Sewage treatment system based on crowd sensing

Abstract

The invention discloses a sewage treatment system based on crowd sensing, and relates to the technical field of pollution prediction.A pollution prediction module is based on a two-dimensional instantaneous pollution source diffusion model and combines sensor point location information provided by a sensor screening module and water depth influence data provided by a water depth measurement and calculation module; the dynamic simulation and prediction module is used for carrying out dynamic simulation and prediction on pollutant diffusion, calculating a water pollutant concentration prediction value of each selected sensor point location, and sending the pollutant concentration prediction value to the feedback determination module; correcting the position of the pollution source according to the deviation by using an inversion algorithm, specifically, continuously correcting the predicted pollution data value again by adjusting the position data of the pollution source and changing the values of x, y and b, and comparing the predicted pollution data value with an actual measurement value until an error requirement is met; therefore, the prediction precision of the pollution condition of the position, which is not covered by the sensor, of the monitoring area can be ensured by continuously calibrating the prediction value and the actual measurement value.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and more specifically, to a sewage treatment system based on crowd intelligence perception. Background Art

[0002] With the rapid development of industrialization and urbanization, water pollution is becoming increasingly serious. Pollutants in water bodies not only affect aquatic ecosystems, but may also threaten human health. Therefore, how to accurately and quickly detect and locate water pollution sources has become an important technical issue in the field of environmental protection. Traditional water pollution detection methods mainly rely on fixed-point sampling and laboratory analysis, but this method is time-consuming, costly, and difficult to reflect the dynamic changes of water pollution conditions in real time. At the same time, in recent years, the rapid development of sensor technology has provided new means for water pollution detection. By arranging multiple sensors in the water body, water quality parameters such as dissolved oxygen, pH value, turbidity, and heavy metal ion concentration can be monitored in real time. This distributed sensor network can capture the water pollution status more comprehensively and realize real-time monitoring of pollution sources.

[0003] The prior art has the following deficiencies:

[0004] Most of the existing pollution source detection and analysis are static. Fixed-point monitoring cannot provide timely warnings when facing changes in pollution sources, and it is difficult to cope with dynamic changes in complex water flow environments, which easily leads to inaccurate prediction results of the pollution status of the entire water area. When dealing with the impact of distant shore sources, existing technologies are often lacking in flexibility and fail to fully consider changes in environmental factors such as water depth and distance between the two banks, resulting in low pollution prediction accuracy for locations in the monitoring area that are not covered by sensors.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a sewage treatment system based on crowd intelligence perception to solve the problems raised in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A sewage treatment system based on crowd intelligence perception includes a sensor screening module, a water depth measurement module, a pollution prediction module and a feedback determination module, and the modules are signal-connected;

[0009] The sensor screening module is used to obtain real-time pollutant concentration data and water flow velocity data from multiple sensors distributed in the water flow area, and screen out the most representative sensors as pollution measurement points by analyzing the data collected by the sensors;

[0010] The water depth calculation module is used to obtain real-time water depth data of the water flow area and perform data preprocessing on the collected water depth data;

[0011] The pollution prediction module is based on a two-dimensional instantaneous pollution source diffusion model, combined with the sensor point information provided by the sensor screening module and the water depth impact data provided by the water depth measurement module, to perform dynamic simulation and prediction of pollutant diffusion, calculate the predicted value of water pollutant concentration at each selected sensor point, and send the predicted value of pollutant concentration to the feedback determination module;

[0012] The feedback determination module is used to compare the pollutant concentration prediction value provided by the pollution prediction module with the actual measurement value of each sensor point and calculate the deviation. If the deviation exceeds the set threshold range, the feedback determination module will determine that there is an abnormality in the pollution source and feedback the abnormal signal to the system, while notifying relevant monitoring personnel or triggering the early warning mechanism; when an abnormal situation is detected, the feedback determination module will feedback the abnormal information to the pollution prediction module, use the inversion algorithm to correct the location of the pollution source according to the deviation, and adjust the model parameters to be closer to the actual situation.

[0013] In a preferred embodiment, the sensor screening module screens out the most representative sensor as the pollution measurement point. The specific process is as follows:

[0014] Determine the false pollution data sensor points in the monitoring area and obtain the expected compliance status of each true pollution data sensor point in the monitoring area;

[0015] Calculate the relative position status of each true pollution data sensor point in the monitoring area;

[0016] The weighted average data fusion algorithm is used to determine the recommended index of each true pollution data sensor point in the monitoring area;

[0017] By comparing the recommended index of each true pollution data sensor point in the monitoring area with the first recommended critical value, representative sensor points are screened out and used as subsequent real-time measurement points.

[0018] In a preferred embodiment, the method for determining the false pollution data sensor points in the monitoring area is as follows:

[0019] The pollution threshold range is set according to the actual pollution situation in the monitoring area. If the measurement data of the sensor point in the monitoring area exceeds the pollution threshold range, the measurement data of the sensor point is identified as a false pollution data sensor point, otherwise it is a true pollution data sensor point;

[0020] The expected pollution data value of the measuring sensor point is obtained by curve fitting of the pollution data values ​​measured at several sensor points before and after the measuring sensor point; the pollution deviation value between the expected pollution data value and the actual pollution data value is calculated; the average jump value is confirmed according to the historical data of the measuring sensor point, and the pollution deviation value is compared with the average jump value. If the pollution deviation value is greater than the average jump value, the measurement data of the sensor point is determined to be a false pollution data sensor point, otherwise it is a true pollution data sensor point.

[0021] In a preferred embodiment, the method for determining the expected compliance state of each true pollution data sensor point in the monitoring area is as follows:

[0022] For the real pollution data sensor point, the difference between the pollution deviation value and the average jump value is taken as the expected compliance value, and the expected compliance state is represented by the expected compliance value;

[0023] The specific method for calculating the relative position state of each true pollution data sensor point in the monitoring area is as follows:

[0024] Obtain the location coordinates of each true pollution data sensor point;

[0025] Calculate the Euclidean distance of each true pollution data sensor point to other true pollution data sensor points;

[0026] According to all the calculated Euclidean distances, the average Euclidean distance corresponding to each true pollution data sensor point is calculated, and it is used as the corresponding relative position value, and the relative position state of the true pollution data sensor point is mathematically represented by the relative position value;

[0027] The expected compliance value and relative position value are taken as the influencing factor features, and the weighted sum is taken to obtain the recommendation index of each true pollution data sensor point.

[0028] In a preferred embodiment, the water depth calculation module can use sonar echo sounding to measure the water depth of the monitoring area. The specific process is as follows:

[0029] Use sonar equipment to measure water depth and record the echo time and sound speed measured by the sonar equipment;

[0030] The sound velocity is calibrated based on temperature, salinity and depth to obtain the sound velocity calibration coefficient k. The specific calculation expression is:

[0031] ;

[0032] Where T is water temperature, S is salinity, and z is measurement depth. is the actual depth;

[0033] K= / c;

[0034] The product of the measured water depth and the calibration coefficient k is taken as the calibrated water depth.

[0035] In a preferred embodiment, the pollution prediction module solves the pollution diffusion status by using the image source method, and the specific method is as follows:

[0036] Step S1, solving the linear equations of boundary conditions and continuous point source problems by image source method and integral variation, and first deriving a two-dimensional instantaneous point source diffusion model with flow;

[0037] The two-dimensional instantaneous point source diffusion model is converted into an ordinary differential equation through Laplace transformation and two-dimensional Fourier transformation, and the obtained ordinary differential equation solution is then subjected to inverse Fourier transformation and inverse Laplace transformation to obtain the two-dimensional instantaneous point source diffusion model;

[0038] Step S2, setting two image sources at distances corresponding to the boundary, namely a nearshore image source and a farshore image source;

[0039] Step S3, determining the distance attenuation coefficient according to the location of the pollution source, the distance to the nearshore boundary and the distance between the two banks, the specific calculation expression is 2b / l, and then determining the comprehensive attenuation coefficient according to the weighted normalized numerical calculation of the distance attenuation coefficient and the actual water depth;

[0040] Step S4, solving a two-dimensional instantaneous pollution source concentration model under boundary conditions according to the comprehensive attenuation coefficient, and predicting the pollution data value at the representative sensor point according to the two-dimensional instantaneous pollution source concentration model under boundary conditions.

[0041] In a preferred embodiment, the feedback determination module uses an inversion algorithm to correct the pollution source position according to the deviation, specifically by adjusting the pollution source position data, changing the values ​​of x, y, and b, and continuously re-correcting the predicted pollution data values ​​at representative sensor points and comparing them with the actual measured values ​​until the error requirements are met.

[0042] The technical effects and advantages of the sewage treatment system based on crowd intelligence perception of the present invention are as follows:

[0043] The present invention can accurately determine the location of pollution sources by reasonably arranging sensors and comprehensively considering various environmental factors, especially in complex waters. This method can effectively avoid errors caused by unreasonable sensor layout. And by comprehensively considering the water depth and the distance between the two banks, the model can better reflect the actual situation, thereby reducing errors in pollutant diffusion prediction, especially in areas with shallow water depth or narrow water areas. The impact of far-shore image sources may be significant. Considering this factor helps to more accurately predict pollutant diffusion. Through dynamic adjustment of the model and real-time data analysis, the system can automatically judge environmental conditions and adjust the parameters of the prediction model according to the conditions. This intelligent processing method reduces dependence on human intervention and improves the autonomy and response speed of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a system block diagram of the sewage treatment system based on crowd intelligence perception of the present invention.

[0045] Figure 2 This is a schematic diagram of the monitoring area scenario of the sewage treatment system based on crowd intelligence perception in the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] In the water flow area with pollution sources and multiple sensors arranged therein, the present invention first selects reasonable sensors as water pollution measurement points at intervals according to the water flow direction, and then determines whether the pollution source is abnormal according to the deviation between the water pollution prediction value and the actual measurement value of each selected sensor point. In addition, when calculating the water pollution prediction value of each selected sensor point, it is comprehensively determined whether to ignore the influence of the far-shore image source according to the water depth of the water flow area and the distance between the two banks.

[0048] The present invention can accurately determine the location of pollution sources by reasonably arranging sensors and comprehensively considering various environmental factors, especially in complex waters. This method can effectively avoid errors caused by unreasonable sensor layout. And by comprehensively considering the water depth and the distance between the two banks, the model can better reflect the actual situation, thereby reducing errors in pollutant diffusion prediction, especially in areas with shallow water depth or narrow water areas. The influence of far-shore image sources may be significant. Considering this factor helps to more accurately predict pollutant diffusion. Through dynamic adjustment of the model and real-time data analysis, the system can automatically judge environmental conditions and adjust the parameters of the prediction model according to the conditions. This intelligent processing method reduces the necessity of human intervention and improves the autonomy and response speed of the system.

[0049] Examples, such as Figure 1 As shown, the present invention discloses a sewage treatment system based on crowd intelligence perception, including: a sensor screening module, a water depth measurement module, a pollution prediction module and a feedback determination module, and the modules are signal connected.

[0050] The functions of each module are as follows:

[0051] The sensor screening module is used to obtain real-time pollutant concentration data, water flow velocity data and other relevant environmental data from multiple sensors distributed in the water flow area. By analyzing the data collected by the sensors, the most representative sensors are selected as pollution measurement points. The screening process comprehensively considers the data stability and reliability of each sensor point to ensure that the selected points can accurately reflect the actual diffusion of pollutants. Finally, the selected sensor point information will be passed to the pollution prediction module and the feedback determination module for subsequent pollution prediction calculation and deviation analysis.

[0052] The water depth measurement module is used to obtain real-time water depth data of the water flow area through water depth sensors or other hydrological measurement equipment, and perform data preprocessing on the collected water depth data, including data cleaning, outlier detection and data smoothing, to ensure the accuracy and stability of the water depth data.

[0053] The pollution prediction module is based on a two-dimensional instantaneous pollution source diffusion model, combined with the sensor point information provided by the sensor screening module and the water depth impact data provided by the water depth measurement module, to perform dynamic simulation and prediction of pollutant diffusion, thereby calculating the predicted value of water pollutant concentration at each selected sensor point, and sending the predicted value of pollutant concentration to the feedback determination module for subsequent deviation analysis and anomaly detection.

[0054] The feedback determination module is used to compare the pollutant concentration prediction value provided by the pollution prediction module with the actual measurement value of each sensor point and calculate the deviation. The size of the deviation can indicate the abnormality of the pollution source. If the deviation exceeds the set threshold range, the feedback determination module will determine that there is an abnormality in the pollution source and feedback the abnormal signal to the system, and notify the relevant monitoring personnel or trigger the early warning mechanism. At the same time, when an abnormality is detected, the feedback determination module can feedback the abnormal information to the pollution prediction module, use the inversion algorithm, correct the location of the pollution source according to the deviation, and adjust the model parameters to be closer to the actual situation.

[0055] Specifically, the sensor screening module selects the most representative sensors as pollution measurement points. The specific process is as follows:

[0056] Determine the false pollution data sensor points in the monitoring area, and obtain the expected compliance status of each true pollution data sensor point in the monitoring area;

[0057] Calculate the relative position status of each true pollution data sensor point in the monitoring area;

[0058] The weighted average data fusion algorithm is used to determine the recommended index of each true pollution data sensor point in the monitoring area;

[0059] By comparing the recommended index of each true pollution data sensor point in the monitoring area with the first recommended critical value, representative sensor points are screened out and used as subsequent real-time measurement points.

[0060] Furthermore, the method for determining the false pollution data sensor points in the monitoring area is as follows:

[0061] The pollution threshold range is set according to the actual pollution situation in the monitoring area. Specifically, the pollution threshold range is determined by comparing and analyzing the historical pollution data of the monitoring area and setting the maximum and minimum water pollution measurement values ​​in the monitoring area. If the measurement data of a sensor point in the monitoring area exceeds the pollution threshold range, the measurement data of the sensor point is identified as a false pollution data sensor point, otherwise it is a true pollution data sensor point.

[0062] Next, by performing curve fitting on the pollution data values ​​measured at several sensor points before and after the measurement sensor point, the expected pollution data value of the measurement sensor point is obtained; then the pollution deviation value between the expected pollution data value and the actual pollution data value is calculated; then the average jump value is confirmed based on the historical data of the measurement sensor point, and the pollution deviation value is compared with the average jump value. If the pollution deviation value is greater than the average jump value, the measurement data of the sensor point is determined to be a false pollution data sensor point, otherwise it is a true pollution data sensor point.

[0063] It should be noted that several sensor points are selected based on actual conditions. For example, based on the water flow direction of the sensor point, the other two sensors closest to the sensor point are confirmed to perform the desired calculation, or based on the four directions or six directions around the sensor point, etc., which will not be elaborated here.

[0064] At the same time, the expected compliance state determination method for each true pollution data sensor point in the monitoring area is as follows:

[0065] For the true pollution data sensor point, the difference between the pollution deviation value and the average jump value is taken as the expected compliance value. The smaller the expected compliance value, the more accurate the pollution data measured by the sensor point is, and the more important the pollution data measured by the sensor point is. The expected compliance state is represented by the expected compliance value.

[0066] This application can eliminate false pollution data generated during the measurement process by judging true and false pollution data at each sensor point in the monitoring area. False pollution data may be caused by errors in the measuring equipment or operating errors, and can be divided into three categories: values ​​outside the normal range, large changes relative to surrounding measurement points, and insignificant changes but outside the normal range. By comparing the changing trends of the data before and after the measurement point, a curve fitting method is used to judge and eliminate false pollution data. The consistency and accuracy of the data are guaranteed.

[0067] Furthermore, the specific method for calculating the relative position state of each true pollution data sensor point in the monitoring area is as follows:

[0068] Obtain the location coordinates of each true pollution data sensor point;

[0069] Calculate the Euclidean distance of each true pollution data sensor point to other true pollution data sensor points;

[0070] According to all calculated Euclidean distances, the average Euclidean distance corresponding to each true pollution data sensor point is calculated and used as the corresponding relative position value. The relative position state of the true pollution data sensor point is mathematically represented by the relative position value.

[0071] It should be noted that when calculating the relative position value of each true pollution data sensor point, it is not to calculate the Euclidean distance of each true pollution data sensor point to all other true pollution data sensor points, but to select other true pollution data sensor points that are closest to the true pollution data sensor point of the research object. That is, the Euclidean distance between the true pollution data sensor point and the closest other true pollution data sensor points in different directions is calculated. For example, in the study of river pollution, the study is generally conducted in the direction perpendicular to the water flow and in the direction parallel to the water flow, such as Figure 2 As shown, with the water flow direction as the x-axis direction and the direction perpendicular to the water flow as the y-axis direction, the four other true pollution data sensor points closest to the true pollution data sensor point as the research object in the positive and negative semi-axis directions of the x-axis and in the positive and negative semi-axis directions of the y-axis are selected to calculate the corresponding Euclidean distances.

[0072] Since the present application first screens the sensor points for true and false pollution data, after the false pollution data sensor points are screened out, the layout of the true pollution data sensor points in the sensor network will no longer be uniform. At this time, the data of the more isolated sensor points is not supplemented by the measurement data of other surrounding sensors, so the measurement data of the true pollution data sensor point is more important. That is, the larger the relative position value of the calculated true pollution data sensor point is, the more isolated the true pollution data sensor point is, and the more important its measurement data is.

[0073] The weighted average data fusion algorithm is used to determine the recommended index of each true pollution data sensor point in the monitoring area. The specific method is as follows:

[0074] The expected compliance value and the relative position value are used as the influencing factor features, and the weighted sum is used to obtain the recommendation index of each true pollution data sensor point, where the expected compliance value has a negative weight and the relative position value has a positive weight, thereby accommodating the importance of the two to the measurement data.

[0075] The absolute value of the fusion weight of the weighted average data fusion algorithm is directly determined by the measurement variance of the sensor. The common traditional variance estimation algorithm regards the average value of the observed data of each sensor as the reference point of the real data information of the measurement target, and regards the difference between the measured value of each sensor and the reference point as the measurement deviation of the sensor to the real data. Then, according to the statistical principle, the variance of the sensor's observation deviation at each measurement point is calculated as the estimated value of the sensor variance. The typical weighted dynamic allocation algorithm is not described here.

[0076] The water depth calculation module can use sonar echo sounding to measure the water depth of the monitoring area. The specific process is as follows:

[0077] Use sonar equipment to measure water depth and record the echo time and sound speed measured by the sonar equipment;

[0078] The sound velocity is calibrated based on temperature, salinity and depth to obtain the sound velocity calibration coefficient k. The specific calculation expression is:

[0079] ;

[0080] Where T is water temperature, S is salinity, and z is measurement depth. is the actual depth.

[0081] K= / c.

[0082] The measured water depth is calibrated using the calibration coefficient k to obtain the calibrated water depth; specifically, the product of the measured water depth and the calibration coefficient k is used as the calibrated water depth.

[0083] It should be noted that in the actual depth calculation formula:

[0084] 1449.2 m / s is the base value of the sound speed measured under standard ocean conditions (usually 0°C and 35 ppt salinity). 4.6 is the coefficient of the first-order temperature term, which means that the sound speed increases by about 4.6 m / s for every 1°C increase in temperature. -0.055 and 0.00029 are the coefficients of the higher-order temperature terms, which correct the linear effect of the first-order temperature term on the sound speed, making the change in the sound speed more in line with the actual situation.

[0085] 1.34 represents the effect of salinity on the speed of sound under standard ocean conditions (temperature 0°C, salinity 35 ppt). This coefficient means that for every 1 ppt increase in salinity, the speed of sound increases by about 1.34 m / s. -0.01T: indicates the degree to which the effect of salinity on the speed of sound decreases when the temperature rises.

[0086] 0.016 m / s / m means that for every 1 meter of depth increase, the speed of sound increases by approximately 0.016 m / s. This linear relationship is based on an empirical formula and is applicable to the common ocean depth range.

[0087] This application is based on the existence of boundaries on both sides of the waters in the monitoring area, and the distance between the two boundaries is l. The boundary closer to the pollution source is called the nearshore boundary, and the boundary farther away is called the farshore boundary. The pollution source is located in the waters, with a distance of b from the nearshore boundary and lb from the farshore boundary. The arrow direction x is the flow direction of the water, and the direction perpendicular to the arrow direction is y. The pollution prediction module solves the pollution diffusion status through the image source method, and the specific method is as follows:

[0088] Step S1, using the image source method and integral variation to solve the linear equations of the boundary conditions and the continuous point source problem, first derive a two-dimensional instantaneous point source flow diffusion model; specifically as follows:

[0089] ;

[0090] In the formula, It is the comprehensive degradation item of pollutants. In uniform waters, the water depth is shallow and the change is not obvious. and are the water flow velocities in the x and y directions, and are the comprehensive diffusion coefficients in the x and y directions, respectively.

[0091] When diffusion begins, at time t=0, if the mass of the pollution source is M, when x and y tend to infinity, it can be concluded that the boundary pollutant concentration is 0. The two-dimensional instantaneous point source diffusion model is as follows:

[0092] ;

[0093] The two-dimensional instantaneous point source diffusion model is converted into an ordinary differential equation through Laplace transform and two-dimensional Fourier transform. The solution of the ordinary differential equation is then subjected to inverse Fourier transform and inverse Laplace transform to obtain the two-dimensional instantaneous point source diffusion model as follows:

[0094] ;

[0095] Step S2, two image sources are set at the distance corresponding to the boundary, namely the nearshore image source and the farshore image source; the nearshore image source is closer to the boundary, and its pollution source diffusion effect is stronger, with the same diffusion concentration as the real source. The farshore image source is farther away from the real source and the boundary, so its diffusion effect is attenuated with the distance between the two and the actual water depth;

[0096] Step S3, determine the distance attenuation coefficient according to the location of the pollution source, the distance from the nearshore boundary and the distance between the two banks. The specific calculation expression is 2b / l, and then determine the comprehensive attenuation coefficient Q based on the numerical calculation after weighted normalization of the distance attenuation coefficient and the actual water depth. Obviously, when the distance between the two banks is infinite, the effect of the far shore diffusion also decreases with the increase of the distance. At this time, the comprehensive attenuation coefficient Q decreases, and the influence of the far shore image source decreases. The deeper the water depth, the larger the volume of the water body in the deep water area, and the pollutants will be diluted more fully during the diffusion process. Therefore, when the pollutants diffuse from the far shore to the deep water area, their concentration will be significantly reduced, and the impact will also be weakened, that is, the comprehensive attenuation coefficient Q will decrease. The weights corresponding to the distance attenuation coefficient and the actual water depth are set according to the actual situation, and this can also be 1, which will not be repeated here.

[0097] Step S4, according to the comprehensive attenuation coefficient Q, solve the two-dimensional instantaneous pollution source concentration model under boundary conditions, as follows:

[0098] ;

[0099] And the pollution data values ​​at representative sensor points are predicted based on the two-dimensional instantaneous pollution source concentration model under boundary conditions.

[0100] The feedback determination module uses an inversion algorithm to correct the location of the pollution source according to the deviation. Specifically, it adjusts the pollution source location data, changes the values ​​of x, y, and b, continuously re-corrects the predicted pollution data value, and compares it with the actual measured value until the error requirements are met.

[0101] The present application can ensure the accuracy of pollution prediction for locations in the monitoring area that are not covered by sensors by continuously calibrating the predicted values ​​and actual measured values.

[0102] The above weighted summation formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain a formula for the most recent actual situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0103] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0104] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0105] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0106] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0107] Finally: 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 in the protection scope of the present invention.

Claims

1. A sewage treatment system based on crowd intelligence perception, characterized by: It includes a sensor screening module, a water depth measurement module, a pollution prediction module and a feedback determination module, and signal connections between the modules; The sensor screening module is used to obtain real-time pollutant concentration data and water flow velocity data from multiple sensors distributed in the water flow area, and screen out the most representative sensors as pollution measurement points by analyzing the data collected by the sensors; The water depth measurement module is used to obtain real-time water depth data of the water flow area and perform data preprocessing on the collected water depth data; The pollution prediction module is based on a two-dimensional instantaneous pollution source diffusion model, combined with the sensor point information provided by the sensor screening module and the water depth impact data provided by the water depth measurement module, to perform dynamic simulation and prediction of pollutant diffusion, calculate the predicted value of water pollutant concentration at each selected sensor point, and send the predicted value of pollutant concentration to the feedback determination module; The feedback determination module is used to compare the pollutant concentration prediction value provided by the pollution prediction module with the actual measurement value of each sensor point and calculate the deviation. If the deviation exceeds the set threshold range, the feedback determination module will determine that there is an abnormality in the pollution source and feedback the abnormal signal to the system, while notifying the monitoring personnel or triggering the early warning mechanism; when an abnormal situation is detected, the feedback determination module will feedback the abnormal information to the pollution prediction module, use the inversion algorithm to correct the location of the pollution source according to the deviation, and adjust the model parameters to be closer to the actual situation.

2. The sewage treatment system based on crowd intelligence perception according to claim 1 is characterized in that: The specific process of the sensor screening module screening out the most representative sensor as the pollution measurement point is as follows: Determine the false pollution data sensor points in the monitoring area, and obtain the expected compliance status of each true pollution data sensor point in the monitoring area; Calculate the relative position status of each true pollution data sensor point in the monitoring area; The weighted average data fusion algorithm is used to determine the recommended index of each true pollution data sensor point in the monitoring area; By comparing the recommended index of each true pollution data sensor point in the monitoring area with the first recommended critical value, representative sensor points are screened out and used as subsequent real-time measurement points.

3. The sewage treatment system based on crowd intelligence perception according to claim 2 is characterized in that: The method for determining the false pollution data sensor points in the monitoring area is as follows: The pollution threshold range is set according to the actual pollution situation in the monitoring area. If the measurement data of the sensor point in the monitoring area exceeds the pollution threshold range, the measurement data of the sensor point is identified as a false pollution data sensor point, otherwise it is a true pollution data sensor point; The expected pollution data value of the measuring sensor point is obtained by curve fitting of the pollution data values ​​measured at several sensor points before and after the measuring sensor point; the pollution deviation value between the expected pollution data value and the actual pollution data value is calculated; the average jump value is confirmed according to the historical data of the measuring sensor point, and the pollution deviation value is compared with the average jump value. If the pollution deviation value is greater than the average jump value, the measurement data of the sensor point is determined to be a false pollution data sensor point, otherwise it is a true pollution data sensor point.

4. The sewage treatment system based on crowd intelligence perception according to claim 2 is characterized in that ； The method for determining the expected compliance status of each true pollution data sensor point in the monitoring area is as follows: For the real pollution data sensor point, the difference between the pollution deviation value and the average jump value is taken as the expected compliance value, and the expected compliance state is represented by the expected compliance value; The specific method for calculating the relative position state of each true pollution data sensor point in the monitoring area is as follows: Obtain the location coordinates of each true pollution data sensor point; Calculate the Euclidean distance of each true pollution data sensor point to other true pollution data sensor points; According to all the calculated Euclidean distances, the average Euclidean distance corresponding to each true pollution data sensor point is calculated, and it is used as the corresponding relative position value, and the relative position state of the true pollution data sensor point is mathematically represented by the relative position value; The expected compliance value and relative position value are taken as the influencing factor features, and the weighted sum is taken to obtain the recommendation index of each true pollution data sensor point.

5. The sewage treatment system based on crowd intelligence perception according to claim 1 is characterized in that: The water depth calculation module can use sonar echo sounding to measure the water depth of the monitoring area. The specific process is as follows: Use sonar equipment to measure water depth and record the echo time and sound speed measured by the sonar equipment; The sound velocity is calibrated based on temperature, salinity and depth to obtain the sound velocity calibration coefficient k. The specific calculation expression is: ； Where T is water temperature, S is salinity, and z is measurement depth. is the actual depth; K= / c; The product of the measured water depth and the calibration coefficient k is taken as the calibrated water depth.

6. The sewage treatment system based on crowd intelligence perception according to claim 1 is characterized in that: The pollution prediction module solves the pollution diffusion status through the image source method. The specific method is as follows: Step S1, solving the linear equations of boundary conditions and continuous point source problems by image source method and integral variation, and first deriving a two-dimensional instantaneous point source diffusion model with flow; The two-dimensional instantaneous point source diffusion model is converted into an ordinary differential equation through Laplace transformation and two-dimensional Fourier transformation, and the obtained ordinary differential equation solution is then subjected to inverse Fourier transformation and inverse Laplace transformation to obtain the two-dimensional instantaneous point source diffusion model; Step S2, setting two image sources at distances corresponding to the boundary, namely a nearshore image source and a farshore image source; Step S3, determining the distance attenuation coefficient according to the location of the pollution source, the distance to the nearshore boundary and the distance between the two banks, the specific calculation expression is 2b / l, and then determining the comprehensive attenuation coefficient according to the weighted normalized numerical calculation of the distance attenuation coefficient and the actual water depth; Step S4, solving a two-dimensional instantaneous pollution source concentration model under boundary conditions according to the comprehensive attenuation coefficient, and predicting the pollution data value at the representative sensor point according to the two-dimensional instantaneous pollution source concentration model under boundary conditions.

7. The sewage treatment system based on crowd intelligence perception according to claim 6 is characterized in that: The feedback determination module uses an inversion algorithm to correct the pollution source position according to the deviation, and by adjusting the pollution source position data, changing the values ​​of x, y, and b, continuously re-corrects the predicted pollution data values ​​at representative sensor points and compares them with the actual measured values ​​until the error requirements are met.

Images