An Internet of Things-based whole-process supervision method and system for sewage treatment

Abstract

The present invention relates to the technical field of water treatment, and discloses a whole-process supervision method and system for sewage treatment based on the Internet of Things, which collects the actual dissolved oxygen data in the biochemical reaction tank during the sewage purification process; determines the dissolved oxygen abnormal points according to the collected actual dissolved oxygen data, and identifies the dissolved oxygen state in the biochemical reaction tank according to the position of the dissolved oxygen abnormal points and the dissolved oxygen abnormal deviation data. In the case of abnormal dissolved oxygen state, taking the center point of the dissolved oxygen abnormality as the reference, circular screening areas are constructed in turn with the dissolved oxygen abnormal points that are gradually far away from the center point of the dissolved oxygen abnormality as the centers and the effective coverage radius of the aeration disk as the radius, and the monitored aeration disks in the multi-layer circular screening areas are overlapped, and the abnormal investigation order of the monitored aeration disks is constructed based on the overlapping data of the monitored aeration disks. The present invention realizes the effective monitoring of dissolved oxygen and locates the aeration disks causing the abnormal dissolved oxygen state during the inspection.

Description

Technical Field

[0001] The present invention relates to the technical field of water treatment, and specifically relates to a whole-process supervision method and system for sewage treatment based on the Internet of Things. Background Art

[0002] Sewage treatment refers to the treatment of wastewater containing various harmful substances to remove or reduce the pollutants therein so that it meets the specified discharge standards. It can effectively reduce the pollution of sewage to water bodies and protect the sustainable utilization of water resources. With the acceleration of the urbanization process, sewage treatment has become an important link in urban environmental protection.

[0003] The biochemical reaction tank is the core unit of a sewage treatment plant, and mainly realizes the degradation of pollutants through the metabolic activities of microorganisms. A pre-anoxic zone, an anaerobic zone, an anoxic zone, and an aerobic zone are arranged inside it to provide a suitable environment for different microorganisms.

[0004] And the dissolved oxygen data is the "regulator" of the biochemical treatment efficiency of the biochemical reaction tank. The dissolved oxygen concentration directly affects the microbial activity and treatment effect. Therefore, it is necessary to effectively monitor the dissolved oxygen in the biochemical reaction tank, and be able to timely detect the abnormality of the aeration disc when the dissolved oxygen concentration does not meet the requirements, so as to improve the sewage treatment stability of the biochemical reaction tank.

[0005] Based on this, the present application provides a whole-process supervision method and system for sewage treatment based on the Internet of Things. Summary of the Invention

[0006] The purpose of the present invention is to provide a whole-process supervision method and system for sewage treatment based on the Internet of Things. By hierarchically dividing the biochemical reaction tank, DO sensors are set on different levels of the biochemical reaction tank, and the dissolved oxygen data in the biochemical reaction tank is comprehensively identified according to the DO sensors, so as to complete the positioning and identification of the dissolved oxygen abnormal area in the biochemical reaction tank, and process the position of the dissolved oxygen abnormal area and the dissolved oxygen abnormal deviation data, so as to realize the determination of the dissolved oxygen state of the biochemical reaction tank from two dimensions of spatial position and data deviation, and complete the positioning inspection of the abnormal aeration disc according to the dissolved oxygen abnormal state of the biochemical reaction tank.

[0007] The purpose of the present invention can be realized by the following technical solutions:

[0008] In the first aspect, the present invention provides a whole-process supervision method for sewage treatment based on the Internet of Things, including the following steps:

[0009] Collect the actual dissolved oxygen data of the biochemical reaction tank during the sewage purification process;

[0010] Determine the dissolved oxygen abnormal points based on the actually collected dissolved oxygen data, and identify the dissolved oxygen state in the biochemical reaction tank according to the positions of the dissolved oxygen abnormal points and the dissolved oxygen abnormal deviation data;

[0011] The dissolved oxygen state in the biochemical reaction tank includes the dissolved oxygen abnormal state and the dissolved oxygen normal state;

[0012] In the dissolved oxygen abnormal state, taking the dissolved oxygen abnormal center point as the benchmark, successively construct circular screening areas with the dissolved oxygen abnormal points gradually away from the dissolved oxygen abnormal center point as the centers and the effective coverage radius of the aeration disk as the radii;

[0013] Perform coincidence processing on the monitored aeration disks within the multi-layer circular screening areas, and construct the abnormal investigation order of the monitored aeration disks based on the coincidence data of the monitored aeration disks.

[0014] In a second aspect, the present invention provides an Internet of Things-based whole-process supervision system for sewage treatment, including:

[0015] A data acquisition module for collecting the actually dissolved oxygen data in the biochemical reaction tank during the sewage purification process;

[0016] A behavior recognition module for determining the dissolved oxygen abnormal points based on the actually collected dissolved oxygen data, and identifying the dissolved oxygen state in the biochemical reaction tank according to the positions of the dissolved oxygen abnormal points and the dissolved oxygen abnormal deviation data;

[0017] The dissolved oxygen state in the biochemical reaction tank includes the dissolved oxygen abnormal state and the dissolved oxygen normal state;

[0018] An abnormal construction module for, in the dissolved oxygen abnormal state, taking the dissolved oxygen abnormal center point as the benchmark, successively constructing circular screening areas with the dissolved oxygen abnormal points gradually away from the dissolved oxygen abnormal center point as the centers and the effective coverage radius of the aeration disk as the radii;

[0019] A positioning and inspection module for performing coincidence processing on the monitored aeration disks within the multi-layer circular screening areas, and constructing the abnormal investigation order of the monitored aeration disks based on the coincidence data of the monitored aeration disks.

[0020] Advantages of the present invention:

[0021] The present invention divides the biochemical reaction tank into levels, sets DO sensors on different levels of the biochemical reaction tank, and comprehensively identifies the dissolved oxygen data in the biochemical reaction tank according to the DO sensors, completes the positioning and identification of the dissolved oxygen abnormal areas in the biochemical reaction tank, and processes the positions and dissolved oxygen abnormal deviation data of the dissolved oxygen abnormal areas, so as to realize the determination of the dissolved oxygen state of the biochemical reaction tank from two dimensions of spatial position and data deviation;

[0022] When the present invention determines that the dissolved oxygen state in the biochemical reaction tank is abnormal, it determines the center point of the dissolved oxygen abnormality, and takes the center point of the dissolved oxygen abnormality as a reference. Sequentially, it constructs circular screening areas with the dissolved oxygen abnormality points that are gradually farther away from the center point of the dissolved oxygen abnormality as the centers and the effective coverage radius of the aeration disk as the radii. It performs overlapping processing on the monitored aeration disks within the multi-layer circular screening areas to obtain the first-level monitoring sequence group of the monitored aeration disks. Then, taking the first-level monitoring sequence group as a reference, it processes each monitored aeration disk in the first-level monitoring sequence group that has adjacent monitored aeration disks to obtain the monitored values of the monitored adjacent aeration disks, and obtains the second-level monitoring sequence group of the monitored aeration disks according to the overlapping times of the monitored aeration disks and the monitored values of the monitored adjacent aeration disks, and constructs the abnormal investigation sequence of the monitored aeration disks. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be further described below with reference to the accompanying drawings.

[0024] Figure 1 is a flowchart of a method for the whole-process supervision of sewage treatment based on the Internet of Things according to an embodiment of the present invention;

[0025] Figure 2 is a block diagram of a system for the whole-process supervision of sewage treatment based on the Internet of Things according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] Please refer to Figure 1 As shown, the present invention is a method for the whole-process supervision of sewage treatment based on the Internet of Things, including the following steps:

[0028] Collect the actual dissolved oxygen data in the biochemical reaction tank during the sewage purification process;

[0029] Determine the dissolved oxygen abnormality points according to the collected actual dissolved oxygen data, and identify the dissolved oxygen state in the biochemical reaction tank according to the positions of the dissolved oxygen abnormality points and the dissolved oxygen abnormality deviation data;

[0030] The dissolved oxygen state in the biochemical reaction tank includes a dissolved oxygen abnormal state and a dissolved oxygen normal state;

[0031] In the dissolved oxygen abnormal state, taking the center point of the dissolved oxygen abnormality as a reference, sequentially construct circular screening areas with the dissolved oxygen abnormality points that are gradually farther away from the center point of the dissolved oxygen abnormality as the centers and the effective coverage radius of the aeration disk as the radii;

[0032] Overlap the monitoring aeration disks within the multi-layer circular screening area, and construct the abnormal troubleshooting sequence of the monitoring aeration disks based on the overlap data of the aeration disks.

[0033] The process of obtaining the actual dissolved oxygen data is as follows:

[0034] Set DO sensors at different spatial positions vertically in the biochemical reaction tank;

[0035] At the top layer position of the biochemical reaction tank, set no less than 2 DO sensors;

[0036] At the middle layer position of the biochemical reaction tank, set no less than 2 DO sensors;

[0037] At the bottom layer position of the biochemical reaction tank, set no less than 2 DO sensors;

[0038] Moreover, the projections of the DO sensors set at the bottom, middle, and top layer positions of the biochemical reaction tank are staggered vertically;

[0039] Based on the Grubbs criterion, for the same layer of DO sensors in each biochemical reaction tank at the same moment:

[0040] That is, through the formula Obtain the standardized distance G of a single data point from the data mean in each layer;

[0041] Among them, Is the dissolved oxygen measurement value of the i-th DO sensor;

[0042] Is the sample mean of the data set, and the data set is the number of DO sensors in the same layer;

[0043] s is the sample standard deviation of the data set;

[0044] Obtain the critical value of the standardized distance of a single data point corresponding to each DO sensor from the data mean ;

[0045] That is, through the formula Obtain the critical value of the standardized distance of a single data point from the data mean ;

[0046] Among them, N is the number of samples in the data set;

[0047] Is the t-partial quantile with degrees of freedom N - 2, corresponding to the significance ;

[0048] If the > If so, it indicates that the dissolved oxygen data of the DO sensor is abnormal, and the DO sensor is marked as an abnormal DO sensor.

[0049] Exemplarily:

[0050] Three DO sensors are set at the top of the biochemical reaction tank. At a certain moment, the dissolved oxygen data of the three DO sensors are measured: X1 = 2.3 mg / L, X2 = 2.5 mg / L, X3 = 1.1 mg / L;

[0051] Calculate the mean and standard deviation:

[0052]

[0053]

[0054] Calculate the G value of each data point

[0055] Such as X3 = 1.1 mg / L

[0056]

[0057] Determine the critical value

[0058] Set a = 0.01, N = 3, degrees of freedom df = N - 2 = 1;

[0059] Look up the Grubbs critical value table. The standard Grubbs critical value table (partial) is shown in Table 1:

[0060]

[0061] When N = 3, regardless of one-sided or two-sided tests, the critical value is 1.155 (because the sample is too small and the statistic distribution is unique);

[0062] The data in Table 1 is from Grubbs' original paper Technometrics 1969 or standard statistical reference books (such as "Statistics in Chemical Analysis");

[0063] Verify through the cumulative distribution function (CDF) formula of the Grubbs distribution:

[0064] For N = 3, Satisfy:

[0065]

[0066] Through Monte Carlo simulation or numerical integration, it can be obtained that when = 1.155, the misjudgment probability of outliers is 1%;

[0067] When G = 1.14 < If = 1.155, the DO sensor corresponding to G3 meets the requirements and is recorded as a normal DO sensor;

[0068] Otherwise, it is recorded as an abnormal DO sensor.

[0069] In this embodiment, the situation where there are multiple dissolved oxygen abnormal points in the biochemical reaction pool is processed. Specifically:

[0070] Identify the states of all DO sensors in the biochemical reaction pool, and mark the positions of all abnormal DO sensors in the biochemical reaction pool, which are recorded as dissolved oxygen abnormal points;

[0071] Process all the dissolved oxygen abnormal points in the biochemical reaction pool to construct a dissolved oxygen abnormal area;

[0072] The processing methods include but are not limited to convex hull, triangulation or voxelization processing;

[0073] Obtain the volume of the dissolved oxygen abnormal area, and perform a ratio processing on the volume of the dissolved oxygen abnormal area and the volume of the biochemical reaction pool to obtain the ratio of the dissolved oxygen abnormal area;

[0074] Obtain the value corresponding to the difference, which is recorded as the abnormal G difference. Perform a ratio calculation on the abnormal G difference of the abnormal DO sensor and the corresponding to obtain the abnormal G difference ratio of the abnormal DO sensor;

[0075] Sum and average the abnormal G difference ratios of all abnormal DO sensors to obtain the total abnormal G difference ratio;

[0076] Establish a state classification model based on a convolutional neural network, and identify the dissolved oxygen state in the biochemical reaction pool through the state classification model. Specifically:

[0077] Obtain a sample data set for identifying the dissolved oxygen state in the biochemical reaction pool, and use 70% of the sample data in the sample data set as the training set and 30% of the sample data as the validation set;

[0078] Construct a state classification model, and input the set of state parameters for identifying the dissolved oxygen state in the biochemical reaction pool in the training set into the convolutional neural network to train the classifier;

[0079] Perform classification through the trained classifier to generate a classification result, and perform an accuracy test based on the classification result to calculate the deviation rate between the classification result and the sample data in the validation set;

[0080] Judge whether the deviation rate is less than the preset deviation rate threshold. If it is less, it proves that the accuracy of the classifier meets the preset standard, and output the state classification model;

[0081] Determine whether the dissolved oxygen state in the biochemical reaction tank belongs to the normal state through the state classification model. If it belongs, record the dissolved oxygen state in the biochemical reaction tank as the normal dissolved oxygen state; if not, record the dissolved oxygen state in the biochemical reaction tank as the abnormal dissolved oxygen state.

[0082] Among them, the process of obtaining the set of state parameters for identifying the dissolved oxygen state of the biochemical reaction tank is as follows:

[0083] Within the historical period, based on any time point, count the ratio of the abnormal dissolved oxygen area and the total ratio of abnormal G differences corresponding to each time point.

[0084] Using the ratio of the abnormal dissolved oxygen area and the total ratio of abnormal G differences at each time point as the source data, construct a set of state parameters for identifying the dissolved oxygen state of the biochemical reaction tank at different time points.

[0085] Among them, it should be noted that: the ratio of the abnormal dissolved oxygen area reflects the position distribution of all abnormal dissolved oxygen points in the biochemical reaction tank. Construct a three-dimensional entity for all abnormal dissolved oxygen points. The larger the volume of the abnormal dissolved oxygen area (three-dimensional entity), the larger the area range where the dissolved oxygen in the biochemical reaction tank does not meet the standard, and the worse the treatment effect of the biochemical reaction tank on sewage.

[0086] The total ratio of abnormal G differences reflects the degree of abnormal dissolved oxygen corresponding to the abnormal area in the biochemical reaction tank. The greater the degree of abnormal dissolved oxygen corresponding to this abnormal area, the more serious the abnormal treatment of the biochemical reaction tank on sewage.

[0087] Based on the abnormal dissolved oxygen state, obtain the difference between the value of the abnormal DO sensor corresponding to it, and record the dissolved oxygen abnormal point of the abnormal DO sensor corresponding to the maximum difference as the center point of dissolved oxygen abnormality. Connect the center point of dissolved oxygen abnormality with the remaining dissolved oxygen abnormal points respectively to obtain several dissolved oxygen abnormal connection segments.

[0088] Project each dissolved oxygen abnormal connection segment vertically onto the bottom of the biochemical reaction tank to obtain a dissolved oxygen abnormal projection segment.

[0089] Sort all the dissolved oxygen abnormal projection segments in ascending order of the segment length, and according to the ascending order of the dissolved oxygen abnormal projection segments in terms of segment length, successively use the dissolved oxygen abnormal points as the centers and the effective coverage radius R of the aeration disk to construct multi-level circular screening areas.

[0090] Specifically:

[0091] Specifically:

[0092] The first-level circular screening area:

[0093] Taking the center point of the abnormal dissolved oxygen as the center and the effective coverage radius R of the aeration disk as the radius, construct a circular screening area to obtain the first-layer circular screening area;

[0094] Mark the aeration disks with their center points within the first-layer circular screening area as monitored aeration disks;

[0095] Integrate all the monitored aeration disks within the first-layer circular screening area to obtain the L1-layer aeration disk group;

[0096] The second-layer circular screening area:

[0097] Taking the one with the shortest length of the abnormal dissolved oxygen projection line segment as the reference, obtain the abnormal dissolved oxygen point corresponding to this abnormal dissolved oxygen projection line segment;

[0098] Taking this abnormal dissolved oxygen point as the center and the effective coverage radius R of the aeration disk as the radius, construct a circular screening area to obtain the second-layer circular screening area;

[0099] Mark the aeration disks with their center points within the second-layer circular screening area as monitored aeration disks;

[0100] Integrate all the monitored aeration disks within the second-layer circular screening area to obtain the L2-layer aeration disk group;

[0101] And so on:

[0102] Construct the nth-layer circular screening area:

[0103] Taking the one with the longest length of the abnormal dissolved oxygen projection line segment as the reference, obtain the abnormal dissolved oxygen point corresponding to this abnormal dissolved oxygen projection line segment;

[0104] Taking this abnormal dissolved oxygen point as the center and the effective coverage radius R of the aeration disk as the radius, construct a circular screening area to obtain the nth-layer circular screening area;

[0105] Mark the aeration disks with their center points within the nth-layer circular screening area as monitored aeration disks;

[0106] Integrate all the monitored aeration disks within the nth-layer circular screening area to obtain the Ln-layer aeration disk group;

[0107] Overlap the L1-layer aeration disk group, L2-layer aeration disk group,..., Ln-layer aeration disk group, and record the number of times the monitored aeration disks are overlapped;

[0108] And sort the monitored aeration disks in descending order according to the number of overlaps to obtain the first-level monitoring sequence group of the monitored aeration disks;

[0109] Mark the adjacent monitoring aeration discs for each monitoring aeration disc in the first-level monitoring sequence group, denoted as the monitored adjacent aeration discs, and sum and average the number of times the monitored adjacent aeration discs are overlapped to obtain the monitored value of the monitored adjacent aeration discs;

[0110] Record the number of overlaps of the monitoring aeration disc as Jzc, and record the monitored value of the monitored adjacent aeration disc adjacent to the monitoring aeration disc as Jxc;

[0111] That is, through the formula Calculate the monitoring base number of the monitoring aeration disc , where k is a preset proportionality coefficient;

[0112] Among them, there are m groups of historical data, and each group of historical data includes the number of overlaps Jzc of the monitoring aeration disc, the monitored value Jxc of the monitored adjacent aeration disc adjacent to the monitoring aeration disc, and the monitoring base number of the monitoring aeration disc ;

[0113] Fit the m groups of historical data using a linear model, and substitute the prepared historical data into the selected fitting model for fitting, and obtain the average value of the fitting coefficients as the preset proportionality coefficient k;

[0114] Sort all the monitoring aeration discs in descending order of the monitoring base number of the monitoring aeration disc to obtain the second-level monitoring sequence group of the monitoring aeration discs;

[0115] Complete the anomaly detection of the aeration discs in the biochemical reaction tank according to the order of the second-level monitoring sequence group of the monitoring aeration discs.

[0116] Among them, the process of obtaining the effective coverage radius R of the aeration disc is as follows:

[0117] Based on the oxygen transfer efficiency (OTE):

[0118] Calculate the effective coverage radius through the standard oxygen transfer rate (SOTR):

[0119]

[0120] Among them, is the standard oxygen transfer rate (kgO2 / h);

[0121] is the ratio of the oxygen transfer coefficient of sewage to that of clear water (usually 0.4 - 0.8);

[0122] is the safety factor (1.2 - 1.5);

[0123] is the volumetric oxygen mass transfer coefficient (h -1 );

[0124] is the saturated dissolved oxygen concentration (mg / L);

[0125] Exemplarily: for a certain microporous aeration disk, SOTR = 0.25 kgO2 / h, = 5 h -1 , = 9 mg / L, then the coverage radius R ≈ 1.2 m, corresponding to the coverage area πR 2 ≈ 4.5 m 2 .

[0126] Example 2

[0127] Please refer to Figure 2 as shown. The present invention is an Internet of Things-based sewage treatment whole-process supervision system, including:

[0128] A data acquisition module, used to acquire the actual dissolved oxygen data in the biochemical reaction tank during the sewage purification process;

[0129] A behavior recognition module, used to determine the dissolved oxygen abnormal points according to the acquired actual dissolved oxygen data, and identify the dissolved oxygen state in the biochemical reaction tank according to the positions of the dissolved oxygen abnormal points and the dissolved oxygen abnormal deviation data;

[0130] The dissolved oxygen state in the biochemical reaction tank includes a dissolved oxygen abnormal state and a dissolved oxygen normal state;

[0131] An abnormal construction module, used in the dissolved oxygen abnormal state, with the dissolved oxygen abnormal center point as the reference, successively construct circular screening areas with the dissolved oxygen abnormal points gradually away from the dissolved oxygen abnormal center point as the centers and the effective coverage radius of the aeration disk as the radius;

[0132] A positioning and inspection module, used to perform coincidence processing on the monitored aeration disks in the multi-layer circular screening areas, and construct the abnormal investigation order of the monitored aeration disks according to the coincidence data of the monitored aeration disks.

[0133] One of the core points of the present invention lies in hierarchically dividing the biochemical reaction tank, setting DO sensors on different levels of the biochemical reaction tank, and comprehensively identifying the dissolved oxygen data in the biochemical reaction tank according to the DO sensors, completing the positioning and identification of the dissolved oxygen abnormal area in the biochemical reaction tank, and processing according to the position of the dissolved oxygen abnormal area and the dissolved oxygen abnormal deviation data, realizing the determination of the dissolved oxygen state of the biochemical reaction tank from two dimensions of spatial position and data deviation, with high accuracy;

[0134] One of the core points of the present invention lies in that when the dissolved oxygen state in the biochemical reaction tank is abnormal, the center point of the abnormal dissolved oxygen is determined. Taking the center point of the abnormal dissolved oxygen as the reference, circular screening areas are constructed in turn with the abnormal dissolved oxygen points that are gradually farther away from the center point of the abnormal dissolved oxygen as the centers and the effective coverage radius of the aeration disk as the radius. The monitored aeration disks within the multi-layer circular screening areas are subjected to overlapping processing to obtain the first-level monitored sequence group of the monitored aeration disks. Then, taking the first-level monitored sequence group as the reference, the adjacent monitored aeration disks of each monitored aeration disk within the first-level monitored sequence group are processed to obtain the monitored values of the monitored adjacent aeration disks. Based on the overlapping times of the monitored aeration disks and the monitored values of the monitored adjacent aeration disks, the second-level monitored sequence group of the monitored aeration disks is obtained, and the abnormal investigation sequence of the monitored aeration disks is constructed, so as to facilitate the staff to conduct an orderly investigation of the aeration disks in the biochemical reaction tank and improve the inspection efficiency of the staff.

[0135] The above has described in detail an embodiment of the present invention, but the content described is only the preferred embodiment of the present invention and cannot be considered as defining the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A method for the whole-process supervision of sewage treatment based on the Internet of Things, characterized in that, It includes the following steps: Collect the actual dissolved oxygen data of the biochemical reaction pool during the sewage purification process; Determine the dissolved oxygen abnormal points based on the collected actual dissolved oxygen data, and identify the dissolved oxygen state in the biochemical reaction pool according to the positions of the dissolved oxygen abnormal points and the dissolved oxygen abnormal deviation data; The dissolved oxygen state in the biochemical reaction pool includes the dissolved oxygen abnormal state and the dissolved oxygen normal state; In the dissolved oxygen abnormal state, taking the dissolved oxygen abnormal center point as the benchmark, circular screening areas are constructed successively with the dissolved oxygen abnormal points that are gradually farther away from the dissolved oxygen abnormal center point as the centers and the effective coverage radius of the aeration disk as the radius; Perform superposition processing on the monitored aeration disks in the multi-layer circular screening areas, and construct the abnormal troubleshooting order of the monitored aeration disks based on the superposition data of the monitored aeration disks; The process of obtaining the actual dissolved oxygen data is as follows: DO sensors at different spatial layers are arranged vertically in the biochemical reaction pool; Based on the Grubbs criterion, for the same layer of DO sensors in each biochemical reaction pool at the same moment, the standardized distance G of each single data point deviating from the data mean value is obtained; Obtain the critical value of the standardized distance by which each single data point corresponding to a DO sensor in the same layer deviates from the data mean ; When the DO value of a single DO sensor > , this DO sensor is marked as an abnormal DO sensor; Based on the abnormal state of dissolved oxygen, obtain the value corresponding to it difference. Denote the dissolved oxygen abnormal point of the abnormal DO sensor corresponding to the maximum difference as the dissolved oxygen abnormal center point; Connect the dissolved oxygen abnormal center point with the remaining dissolved oxygen abnormal points respectively to obtain a number of dissolved oxygen abnormal connection segments; Vertically project each dissolved oxygen abnormal connection segment onto the bottom of the biochemical reaction pool to obtain the dissolved oxygen abnormal projection segment; Sort all the dissolved oxygen abnormal projection segments in ascending order of the segment length, and successively construct multi-layer circular screening areas with the dissolved oxygen abnormal points as the centers and the effective coverage radius R of the aeration disk according to the ascending order of the dissolved oxygen abnormal projection segments by segment length; The aeration disks located within any circular screening area are recorded as the monitored aeration disks; Obtain the number of times the monitored aeration disks in the multi-layer circular screening areas are superimposed, and sort the monitored aeration disks in descending order of the number of superimpositions to obtain the first-level monitoring sequence group of the monitored aeration disks; Mark the monitored aeration disks that have adjacent monitored aeration disks in the first-level monitoring sequence group, denoted as the monitored adjacent aeration disks, and sum and average the number of times the monitored adjacent aeration disks are superimposed to obtain the monitored values of the monitored adjacent aeration disks; Process the number of times the monitored aeration disks are superimposed and the monitored values of the monitored adjacent aeration disks adjacent to the monitored aeration disks to obtain the monitoring base number of the monitored aeration disks; Complete the abnormal troubleshooting of the aeration disks in the biochemical reaction pool according to the order of the second-level monitoring sequence group of the monitored aeration disks.

2. The whole-process supervision method for sewage treatment based on the Internet of Things according to claim 1, characterized in that The dissolved oxygen abnormal points are the position points where the abnormal DO sensors are located in the biochemical reaction pool.

3. The method for the whole-process supervision of sewage treatment based on the Internet of Things according to claim 2, characterized in that, The process of identifying the dissolved oxygen state in the biochemical reaction pool is as follows: Establish a state classification model based on the convolutional neural network; Input the set of state parameters for identifying the dissolved oxygen state in the biochemical reaction pool in the training set into the convolutional neural network to train the classifier; Perform classification through the trained classifier to generate a classification result, conduct accuracy verification according to the classification result, and calculate the deviation rate between the classification result and the sample data in the validation set; Judge whether the deviation rate is less than the preset deviation rate threshold. If it is less, it proves that the accuracy of the classifier meets the preset standard, and output the state classification model; Judge whether the dissolved oxygen state in the biochemical reaction pool belongs to the normal state through the state classification model. If it belongs, record the dissolved oxygen state in the biochemical reaction pool as the normal dissolved oxygen state; if it does not belong, record the dissolved oxygen state in the biochemical reaction pool as the abnormal dissolved oxygen state.

4. The whole-process supervision method for sewage treatment based on the Internet of Things according to claim 3, characterized in that, The acquisition process of the set of state parameters for identifying the dissolved oxygen state of the biochemical reaction pool is as follows: During the historical period, based on any time point, count the ratio of the abnormal dissolved oxygen area and the total ratio of abnormal G differences corresponding to each time point; Using the ratio of the abnormal dissolved oxygen area and the total ratio of abnormal G differences at each time point as source data, construct a set of state parameters for identifying the dissolved oxygen state of the biochemical reaction pool at different time points; The ratio of the abnormal dissolved oxygen area is the ratio of the volume of the abnormal dissolved oxygen area to the volume of the biochemical reaction pool; The value corresponding to the abnormal DO sensor is subtracted to calculate the abnormal G difference. The abnormal G difference of the abnormal DO sensor is used for ratio calculation to obtain the abnormal G difference ratio of the abnormal DO sensor; Sum up the abnormal G difference ratios of all abnormal DO sensors and take the average value to obtain the total ratio of abnormal G differences.

5. An Internet of Things-based sewage treatment whole-process supervision system, characterized in that It includes: A data acquisition module for collecting the actual dissolved oxygen data in the biochemical reaction pool during the sewage purification process; A behavior recognition module for determining the dissolved oxygen abnormal points according to the collected actual dissolved oxygen data, and identifying the dissolved oxygen state in the biochemical reaction pool based on the positions of the dissolved oxygen abnormal points and the dissolved oxygen abnormal deviation data; The dissolved oxygen state in the biochemical reaction pool includes the abnormal dissolved oxygen state and the normal dissolved oxygen state; An abnormal construction module for, in the case of an abnormal dissolved oxygen state, taking the center point of the abnormal dissolved oxygen as a reference, and successively constructing circular screening areas with the dissolved oxygen abnormal points gradually far away from the center point of the abnormal dissolved oxygen as the centers and the effective coverage radius of the aeration disk as the radii; A positioning and inspection module for overlapping the monitored aeration disks in the multi-layer circular screening areas, and constructing an abnormal inspection order of the monitored aeration disks based on the overlapping data of the monitored aeration disks; The acquisition process of the actual dissolved oxygen data is as follows: Set DO sensors at different spatial layers in the vertical direction in the biochemical reaction pool; Based on the Grubbs criterion, process the DO sensors at the same layer in each biochemical reaction pool at the same moment to obtain the standardized distance G of each single data point deviating from the data mean value in each layer; Obtain the critical value of the standardized distance by which each single data point corresponding to a DO sensor in the same layer deviates from the data mean ; When the DO value of a single DO sensor > it is recorded as an abnormal DO sensor; Based on the abnormal state of dissolved oxygen, obtain the value corresponding to it difference, and record the dissolved oxygen abnormal point of the abnormal DO sensor corresponding to the maximum difference as the dissolved oxygen abnormal center point; Connect the center point of the abnormal dissolved oxygen to the remaining abnormal dissolved oxygen points respectively to obtain several abnormal dissolved oxygen line segments; Vertically project each abnormal dissolved oxygen line segment onto the bottom of the biochemical reaction pool to obtain the abnormal dissolved oxygen projection line segments; Sort all the abnormal dissolved oxygen projection line segments in ascending order of the line segment length, and successively construct multi-layer circular screening areas with the abnormal dissolved oxygen points as the centers and the effective coverage radius R of the aeration disk according to the ascending order of the line segment length of the abnormal dissolved oxygen projection line segments; Mark the aeration disks located within any circular screening area as the monitored aeration disks; Obtain the number of times the monitored aeration disks in the multi-layer circular screening areas are overlapped, and sort the monitored aeration disks in descending order of the number of overlaps to obtain the first-level monitoring sequence group of the monitored aeration disks; Mark the monitored aeration disks that have adjacent monitored aeration disks in the first-level monitoring sequence group, record them as the monitored adjacent aeration disks, and sum up and take the average value of the number of times the monitored adjacent aeration disks are overlapped to obtain the monitored value of the monitored adjacent aeration disks; The coincidence times of the monitoring aeration disk are processed with the monitored values of the monitoring adjacent aeration disk adjacent to the monitoring aeration disk to obtain the monitoring base number of the monitoring aeration disk. Abnormalities of the aeration disks in the biochemical reaction tank are checked according to the order of the secondary monitoring sequence groups of the monitoring aeration disks.

Images