Sewage nitrogen and phosphorus removal treatment method and system based on data driving

Abstract

The invention discloses a sewage nitrogen and phosphorus removal treatment method and system based on data driving, and relates to the technical field of sewage treatment.The sewage is subjected to nitrogen and phosphorus removal treatment according to a preset nitrogen and phosphorus removal treatment pond combination scheme when a sensor arranged at the inlet end of a sewage pipe detects that sewage enters the sewage pipe, and after one-time sewage treatment is completed, the sewage is subjected to nitrogen and phosphorus removal treatment according to the preset nitrogen and phosphorus removal treatment pond combination scheme; the method comprises the following steps: acquiring operation data of each denitrification and dephosphorization treatment pond, analyzing an abnormal condition of each denitrification and dephosphorization treatment pond through an abnormal model, performing clustering processing according to the abnormal condition, generating a plurality of combination schemes based on a clustering result, evaluating the performance of each combination scheme by using a performance function, and performing secondary optimization on the combination schemes to obtain a plurality of denitrification and dephosphorization treatment ponds. And applying the combined scheme after secondary optimization to the next sewage treatment. The treatment system can dynamically adjust a combination scheme of the anoxic tank, the anaerobic tank and the aerobic tank in combination with an actual treatment scene, so that the sewage treatment efficiency and treatment effect are improved.

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment technology, and specifically to a data-driven wastewater denitrification and phosphorus removal method and system. Background Technology

[0002] Wastewater denitrification and phosphorus removal is a key part of urban wastewater treatment. Its main purpose is to remove nitrogen and phosphorus from the water to prevent them from entering natural water bodies and causing eutrophication, which in turn leads to water quality deterioration and ecological problems. Nitrogen and phosphorus are the main substances that cause eutrophication. Excessive nitrogen and phosphorus can promote algal growth, leading to algal blooms, affecting the dissolved oxygen content in the water, which in turn affects the growth of aquatic organisms and may even damage the ecosystem.

[0003] To improve the efficiency of nitrogen and phosphorus removal in wastewater treatment, wastewater treatment plants typically employ a combination of anoxic, anaerobic, and aerobic tanks. Large-scale wastewater treatment plants, due to the large volume of wastewater they treat, usually have multiple anoxic, anaerobic, and aerobic tanks connected in a multi-tank configuration to treat wastewater simultaneously.

[0004] The existing technology has the following drawbacks:

[0005] In practical applications, treatment systems typically select a fixed combination of anoxic, anaerobic, and aerobic tanks based on a pre-set wastewater treatment route. However, due to the varying health conditions of each anoxic, anaerobic, and aerobic tank, this combination may not be optimal in practice. This could lead to a decrease in overall wastewater treatment efficiency, or the inability to use a particular tank, resulting in wastewater treatment interruption and compromising treatment efficiency and effectiveness.

[0006] Based on this, the present invention proposes a data-driven wastewater denitrification and phosphorus removal method and system, which can dynamically adjust the combination scheme of anoxic tank, anaerobic tank and aerobic tank according to the actual treatment scenario, thereby improving wastewater treatment efficiency and treatment effect. Summary of the Invention

[0007] The purpose of this invention is to provide a data-driven wastewater denitrification and phosphorus removal method and system to address the shortcomings of the prior art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a data-driven wastewater denitrification and phosphorus removal method, the method comprising the following steps:

[0009] The data acquisition terminal obtains information on all denitrification and phosphorus removal treatment tanks in the wastewater treatment plant through the API interface of the wastewater treatment plant management platform, numbers and marks all denitrification and phosphorus removal treatment tanks, and when the sensor installed at the wastewater pipe inlet detects that wastewater is entering, it performs denitrification and phosphorus removal treatment on the wastewater according to the preset denitrification and phosphorus removal treatment tank combination scheme.

[0010] After completing a wastewater treatment cycle, the treatment unit obtains the operating data of each nitrogen and phosphorus removal treatment tank, analyzes the abnormal conditions of each nitrogen and phosphorus removal treatment tank through anomaly model, and performs clustering based on the abnormal conditions.

[0011] Several combination schemes are generated based on the clustering results. The performance of each combination scheme is evaluated using a performance function, and then the combination schemes are optimized in the second stage. The optimized combination schemes are then applied to the next wastewater treatment.

[0012] Preferably, after completing one wastewater treatment cycle, the treatment unit acquires operational data from each nitrogen and phosphorus removal treatment tank, including the following steps:

[0013] The denitrification and phosphorus removal treatment ponds include anoxic ponds, anaerobic ponds, and aerobic ponds;

[0014] Obtain operational data for the anoxic tank, including aeration fluctuation amplitude, dissolved oxygen concentration rise rate, and denitrifying bacteria decay index;

[0015] Obtain operational data for the anaerobic tank, including aeration fluctuation amplitude, water flow distribution fluctuation amplitude, and anaerobic fermentation bacteria attenuation index;

[0016] Obtain operational data for the aerobic tank, including aeration fluctuation amplitude, dissolved oxygen concentration decrease rate, and nitrifying bacteria decay index.

[0017] Preferably, the abnormal conditions of each nitrogen and phosphorus removal treatment tank are analyzed using anomaly models, including the following steps:

[0018] The operating data of each nitrogen and phosphorus removal treatment tank is substituted into the anomaly model, and the anomaly model outputs the anomaly coefficient for each nitrogen and phosphorus removal treatment tank. The expression of the anomaly model is as follows:

[0019] In the formula, The outlier coefficient, As variables, These are the regression coefficients of each variable, and the regression coefficients are greater than 0;

[0020] The obtained anomaly coefficients are compared with preset anomaly thresholds. The anomaly thresholds are used to analyze whether the denitrification and phosphorus removal treatment tanks can support operation.

[0021] If the abnormality coefficient of the denitrification and phosphorus removal treatment tank is less than or equal to the abnormality threshold, the analysis shows that there is no abnormality in the denitrification and phosphorus removal treatment tank and operation is supported. If the abnormality coefficient of the denitrification and phosphorus removal treatment tank is greater than the abnormality threshold, the analysis shows that there is an abnormality in the denitrification and phosphorus removal treatment tank and operation is not supported.

[0022] Preferably, clustering based on anomalies includes the following steps:

[0023] Close the valves corresponding to the denitrification and phosphorus removal treatment tanks that do not support operation, obtain the abnormality coefficients of the remaining denitrification and phosphorus removal treatment tanks, and then cluster the remaining anoxic tanks, remaining anaerobic tanks, and remaining aerobic tanks.

[0024] Clustering of the remaining anoxic pools includes the following steps:

[0025] S1: Suppose that the remaining anoxic pools are clustered into K clusters, and randomly select the anomaly coefficients of K remaining anoxic pools as the initial cluster center values;

[0026] S2: Calculate the difference between the anomaly coefficient of each anoxic pool and the initial cluster center value, and assign the anoxic pool to the cluster with the smallest difference;

[0027] S3: After all anoxic pools have been allocated, calculate the mean anomaly coefficient of each cluster as the new cluster center value;

[0028] S4: Repeat steps S2 and S3. If the difference between the new cluster center value of the current iteration and the new cluster center value of the previous iteration is less than or equal to the difference threshold, it is determined that the convergence condition is met, and all clusters are output.

[0029] Preferably, several combination schemes are generated based on the clustering results, including the following steps:

[0030] After obtaining the clusters in the anoxic, anaerobic, and aerobic pools, calculate the mean anomaly coefficients of the K anoxic pool clusters, the mean anomaly coefficients of the K anaerobic pool clusters, and the mean anomaly coefficients of the K aerobic pool clusters.

[0031] Sort all anoxic pool clusters in ascending order of their mean anomaly coefficients; sort all anaerobic pool clusters in ascending order of their mean anomaly coefficients; sort all aerobic pool clusters in ascending order of their mean anomaly coefficients.

[0032] Based on the ranking results, select the corresponding anoxic pool cluster, anaerobic pool cluster, and aerobic pool cluster to combine them, resulting in several combination schemes.

[0033] Preferably, after evaluating the performance of each combination scheme using a performance function, a secondary optimization of the combination scheme is performed, including the following steps:

[0034] After obtaining all the combined schemes, obtain the distance index of the denitrification and phosphorus removal treatment tank and the water flow transmission time in each combined scheme, and calculate the performance value of each combined scheme through the performance function;

[0035] The performance values ​​of the combined schemes with the same denitrification and phosphorus removal treatment tank number are compared. The combined scheme with the larger performance value is retained, and the combined scheme with the smaller performance value is deleted.

[0036] Preferably, the distance index of the denitrification and phosphorus removal treatment tanks and the water flow transmission time in each combination scheme are obtained, and the performance value of each combination scheme is calculated using a performance function, the expression of which is: In the formula, For performance values, The distance index for nitrogen and phosphorus removal treatment ponds. For the duration of water flow transmission, , It is the proportionality coefficient, and the proportionality coefficient , Greater than 0.

[0037] Preferably, the calculation logic for the distance index of the denitrification and phosphorus removal treatment tanks is as follows: The distance between any two denitrification and phosphorus removal treatment tanks in the combined scheme is calculated using the following expression:

[0038] In the formula, The distance between two nitrogen and phosphorus removal treatment tanks. Let's define the coordinates of a nitrogen and phosphorus removal treatment tank. Let the coordinates of another nitrogen and phosphorus removal treatment tank be used, and mark the distance between the anoxic tank and the anaerobic tank as follows: The distance between the anoxic tank and the aerobic tank was marked as... The distance between the anaerobic tank and the aerobic tank was marked as... The formula for calculating the distance index of the denitrification and phosphorus removal treatment tank is: .

[0039] Preferably, when a sensor installed at the inlet of the sewage pipe detects the entry of sewage, the sewage is treated for nitrogen and phosphorus removal according to a preset denitrification and phosphorus removal treatment tank combination scheme, including the following steps:

[0040] The preset combination scheme defines the combination of anoxic, anaerobic, and aerobic tanks with various numbers;

[0041] According to the selected scheme, the corresponding anoxic tank is started. In the anoxic tank, denitrifying bacteria consume nitrates in the wastewater and remove nitrogen. The corresponding anaerobic tank is started. In the anaerobic tank, polyphosphate-accumulating bacteria release phosphorus and absorb organic matter in the wastewater. The corresponding aerobic tank is started. In the aerobic tank, nitrifying bacteria convert ammonia nitrogen in the wastewater into nitrates. At the same time, polyphosphate-accumulating bacteria reabsorb phosphorus in an aerobic environment.

[0042] A data-driven wastewater nitrogen and phosphorus removal system includes a labeling module, a data acquisition module, a cluster analysis module, and a solution application module.

[0043] Tagging module: Obtain information on all nitrogen and phosphorus removal treatment tanks in the wastewater treatment plant through the API interface of the wastewater treatment plant management platform, and number and tag all nitrogen and phosphorus removal treatment tanks;

[0044] Data Acquisition Module: When the sensor installed at the inlet of the sewage pipe detects the entry of sewage, it performs denitrification and phosphorus removal treatment on the sewage according to the preset denitrification and phosphorus removal treatment tank combination scheme. After completing one sewage treatment cycle, it acquires the operating data of each denitrification and phosphorus removal treatment tank.

[0045] Clustering analysis module: Analyzes the abnormal conditions of each nitrogen and phosphorus removal treatment tank through anomaly model, and performs clustering based on the abnormal conditions;

[0046] Solution application module: Based on the clustering results, several combined solutions are generated. After evaluating the performance of each combined solution using a performance function, the combined solutions are optimized a second time. The optimized combined solutions are then applied to the next wastewater treatment.

[0047] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0048] This invention utilizes sensors installed at the inlet of the sewage pipe to detect sewage entering. Based on a pre-defined combination scheme of denitrification and phosphorus removal treatment tanks, the sewage undergoes denitrification and phosphorus removal treatment. After completing one sewage treatment cycle, operational data from each denitrification and phosphorus removal treatment tank is acquired. Anomaly models are used to analyze the abnormal conditions of each tank, and clustering is performed based on these anomalies. Several combination schemes are generated based on the clustering results. The performance of each combination scheme is evaluated using a performance function, and then the schemes are optimized a second time. The optimized combination schemes are then applied to the next sewage treatment cycle. The treatment system can dynamically adjust the combination scheme of anoxic, anaerobic, and aerobic tanks according to the actual treatment scenario, thereby improving sewage treatment efficiency and effectiveness. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0050] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1: Please refer to Figure 1 As shown in this embodiment, a data-driven wastewater nitrogen and phosphorus removal method includes the following steps:

[0053] The data acquisition unit obtains information on all nitrogen and phosphorus removal treatment tanks within the wastewater treatment plant through the API interface of the wastewater treatment plant management platform. This information includes the number and location of anoxic, anaerobic, and aerobic tanks. All nitrogen and phosphorus removal treatment tanks are numbered and marked. When a sensor installed at the wastewater inlet detects wastewater entering, the wastewater is treated for nitrogen and phosphorus removal according to a preset combination scheme of nitrogen and phosphorus removal treatment tanks. After completing one wastewater treatment cycle, the processing unit obtains the operating data of each nitrogen and phosphorus removal treatment tank, analyzes the abnormal conditions of each tank using an anomaly model, and performs clustering based on the anomalies. Several combination schemes are generated based on the clustering results. The performance of each combination scheme is evaluated using a performance function, and the combination schemes are then optimized a second time. The optimized combination schemes are then applied to the next wastewater treatment cycle.

[0054] This application utilizes sensors installed at the inlet of the sewage pipe to detect sewage entering. Based on a pre-defined combination scheme of nitrogen and phosphorus removal treatment tanks, the sewage undergoes nitrogen and phosphorus removal treatment. After completing one sewage treatment cycle, operational data from each nitrogen and phosphorus removal treatment tank is acquired. Anomaly analysis is performed on the abnormal conditions of each tank using an anomaly model, and clustering is conducted based on these anomalies. Several combination schemes are generated based on the clustering results. The performance of each combination scheme is evaluated using a performance function, and then the scheme is optimized a second time. The optimized combination scheme is then applied to the next sewage treatment cycle. The treatment system can dynamically adjust the combination scheme of anoxic, anaerobic, and aerobic tanks according to the actual treatment scenario, thereby improving sewage treatment efficiency and effectiveness.

[0055] Example 2: The data acquisition terminal obtains information on all nitrogen and phosphorus removal treatment tanks within the wastewater treatment plant through the API interface of the wastewater treatment plant management platform. This information includes the number and location of anoxic, anaerobic, and aerobic tanks. All nitrogen and phosphorus removal treatment tanks are then numbered and marked, including the following steps:

[0056] Request API: The data acquisition terminal sends a request to the wastewater treatment plant management platform to call the API interface provided by it.

[0057] Interface address: The URL of the interface (e.g., http: / / api.swwtp.com / get_pools).

[0058] Request method: Depending on the platform's supported methods, it is usually a GET request.

[0059] Request parameters include authorization information (such as API key, token) and query conditions (such as wastewater treatment plant number, region, treatment type, etc.).

[0060] Example request:

[0061] GET-http: / / api.swwtp.com / get_pools?plant_id=12345&filter=denitrification-Authorization:Bearer <token>

[0062] The API interface returns detailed information about all nitrogen and phosphorus removal treatment tanks in the wastewater treatment plant, including the number, location, and type of anoxic, anaerobic, and aerobic tanks. The returned data is typically in JSON format.

[0063] {

[0064] "status":"success",

[0065] "data":[

[0066] {

[0067] "pool_id":1,

[0068] "type":"anoxic",

[0069] "location":"SectionA,Zone1"

[0070] },

[0071] {

[0072] "pool_id":2,

[0073] "type":"anaerobic",

[0074] "location":"SectionB,Zone3"

[0075] },

[0076] {

[0077] "pool_id":3,

[0078] "type":"aerobic",

[0079] "location":"SectionC,Zone2"

[0080] } ]

[0082] }

[0083] Parse the pool information returned by the API, extract the basic data of each pool (such as pool ID, type, and location), and classify and organize the pool information according to pool type (anoxic pool, anaerobic pool, and aerobic pool) for easy subsequent processing and numbering.

[0084] Numbering rules: Based on the internal standards of the wastewater treatment plant, each pool will be assigned a unique number. The numbering format can consider the following methods:

[0085] Example number: P-01-AN (P represents the pool, 01 is the pool number, and AN represents the pool type, such as "anoxic pool").

[0086] Pool Number: Assign a unique number to each pool to avoid duplication and ensure that each pool has a unique identifier.

[0087] Assigning values ​​to numbers:

[0088] Anoxic tanks: Number them sequentially according to their number, such as P-001-AN, P-002-AN.

[0089] Anaerobic tanks: Number them sequentially according to the number of anaerobic tanks, such as P-001-AA, P-002-AA.

[0090] Aerobic tanks: Number them sequentially according to the number of aerobic tanks, such as P-001-AO, P-002-AO.

[0091] Example results:

[0092] {

[0093] "status":"success",

[0094] "data":[

[0095] {

[0096] "pool_id":"P-001-AN",

[0097] "type":"anoxic",

[0098] "location":"SectionA,Zone1"

[0099] },

[0100] {

[0101] "pool_id":"P-001-AA",

[0102] "type":"anaerobic",

[0103] "location":"SectionB,Zone3"

[0104] },

[0105] {

[0106] "pool_id":"P-001-AO",

[0107] "type":"aerobic",

[0108] "location":"SectionC,Zone2"

[0109] } ]

[0111] }

[0112] The pool number and pool information are combined to generate a final pool information list. This list will include information such as pool number, type, and location, facilitating subsequent use by the data acquisition end.

[0113] {

[0114] "status":"success",

[0115] "data":[

[0116] {

[0117] "pool_id":"P-001-AN",

[0118] "type":"anoxic",

[0119] "location":"SectionA,Zone1"

[0120] },

[0121] {

[0122] "pool_id":"P-002-AN",

[0123] "type":"anoxic",

[0124] "location":"SectionA,Zone2"

[0125] },

[0126] {

[0127] "pool_id":"P-003-AA",

[0128] "type":"anaerobic",

[0129] "location":"SectionB,Zone1"

[0130] },

[0131] {

[0132] "pool_id":"P-004-AO",

[0133] "type":"aerobic",

[0134] "location":"SectionC,Zone1"

[0135] } ]

[0137] }

[0138] Information about the nitrogen and phosphorus removal treatment tanks, including their numbers and locations, is sent to relevant systems, such as databases and monitoring platforms, for subsequent monitoring and management.

[0139] When a sensor installed at the inlet of the sewage pipe detects the entry of sewage, the sewage is treated for nitrogen and phosphorus removal according to a pre-set denitrification and phosphorus removal treatment tank combination scheme, including the following steps:

[0140] Sensors installed at the inlet of the sewage pipe (e.g., flow rate, ammonia nitrogen, turbidity, pH sensors) monitor the sewage flow rate, mass, and pollutant concentration in real time. The sensors determine if sewage is flowing in, and if the pollutant concentration (e.g., ammonia nitrogen, phosphate) exceeds a predetermined threshold, triggering subsequent treatment processes. Wastewater sensor data is transmitted to the treatment system in real time via a network or control system. The data acquisition system obtains real-time data about the sewage from the sensors, including parameters such as flow rate, ammonia nitrogen, phosphate, pH, and dissolved oxygen. The system analyzes the pollutant load of the sewage to determine whether denitrification and phosphorus removal treatment are necessary.

[0141] Pool combination scheme details: Pre-defined schemes typically define combinations of anoxic, anaerobic, and aerobic pools with specific numbering. For example:

[0142] Option 1: Use anoxic tank 1, anaerobic tank 2, and aerobic tank 3.

[0143] Option 2: Use anoxic tank 2, anaerobic tank 1, and aerobic tank 2.

[0144] Option 3: Use 3 anoxic tanks, 3 anaerobic tanks, and 1 aerobic tank.

[0145] The system automatically selects the most suitable pool combination scheme based on the current sewage treatment needs.

[0146] Start the anoxic tank: According to the selected scheme, start the corresponding anoxic tank (e.g., tank number P-001-AN). Inside the anoxic tank, denitrifying bacteria will consume nitrates in the wastewater and remove nitrogen.

[0147] Start the anaerobic tank: According to the plan, start the corresponding anaerobic tank (e.g., tank number P-001-AA). In the anaerobic tank, polyphosphate-accumulating bacteria will release phosphorus and absorb organic matter in the wastewater.

[0148] Start the aerobic tank: According to the selected scheme, start the corresponding aerobic tank (e.g., tank number P-001-AO). In the aerobic tank, nitrifying bacteria will convert ammonia nitrogen in the wastewater into nitrate, while polyphosphate-accumulating bacteria will reabsorb phosphorus in an aerobic environment.

[0149] Water flow scheduling: Based on the sewage flow rate and pool capacity, the flow rate of sewage through each pool is reasonably adjusted to ensure that each pool operates under optimal conditions.

[0150] Reflux ratio adjustment: Based on the nitrogen and phosphorus concentrations in the wastewater, control the reflux flow rate (e.g., reflux from the aerobic tank to the anoxic or anaerobic tank) to ensure the optimal working environment for microorganisms in the tank.

[0151] Inter-pool connection: Ensure smooth flow of sewage between pools. The treatment effect can be optimized by adjusting the water flow rate and changing the size of the pool.

[0152] After completing one wastewater treatment cycle, the treatment unit acquires operational data from each nitrogen and phosphorus removal tank, including the following steps:

[0153] The denitrification and phosphorus removal treatment ponds include anoxic ponds, anaerobic ponds, and aerobic ponds.

[0154] Obtain operational data for the anoxic tank, including aeration fluctuation amplitude, dissolved oxygen concentration rise rate, and denitrifying bacteria decay index.

[0155] Obtain operational data for the anaerobic tank, including aeration fluctuation amplitude, water flow distribution fluctuation amplitude, and anaerobic fermentation bacteria decay index.

[0156] Obtain operational data for the aerobic tank, including aeration fluctuation amplitude, dissolved oxygen concentration decrease rate, and nitrifying bacteria decay index.

[0157] The aeration fluctuation amplitude and the water flow distribution fluctuation amplitude are both calculated using the general parameter standard deviation calculation formula. The expression for the general parameter standard deviation is: In the formula, For the standard deviation of the parameter, For the number of data points, For the first Parameter values ​​at each data point This represents the mean of the parameters.

[0158] The logic for obtaining the aeration fluctuation amplitude is as follows: The aeration volume is obtained at multiple time points within the monitoring period, and the aeration fluctuation amplitude is calculated based on the aeration volume at these multiple time points. For example, in the formula for the standard deviation of a general parameter... This represents the aeration fluctuation amplitude. For the number of time points, For the first Aeration volume at each time point This represents the average aeration rate.

[0159] The logic for obtaining the amplitude of water flow distribution fluctuation is as follows: Multiple monitoring points are set up in the anaerobic tank, and the water flow velocity at each point is acquired. The amplitude of water flow distribution fluctuation is then calculated based on the water flow velocity at each monitoring point. For example, in the formula for the standard deviation of a general parameter... The amplitude of water flow distribution fluctuation. For the number of monitoring points, For the first Water flow velocity at each monitoring point This represents the average water flow velocity.

[0160] The logic for obtaining the dissolved oxygen concentration rise rate is as follows: obtain the dissolved oxygen concentration at the current time point and the dissolved oxygen concentration at the previous time point; obtain the concentration rise value by subtracting the dissolved oxygen concentration at the previous time point from the dissolved oxygen concentration at the current time point; obtain the monitoring duration by subtracting the previous time point from the current time point; and obtain the dissolved oxygen concentration rise rate by dividing the concentration rise value by the monitoring duration.

[0161] The logic for obtaining the dissolved oxygen concentration decrease rate is as follows: obtain the dissolved oxygen concentration at the current time point and the dissolved oxygen concentration at the previous time point; obtain the concentration decrease value by subtracting the dissolved oxygen concentration at the current time point from the dissolved oxygen concentration at the previous time point; obtain the monitoring duration by subtracting the previous time point from the current time point; and obtain the dissolved oxygen concentration decrease rate by dividing the concentration decrease value by the monitoring duration.

[0162] The attenuation indices of denitrifying bacteria, anaerobic fermentation bacteria, and nitrifying bacteria are obtained by calculating the microbial attenuation index, expressed as: In the formula, The microbial attenuation index, This represents the initial number of microorganisms. The attenuation constant is For time.

[0163] Anomaly analysis of each nitrogen and phosphorus removal treatment tank was conducted using anomaly models, including the following steps:

[0164] The operating data of each nitrogen and phosphorus removal treatment tank is substituted into the anomaly model, and the anomaly model outputs the anomaly coefficient for each nitrogen and phosphorus removal treatment tank. The expression of the anomaly model is as follows:

[0165] In the formula, The outlier coefficient, For variables (various runtime data). These are the regression coefficients of each variable, and the regression coefficients are greater than 0;

[0166] The design of the anomaly coefficients is based on the following logic:

[0167] Exponential function: used to amplify the sensitivity of abnormal indicators, making it easier to distinguish between normal and abnormal states.

[0168] Regression coefficients: determined through training with historical operational data. The specific method is as follows:

[0169] Collect historical operational data (including aeration fluctuation amplitude, dissolved oxygen concentration change rate, bacterial decay index, etc.) and corresponding treatment tank abnormality labels (normal / abnormal).

[0170] Logistic regression is used for training to determine the regression coefficients of each variable. During training, the coefficients are optimized using maximum likelihood estimation to ensure the model's accuracy in predicting abnormal states.

[0171] Example training results: If the aeration fluctuation amplitude of a certain treatment tank has the greatest impact on anomalies, then The value is significantly higher than other coefficients.

[0172] The anomaly threshold is determined using statistical methods: distribution analysis (e.g., normal distribution) is performed on the anomaly coefficients under historical normal conditions, and the upper limit of the confidence interval at 95% is taken as the threshold. For example, if the C value under normal conditions is all less than 2.5, then the threshold is set to 2.5.

[0173] The larger the anomaly coefficient of the denitrification and phosphorus removal treatment tank, the more serious the anomaly and the faster the performance decline. The obtained anomaly coefficient is compared with the preset anomaly threshold. The anomaly threshold is used to analyze whether the denitrification and phosphorus removal treatment tank can support operation. If the anomaly coefficient of the denitrification and phosphorus removal treatment tank is less than or equal to the anomaly threshold, it is analyzed that the denitrification and phosphorus removal treatment tank has no anomaly and can support operation. If the anomaly coefficient of the denitrification and phosphorus removal treatment tank is greater than the anomaly threshold, it is analyzed that the denitrification and phosphorus removal treatment tank has an anomaly and cannot support operation.

[0174] The logical components of the anomaly coefficient used in this invention are as follows: taking the impact of operating data on the denitrification and phosphorus removal treatment tank as an example, the components are: first, indicators, i.e., factors that cause changes in the performance of the denitrification and phosphorus removal treatment tank (in this invention, the impact of operating data on the denitrification and phosphorus removal treatment tank); second, the weights of these indicators, i.e., the proportion of each type of operating data when it is generated; and third, the calculation equation, i.e., the mathematical calculation process used to obtain the result, and the anomaly coefficient obtained by calculating the indicators with their respective weights through the calculation equation.

[0175] The operational data obtained from the sample were transformed and processed into a data language recognizable by computer software. Secondly, these evaluation factors were analyzed using SPSS software through Logistic Regression to identify factors and their weights that were significantly correlated with the results. Thirdly, the evaluation factors and weights were substituted into the Logistic Regression equation to obtain the results, specifically:

[0176] First, ensure the integrity of the running data, handling missing and outlier values. Convert the data to a format that SPSS software can recognize, typically storing it as .csv, .xlsx, etc., and then import it into SPSS. Open SPSS, import the processed data file, and transform the variables as needed. For example, for continuous variables, perform standardization or normalization. Select the "Analyze" menu, then select the "Regression" option and choose the "Bivariate Logistic" option. In the dialog box, add the dependent variable (outcome) and independent variables (running data) to the corresponding boxes. SPSS will then fit a Logistic regression model based on the selected variables. In the output results, you will see information such as the model's coefficients, standard errors, and p-values. Examining the coefficients and p-values ​​helps determine which variables are significantly correlated with the results. Generally, a p-value less than 0.05 is considered significant. While fitting the model, variable selection methods, such as stepwise regression, are used to help screen the most relevant factors. Based on the coefficients of the Logistic regression model, the magnitude of the coefficients reflects the degree of influence of each factor on the result, and the sign of the coefficients indicates the direction of influence. After obtaining the significant factors and their coefficients, the Logistic regression equation is obtained. This equation is used to calculate the probability of each sample and thus predict the results.

[0177] Clustering based on outliers includes the following steps:

[0178] After closing the valves corresponding to the denitrification and phosphorus removal treatment tanks that do not support operation, and obtaining the abnormality coefficients of the remaining denitrification and phosphorus removal treatment tanks, the remaining anoxic tanks, remaining anaerobic tanks, and remaining aerobic tanks are clustered.

[0179] Clustering of the remaining anoxic pools includes the following steps:

[0180] S1: Suppose that the remaining anoxic pools are clustered into K clusters, and randomly select the anomaly coefficients of the K remaining anoxic pools as the initial cluster center values.

[0181] S2: Calculate the difference between the anomaly coefficient of each anoxic pool and the initial cluster center value, and assign the anoxic pool to the cluster with the smallest difference.

[0182] S3: After all anoxic pools have been allocated, calculate the mean of the anomaly coefficients of the clusters as the new cluster center values.

[0183] S4: Repeat steps S2 and S3. If the difference between the new cluster center value of the current iteration and the new cluster center value of the previous iteration is less than or equal to the difference threshold (usually 0.1), it is determined that the convergence condition is met, and all clusters are output.

[0184] In this application, the steps for clustering the remaining anaerobic tanks and the remaining aerobic tanks are the same as the steps for clustering the remaining anoxic tanks, and will not be repeated here.

[0185] Several combination schemes are generated based on the clustering results, including the following steps:

[0186] After obtaining the clusters in the anoxic, anaerobic, and aerobic pools, calculate the mean anomaly coefficients of the K anoxic pool clusters, the mean anomaly coefficients of the K anaerobic pool clusters, and the mean anomaly coefficients of the K aerobic pool clusters.

[0187] Sort all anoxic pool clusters in ascending order of their mean anomaly coefficients. The higher the cluster's ranking, the better the anoxic pool performance within that cluster. Sort all anaerobic pool clusters in ascending order of their mean anomaly coefficients. Sort all aerobic pool clusters in ascending order of their mean anomaly coefficients.

[0188] Assuming there are 3 anoxic pool clusters, 3 anaerobic pool clusters, and 3 aerobic pool clusters, with each cluster containing multiple pools, and the anomaly coefficient of each pool known, we can calculate and sort the pool clusters using the following steps.

[0189] The mean anomaly coefficient for cluster 1 in the anoxic pool was 1.15, the mean anomaly coefficient for cluster 2 in the anoxic pool was 1.0, and the mean anomaly coefficient for cluster 3 in the anoxic pool was 1.6.

[0190] The mean anomaly coefficient for anaerobic cluster 1 was 1.0, the mean anomaly coefficient for anaerobic cluster 2 was 1.35, and the mean anomaly coefficient for anaerobic cluster 3 was 1.2.

[0191] The mean anomaly coefficient for aerobic cluster 1 was 0.68, the mean anomaly coefficient for aerobic cluster 2 was 0.9, and the mean anomaly coefficient for aerobic cluster 3 was 1.15.

[0192] Anoxic pool clusters are sorted by mean anomaly coefficient from smallest to largest:

[0193] Cluster ratio in the anoxic pool: 2:1.0;

[0194] Cluster ratio in the anoxic pool: 1:1.15;

[0195] The ratio of hypoxic pool clusters is 3:1.6.

[0196] Anaerobic tank clusters are sorted by mean anomaly coefficient from smallest to largest:

[0197] Anaerobic tank cluster ratio 1:1.0;

[0198] Anaerobic pool cluster ratio 3:1.2;

[0199] Anaerobic pool cluster ratio 2:1.35.

[0200] Aerobic tank clusters are sorted by mean anomaly coefficient from smallest to largest:

[0201] The aerobic pool cluster ratio was 1:0.68;

[0202] The aerobic pool cluster ratio was 2:0.9;

[0203] The aerobic pool cluster ratio is 3:1.15.

[0204] The combination clusters are anoxic pool cluster 2, anaerobic pool cluster 1, and the anoxic, anaerobic, and aerobic pools in aerobic pool cluster 1.

[0205] The anoxic pools, anaerobic pools, and aerobic pools in anoxic pool cluster 1, anaerobic pool cluster 3, and aerobic pool cluster 2 are combined.

[0206] The anoxic pool cluster 3, the anaerobic pool cluster 2, and the anoxic pool, anaerobic pool, and aerobic pool in the aerobic pool cluster 3 are combined.

[0207] Anoxic pool cluster 2, anaerobic pool cluster 1, and the anoxic, anaerobic, and aerobic pools in aerobic pool cluster 1 are combined. When combining, any anoxic pool in anoxic pool cluster 2 is combined with any anaerobic pool in anaerobic pool cluster 1 and any aerobic pool in aerobic pool cluster 1 to obtain several combination schemes.

[0208] Assuming that anoxic pool cluster 2 contains anoxic pool 1, anoxic pool 2, and anoxic pool 3; anaerobic pool cluster 1 contains anaerobic pool 1, anaerobic pool 2, and anaerobic pool 3; and aerobic pool cluster 1 contains aerobic pool 1, aerobic pool 2, and aerobic pool 3, then there are a total of 27 possible combinations:

[0209] Anoxic tank 1 + anaerobic tank 1 + aerobic tank 1; Anoxic tank 1 + anaerobic tank 1 + aerobic tank 2; Anoxic tank 1 + anaerobic tank 1 + aerobic tank 3; Anoxic tank 1 + anaerobic tank 2 + aerobic tank 1; Anoxic tank 1 + anaerobic tank 2 + aerobic tank 2; Anoxic tank 1 + anaerobic tank 2 + aerobic tank 3; Anoxic tank 1 + anaerobic tank 3 + aerobic tank 1; Anoxic tank 1 + anaerobic tank 3 + aerobic tank 2; Anoxic tank 1 + anaerobic tank 3 + aerobic tank 3; Anoxic tank 2 + anaerobic tank 1 + aerobic tank 1; Anoxic tank 2 + anaerobic tank 1 + aerobic tank 2; Anoxic tank 2 + anaerobic tank 1 + aerobic tank 3; Anoxic tank 2 + anaerobic tank 2 + aerobic tank 1; Anoxic tank 2 + anaerobic tank 2 Pool 2 + aerobic pool 2; anoxic pool 2 + anaerobic pool 2 + aerobic pool 3; anoxic pool 2 + anaerobic pool 3 + aerobic pool 1; anoxic pool 2 + anaerobic pool 3 + aerobic pool 2; anoxic pool 2 + anaerobic pool 3 + aerobic pool 3; anoxic pool 3 + anaerobic pool 1 + aerobic pool 1; anoxic pool 3 + anaerobic pool 1 + aerobic pool 2; anoxic pool 3 + anaerobic pool 1 + aerobic pool 3; anoxic pool 3 + anaerobic pool 2 + aerobic pool 1; anoxic pool 3 + anaerobic pool 2 + aerobic pool 2; anoxic pool 3 + anaerobic pool 2 + aerobic pool 3; anoxic pool 3 + anaerobic pool 3 + aerobic pool 1; anoxic pool 3 + anaerobic pool 3 + aerobic pool 2; anoxic pool 3 + anaerobic pool 3 + aerobic pool 3.

[0210] Regardless of the number of denitrification and phosphorus removal treatment tanks in the denitrification and phosphorus removal treatment tank cluster, they should be combined according to the above combination method.

[0211] Selecting the best-performing tanks within each cluster ensures optimal performance at every stage of the treatment process. For instance, combining the best-performing anoxic, anaerobic, and aerobic tank clusters minimizes the impact of suboptimal tanks throughout the wastewater treatment process, thereby improving the efficiency and stability of nitrogen and phosphorus removal. Choosing tank combinations based on their performance optimizes resource allocation. Furthermore, selecting the best-performing clusters avoids overusing suboptimal tanks, improving system resource utilization efficiency and reducing unnecessary energy and material consumption.

[0212] After evaluating the performance of each combination scheme using a performance function, the combination scheme is then optimized a second time. The optimized combination scheme is then applied to the next wastewater treatment process, including the following steps:

[0213] Since wastewater denitrification and phosphorus removal treatment requires that the combined scheme include at least one anoxic tank, an anaerobic tank, and an aerobic tank, after obtaining all the combined schemes, the distance index of the denitrification and phosphorus removal treatment tanks and the water flow transmission time in each combined scheme are obtained. The performance value of each combined scheme is calculated using a performance function, the expression of which is: In the formula, For performance values, The distance index for nitrogen and phosphorus removal treatment ponds. For the duration of water flow transmission, , It is the proportionality coefficient, and the proportionality coefficient , Greater than 0.

[0214] proportionality coefficient , Optimization:

[0215] Objective: Minimize total processing cost, including energy consumption (related to distance exponent) and time cost (related to water flow transmission duration). Optimization method:

[0216] Collect multiple sets of historical data and calculate different and Total cost under the combination.

[0217] The optimal scaling factor is determined using gradient descent or grid search. For example, the optimization result for a wastewater treatment plant is... , This indicates that distance has a slightly greater impact on cost than time.

[0218] The physical meaning of the distance index: The distance index reflects the spatial layout efficiency between treatment pools. The shorter the distance, the lower the pipe resistance and the lower the energy consumption. The accuracy of the physical meaning is ensured by calculating the Euclidean distance using three-dimensional coordinates.

[0219] The higher the performance value, the better the overall performance of the combined scheme. The performance values ​​of combined schemes with the same denitrification and phosphorus removal treatment tank number are compared. The combined scheme with the higher performance value is retained, and the combined scheme with the lower performance value is deleted.

[0220] For example, if the performance value of the combination scheme of anoxic tank 1 + anaerobic tank 1 + aerobic tank 1 is 1.5, and the performance value of the combination scheme of anoxic tank 1 + anaerobic tank 1 + aerobic tank 2 is 1.2, then the combination scheme of anoxic tank 1 + anaerobic tank 1 + aerobic tank 1 is retained, and the combination scheme of anoxic tank 1 + anaerobic tank 1 + aerobic tank 2 is deleted.

[0221] The calculation logic for the distance index between nitrogen and phosphorus removal treatment tanks is as follows: The distance between any two nitrogen and phosphorus removal treatment tanks in the combined scheme is calculated using the following expression:

[0222] In the formula, The distance between two nitrogen and phosphorus removal treatment tanks. Let's define the coordinates of a nitrogen and phosphorus removal treatment tank. Let the coordinates of another nitrogen and phosphorus removal treatment tank be used, and mark the distance between the anoxic tank and the anaerobic tank as follows: The distance between the anoxic tank and the aerobic tank was marked as... The distance between the anaerobic tank and the aerobic tank was marked as... The formula for calculating the distance index of the denitrification and phosphorus removal treatment tank is: .

[0223] Supplementary Examples of Information Processing Efficiency:

[0224] Dynamic adjustment case:

[0225] Scenario: A wastewater treatment plant originally used a fixed combination of "2 anoxic tanks + 1 anaerobic tank + 3 aerobic tanks", with an average nitrogen removal rate of 85%.

[0226] After implementing this method:

[0227] An abnormal nitrifying bacteria attenuation index was detected in aerobic tank 3 using an anomaly model. The tank was then shut down, and a new combination of "anoxic tank 2 + anaerobic tank 1 + aerobic tank 1" was generated. The nitrogen removal rate increased to 92%, and the phosphorus removal rate increased from 88% to 94%.

[0228] Parallel computing technology is used: cluster analysis and performance evaluation are completed in a multi-threaded environment, reducing the time for a single optimization from 30 minutes to 5 minutes.

[0229] Example 3: The data-driven wastewater nitrogen and phosphorus removal system described in this example includes a labeling module, a data acquisition module, a cluster analysis module, and a solution application module;

[0230] The tagging module obtains information on all nitrogen and phosphorus removal treatment tanks within the wastewater treatment plant through the API interface of the wastewater treatment plant management platform. The information includes the number and location of anoxic, anaerobic, and aerobic tanks. All nitrogen and phosphorus removal treatment tanks are numbered and tagged, and the tagging information is sent to the acquisition module.

[0231] Data Acquisition Module: When the sensor installed at the inlet of the sewage pipe detects the entry of sewage, it performs denitrification and phosphorus removal treatment on the sewage according to the preset denitrification and phosphorus removal treatment tank combination scheme. After completing one sewage treatment cycle, it acquires the operating data of each denitrification and phosphorus removal treatment tank and sends the operating data to the cluster analysis module.

[0232] Clustering analysis module: Analyzes the abnormal conditions of each nitrogen and phosphorus removal treatment tank through anomaly model, performs clustering based on the abnormal conditions, and sends the clustering results to the solution application module.

[0233] Solution application module: Based on the clustering results, several combined solutions are generated. After evaluating the performance of each combined solution using a performance function, the combined solutions are optimized a second time. The optimized combined solutions are then applied to the next wastewater treatment.

[0234] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0235] In the description of this specification, references to terms such as "an embodiment," "example," and "specific example" indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0236] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.< / token>

Claims

1. A data-driven wastewater nitrogen and phosphorus removal treatment method, characterized in that: The processing method includes the following steps: The data acquisition terminal obtains information on all denitrification and phosphorus removal treatment tanks in the wastewater treatment plant through the API interface of the wastewater treatment plant management platform, numbers and marks all denitrification and phosphorus removal treatment tanks, and when the sensor installed at the wastewater pipe inlet detects that wastewater is entering, it performs denitrification and phosphorus removal treatment on the wastewater according to the preset denitrification and phosphorus removal treatment tank combination scheme. After completing a wastewater treatment cycle, the treatment unit obtains the operating data of each nitrogen and phosphorus removal treatment tank, analyzes the abnormal conditions of each nitrogen and phosphorus removal treatment tank through anomaly model, and performs clustering based on the abnormal conditions. Several combination schemes are generated based on the clustering results. The performance of each combination scheme is evaluated using a performance function, and then the combination schemes are optimized in the second stage. The optimized combination schemes are then applied to the next wastewater treatment.

2. The data-driven wastewater denitrification and phosphorus removal method according to claim 1, characterized in that: After completing one wastewater treatment cycle, the treatment unit acquires operational data from each nitrogen and phosphorus removal tank, including the following steps: The denitrification and phosphorus removal treatment ponds include anoxic ponds, anaerobic ponds, and aerobic ponds; Obtain operational data for the anoxic tank, including aeration fluctuation amplitude, dissolved oxygen concentration rise rate, and denitrifying bacteria decay index; Obtain operational data for the anaerobic tank, including aeration fluctuation amplitude, water flow distribution fluctuation amplitude, and anaerobic fermentation bacteria attenuation index; Obtain operational data for the aerobic tank, including aeration fluctuation amplitude, dissolved oxygen concentration decrease rate, and nitrifying bacteria decay index.

3. The data-driven wastewater denitrification and phosphorus removal method according to claim 2, characterized in that: Anomaly analysis of each nitrogen and phosphorus removal treatment tank was conducted using anomaly models, including the following steps: The operating data of each nitrogen and phosphorus removal treatment tank is substituted into the anomaly model, and the anomaly model outputs the anomaly coefficient for each nitrogen and phosphorus removal treatment tank. The expression of the anomaly model is as follows: In the formula, The outlier coefficient, As variables, These are the regression coefficients of each variable, and the regression coefficients are greater than 0; The obtained anomaly coefficients are compared with preset anomaly thresholds. The anomaly thresholds are used to analyze whether the denitrification and phosphorus removal treatment tanks can support operation. If the abnormality coefficient of the denitrification and phosphorus removal treatment tank is less than or equal to the abnormality threshold, the analysis shows that there is no abnormality in the denitrification and phosphorus removal treatment tank and operation is supported. If the abnormality coefficient of the denitrification and phosphorus removal treatment tank is greater than the abnormality threshold, the analysis shows that there is an abnormality in the denitrification and phosphorus removal treatment tank and operation is not supported.

4. The data-driven wastewater denitrification and phosphorus removal method according to claim 3, characterized in that: Clustering based on outliers includes the following steps: Close the valves corresponding to the denitrification and phosphorus removal treatment tanks that do not support operation, obtain the abnormality coefficients of the remaining denitrification and phosphorus removal treatment tanks, and then cluster the remaining anoxic tanks, remaining anaerobic tanks, and remaining aerobic tanks. Clustering of the remaining anoxic pools includes the following steps: S1: Suppose that the remaining anoxic pools are clustered into K clusters, and randomly select the anomaly coefficients of K remaining anoxic pools as the initial cluster center values; S2: Calculate the difference between the anomaly coefficient of each anoxic pool and the initial cluster center value, and assign the anoxic pool to the cluster with the smallest difference; S3: After all anoxic pools have been allocated, calculate the mean anomaly coefficient of each cluster as the new cluster center value; S4: Repeat steps S2 and S3. If the difference between the new cluster center value of the current iteration and the new cluster center value of the previous iteration is less than or equal to the difference threshold, it is determined that the convergence condition is met, and all clusters are output.

5. The data-driven wastewater denitrification and phosphorus removal method according to claim 4, characterized in that: Several combination schemes are generated based on the clustering results, including the following steps: After obtaining the clusters in the anoxic, anaerobic, and aerobic pools, calculate the mean anomaly coefficients of the K anoxic pool clusters, the mean anomaly coefficients of the K anaerobic pool clusters, and the mean anomaly coefficients of the K aerobic pool clusters. Sort all anoxic pool clusters in ascending order of their mean anomaly coefficients; sort all anaerobic pool clusters in ascending order of their mean anomaly coefficients; sort all aerobic pool clusters in ascending order of their mean anomaly coefficients. Based on the ranking results, select the corresponding anoxic pool cluster, anaerobic pool cluster, and aerobic pool cluster to combine them, resulting in several combination schemes.

6. The data-driven wastewater denitrification and phosphorus removal method according to claim 5, characterized in that: After evaluating the performance of each combination scheme using a performance function, a secondary optimization of the combination scheme is performed, including the following steps: After obtaining all the combined schemes, obtain the distance index of the denitrification and phosphorus removal treatment tank and the water flow transmission time in each combined scheme, and calculate the performance value of each combined scheme through the performance function; The performance values ​​of the combined schemes with the same denitrification and phosphorus removal treatment tank number are compared. The combined scheme with the larger performance value is retained, and the combined scheme with the smaller performance value is deleted.

7. The data-driven wastewater denitrification and phosphorus removal method according to claim 6, characterized in that: Obtain the distance index of the nitrogen and phosphorus removal treatment tanks and the water flow transmission time for each combination scheme, and calculate the performance value of each combination scheme using the performance function, the expression of which is: In the formula, For performance values, The distance index for nitrogen and phosphorus removal treatment ponds. For the duration of water flow transmission, , It is the proportionality coefficient, and the proportionality coefficient , Greater than 0.

8. The data-driven wastewater denitrification and phosphorus removal method according to claim 7, characterized in that: The calculation logic for the distance index between the nitrogen and phosphorus removal treatment tanks is as follows: The distance between any two nitrogen and phosphorus removal treatment tanks in the combined scheme is calculated using the following expression: In the formula, The distance between two nitrogen and phosphorus removal treatment tanks. Let's define the coordinates of a nitrogen and phosphorus removal treatment tank. Let the coordinates of another nitrogen and phosphorus removal treatment tank be used, and mark the distance between the anoxic tank and the anaerobic tank as follows: The distance between the anoxic tank and the aerobic tank was marked as... The distance between the anaerobic tank and the aerobic tank was marked as... The formula for calculating the distance index of the denitrification and phosphorus removal treatment tank is: .

9. The data-driven wastewater denitrification and phosphorus removal method according to claim 8, characterized in that: When a sensor installed at the inlet of the sewage pipe detects the entry of sewage, the sewage is treated for nitrogen and phosphorus removal according to a pre-set denitrification and phosphorus removal treatment tank combination scheme, including the following steps: The preset combination scheme defines the combination of anoxic, anaerobic, and aerobic tanks with various numbers; According to the selected scheme, the corresponding anoxic tank is started. In the anoxic tank, denitrifying bacteria consume nitrates in the wastewater and remove nitrogen. The corresponding anaerobic tank is started. In the anaerobic tank, polyphosphate-accumulating bacteria release phosphorus and absorb organic matter in the wastewater. The corresponding aerobic tank is started. In the aerobic tank, nitrifying bacteria convert ammonia nitrogen in the wastewater into nitrates. At the same time, polyphosphate-accumulating bacteria reabsorb phosphorus in an aerobic environment.

10. A data-driven wastewater nitrogen and phosphorus removal system for implementing the treatment method according to any one of claims 1-9, characterized in that: Includes a labeling module, a data acquisition module, a cluster analysis module, and a solution application module: Tagging module: Obtain information on all nitrogen and phosphorus removal treatment tanks in the wastewater treatment plant through the API interface of the wastewater treatment plant management platform, and number and tag all nitrogen and phosphorus removal treatment tanks; Data Acquisition Module: When the sensor installed at the inlet of the sewage pipe detects the entry of sewage, it performs denitrification and phosphorus removal treatment on the sewage according to the preset denitrification and phosphorus removal treatment tank combination scheme. After completing one sewage treatment cycle, it acquires the operating data of each denitrification and phosphorus removal treatment tank. Clustering analysis module: Analyzes the abnormal conditions of each nitrogen and phosphorus removal treatment tank through anomaly model, and performs clustering based on the abnormal conditions; Solution application module: Based on the clustering results, several combined solutions are generated. After evaluating the performance of each combined solution using a performance function, the combined solutions are optimized a second time. The optimized combined solutions are then applied to the next wastewater treatment.

Images