Method and system for performing intelligent regulation for sewage treatment on basis of microbial activity

Abstract

The present invention belongs to the technical field of intelligent operation of sewage treatment, and specifically relates to a method and system for performing intelligent regulation for sewage treatment on the basis of microbial activity. The method in the present invention comprises the following steps: acquiring value ranges of input features, wherein the input features comprise microbial activity data; within the value ranges of the input features, and with the objective being that a sewage treatment result is in compliance with standards and has the lowest cost, finding the optimal feature combination of the input features; and on the basis of the optimal feature combination, determining regulation parameters and applying same to a sewage treatment system, wherein the regulation parameters comprise an airflow rate and a carbon source dosage, and the cost calculation formula is as follows: cost=(airflow rate-700) / 100+5×(carbon source dosage-3000) / 1000. The present invention can reduce the airflow rate and the carbon source dosage while ensuring compliant effluent, thereby reducing the process cost; and the present invention has good application prospects.

Description

An intelligent regulation method and system for sewage treatment based on microbial activity Technical Field

[0001] The present invention belongs to the technical field of intelligent operation of sewage treatment, and specifically relates to an intelligent regulation method and system for sewage treatment based on microbial activity. Background Art

[0002] Wastewater treatment is one of the top 10 carbon-emitting industries. Reducing operating costs (such as energy and chemical consumption) and improving pollutant removal performance are constant themes in wastewater treatment plant regulation. Currently, management of system operating parameters in wastewater treatment processes is relatively extensive, often relying on manual experience. However, biological wastewater treatment is a nonlinear process with significant time lags and strong coupling. Manual management methods often suffer from lags and uncertainties, making it difficult to meet the requirements for stable, economical, and efficient operation of wastewater treatment systems.

[0003] In wastewater biological treatment systems, microorganisms are the primary functional entities, and their metabolic activity plays a decisive role in pollutant treatment performance. Specific oxygen consumption respiration rate (SOUR) and adenosine triphosphate (ATP) are rapid and sensitive tests. Previous studies have shown that SOUR can characterize the activity of denitrifying microorganisms in activated sludge. As the primary carrier of available energy in cellular metabolism, ATP can characterize the number and metabolic activity of microorganisms. Therefore, using SOUR and ATP to characterize microbial status and metabolic activity and correlate this with pollutant degradation is a promising approach for optimizing and regulating wastewater treatment systems.

[0004] In addition, due to the powerful nonlinear capabilities of artificial intelligence, using artificial intelligence methods to solve the problems of early warning and precise operation management of sewage treatment systems is a feasible solution. It is expected to solve the lag and uncertainty of manually controlled sewage treatment systems. The Chinese invention patent application "CN116798530A A method and device for intelligent prediction of sewage treatment performance" discloses a method based on SOUR and ATP, which uses an artificial intelligence model to predict water treatment effects, which can provide key reference parameters for the regulation of sewage treatment systems. However, the technical solution of this document can only predict the effect of sewage treatment under specific parameter conditions, and cannot directly regulate process parameters to meet the requirements of effluent standards and reduce process costs. Therefore, there is still an urgent need in this field to develop a method that can automatically regulate sewage treatment process parameters based on specific input features and algorithms. Summary of the Invention

[0005] In response to the problems of the prior art, the present invention provides a method and system for intelligent regulation of sewage treatment based on microbial activity.

[0006] An intelligent control method for sewage treatment based on microbial activity, obtaining a value range of an input feature, wherein the input feature includes microbial activity data;

[0007] Finding the optimal feature combination of input features within the value range of the input features with the goal of meeting the sewage treatment standards at the lowest cost;

[0008] Determining control parameters based on the optimal feature combination and applying them to the sewage treatment system;

[0009] Wherein, the control parameters include: air flow rate and carbon source dosage;

[0010] The cost calculation formula is as follows:

[0011] Cost = (air flow rate - 700) / 100 + 5 × (carbon source dosage - 3000) / 1000.

[0012] Preferably, the method for finding the best feature combination of input features is a particle swarm optimization algorithm.

[0013] Preferably, the method comprises the following steps:

[0014] Step 1: traverse at least one input feature to obtain several groups of input feature combinations;

[0015] Step 2: input the feature combination into a machine learning model, and obtain a predicted value of the sewage treatment result indicator based on the model output;

[0016] Step 3: Determine corresponding control parameters based on the feature combination;

[0017] Step 4: Calculate the cost of each feature combination:

[0018] Step 5: Select the best feature combination with the lowest cost and the best indicator prediction value according to the calculation results of step 4, and apply the control parameters of the best feature combination calculated in step 3 to the sewage treatment system.

[0019] Preferably, the machine learning model is a random forest model.

[0020] Preferably, in step 3, the control parameter is determined in one of the following ways:

[0021] Mode 1: The input features also include air flow rate and / or carbon source dosage, and the values ​​of air flow rate and / or carbon source dosage in the feature combination are directly used as corresponding control parameters;

[0022] Mode 2: The input features do not include air flow or carbon source dosage, and the air flow or carbon source dosage is calculated by taking the values ​​of each input feature in the feature combination, and the calculation formula is obtained by fitting historical data.

[0023] Preferably, the microbial activity data include: aerobic tank SOUR ATP 、Anoxic pool SOUR ATP_DB and hypoxic pool SOUR MLSS_DB ;

[0024] The process of finding the best feature combination of input features is to ATP and hypoxic pool SOUR MLSS_DB Make changes;

[0025] Aerobic Pool SOUR ATP The value range is [15,32], and the value change step of different feature combinations is 1;

[0026] Hypoxic pool SOUR MLSS_DB The value range is [0.7, 1.2], and the value change step of different feature combinations is 0.1.

[0027] Preferably, the input characteristics also include air flow and / or carbon source dosage;

[0028] The air flow rate range is [700m 3 / h,3000m 3 / h], in the process of finding the best feature combination of input features, the value change step of different feature combinations is 200m 3 / h;

[0029] The range of carbon source dosage is [3000Kg, 7000Kg]. In the process of finding the best feature combination of input features, the value change step of different feature combinations is 200Kg.

[0030] Preferably, the indicator used to evaluate whether the sewage treatment results meet the standards is the effluent total nitrogen (the effluent total nitrogen is calculated from the influent total nitrogen and the predicted total nitrogen removal rate).

[0031] The present invention also provides an intelligent control system for sewage treatment of microbial activity, comprising:

[0032] Input module, used to input data;

[0033] A calculation module, used to calculate the control parameters applied to the sewage treatment system according to the above-mentioned intelligent control method for sewage treatment;

[0034] The output module is used to output the control parameters to the sewage treatment system.

[0035] The present invention also provides a computer-readable storage medium storing thereon: a computer program for implementing the above-mentioned intelligent control method for sewage treatment.

[0036] In the present invention, the abbreviations of terms related to water treatment have the following meanings:

[0037] The present invention provides a method that uses microbial activity data to predict sewage treatment effectiveness and calculate treatment costs, thereby optimizing the optimal control parameters for the sewage treatment system. This method uses sewage quality-related parameters as input data to automatically provide control parameters that balance treatment effectiveness and cost reduction. This avoids the waste caused by manual, empirical control and ensures stable, economical, and efficient operation of the sewage treatment process. Therefore, the present invention has promising application prospects.

[0038] Obviously, based on the above contents of the present invention, according to common technical knowledge and customary means in this field, without departing from the above basic technical ideas of the present invention, other various forms of modifications, replacements or changes can be made.

[0039] The following further describes the above content of the present invention in detail through specific embodiments in the form of examples. However, this should not be construed as limiting the scope of the above subject matter of the present invention to the following examples. All technologies implemented based on the above content of the present invention fall within the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] FIG1 is a comparison of operating parameters before and after regulation according to the method of Example 1;

[0041] FIG2 is a comparison of operating parameters before and after regulation according to the method of Example 2;

[0042] FIG3 is a comparison of operating parameters before and after regulation according to the method of Example 3;

[0043] FIG4 is a comparison of operating parameters before and after adjustment according to the method of Comparative Example 1;

[0044] Figure 5 is a heat map showing the correlation between operating parameters and microbial activity in Experimental Example 2;

[0045] Figure 6 shows the data fitting of some microbial activities and operating parameters in Experimental Example 2;

[0046] FIG7 is a data fitting of some microbial activities and operating parameters in Experimental Example 2. DETAILED DESCRIPTION

[0047] It should be noted that the algorithms for data collection, transmission, storage and processing steps not specifically described in the embodiments, as well as the hardware structures, circuit connections, etc. not specifically described can all be implemented through the disclosed content of the prior art.

[0048] Example 1 Intelligent Control Method and System for Wastewater Treatment Based on Microbial Activity

[0049] The system of this embodiment includes:

[0050] An input module, used to input data; the input data includes water quality related parameters, data sets for training models, etc.;

[0051] A calculation module, used to calculate the control parameters applied to the sewage treatment system through the sewage treatment intelligent control method;

[0052] The output module is used to output the control parameters to the sewage treatment system.

[0053] Through the above system, the method for intelligent regulation of sewage treatment includes the following steps:

[0054] Step 1: traverse each input feature to obtain several sets of input feature combinations; the input features include microbial activity data. In this embodiment, the microbial activity data is selected from the aerobic pool SOUR ATP (the total specific oxygen respiration rate of the aerobic pool with ATP as the biomass), the anoxic pool SOUR ATP_DB (Specific oxygen consumption respiration rate of heterotrophic bacteria in anoxic pool with ATP as biomass) and SOUR MLSS_DB (The specific oxygen consumption respiration rate of heterotrophic bacteria in anoxic pools is based on MLSS as biomass.) When traversing microbial activity data, the parameters are set as follows:

[0055] Aerobic Pool SOUR ATP The value range is [15,32], and the step size is 1;

[0056] Hypoxic pool SOUR MLSS_DB The value range is [0.7, 1.2], with a step size of 0.1;

[0057] In this embodiment, SOUR ATP_DB The value of the aerobic pool SOUR remains unchanged, only the ATP and hypoxic pool SOUR MLSS_DB Because the latter two are more related to the operating parameters of the aerobic pool air flow and carbon source dosage.

[0058] Step 2: Input the feature combination into a machine learning model, and obtain a predicted value for the sewage treatment result indicator based on the model output. In this embodiment, the machine learning model uses a random forest algorithm, and the predicted indicator is the TN removal rate. Furthermore, the predicted TN concentration in the sewage treatment effluent can be calculated using the TN concentration in the influent water quality data and the predicted TN removal rate.

[0059] Step 3: Determine the corresponding control parameters based on the feature combination; the control parameters include: air flow rate and carbon source dosage. In this embodiment, air flow rate and carbon source dosage are not input into the machine learning model as input features. Therefore, it is necessary to calculate the air flow rate and carbon source dosage corresponding to each feature combination based on microbial activity data. The calculation formula is obtained by fitting historical data. In this embodiment, a set of specific calculation formulas is provided:

[0060] Air flow (unit: m 3 / h)=9.05+0.00774×aerobic pool SOUR ATP (mgO2 / (nmATP·h));

[0061] Carbon source dosage (unit: Kg) = 8521.91-4519.99 × anoxic tank SOUR MLSS_DB (mgO2 / (gMLSS·h)).

[0062] Step 4: Calculate the cost of each feature combination. The cost calculation formula is as follows:

[0063] Cost = (air flow rate - 700) / 100 + 5 × (carbon source dosage - 3000) / 1000;

[0064] Step 5: Select the best feature combination with the lowest cost and the best indicator prediction value according to the calculation results of step 4, and apply the control parameters of the best feature combination calculated in step 3 to the sewage treatment system.

[0065] Example 2 Intelligent Control Method and System for Wastewater Treatment Based on Microbial Activity

[0066] The technical solution of this embodiment is basically the same as that of embodiment 1, except that:

[0067] The input features also include: air flow rate and carbon source dosage, and the parameters of the traversal process include:

[0068] The air flow rate range is [700m 3 / h,3000m 3 / h], with a step length of 200m 3 / h;

[0069] The range of carbon source dosage is [3000Kg, 7000Kg], with a step size of 200Kg.

[0070] In addition, in step 3, there is no need to calculate the air flow rate and the carbon source dosage through formulas, and the values ​​of the air flow rate and the carbon source dosage in the characteristic combination can be directly used as the corresponding control parameters.

[0071] Example 3 Intelligent Control Method and System for Wastewater Treatment Based on Microbial Activity

[0072] The technical solution of this embodiment is basically the same as that of Example 2, except that the input features are input into the random forest model, and the PSO optimization algorithm is used to optimize the operating parameters to obtain the optimal air flow rate and carbon source dosage. In the PSO optimization algorithm, the fitness function (i.e., cost) is formulated as follows:

[0073] Cost = (air flow rate - 700) / 100 + 5 × (carbon source dosage - 3000) / 1000;

[0074] The remaining parameter settings not specifically described can be performed according to the same parameters as in Example 1 or Example 2.

[0075] Comparative Example 1: Intelligent Control Method and System for Sewage Treatment Based on Operating Parameters

[0076] The technical solution of this comparative example is similar to that of Example 2, except that the input features of the machine learning model do not include microbial activity data (i.e., the aerobic tank SOUR ATP 、Anoxic pool SOUR ATP_DB and hypoxic pool SOUR MLSS_DB ), only including the operating parameter data (i.e., only the anoxic tank MLSS, air flow rate and carbon source dosage are used as input features, and only the air flow rate and carbon source dosage are changed during the optimization process, and only these two operating parameter values ​​are optimized).

[0077] The technical solution of the present invention is further illustrated by experiments below.

[0078] Experimental Example 1 Comparison of control results of different control methods

[0079] 1. Experimental Methods

[0080] This experiment collects the operation data of a sewage treatment plant over a period of time as a data set, including the following parameters during actual operation: influent water quality (including TN concentration), effluent water quality (including TN concentration), microbial activity data during operation (i.e., aerobic tank SOUR ATP 、Anoxic pool SOUR ATP_DB and hypoxic pool SOUR MLSS_DB), the control parameters during operation (i.e. air flow, carbon source dosage and anoxic tank MLSS).

[0081] The cost calculation formula provided in Example 1 was used to calculate the cost during the actual operation of the data set, which is recorded as the cost "before regulation" in the following experimental data presentation.

[0082] Then, the optimal control parameters and the corresponding costs during the operation are calculated respectively using the methods of Example 1, Example 2, Example 3 and Comparative Example 1.

[0083] In the above embodiments and comparative examples, the performance of the established random forest model is as follows:

[0084] The above results show that the random forest model established in this experimental case has good prediction performance.

[0085] 2. Experimental Results

[0086] The operating data after regulation according to the method of Example 1 and the comparison of the original operating data (before regulation) are shown in Figure 1. The operating data after regulation according to the method of Example 2 and the comparison of the original operating data (before regulation) are shown in Figure 2. The operating data after regulation according to the method of Example 3 and the comparison of the original operating data (before regulation) are shown in Figure 3. The operating data after regulation according to the method of Comparative Example 1 and the comparison of the original operating data (before regulation) are shown in Figure 4. The compared data include: air flow rate, carbon source dosage, effluent TN content and cost.

[0087] It can be seen that after regulation according to the method of Examples 1-3, while ensuring that the TN content of the effluent meets the standard, the air flow rate, the amount of carbon source added and the operating cost (the cost can reflect the total cost of operation) can be effectively reduced, thereby effectively achieving the purpose of reducing the operating cost of the sewage treatment process.

[0088] The total air flow rate, total carbon source dosage used in Examples 1-3 and Comparative Example 1, and the cost calculated using the total air flow rate and total carbon source dosage, are further compared. The results are shown in the following table:

[0089] As can be seen, using both microbial activity and operating parameters as input data can significantly reduce air flow, carbon source dosage, and cost, leading to better cost reduction. Furthermore, the PSO optimization algorithm has a stronger optimization capability than the traversal optimization algorithm, resulting in lower costs.

[0090] Experimental Example 2 Analysis of the relationship between microbial activity and sewage treatment operating parameters

[0091] Taking the operation data of a sewage treatment plant over a period of time as a data set, the correlation between the operation parameters and microbial activity in the sewage treatment system is shown in Figure 5 below. MLSS Negatively correlated with operating parameters; SOUR ATP There is a positive correlation with the operating parameters, and most of the correlations are extremely significant.

[0092] To further analyze the intrinsic correlation between microbial activity and operating parameters, the extremely significant correlation relationships were visualized separately and data fitting was performed to clarify the intrinsic mechanism between microbial activity and operating parameters.

[0093] Aerobic Pool SOUR ATP With aerobic pool MLSS, aerobic pool SOUR ATP With aerobic pool air flow, anoxic pool SOUR ATP_DB The MLSS of the anoxic pool, the ATP of the anoxic pool and the DO of the anoxic pool all showed a very significant correlation. Therefore, the data were fitted and the data fitting results are shown in Figure 6. ATP It showed a linear positive correlation with the MLSS of the aerobic pool, with a correlation coefficient r of 0.74 and a fitted linear equation of y = -2.0522 + 0.0023x. As the MLSS of the aerobic pool increased, the total microbial activity SOUR ATP The respiratory activity of denitrifying bacteria in anoxic pools SOUR ATP_DB It also showed a linear positive correlation with the MLSS of the anoxic pool, with a Pearson correlation coefficient r of 0.86 and a fitted linear equation of y = 1.65 + 0.0019x. ATP It also showed a positive linear correlation with a correlation coefficient of 0.63, and the linear equation was y = 9.05 + 0.00774x. The DO of the hypoxic pool and the ATP of the hypoxic pool showed a nonlinear negative correlation. When the DO of the hypoxic pool increased from 3.7 to 0.43, the ATP of the hypoxic pool remained basically unchanged. In order to save energy and maintain high ATP, the DO of the hypoxic pool is recommended to be kept at around 0.3. The fitting r is 0.67, and the fitting nonlinear equation is y = 386.12 + 44.51 / (sqrt(2*pi)*0.2x)*exp(-(ln(x / 0.2386))^2 / 0.08).

[0094] The ATP in the anoxic pool is negatively correlated with the amount of carbon source added, and the data fitting results are shown in Figure 7. When the ATP in the anoxic pool is within the range of 350-650nm, the ATP in the anoxic pool increases as the amount of carbon source added decreases. Therefore, in order to maintain high ATP of microorganisms in the anoxic pool, the drug consumption can be appropriately reduced, and the amount of carbon source added can be reduced as much as possible to between 3000-3500Kg / d. The Pearson correlation coefficient r of the linear fitting is -0.67, and the linear equation of the fitting is y=8452-9.18x. The amount of carbon source added is related to the SOUR of denitrifying bacteria in the anoxic pool. MLSS_DB There is also a good linear negative correlation between the two, with a correlation coefficient r of -0.69 and a fitted linear equation of y = 8521.91-4519.99x. MLSS It also showed a linear negative correlation, with a correlation coefficient r of -0.56, and the fitting equation was y=1.025-0.075x.

[0095] Through the above linear and nonlinear fitting results between microbial activity and operating parameters, it can be found that the SOUR ATP It can better characterize the MLSS of aerobic pool, the air flow of aerobic pool, and the SOUR of aerobic pool. MLSS Can better characterize SV30, anoxic pool SOUR ATP_DB It can better characterize the MLSS of the anoxic pool, and the ATP of the anoxic pool can better characterize the DO and carbon source dosage of the anoxic pool. MLSS_DB In view of the inherent mechanism relationship between the above-mentioned microbial activity and the operating parameters of the sewage treatment system, the present invention selects the aerobic tank SOUR ATP 、Anoxic pool SOUR ATP_DB and hypoxic pool SOUR MLSS_DB Three microbial activity data are used to regulate the sewage treatment system.

[0096] As can be seen from the above embodiments and experimental examples, the present invention provides a method and system for automatically controlling wastewater treatment process parameters, which can reduce air flow and carbon source dosage while ensuring that effluent meets standards, thereby reducing process costs. Therefore, the present invention has great application prospects.

Claims

1. An intelligent control method for sewage treatment based on microbial activity, characterized by: obtaining a value range of an input feature, wherein the input feature includes microbial activity data; Finding the optimal feature combination of input features within the value range of the input features with the goal of meeting the sewage treatment standards at the lowest cost; Determining control parameters based on the optimal feature combination and applying them to the sewage treatment system; Wherein, the control parameters include: air flow rate and carbon source dosage; The cost calculation formula is as follows: Cost = (air flow rate - 700) / 100 + 5 × (carbon source dosage - 3000) / 1000.

2. The intelligent control method for sewage treatment according to claim 1, characterized in that: The method to find the best feature combination of input features is the particle swarm optimization algorithm.

3. The intelligent control method for sewage treatment according to claim 1, characterized in that: The steps include: Step 1: traverse at least one input feature to obtain several groups of input feature combinations; Step 2: input the feature combination into a machine learning model, and obtain a predicted value of the sewage treatment result indicator based on the model output; Step 3: Determine corresponding control parameters based on the feature combination; Step 4: Calculate the cost of each feature combination: Step 5: Select the best feature combination with the lowest cost and the best indicator prediction value according to the calculation results of step 4, and apply the control parameters of the best feature combination calculated in step 3 to the sewage treatment system.

4. The intelligent control method for sewage treatment according to claim 3, characterized in that: The machine learning model is a random forest model.

5. The intelligent control method for sewage treatment according to claim 3, characterized in that: In step 3, the control parameters are determined in one of the following ways: Mode 1: The input features also include air flow rate and / or carbon source dosage, and the values ​​of air flow rate and / or carbon source dosage in the feature combination are directly used as corresponding control parameters; Mode 2: The input features do not include air flow or carbon source dosage, and the air flow or carbon source dosage is calculated by taking the values ​​of each input feature in the feature combination, and the calculation formula is obtained by fitting historical data.

6. The intelligent control method for sewage treatment according to claim 1, characterized in that: The microbial activity data include: aerobic pool SOUR ATP 、Anoxic pool SOUR ATP_DB and hypoxic pool SOUR MLSS_DB ; The process of finding the best feature combination of input features is to ATP and hypoxic pool SOUR MLSS_DB Make changes; Aerobic Pool SOUR ATP The value range is [15,32], and the value change step of different feature combinations is 1; Hypoxic pool SOUR MLSS_DB The value range is [0.7, 1.2], and the value change step of different feature combinations is 0.

1.

7. The intelligent control method for sewage treatment according to claim 1, characterized in that: The input characteristics also include air flow and / or carbon source dosage; The air flow rate range is [700m 3 / h,3000m 3 / h], in the process of finding the best feature combination of input features, the value change step of different feature combinations is 200m 3 / h; The range of carbon source dosage is [3000Kg, 7000Kg]. In the process of finding the best feature combination of input features, the value change step of different feature combinations is 200Kg.

8. The intelligent control method for sewage treatment according to claim 1, characterized in that: The indicator used to evaluate the compliance of sewage treatment results is the total nitrogen in the effluent.

9. An intelligent control system for sewage treatment based on microbial activity, characterized in that: include: Input module, used to input data; a calculation module, configured to calculate control parameters applied to a sewage treatment system according to the intelligent control method for sewage treatment according to any one of claims 1 to 8; The output module is used to output the control parameters to the sewage treatment system.

10. A computer-readable storage medium, characterized in that Stored thereon is: a computer program for implementing the intelligent control method for sewage treatment according to any one of claims 1-8.

Images