values in the theoretical oxygen demand data set and the biochemical tank dissolved oxygen in the historical data are removed to obtain the optimized theoretical oxygen demand data set and the dissolved oxygen sample set.

[0068] In step S3, the IQR criterion is used to detect the abnormal values of each column of data in the data sample set, the characteristics are the theoretical oxygen demand and the biochemical tank dissolved oxygen, and the label is the actual air volume. The interquartile range is calculated, and the upper and lower limits of the critical value are calculated. The abnormal values greater than the upper limit of the critical value and less than the lower limit of the critical value are removed.

[0069] Step S4. Data clustering: the optimized theoretical oxygen demand data set and the biochemical tank dissolved oxygen sample set are classified into different categories by using a clustering algorithm.

[0070] In step S4, the optimized and arranged theoretical oxygen demand and biochemical pool dissolved oxygen sample set is subjected to cluster analysis, is divided into K categories, and the elbow rule is applied to select the K value with the most obvious change in the slope of the descending curve as the optimal clustering number. The cluster algorithm is used to classify the theoretical oxygen demand + reaction pool dissolved oxygen data sample set (not containing the actual air volume). The data sample set is divided into different categories.

[0071] Step S5. A BP neural network model is established, the data of the same category are respectively divided into a training set, a validation set and a test set; the training set is input into the BP neural network model, the BP neural network model is trained, a trained BP neural network model is obtained, and the BP neural network model is verified and tested by using the validation set and the test set.

[0072] In step S5, the BP neural network model system trains and verifies the samples after being divided into K categories, the accuracy rate of each category of the test set reaches the total accuracy rate after weighted averaging, and the total accuracy rate reaches the standard. After the total accuracy rate reaches the standard, the BP neural network model is saved as a trained BP neural network model. Subsequently, the controller only needs to collect the latest influent water quality, influent water quantity and biochemical pool dissolved oxygen data of the sewage treatment plant, and an accurate predicted air volume value is obtained after the theoretical oxygen demand + biochemical pool DO is automatically converted by the programming program code background through the neural network fixed model, which is output to the equipment control.

[0073] Step S5 includes:

[0074] The number of generations of the training set and the validation set, the input layer, the hidden layer, the output layer and the number of neurons are set;

[0075] A nonlinear activation is adopted, and the predicted air volume of each group is obtained by the BP neural network model;

[0076] The loss between the predicted air volume and the actual air volume is calculated by using the built-in loss function, and then the gradient of the loss function with respect to the model parameter weight is calculated by back propagation, and the gradient of each parameter is updated;

[0077] The built-in function optimizer of the neural network is used to optimize the weight and bias according to the obtained gradient, so that the loss function is minimized to reduce the error;

[0078] The average error of each generation of the training set and the validation set is obtained by iteration according to the above steps, and finally the BP neural network model of the generation with the minimum average error is selected, and the test set accuracy rate is used;

[0079] When the accuracy rate of each category of the test set is obtained after weighted averaging, the total accuracy rate reaches the standard, and the predicted aeration amount.

[0080] Step S7. The device controller saves the BP neural network model as a trained BP neural network model.

[0081] like Fig. 2 As shown, during the training process of the BP neural network, the learning rate, number of iterations, hidden layers, etc. of the BP neural network model are first set. When the error is less than the specified error or the number of iterations is reached, the predicted air volume is output. When the weighted average accuracy of the air volume is less than the specified error, the trained BP neural network model is obtained. Otherwise, the number of classification clusters is readjusted.

[0082] Step S6. Aeration Volume Prediction: Input the data from the wastewater tank to be tested into the trained BP neural network model to obtain the aeration volume prediction. The control system then controls the aeration equipment in the wastewater tank according to the predicted aeration volume to achieve precise aeration. The equipment control system has multiple operating modes, including manual control, automatic control, and semi-automatic control, to meet the needs of different scenarios. The automatic and semi-automatic control modes can intelligently adjust the operating parameters of the blower and aeration valve in the biological treatment tank based on the latest predicted precise airflow, achieving precise aeration in the biological treatment tank.

[0083] This embodiment of a wastewater treatment aeration control method involves using collected historical data to generate an accurate sample dataset through intelligent calculation and optimization correction to remove outliers. This dataset is then classified using a clustering model. Each classification undergoes training and validation testing to derive a trained BP neural network model. Subsequently, the controller uses the latest data and the precise aeration volume output by the BP neural network model to control each aeration device, automatically adjusting the increase / decrease of airflow in the biological treatment tank. Through process calculation, parameters related to airflow, such as influent water quality, are transformed into a single key output parameter by establishing a process mechanism calculation model. This is followed by clustering model classification and neural network computation to reduce the number of "features," rationally allocate the weights of each "feature," and improve the accuracy of the machine learning algorithm model. This achieves a combined application of the process mechanism calculation model and the BP neural network model. This embodiment controls the aeration volume of the biological treatment tank based on the influent water quality and quantity, and controls the aeration volume to match the operating conditions. This can effectively improve the control accuracy of aeration volume in the sewage treatment process, reduce energy waste, improve process control efficiency, ensure that the aeration volume of the biological treatment tank is at the optimal value under different operating conditions, guarantee the stability and compliance rate of effluent water quality, and reduce power consumption costs. It is applicable to sewage treatment plants of various types and sizes and has a wide range of market application prospects.

[0084] Example 2

[0085] This embodiment is a first embodiment of a wastewater treatment aeration control system, including:

[0086] Data acquisition module: for collecting historical data in a set period of time, forming data samples including different water inflow conditions in rainy season and dry season. Through programming program code, automatically collect historical data of influent chemical oxygen demand, influent biochemical oxygen demand, influent total nitrogen, effluent total nitrogen, influent quantity, biochemical pool dissolved oxygen quantity, actual air quantity at every even integer point in time in a set period of time, form a large number of data sample sets including different water inflow conditions in rainy season and dry season, sort out and save to Excel table for subsequent use.

[0087] Process calculation module: for establishing process mechanism calculation model, calculating theoretical oxygen demand under different water inflow conditions according to collected historical data, obtaining theoretical oxygen demand data set. The process calculation system adopts actual influent and effluent water quality (chemical oxygen demand, biochemical oxygen demand, total nitrogen), influent quantity operation basic algorithm to calculate theoretical oxygen demand, laying a foundation for subsequent neural network fixed model.

[0088] Theoretical oxygen demand is calculated by the following method:

[0089] O2=0.001aQ(S o -S e )-cΔX v +b[0.001Q(N k -N ke )-0.12ΔX v ]

[0090] -0.62[0.001Q(N t -N ke -N oe )-0.12ΔX v ]

[0091] In the formula: O2 is the theoretical oxygen demand, kgO2 / d; ɑ is the oxygen equivalent of carbon, taking 1.47 when the carbon-containing substance is calculated in terms of biochemical oxygen demand; b is a constant, the oxygen demand of oxidase per kilogram of ammonia nitrogen, kgO2 / kgN, taking 4.57; c is a constant, the oxygen equivalent of bacteria cells, taking 1.42; Q is the treatment water quantity, m 3 / d; S0 is the influent biochemical oxygen demand, mg / L; Se is the effluent biochemical oxygen demand, mg / L; Nk is the influent total nitrogen, mg / L; Nke is the effluent total nitrogen, mg / L; Nt is the total nitrogen concentration of the biological reaction tank influent, mg / L; Noe is the nitrate nitrogen concentration of the biological reaction tank effluent, taking the empirical value of the plant area, mg / L; ΔXv is the microbial quantity discharged from the biological reaction tank system, kg / d; 0.12ΔXv is the nitrogen content of the microorganisms discharged from the biological reaction tank system, kg / d;

[0092] S0=BOD / COD×COD 进

[0093] Se = BOD / COD x COD 出

[0094] Wherein, BOD / COD is the ratio of biochemical oxygen demand and chemical oxygen demand in sewage, and the monthly average value is taken;

[0095] COD 进 is the influent chemical oxygen demand, mg / L; COD 出 is the effluent chemical oxygen demand, mg / L.

[0096] Data optimization module: used for removing outliers of biochemical pool dissolved oxygen in theoretical oxygen demand data set and historical data, to obtain optimized theoretical oxygen demand data set and dissolved oxygen sample set. The IQR (interquartile range) criterion is used to detect outliers of each column data in the data sample set, the features are theoretical oxygen demand and biochemical pool dissolved oxygen, and the label is actual air volume. The IQR is calculated, and the upper and lower critical values are calculated. The outliers greater than the upper critical value and less than the lower critical value are removed.

[0097] Data clustering: using clustering algorithm to classify the optimized theoretical oxygen demand data set and biochemical pool dissolved oxygen sample set into different categories. The optimized and sorted theoretical oxygen demand and biochemical pool dissolved oxygen sample set are clustered and classified into K categories, and the elbow rule is applied to select the K value with the most obvious change in the slope of the descending curve as the best clustering number. The clustering algorithm is used to classify the theoretical oxygen demand + reaction pool dissolved oxygen data sample set (not including the label actual air volume). The data sample set is divided into different categories.

[0098] BP neural network model module: used for establishing BP neural network model, dividing the data of the same category into training set, validation set and test set; inputting the training set into the BP neural network model, training the BP neural network model, obtaining the trained BP neural network model, and verifying and testing the BP neural network model with the validation set and test set. Subsequently, the controller only needs to collect the latest influent water quality, influent water quantity and biochemical pool dissolved oxygen data of the sewage treatment plant, and through the background automatic conversion of the programming program code, the theoretical oxygen demand + biochemical pool DO is obtained after the neural network fixed model, and an accurate predicted air volume value is output to the equipment control.

[0099] The processing process includes: setting the algebra of the training set, the validation set, the input layer, the hidden layer, the output layer, and the number of neurons; using a nonlinear activation, obtaining the predicted air volume of each group through the BP neural network model; calculating the loss between the predicted air volume and the actual air volume through the built-in loss function, then calculating the gradient of the loss function with respect to the model parameter weight through back propagation, and updating the gradient of each parameter; using the built-in function optimizer of the neural network and adjusting the weight and bias according to the obtained gradient to minimize the loss function to reduce the error; obtaining the average error of each generation of the training set and the validation set according to the above steps through the iterative method, and finally selecting the BP neural network model of the generation with the minimum average error, and using the test set to test the accuracy; when the total accuracy of the weighted average of the accuracy of each test set is up to standard, the BP neural network model is saved as the trained BP neural network model.

[0100] The aeration amount prediction module is used for inputting the data of the wastewater pool to be tested into the trained BP neural network model to obtain the predicted aeration amount.

[0101] The device control module is used for controlling the aeration device of the wastewater pool according to the predicted aeration amount to realize accurate aeration. The device control has switching of multiple operation modes, including manual control mode, automatic control mode, and semi-automatic control mode, etc. to meet the needs of different scenes. The automatic control mode and the semi-automatic control mode can intelligently regulate and control the operation parameters of the biochemical pool blower and the aeration valve according to the latest predicted accurate air volume to realize accurate aeration of the biochemical pool.

[0102] The wastewater treatment aeration control system of the embodiment includes a data acquisition module, a process calculation module, an optimization correction module, a clustering algorithm module, a neural network training module, an aeration amount prediction module, and a device control module. The theoretical oxygen demand is calculated by collecting the influent data, using the classical process algorithm, introducing the correction parameter correction, and other technical means. After the clustering algorithm classifies the theoretical oxygen demand and the corresponding reaction pool DO historical data sample set, the historical data sample is combined with the "label" actual air volume of the same time period. After training and verification by the BP neural network model module, the hidden relationship between the data is mined, and finally a fixed model is obtained. The subsequent controller only needs to collect the latest influent quality, quantity, biochemical pool DO, and air volume to obtain the accurate predicted air volume through the programming program conversion. The system can be directly involved in the blower control by establishing communication connection with the original blower aeration system of the wastewater treatment plant.

[0103] Embodiment three

[0104] The embodiment is an embodiment of a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the steps of the wastewater treatment aeration control method of the above embodiment one when executing the computer program.

[0105] The application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the wastewater treatment aeration control method.

[0106] Embodiment four

[0107] The embodiment is an embodiment of a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the wastewater treatment aeration control method in the above embodiment one.

[0108] In the specific contents of the above specific embodiments, each technical feature can be combined arbitrarily without contradiction, and for the sake of brevity, all possible combinations of the above technical features are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the description.

[0109] Obviously, the above embodiments of the application are only examples for clearly illustrating the application, and are not intended to limit the implementation modes of the application. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the implementation modes are not required or possible to be exhausted. Any modification, equivalent replacement and improvement made within the spirit and principle of the application should be included in the protection scope of the claims of the application.

Claims

1. A method of controlling aeration of sewage treatment, characterized by, Comprise the following steps: S1. Data collection: collect historical data in a set period of time, form a data sample set including different water inflow conditions in the rainy season and the dry season; automatically collect the historical data of influent chemical oxygen demand, influent biochemical oxygen demand, influent total nitrogen, effluent total nitrogen, influent quantity, biochemical pool dissolved oxygen quantity and actual air quantity at every even integer point in a set time period through programming program code, form a large number of data sample sets including different water inflow conditions in the rainy season and the dry season, and save them in an Excel table after sorting for subsequent use; S2. Process calculation: establish a process mechanism calculation model, calculate the theoretical oxygen demand under different water inflow conditions according to the collected historical data by using the process mechanism calculation model, and obtain a theoretical oxygen demand data set; the theoretical oxygen demand is calculated by the following method: wherein: O2 is the theoretical oxygen demand; ɑ is the oxygen equivalent of carbon, taken as 1.47 when carbonaceous material is measured as biochemical oxygen demand; b is a constant, the oxygen demand of oxidase per kilogram of ammonia nitrogen, taken as 4.57; c is a constant, the oxygen equivalent of bacterial cells, taken as 1.42; Q is the quantity of water treated; So is the influent biochemical oxygen demand; S e is the effluent biochemical oxygen demand; N k is the influent total nitrogen; N ke is the effluent total nitrogen; Nt is the total nitrogen concentration of the influent to the biological reaction tank; Noe is the nitrate nitrogen concentration of the effluent from the biological reaction tank; ΔXv is the quantity of microorganisms discharged from the biological reaction tank system; and 0.12ΔXv is the nitrogen content of the microorganisms discharged from the biological reaction tank system. Wherein, BOD / COD is the ratio of biochemical oxygen demand and chemical oxygen demand in sewage, taking the monthly average; COD 进 is the influent chemical oxygen demand; COD 出 is the effluent chemical oxygen demand; S3. Data optimization: remove outliers in the theoretical oxygen demand data set and the biochemical pool dissolved oxygen quantity in the historical data, and obtain an optimized theoretical oxygen demand data set and a dissolved oxygen quantity sample set; S4. Data clustering: classify the optimized theoretical oxygen demand data set and the biochemical pool dissolved oxygen quantity sample set by using a clustering algorithm, and divide them into different categories; perform clustering analysis on the optimized theoretical oxygen demand and the biochemical pool dissolved oxygen quantity sample set, divide them into K categories, and select the K value with the most obvious change in the slope of the descending curve as the best clustering number by applying the elbow rule; S5. Neural network training: divide the data of the same category into a training set, a validation set and a test set; input the training set into the BP neural network model, train the BP neural network model, obtain a trained BP neural network model, and verify and test the BP neural network model by using the validation set and the test set; wherein, the BP neural network model system trains and verifies the samples after being divided into K categories, and when the accuracy rate of each category reaches the total accuracy rate after weighted averaging, the total accuracy rate meets the standard, and the BP neural network model is saved as the trained BP neural network model; S6. Aeration quantity prediction: input the data of the wastewater pool to be tested into the trained BP neural network model, and obtain the predicted aeration quantity; S7. Equipment control: control the aeration equipment of the wastewater pool according to the predicted aeration quantity, and realize accurate aeration.

2. The wastewater treatment aeration control method according to claim 1, characterized by, In step S3, the IQR rule is used to detect outliers in each column of the data sample set, the features are the theoretical oxygen demand and the biochemical pool dissolved oxygen quantity, the label is the actual air quantity, the interquartile range is calculated, and the upper and lower critical values are calculated; the outliers greater than the upper critical value and less than the lower critical value are removed.

3. The wastewater treatment aeration control method according to claim 1, characterized by, The step S5 comprises: Setting the algebra of the training set and the validation set, the input layer, the hidden layer, the output layer, and the number of neurons; Using a nonlinear activation to obtain the predicted air quantity of each group by the BP neural network model; Calculating the loss between the predicted air quantity and the actual air quantity by using the built-in loss function, and then calculating the gradient of the loss function with respect to the model parameter weight by back propagation, and updating the gradient of each parameter; The built-in function optimizer of the neural network is used to adjust the optimization weight and bias according to the obtained gradient, so that the loss function is minimized to reduce the error; The average error of each generation training set and validation set is obtained through iteration according to the above steps, and finally the BP neural network model with the minimum average error is selected, and the test set is used to test the accuracy; When the total accuracy rate obtained by weighted averaging the accuracy rate of each test set reaches the standard, the BP neural network model is saved as the trained BP neural network model.

4. A wastewater treatment aeration control system characterized by, It comprises: A data acquisition module is used to collect historical data in a set time period to form data samples including different water inflow conditions in the rainy season and the dry season; Through programming program code, historical data of influent chemical oxygen demand, influent biochemical oxygen demand, influent total nitrogen, effluent total nitrogen, influent quantity, biochemical pool dissolved oxygen quantity and actual air quantity are automatically collected at every even integer point in a set time period to form a large number of data sample sets including different water inflow conditions in the rainy season and the dry season, which are sorted and saved in an Excel table for subsequent use; A process calculation module is used to establish a process mechanism calculation model, calculate theoretical oxygen demand under different water inflow conditions according to the collected historical data, and obtain a theoretical oxygen demand data set; the theoretical oxygen demand is calculated by the following formula: wherein: O2 is the theoretical oxygen demand; ɑ is the oxygen equivalent of carbon, taken as 1.47 when carbonaceous material is measured in terms of biochemical oxygen demand; b is a constant, the oxygen demand of oxidase per kilogram of ammonia nitrogen, taken as 4.57; c is a constant, the oxygen equivalent of bacterial cells, taken as 1.42; Q is the quantity of water treated; So is the influent biochemical oxygen demand; S e is the effluent biochemical oxygen demand; N k is the influent total nitrogen; N ke is the effluent total nitrogen; Nt is the total nitrogen concentration of the influent to the biological reaction tank; Noe is the nitrate nitrogen concentration of the effluent from the biological reaction tank; ΔXv is the quantity of microorganisms discharged from the biological reaction tank system; and 0.12ΔXv is the nitrogen content of the microorganisms discharged from the biological reaction tank system. Wherein, BOD / COD is the ratio of biochemical oxygen demand and chemical oxygen demand in sewage, taking the monthly average; COD 进 is the influent chemical oxygen demand; COD 出 is the effluent chemical oxygen demand; A data optimization module is used to remove outliers from the theoretical oxygen demand data set and the biochemical pool dissolved oxygen quantity in the historical data to obtain an optimized theoretical oxygen demand data set and a dissolved oxygen quantity sample set; Data clustering: the optimized theoretical oxygen demand data set and the biochemical pool dissolved oxygen quantity sample set are classified into different categories by using a clustering algorithm; the optimized theoretical oxygen demand and the biochemical pool dissolved oxygen quantity sample set are clustered and analyzed to divide them into K categories, and the K value with the most obvious change in the slope of the descending curve is selected as the best clustering number by applying the elbow rule; A BP neural network model module is used to establish a BP neural network model, divide data of the same category into a training set, a validation set and a test set, input the training set into the BP neural network model, train the BP neural network model to obtain a trained BP neural network model, and verify and test the BP neural network model with the validation set and the test set; wherein, the BP neural network model system trains and verifies the samples after being divided into K categories, and when the accuracy rate of each test set reaches the total accuracy rate obtained by weighted averaging, the total accuracy rate reaches the standard, and the BP neural network model is saved as the trained BP neural network model; An aeration quantity prediction module is used to input data of a wastewater pool to be tested into the trained BP neural network model to obtain predicted aeration quantity; A device control module is used to control the aeration device of the wastewater pool according to the predicted aeration quantity to realize accurate aeration.

5. A computer device comprising a memory and a processor, said memory storing a computer program, characterized in that, The processor executes the computer program to realize the steps of the wastewater treatment aeration control method in any one of claims 1 to 3.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the wastewater treatment aeration control method in any one of claims 1 to 3.

Images