Sewage treatment monitoring and diagnosis system and method based on AI intelligence

Abstract

The invention discloses a sewage treatment monitoring and diagnosis system and method based on AI intelligence, and relates to the technical field of sewage treatment diagnosis, and the system comprises a monitoring center, a data acquisition module, a data processing module, a data analysis module and an execution module. A data acquisition module acquires process parameters of each treatment process in the whole sewage treatment process and respective water quality detail parameters of sewage after execution of each treatment process, the process parameters and the water quality detail parameters are integrated to serve as a sewage data set of the whole sewage treatment process, and a data processing module performs data mining on the sewage data set to obtain a data mining result. The process execution efficiency of each treatment process and the key characteristic index of each treatment process are obtained through a data analysis module, a sewage treatment monitoring model is constructed by a data analysis module through an AI technology and is used for analyzing the process execution efficiency and the key characteristic index of each treatment process, then the abnormal treatment process is diagnosed, and finally, the abnormal treatment process is analyzed. And the execution module performs process optimization on the abnormal processing process.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment diagnosis, and specifically to an AI-based sewage treatment monitoring and diagnosis system and method. Background Art

[0002] Wastewater treatment is an environmental engineering process that aims to remove or degrade suspended and dissolved solid matter and harmful components in wastewater through physical, chemical or biological methods to reduce its pollution to the environment. The purpose of wastewater treatment is to ensure that the treated wastewater meets the discharge standards, reduce pollution to rivers, lakes and groundwater, and reduce the negative impact on the ecosystem.

[0003] The existing sewage treatment monitoring and diagnosis system can often only monitor the final water quality at the end of the entire sewage treatment process to determine whether the water quality after sewage treatment meets the discharge standards. It lacks effective supervision of the entire sewage treatment process and cannot promptly discover which step of the sewage treatment has a problem, resulting in sewage treatment that is not as effective as expected. Therefore, how to achieve effective supervision of the entire sewage treatment process has become a key and difficult issue in the field. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a sewage treatment monitoring and diagnosis system and method based on AI intelligence.

[0005] The object of the present invention can be achieved by the following technical solutions: A sewage treatment monitoring and diagnosis system based on AI intelligence includes a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a data processing module, a data analysis module and an execution module; The data collection module is used to collect the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrate them into a sewage data set for the whole process of sewage treatment; The data processing module is used to perform data mining on the sewage data set, thereby obtaining the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process; The data analysis module constructs a sewage treatment monitoring model through AI technology, and analyzes the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnoses the abnormal treatment process; The execution module is used to optimize the process of the abnormal processing process.

[0006] Furthermore, the process of collecting the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the water quality detailed parameters corresponding to each treatment process after the sewage is completed, and integrating them as the sewage data set of the whole process of sewage treatment includes: Monitoring points are set up at each treatment process of sewage treatment in the sewage treatment plant, and process parameters of the corresponding treatment process are collected at each monitoring point. The treatment process includes pretreatment process, biochemical treatment process and deep treatment process; Configure the respective operating steps of the pretreatment process, biochemical treatment process and deep treatment process. According to the respective operating steps, the sewage to be treated is pretreated, biochemically treated and deeply treated in turn, and the corresponding process parameters of each treatment process are collected. The process parameters of the pretreatment process include the relevant parameters of the grille decontamination, the relevant parameters of the grit chamber, the relevant parameters of the primary sedimentation tank and the relevant parameters of the regulating tank; the process parameters of the biochemical treatment process include the aerobic treatment process parameters, the anaerobic treatment process parameters and the nutrient removal process parameters; the process parameters of the deep treatment process include the coagulation sedimentation parameters, the filtration parameters, the membrane treatment parameters and the disinfection parameters; Obtain water quality indicators before each treatment process is performed; Obtain water quality indicators after each treatment process is completed; The time periods before and after each sewage treatment process are marked as periodic monitoring periods, and the periodic change data of water quality characterization indicators within each periodic monitoring period are obtained as the detailed water quality parameters of the sewage after the corresponding treatment process is completed. The detailed water quality parameters corresponding to all treatment processes are integrated as the sewage data set corresponding to the entire sewage treatment process.

[0007] Furthermore, data mining is performed on the sewage data set to obtain the efficient execution process of each treatment process, including: Construct a data processing stack, input the sewage data set into the data processing stack, and divide the sewage data set into segmented data corresponding to each treatment process by the data processing stack, and set the theoretical sewage composition index corresponding to the sewage after each treatment process is completed in the whole sewage treatment process; According to the preset sewage composition index reference table, data mining is performed on the segmented data corresponding to each treatment process, and then the actual sewage composition index of each treatment process is obtained, wherein the theoretical sewage composition index is composed of several single sewage indexes under ideal conditions, and the actual sewage composition index is composed of several single sewage indexes under actual treatment conditions; The actual sewage composition index and the theoretical sewage composition index are compared to obtain the efficiency of each treatment process.

[0008] Furthermore, the process of obtaining the key characteristic indicators corresponding to each processing step includes: A bag-of-words analysis model is constructed through deep learning technology, and historical sewage treatment data is obtained. The historical sewage treatment data is processed through text analysis technology to obtain a characteristic data set of each treatment process corresponding to the sewage in the entire sewage treatment process; Each treatment process corresponds to a feature data set in the historical sewage treatment data, which is composed of several feature indicators. The feature data set of each treatment process is input into the bag-of-words analysis model, and the bag-of-words analysis model counts the contribution of different feature indicators to the efficiency of process execution in the feature data set corresponding to each treatment process. The feature indicator with the highest contribution is taken as the key feature indicator of the current treatment process. The bag-of-words analysis model is used to repeatedly analyze and count the feature data sets of each treatment process, so as to obtain the key feature indicators corresponding to each treatment process during the whole sewage treatment process.

[0009] Furthermore, the process of building a sewage treatment monitoring model through AI technology includes: Obtain the known actual data of sewage under each treatment process, which records the content of each component index in the sewage, and obtain the actual execution operation log of each treatment process, which is used to record the actual entire operation process of each treatment process; A preliminary monitoring model is constructed through AI technology, and the known sewage real-time data of each treatment process in the whole sewage treatment process and the actual execution operation log of each treatment process are integrated as a monitoring item. All the monitoring items are input into the preliminary monitoring model, and the monitoring model is used to obtain the treatment completion degree of each treatment process. Compare the processing completion degree of each processing step obtained by the monitoring model with the processing completion degree of the processing step in actual conditions, thereby screening out processing steps with inconsistent processing completion degrees and increasing the amount of data processed by the monitoring model for the corresponding processing steps; Obtain the model prediction success rate of the monitoring model and record it as Sc; Set the lower threshold of the model prediction success rate and record it as Γ; When Sc ≥ Γ, the preliminary monitoring model is converted into the final sewage treatment monitoring model; When Sc<Γ, configure the AI ​​computing power of the monitoring model, and train the monitoring model until the model prediction success rate of the monitoring model meets the requirements, and build a sewage treatment monitoring model.

[0010] Furthermore, the process of analyzing the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnosing the abnormal treatment process includes: Input the process execution efficiency and key characteristic indicators of each treatment process in the whole process of sewage treatment into the sewage treatment monitoring model, set the effective interval corresponding to the process execution efficiency, set the value fluctuation range of the key characteristic indicators, and input the effective interval and value fluctuation range of each treatment process into the sewage treatment monitoring model; The sewage treatment monitoring model is used to analyze whether the efficiency of each treatment process is within the effective range, and whether the fluctuation of the key characteristic indicators corresponding to each treatment process is within the corresponding numerical fluctuation range; If the process execution efficiency of the processing process is within the effective range, and the indicator fluctuation size of the key characteristic indicator of the processing process is within the numerical fluctuation range, then the current processing process is diagnosed to be normal; otherwise, the corresponding processing process is diagnosed to be abnormal.

[0011] Furthermore, the process of optimizing the abnormal processing process includes: Construct an expert knowledge base, which stores the standard process steps of each treatment process for the whole process of sewage treatment, the abnormal operation events that occur when executing each treatment process, and the corresponding solutions for abnormal operation events; The abnormal processing steps are input into the expert knowledge base, and then the abnormal operation events corresponding to the current processing steps and the solutions for the abnormal operation events are matched from the expert knowledge base. The administrator arranges relevant staff to implement the solutions and intervene in the abnormal processing steps.

[0012] Furthermore, the diagnostic method of the sewage treatment monitoring and diagnostic system based on AI intelligence includes the following steps: Step S1: Collect the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrate them as a sewage data set for the whole process of sewage treatment; Step S2: Perform data mining on the sewage data set to obtain the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process; Step S3: constructing a sewage treatment monitoring model through AI technology, and analyzing the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnosing the abnormal treatment process; Step S4: Optimize the abnormal processing steps.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: the process parameters of each treatment process in the whole process of sewage treatment and the corresponding water quality detailed parameters of the sewage after each treatment process are completed are collected through the data acquisition module, and integrated as the sewage data set of the whole process of sewage treatment; the data processing module performs data mining on the sewage data set to obtain the process execution efficiency and key characteristic indicators of each treatment process; a sewage treatment monitoring model is constructed through the data analysis module to analyze the process execution efficiency and key characteristic indicators of each treatment process, and then diagnose the abnormal treatment process; finally, the execution module optimizes the process of the abnormal treatment process, thereby realizing effective supervision of each treatment process in the whole process of sewage treatment, facilitating the diagnosis of the treatment process under abnormal conditions, and timely optimizing the process, ensuring the efficient implementation of sewage treatment, and achieving global supervision of sewage treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0015] like Figure 1 As shown, the AI-based sewage treatment monitoring and diagnosis system includes a monitoring center, which is communicatively connected to a data acquisition module, a data processing module, a data analysis module, and an execution module; The data collection module is used to collect the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrate them into a sewage data set for the whole process of sewage treatment; The data processing module is used to perform data mining on the sewage data set, thereby obtaining the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process; The data analysis module constructs a sewage treatment monitoring model through AI technology, and analyzes the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnoses the abnormal treatment process; The execution module is used to optimize the process of the abnormal processing process.

[0016] It should be further explained that, in the specific implementation process, the process of collecting the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the completion of each treatment process, and integrating the sewage data set as the whole process of sewage treatment includes: Monitoring points are set up at each treatment process of sewage treatment in the sewage treatment plant, and the process parameters of the corresponding treatment process are collected at each monitoring point. The treatment processes corresponding to the whole sewage treatment process include pretreatment process, biochemical treatment process and deep treatment process; Configure the operation steps of the pretreatment process, biochemical treatment process and deep treatment process; According to the respective operation steps, the sewage to be treated is subjected to pretreatment, biochemical treatment and deep treatment in turn, and the process parameters corresponding to each treatment process are collected. The process parameters of the pretreatment process include parameters related to screen decontamination, parameters related to grit chamber, parameters related to primary sedimentation tank and parameters related to regulating tank; the process parameters of the biochemical treatment process include aerobic treatment process parameters, anaerobic treatment process parameters and nutrient removal process parameters; the process parameters of the deep treatment process include coagulation sedimentation parameters, filtration parameters, membrane treatment parameters and disinfection parameters; Obtain water quality indicators before each treatment process is performed; Obtain water quality indicators after each treatment process is completed; The time periods before and after each sewage treatment process are marked as periodic monitoring periods, and the periodic change data of water quality characterization indicators within each periodic monitoring period are obtained as the detailed water quality parameters of the sewage after the corresponding treatment process is completed. The detailed water quality parameters corresponding to all treatment processes are integrated as the sewage data set corresponding to the entire sewage treatment process.

[0017] It should be noted that the parameters related to the screen pollution removal include the size of the screen gap, the water level difference before and after the screen, and the amount of solid matter intercepted by the screen; the parameters related to the grit chamber include the residence time of the grit chamber, the surface load of the grit chamber, and the discharge of grit; the parameters related to the primary settling tank include the residence time, surface load, sludge settling ratio, sludge concentration, and the discharge of sediment; the parameters related to the equalization tank include the capacity of the equalization tank, the hydraulic retention time of the equalization tank, the pH value of the sewage, and the temperature.

[0018] It should be noted that aerobic treatment process parameters include sludge volume index, dissolved oxygen (DO) concentration, mixed liquor reflow ratio, foam control parameters, aerobic tank residence time and load rate; anaerobic treatment process parameters include anaerobic sludge concentration, hydraulic retention time, upflow velocity and organic loading rate; nutrient removal process parameters include ammonia nitrogen concentration, nitrite nitrogen concentration, nitrate nitrogen concentration, total nitrogen concentration, total phosphorus concentration and sludge age.

[0019] It should be noted that the coagulation and sedimentation parameters include the type and dosage of coagulant, mixing time and residence time in the sedimentation tank; the filtration parameters include the type of filtration medium, filtration speed, backwash cycle and time, and backwash intensity; the membrane treatment parameters include membrane material type, transmembrane pressure difference, water production, recovery rate and membrane cleaning cycle; the disinfection parameters include the type and dosage of disinfectant, contact time and residual disinfectant concentration.

[0020] It should be further explained that, in the specific implementation process, data mining is performed on the sewage data set to obtain the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process, including: Construct a data processing stack, input the sewage data set into the data processing stack, and divide the sewage data set into segmented data corresponding to each treatment process by the data processing stack, and set the theoretical sewage composition index corresponding to the sewage after each treatment process is completed in the whole sewage treatment process; According to the preset sewage composition index reference table, data mining is performed on the segmented data corresponding to each treatment process, and then the actual sewage composition index of each treatment process is obtained, wherein the theoretical sewage composition index is composed of several single sewage indexes under ideal conditions, and the actual sewage composition index is composed of several single sewage indexes under actual treatment conditions; According to the actual sewage composition index and the theoretical sewage composition index, the same item is compared, and then the process execution efficiency of each treatment process is obtained. The process execution efficiency of the pretreatment process, the biochemical treatment process and the deep treatment process are recorded as γ1, γ2 and γ3 respectively; Then γ1, γ2 and γ3 are expressed as follows: γ1=a / A, in the pretreatment process, a represents the number of single sewage indicators under actual treatment conditions that meet the same single sewage indicators under ideal conditions, and A represents the total number of corresponding single sewage indicators after the pretreatment process is completed; γ2=b / B. In the biochemical treatment process, b represents the number of single sewage indicators under the actual treatment conditions that meet the same single sewage indicators under ideal conditions. B represents the total number of single sewage indicators corresponding to the completion of the biochemical treatment process. γ3=c / C, in the deep treatment process, c represents the number of single sewage indicators under actual treatment conditions and the number of single sewage indicators under ideal conditions that are the same as the single sewage indicators under actual treatment conditions, and C represents the total number of single sewage indicators corresponding to the deep treatment process after completion; Among them, the value ranges of γ1, γ2 and γ3 are all real numbers between 0 and 1.

[0021] A bag-of-words analysis model is constructed through deep learning technology, and historical sewage treatment data is obtained. The historical sewage treatment data is processed through text analysis technology to obtain a characteristic data set of each treatment process corresponding to the sewage in the entire sewage treatment process; Each treatment process corresponds to a feature data set in the historical sewage treatment data, which is composed of several feature indicators. The feature data set of each treatment process is input into the bag-of-words analysis model, and the bag-of-words analysis model counts the contribution of different feature indicators to the efficiency of process execution in the feature data set corresponding to each treatment process. For any processing step, when a certain characteristic index of the processing step is input into the word bag analysis model, the process execution efficiency of the current processing step is obtained in real time, and the change amplitude of the process execution efficiency before and after the characteristic index is input into the preliminary word bag analysis model is counted, and the change amplitude is used as the contribution of the current characteristic index to the current processing step; The feature indicator with the highest contribution is taken as the key feature indicator of the current treatment process. The bag-of-words analysis model is used to repeatedly analyze and count the feature data sets of each treatment process, so as to obtain the key feature indicators corresponding to each treatment process during the whole sewage treatment process.

[0022] It should be further explained that in the specific implementation process, the process of building a sewage treatment monitoring model through AI technology includes: Obtain the actual data of known sewage corresponding to each treatment process during the whole sewage treatment process. The actual data of known sewage records the content of each component index in the sewage. Obtain the actual execution operation log of each treatment process during the whole sewage treatment process. The actual execution operation log is used to record the actual operation process of each treatment process. A preliminary monitoring model is constructed through AI technology, and the known sewage real-time data of each treatment process in the whole sewage treatment process and the actual execution operation log of each treatment process are integrated as a monitoring item. The monitoring items of all treatment processes are input into the preliminary monitoring model, and then the monitoring model is used to derive the treatment completion degree of each treatment process. Compare the processing completion degree of each processing step obtained by the monitoring model with the processing completion degree of the processing step in actual conditions, thereby screening out processing steps with inconsistent processing completion degrees and increasing the amount of data processed by the monitoring model for the corresponding processing steps; Obtain the model prediction success rate of the monitoring model, and record the model prediction success rate as Sc; Set the lower limit threshold of the model prediction success rate, and record the lower limit threshold as Γ; When Sc ≥ Γ, the preliminary monitoring model is converted into the final sewage treatment monitoring model; When Sc<Γ, configure the AI ​​computing power of the monitoring model, and train the monitoring model until the model prediction success rate of the monitoring model meets the requirements, and build a sewage treatment monitoring model.

[0023] It should be further explained that, in the specific implementation process, the process of analyzing the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnosing the abnormal treatment process includes: Input the process execution efficiency and key characteristic indicators of each treatment process in the whole process of sewage treatment into the sewage treatment monitoring model, set the effective interval corresponding to the process execution efficiency, set the value fluctuation range of the key characteristic indicators, and input the effective interval and value fluctuation range of each treatment process into the sewage treatment monitoring model; The sewage treatment monitoring model is used to analyze whether the efficiency of each treatment process is within the effective range, and whether the fluctuation of the key characteristic indicators corresponding to each treatment process is within the corresponding numerical fluctuation range; If the process execution efficiency of the processing process is within the effective range, and the indicator fluctuation size of the key characteristic indicator of the processing process is within the numerical fluctuation range, then the current processing process is diagnosed to be normal; otherwise, the corresponding processing process is diagnosed to be abnormal.

[0024] It should be further explained that, in the specific implementation process, the process of optimizing the abnormal processing process includes: Construct an expert knowledge base, which stores the standard process steps of each treatment process for the whole process of sewage treatment, the abnormal operation events that occur when executing each treatment process, and the corresponding solutions for abnormal operation events; The abnormal processing steps are input into the expert knowledge base, and then the abnormal operation events corresponding to the current processing steps and the solutions for the abnormal operation events are matched from the expert knowledge base. The administrator arranges relevant staff to implement the solutions and intervene in the abnormal processing steps.

[0025] The present invention also provides a diagnostic method for a sewage treatment monitoring and diagnostic system based on AI intelligence, comprising the following steps: Step S1: Collect the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrate them as a sewage data set for the whole process of sewage treatment; Step S2: Perform data mining on the sewage data set to obtain the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process; Step S3: constructing a sewage treatment monitoring model through AI technology, and analyzing the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnosing the abnormal treatment process; Step S4: Optimize the abnormal processing steps.

[0026] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The AI-based sewage treatment monitoring and diagnosis system includes a monitoring center, which is characterized by: The monitoring center is communicatively connected to a data acquisition module, a data processing module, a data analysis module and an execution module; The data collection module is used to collect the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrate them into a sewage data set for the whole process of sewage treatment; The data processing module is used to perform data mining on the sewage data set, thereby obtaining the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process; The data analysis module constructs a sewage treatment monitoring model through AI technology, and analyzes the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnoses the abnormal treatment process; The execution module is used to optimize the process of the abnormal processing process.

2. The AI-based sewage treatment monitoring and diagnosis system according to claim 1 is characterized in that: The process of collecting the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrating them as the sewage data set of the whole process of sewage treatment includes: Monitoring points are set up at each treatment process of sewage treatment in the sewage treatment plant, and process parameters of the corresponding treatment process are collected at each monitoring point. The treatment process includes pretreatment process, biochemical treatment process and deep treatment process; Configure the respective operating steps of the pretreatment process, biochemical treatment process and deep treatment process. According to the respective operating steps, the sewage to be treated is pretreated, biochemically treated and deeply treated in turn, and the corresponding process parameters of each treatment process are collected. The process parameters of the pretreatment process include the relevant parameters of the grille decontamination, the relevant parameters of the grit chamber, the relevant parameters of the primary sedimentation tank and the relevant parameters of the regulating tank; the process parameters of the biochemical treatment process include the aerobic treatment process parameters, the anaerobic treatment process parameters and the nutrient removal process parameters; the process parameters of the deep treatment process include the coagulation sedimentation parameters, the filtration parameters, the membrane treatment parameters and the disinfection parameters; Obtain water quality indicators before each treatment process is performed; Obtain water quality indicators after each treatment process is completed; The time periods before and after each sewage treatment process are marked as periodic monitoring periods, and the periodic change data of water quality characterization indicators within each periodic monitoring period are obtained as the detailed water quality parameters of the sewage after the corresponding treatment process is completed. The detailed water quality parameters corresponding to all treatment processes are integrated as the sewage data set corresponding to the entire sewage treatment process.

3. The AI-based sewage treatment monitoring and diagnosis system according to claim 2 is characterized in that: The process of mining the sewage data set and finding out the efficiency of each treatment process includes: Construct a data processing stack, input the sewage data set into the data processing stack, and divide the sewage data set into segmented data corresponding to each treatment process by the data processing stack, and set the theoretical sewage composition index corresponding to the sewage after each treatment process is completed in the whole sewage treatment process; According to the preset sewage composition index reference table, data mining is performed on the segmented data corresponding to each treatment process, and then the actual sewage composition index of each treatment process is obtained, wherein the theoretical sewage composition index is composed of several single sewage indexes under ideal conditions, and the actual sewage composition index is composed of several single sewage indexes under actual treatment conditions; The actual sewage composition index and the theoretical sewage composition index are compared to obtain the efficiency of each treatment process.

4. The AI-based sewage treatment monitoring and diagnosis system according to claim 3 is characterized in that: The process of deriving the key characteristic indicators corresponding to each processing step includes: A bag-of-words analysis model is constructed through deep learning technology, and historical sewage treatment data is obtained. The historical sewage treatment data is processed through text analysis technology to obtain a characteristic data set of each treatment process corresponding to the sewage in the entire sewage treatment process; Each treatment process corresponds to a feature data set in the historical sewage treatment data, which is composed of several feature indicators. The feature data set of each treatment process is input into the bag-of-words analysis model, and the bag-of-words analysis model counts the contribution of different feature indicators to the efficiency of process execution in the feature data set corresponding to each treatment process. The feature indicator with the highest contribution is taken as the key feature indicator of the current treatment process. The bag-of-words analysis model is used to repeatedly analyze and count the feature data sets of each treatment process, so as to obtain the key feature indicators corresponding to each treatment process during the whole sewage treatment process.

5. The AI-based sewage treatment monitoring and diagnosis system according to claim 4 is characterized in that: The process of building a sewage treatment monitoring model through AI technology includes: Obtain the known actual data of sewage under each treatment process, which records the content of each component index in the sewage, and obtain the actual execution operation log of each treatment process, which is used to record the actual entire operation process of each treatment process; A preliminary monitoring model is constructed through AI technology, and the known sewage real-time data of each treatment process in the whole sewage treatment process and the actual execution operation log of each treatment process are integrated as a monitoring item. All the monitoring items are input into the preliminary monitoring model, and the monitoring model is used to obtain the treatment completion degree of each treatment process. Compare the processing completion degree of each processing step obtained by the monitoring model with the processing completion degree of the processing step in actual conditions, thereby screening out processing steps with inconsistent processing completion degrees and increasing the amount of data processed by the monitoring model for the corresponding processing steps; Obtain the model prediction success rate of the monitoring model and record it as Sc; Set the lower threshold of the model prediction success rate and record it as Γ; When Sc ≥ Γ, the preliminary monitoring model is converted into the final sewage treatment monitoring model; When Sc<Γ, configure the AI ​​computing power of the monitoring model, and train the monitoring model until the model prediction success rate of the monitoring model meets the requirements, and build a sewage treatment monitoring model.

6. The AI-based sewage treatment monitoring and diagnosis system according to claim 5 is characterized in that: The process of analyzing the efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model and then diagnosing the abnormal treatment process includes: Input the process execution efficiency and key characteristic indicators of each treatment process in the whole process of sewage treatment into the sewage treatment monitoring model, set the effective interval corresponding to the process execution efficiency, set the value fluctuation range of the key characteristic indicators, and input the effective interval and value fluctuation range of each treatment process into the sewage treatment monitoring model; The sewage treatment monitoring model is used to analyze whether the efficiency of each treatment process is within the effective range, and whether the fluctuation of the key characteristic indicators corresponding to each treatment process is within the corresponding numerical fluctuation range; If the process execution efficiency of the processing process is within the effective range, and the indicator fluctuation size of the key characteristic indicator of the processing process is within the numerical fluctuation range, then the current processing process is diagnosed to be normal; otherwise, the corresponding processing process is diagnosed to be abnormal.

7. The AI-based sewage treatment monitoring and diagnosis system according to claim 6 is characterized in that: The process of optimizing the abnormal processing process includes: Construct an expert knowledge base, which stores the standard process steps of each treatment process for the whole process of sewage treatment, the abnormal operation events that occur when executing each treatment process, and the corresponding solutions for abnormal operation events; The abnormal processing steps are input into the expert knowledge base, and then the abnormal operation events corresponding to the current processing steps and the solutions for the abnormal operation events are matched from the expert knowledge base. The administrator arranges relevant staff to implement the solutions and intervene in the abnormal processing steps.

8. The diagnostic method of the sewage treatment monitoring and diagnostic system based on AI intelligence according to any one of claims 1 to 7, characterized in that: The following steps are involved: Step S1: Collect the process parameters corresponding to each treatment process in the whole process of sewage treatment, as well as the detailed water quality parameters corresponding to each treatment process after the sewage is completed, and integrate them as a sewage data set for the whole process of sewage treatment; Step S2: Perform data mining on the sewage data set to obtain the process execution efficiency corresponding to each treatment process and the key characteristic indicators corresponding to each treatment process; Step S3: constructing a sewage treatment monitoring model through AI technology, and analyzing the process execution efficiency and key characteristic indicators of each treatment process through the sewage treatment monitoring model, and then diagnosing the abnormal treatment process; Step S4: Optimize the abnormal processing steps.

Images