A NOB Suppression Method Based on SVM-based FA / FNA Intelligent Switching

Abstract

This invention discloses an intelligent switching NOB suppression method based on SVM (Self-Manufacturing Analyzer) for FA / FNA (Fluorescent Nanoparticle) switching. It utilizes an SVM classifier to analyze the operating parameters and influent / effluent water quality data of a continuous flow short-cut nitrification reactor to determine whether the nitrite nitrogen accumulation rate in the effluent is below 96%. Based on this, different suppression methods are adopted to achieve long-term effective NOB suppression. This SVM-based intelligent switching NOB suppression method for FA / FNA switching overcomes the shortcomings of existing alternating FA / FNA suppression methods. It enables the control system to automatically determine whether strategy switching is necessary based on the reaction operation status and achieves automatic switching between FA and FNA suppression. Long-term effective NOB suppression can be achieved without manual operation, making the automated operation of alternating FA / FNA suppression methods possible.

Description

Technical Field

[0001] This invention relates to wastewater treatment technology, and in particular to a NOB suppression method based on support vector machine (SVM) intelligent switching between FA and FNA. Background Technology

[0002] Short-cut nitrification anaerobic ammonium oxidation (PN / A) is a newly emerging biological nitrogen removal process in recent years. Compared with traditional biological treatment processes, it can reduce carbon source addition by 100%, residual sludge production by 90%, and aeration energy consumption by 60%, and is regarded as an alternative to traditional biological nitrogen removal processes.

[0003] In short-cut nitrification (PN) processes, aerobic ammonia-oxidizing bacteria (AOBs) oxidize ammonia nitrogen to nitrite nitrogen, which, along with ammonia nitrogen, serves as substrates in subsequent anaerobic ammonia oxidation reactions. However, during short-cut nitrification, the presence of nitrite-oxidizing bacteria (NOBs) further oxidizes nitrite nitrogen to nitrate nitrogen, thus reducing the substrate supply required for anaerobic ammonia oxidation. This poses a threat to the stable operation of the PN / A process. Therefore, stably suppressing NOB activity is crucial for the stable operation of the PN / A process.

[0004] Recent studies have shown that alternating inhibition of free ammonia (FA) and free nitrite (FNA) is an effective means of suppressing NOB growth. Switching between these two inhibition methods can prevent NOB from adapting to a single method over a long period, thus avoiding long-term inhibition of NOB activity. However, current FA / FNA alternating inhibition methods rely heavily on manual operation, with the decision to switch methods based on whether the effluent quality meets requirements. The main criteria for switching strategies are the effluent nitrate nitrogen concentration or nitrite nitrogen accumulation rate (NAR). However, the high cost of online nitrate nitrogen analyzers makes it difficult to measure these criteria online, hindering the automated implementation of FA / FNA alternating inhibition. However, these evaluation indicators can be reflected not only directly by measuring nitrate nitrogen concentration but also indirectly by measuring other water quality indicators. These indicators are relatively easy to obtain using conventional online monitoring instruments, such as pH, DO, and aeration rate. The complex correlation between these indicators and effluent NAR and other indicators makes accurate judgment difficult using human experience or traditional control systems.

[0005] In recent years, thanks to its powerful data processing capabilities and ability to understand variable relationships, machine learning has been increasingly used in water quality prediction and process decision-making and early warning. Among these, Support Vector Machine (SVM) is a machine learning method used for classification and regression analysis. Based on statistical learning theory and the principle of maximizing data margins, it separates sample points of different categories as much as possible. For linearly separable problems, SVM exhibits excellent classification performance and has wide applications in decision-making and early warning, such as determining whether effluent water quality is abnormal and whether process operations should be performed. Therefore, using SVM to determine whether switching suppression methods is necessary is highly feasible and has practical significance for automating alternating FA / FNA suppression. This can not only improve process efficiency and stability but also further reduce operating costs, promoting the widespread application of PN / A processes in practice. Summary of the Invention

[0006] To address the difficulty in directly measuring the switching criteria for NOB suppression methods during alternating FA / FNA suppression, this invention provides an SVM-based intelligent switching method for FA / FNA NOB suppression. An SVM classification model is trained using 207 days of reactor operation data, and the model structure is optimized using an optimization algorithm. This allows the trained model to accurately determine whether the effluent NAR is below 96%, thus deciding whether to switch suppression methods. This ensures long-term stable suppression of NOB during short-cut nitrification and provides technical support for the automated operation of alternating FA / FNA suppression methods.

[0007] To achieve the above objectives, this invention provides a method for intelligent switching NOB suppression based on SVM between FA and FNA, comprising the following steps:

[0008] A continuous flow short-cut nitrification reactor was constructed and operated at different FA and FNA concentrations. The calculation formulas for FA and FNA are as follows. Therefore, the pH of the reactor can be controlled by adding an alkali pump, thereby controlling the FA and FNA concentrations of the reactor. The operating parameters of the reactor and the influent and effluent water quality data at different FA and FNA concentrations were collected to train an SVM classifier.

[0009]

[0010]

[0011] Furthermore, operational data such as those from the online pH meter, online DO detector, and influent pump frequency are transmitted to the client database via the control unit;

[0012] Furthermore, the concentration of nitrate nitrogen in the water was measured using a spectrophotometer, and the nitrite nitrogen accumulation rate (NAR) was calculated.

[0013] Furthermore, based on the NAR obtained in the previous step, the collected 207 days of data were labeled, with the sample set with NAR higher than 96% labeled as 0 and the sample set with NAR lower than 96% labeled as 1.

[0014] Furthermore, aeration rate, aeration coefficient, dissolved oxygen (DO), influent ammonia nitrogen concentration, ammonia oxidation rate, and effluent nitrite nitrogen concentration are selected as inputs to the model. The SVM classifier is trained using the data classified in the previous step, and the model is optimized using a grid search cross-validation optimization algorithm to improve the model's prediction accuracy.

[0015] Furthermore, a control program is written in the control unit so that the control program corresponding to the output of the SVM model being 0 is FNA suppression, and the program corresponding to the output being 1 is FA suppression;

[0016] Furthermore, a trained SVM model is used to make real-time decisions on the alternating FA / FNA suppression method. Data transmission and control of FA and FNA concentrations are achieved through a control unit, enabling intelligent switching between the FA / FNA suppression methods.

[0017] Technical effect

[0018] This invention proposes an intelligent switching NOB suppression method for FA / FNA based on SVM. The operating parameters of the reactor and the water quality data of the influent and effluent are input into the trained SVM classification model through the control unit. The model can judge the NOB effluent NAR using conventional online monitoring instrument data, and then decide whether to switch the suppression method, so as to achieve long-term effective suppression of NOB and long-term stable operation of the short-cut nitrification process.

[0019] (1) The SVM classification model introduced in this invention can analyze and learn the relationship between different data variables, use easily measurable water quality data and operating parameters to judge the effluent NAR, and then decide whether to switch the suppression method. After optimization by grid search cross-validation, the SVM classifier achieved an accuracy of 0.90 and an F1 score of 0.92, achieving excellent prediction results. Compared with the traditional FA / FNA alternating suppression method that relies on human judgment, the intelligent switching suppression method proposed in this invention can predict whether the effluent NAR is below 96% without additional equipment such as an online nitrate nitrogen analyzer, and automatically determine whether to switch the suppression method, making it possible for wastewater treatment plants to achieve FA / FNA alternating suppression based on existing equipment.

[0020] (2) The control unit of this invention can automatically execute different inhibition methods based on the output of the SVM classifier, realizing automatic switching between FA inhibition and FNA inhibition. By presetting the inhibition concentrations of FA and FNA in the control program, the control unit controls the alkali pump to adjust the pH, thereby achieving the predetermined inhibition concentration. Considering the alkalinity consumption during short-range nitrification, this invention uses FNA inhibition as a long-term inhibition method and FA inhibition as a short-term inhibition method. This reduces the amount of alkali added while ensuring NOB inhibition, greatly reducing operating costs.

[0021] Combining the above advantages, the SVM-based intelligent switching NOB suppression method for FA / FNA proposed in this invention has significant technical effects such as high accuracy, widespread applicability, and reduced operating costs. It can effectively solve the problem that existing FA / FNA alternating suppression methods require human intervention, and can realize the decision-making and automatic switching of the suppression method, providing important technical support for the automated operation of FA / FNA alternating suppression methods. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the operating device for an SVM-based FA / FNA intelligent switching NOB suppression method according to the present invention.

[0023] Figure 2 This is a real-time running effect diagram of a preferred embodiment of the FA / FNA intelligent switching NOB suppression method based on SVM of the present invention.

[0024] Figure 3 This is a real-time running effect diagram of another preferred embodiment of the FA / FNA intelligent switching NOB suppression method based on SVM of the present invention. Detailed Implementation

[0025] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0026] In the following description, specific details, such as particular internal procedures and techniques, are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will appreciate that the invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of the invention with unnecessary detail.

[0027] A method for intelligent switching of NOB suppression based on SVM and FA / FNA, comprising the following steps:

[0028] Construct a short-path nitrification reactor and its control unit, such as... Figure 1 As shown. The reactor bottom is equipped with aeration pipes, and the aeration rate of the air pump into the reactor is controlled by a rotor flow meter. The reactor is equipped with online pH monitors, online DO monitors, and online ammonia nitrogen monitors to monitor changes in pH, DO, and ammonia nitrogen concentrations. An alkali addition pump adjusts the pH by adding sodium bicarbonate to the reactor, ensuring that FA or FNA reaches the set concentration while maintaining a pH not lower than 6.7. A heating rod is located inside the reactor to maintain the reaction system at 34°C, ensuring that AOB is under suitable environmental conditions.

[0029] The control unit interacts with the programming platform via the Modbus TCP protocol. The control unit transmits the collected operating parameters and water quality data to the SVM classification model in the programming platform, and then switches different suppression methods based on the results returned by the model.

[0030] Specifically as follows:

[0031] First, a short-cut nitrification reactor was constructed and operated stably for 207 days. During these 207 days, aeration rate, aeration coefficient, dissolved oxygen (DO), influent ammonia nitrogen concentration, ammonia oxidation rate, effluent nitrite nitrogen concentration, and nitrogen angiogenesis (NAR) were collected. Then, data sets were labeled based on whether the NAR exceeded 96%; samples with an NAR below 96% were labeled as 1, and those with an NAR above 96% were labeled as 0. These 207 labeled data sets were then used to train an SVM classification model. At this stage, the model's prediction accuracy was 0.81, and its F1 score was 0.79. To further improve the model's prediction accuracy and robustness, this invention employs grid search to try different model hyperparameters and uses 5-fold cross-validation to determine the optimal combination of model hyperparameters. After optimization, the SVM classification model's accuracy on the validation set improved to 0.90, and its F1 score improved to 0.92, indicating a significant improvement in the model's accuracy and reliability.

[0032] The target wastewater treated by this invention exhibits insufficient alkalinity, and short-cut nitrification consumes a certain amount of alkalinity. Therefore, long-term FNA inhibition is easier to achieve under these conditions, while FA inhibition requires the addition of a certain amount of additional alkalinity. To reduce operating costs, this invention uses FNA inhibition as the long-term inhibition method; that is, FNA inhibition is employed when the SVM classification model output is 0, and FA inhibition is employed when the output is 1. A corresponding control program is then written in the control unit, allowing modification of the set inhibition concentrations of FA and FNA via a human-machine interface. The control unit automatically adjusts the reactor pH based on the collected effluent ammonia nitrogen and nitrite nitrogen concentrations to ensure that FA and FNA reach the set inhibition concentrations. Furthermore, to ensure sufficient time for the switched strategies to inhibit NOB, this invention sets the time interval between switching between different inhibition methods to 4 days.

[0033] At this point, the reactor can be controlled in real time using the trained model and the written control program. The control unit only needs to collect readily available indicators such as aeration rate, aeration coefficient, DO, influent ammonia nitrogen concentration, ammonia oxidation rate, and effluent nitrite nitrogen concentration to complete the intelligent switching of FA / FNA inhibition methods. The alternating inhibition method can be automatically operated without manual operation.

[0034] Example 1

[0035] This embodiment provides a NOB suppression method based on SVM-based intelligent switching FA / FNA, wherein the reaction device is as follows: Figure 1 As shown. In this embodiment, the Python platform is used to realize data interaction with the control unit and the training and application of the SVM model. The data collected by the platform is classified according to whether the effluent NAR is below 96%; data above 96% is labeled 0, and data below 96% is labeled 1. Aeration rate, aeration coefficient, DO, influent ammonia nitrogen concentration, ammonia oxidation rate, and effluent nitrite nitrogen concentration are selected as inputs to train the SVM classification model. Grid search cross-validation is used to optimize the model structure and improve the model's prediction accuracy. Then, a control program is written for the control unit so that a model output of 0 corresponds to the FNA suppression method, and an output of 1 corresponds to the FA suppression method. The FA and FNA suppression concentrations are set in the program, and these concentrations can be adjusted at any time via a touchscreen. Subsequently, the control unit controls the alkali pump to adjust the pH according to the ammonia nitrogen and nitrite nitrogen concentrations in the reactor, so that the FA and FNA concentrations in the reactor reach the set suppression concentrations.

[0036] On day 211 of the experiment, the influent ammonia nitrogen concentration was 1618 mg / L, the effluent nitrite nitrogen concentration was 1043.05 mg / L, the aeration rate was 10 L / min, the aeration coefficient was 99.51 L / g, the dissolved oxygen (DO) concentration was 0.25 mg / L, and the ammonia oxidation rate was 83.58%. At this point, the SVM output was 1, indicating that the model considered switching to the FA suppression method. Simultaneously, the control unit automatically increased the pH to suppress FA. To verify the effectiveness of the intelligent switching model, the effluent nitrite nitrogen concentration was measured using a spectrophotometer, yielding a concentration of 40.38 mg / L and a NAR of 96.27%. On day 222 of the experiment, the influent ammonia nitrogen concentration was 1618 mg / L, the effluent nitrite nitrogen concentration was 843.03 mg / L, the aeration rate was 10 L / min, the aeration coefficient was 99.51 L / g, the DO concentration was 0.15 mg / L, and the ammonia oxidation rate was 75.36%. At this time, the measured effluent nitrate nitrogen concentration was 40.57 mg / L, and the NAR was 95.41%, lower than the training switching criterion. Although the SVM classification result was 0 at this point, FA suppression was still performed because the program was set to switch between different strategies at least every 4 days. This example demonstrates that the trained SVM classification model has good robustness and generalization ability, and can make timely and accurate decision-making warnings.

[0037] Example 2

[0038] The apparatus and experimental procedures in this embodiment are the same as in Embodiment 1, and will not be repeated here. On day 226 of the experiment, the influent ammonia nitrogen concentration was 1600.4 mg / L, the effluent nitrite nitrogen concentration was 722.5 mg / L, the aeration rate was 4.6 L / min, the aeration coefficient was 99.88 L / g, the DO concentration was 0.13 mg / L, the ammonia oxidation rate was 58.82%, the measured effluent nitrate nitrogen concentration was 61.5 mg / L, and the NAR was 92.16%. At this time, the output result of the SVM classifier was 1. For the next 6 consecutive days, the SVM decision result was the FA suppression method. On day 231 of the experiment, the control unit switched the suppression method back to the FNA suppression method based on the SVM classification result. Possibly due to the influence of AOB activity at this time, NAR did not reach the set switching standard, but the measured effluent nitrate nitrogen concentration had dropped to 32.07 mg / L, proving that NOB activity had been significantly suppressed, thus demonstrating the effectiveness of the SVM-based FA / FNA intelligent switching NOB suppression method of this invention.

[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A SVM-based FA / FNA intelligent switching NOB suppression method, characterized in that, The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process.

2. The SVM-based FA / FNA intelligent switching NOB suppression method of claim 1, wherein, The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process.

3. The SVM-based FA / FNA intelligent switching NOB suppression method according to claim 1 or 2, characterized in that, The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process.

4. The SVM-based FA / FNA intelligent switching NOB suppression method of claim 3, wherein, The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process.

5. The SVM-based FA / FNA intelligent switching NOB suppression method of claim 1, wherein, The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process.

6. The SVM-based FA / FNA intelligent switching NOB suppression method of claim 5, wherein, The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic switching of the NOB inhibition method in a short-cut nitrification process. The application discloses a method for realizing decision warning of NOB inhibition in a continuous flow short-cut nitrification reactor by using an SVM classifier, and automatic

Images