A highly efficient method and system for tracing the source of water pollution

Abstract

This invention discloses an efficient method and system for tracing water pollution sources, belonging to the field of water pollution source tracing technology. The initial setup of this invention uses the water flow velocity of the main system to correct the local concentration gradient vector to determine the direction of the pollution source. Subsystems employ a pilot iteration for dynamic source tracing, selecting the subsystem with the highest concentration as the pilot subsystem. Other subsystems follow the pilot subsystem, evenly distributed in a fan shape with the pilot subsystem as the center. The local concentration gradient vector of the region where the subsystem at the end of the iteration is located is calculated. The local concentration gradient vector is corrected based on the water flow velocity of the pilot subsystem to obtain the pollution source direction again. A fine search is then performed around subsystems that meet the local fine search conditions to determine the location of the pollution source. This invention enables rapid initial orientation, has high search efficiency, strong adaptability, and strong anti-interference capability.

Description

Technical Field

[0001] This invention relates to the field of water pollution source tracing technology, and in particular to a highly efficient water pollution source tracing method and system. Background Technology

[0002] Patent application CN202510249772.7 discloses a water pollution source tracing system and method, including a main system and a subsystem. The main system is fixedly deployed on a river. In non-source tracing situations, the subsystem docks with the main system and has the ability to move in the river. Both the subsystem and the main system have the function of monitoring water quality parameters and pollutants at their respective locations. The main system is used for: communicating with a cloud platform that has official pollution monitoring data; collecting and analyzing monitoring data from the main system and subsystems as well as official pollution monitoring data; issuing warnings based on set warning conditions; planning source tracing paths; launching and recovering the subsystems; and controlling the subsystems to trace the source according to the source tracing paths.

[0003] The above-mentioned technologies mainly have the following problems: 1. Low source tracing efficiency and slow response: A single subsystem needs to locate the pollution source by using circular inspection and path fitting. However, in environments with complex pollutant diffusion and drastic water flow changes, this method requires frequent circular inspections, resulting in low efficiency and possible misjudgment; 2. Limited search range: The inspection coverage area of ​​a single subsystem is small, and multiple round trips are required for large water areas; 3. Weak anti-interference ability: Single-point data is easily affected by instantaneous fluctuations, gradient estimation is unstable, and reliability needs to be improved. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a method and system for efficient source tracing of water pollution.

[0005] To solve the above-mentioned technical problems, the present invention provides a highly efficient water pollution source tracing method, which employs a main system and a subsystem for mobile source tracing on a river. The main system is fixedly deployed in the middle reaches of the river. In the absence of source tracing, the subsystem docks with the main system. There are multiple subsystems, and all subsystems and the main system can communicate with each other. The source tracing method includes the following steps: (1) After the main system is triggered by the pollution warning, the main system preliminarily infers the possible pollution source area based on the water flow velocity and official monitoring data, and determines the initial sampling point of the subsystem; (2) The main system releases at least four subsystems, and calculates the local concentration gradient vector of the region where the subsystem is located based on the initial sampling data of the subsystems. ; (3) Correct the local concentration gradient vector based on the water flow velocity of the main system. Determine the direction of the pollution source ; (4) Subsystem Dynamic Source Tracing: (4.1) Navigation Iteration: Select the subsystem with the highest concentration as the navigation subsystem. The main system moves in a direction, with other subsystems following behind the navigation subsystem, arranged in a fan shape centered on the navigation subsystem. The fan shape is symmetrical about the radial direction of the water flow. After the navigation subsystem moves a distance L, the main system calculates the local concentration gradient vector of the region where the iteration end point subsystem is located using the sampling data of the iteration end point subsystem. The local concentration gradient vector is corrected based on the water flow velocity of the navigation subsystem. Once again, the direction of the pollution source was determined. (4.2) Determine whether the local fine search conditions are met. If not, repeat step (4.1) until a certain subsystem meets the local fine search conditions. (5) Conduct a fine search around the subsystem that meets the local fine search conditions to determine the location of the pollution source.

[0006] Preferably, step (5) includes: (5.1) The subsystem that meets the local fine search conditions keeps its position unchanged and is set as the center. Other subsystems are arranged at equal intervals along the same radial direction and are sampled in a circular scanning mode. Local fine sampling is carried out. After each complete scan, the scanning radius is reduced to the center direction and circular fine sampling is continued until the distance between the subsystem closest to the center and the center is less than or equal to the safe distance of the subsystem sampling. When sampling stops. (5.2) The point with the highest concentration obtained from the scan is selected as the candidate location of the pollution source; (5.3) Verify candidate points; candidate points that meet the verification conditions are determined as pollution source locations. (5.4) For candidate points that do not meet the verification conditions, repeat steps (5.1) to (5.4) with the candidate point as the center.

[0007] Preferably, the local fine search condition is: a) the concentration change rate Y of a certain subsystem at the end of each round of navigation iteration, 0≤Y≤0.1; b. The concentration of this subsystem at the end of the current navigation iteration is more than 20% higher than the concentration measured by all other subsystems; or the concentration at this point is the highest value among all current subsystem sampling points and the difference between it and the second highest value exceeds the set threshold Z.

[0008] Preferably, the method for determining the initial sampling points of the subsystem is as follows: If the water flow velocity is greater than or equal to the threshold X, the main system releases N subsystems, which are then deployed at equal intervals d along the counter-current direction, with one subsystem on each side of the main system adjacent to the riverbank. If the water flow velocity < threshold X and the official data does not specify the pollution source, then a radial sampling layout is adopted: with the main system as the center, N sub-systems are evenly distributed on a circle with a radius of R, where 1 / 3W < R < 1 / 2W and W is the river width; If the water flow velocity < threshold X and the official data indicates that there are abnormal sites in a certain direction, the main system releases N sub-systems, which are arranged at equal intervals d along the abnormal direction, and one sub-system is arranged on each side of the main system adjacent to the riverbank; N≥2, and the arranged positions are the initial sampling points of the sub-systems.

[0009] Determine the initial sampling points according to different situations of the water flow magnitude and official data, so as to quickly obtain the initial leading iteration direction, sample comprehensively, and not easily cause misjudgment in source tracing.

[0010] Preferably, X = 0.1m / s and 1 / 5W < d < 1 / 2W.

[0011] Preferably, when the water flow velocity ≥ threshold X, the calculation method is: ; When the water flow velocity < threshold X, ; where is an empirical coefficient; : local concentration gradient vector; : water flow velocity vector.

[0012] Preferably, 1 / 5W < L ≤ W, and the moving distance L of the leading sub-system in each round of iteration is not greater than the moving distance L of the leading sub-system in the previous round of iteration.

[0013] Preferably, during the source tracing process, when any sub-system is ≤ 1m away from the riverbank, it automatically adjusts to travel along the riverbank and resumes the set route when the set route does not meet this condition.

[0014] Preferably, the verification conditions in step (5.3) are: a Supplementary sampling is carried out in multiple directions within a range less than or equal to 1 / 5 of the river width around the candidate point, and the concentrations at each point are lower than that at this point; b The fluctuation of the pollutant concentration obtained by sampling the candidate point multiple times is less than the set threshold.

[0015] The high-efficiency water pollution source tracing system of the present invention is used to implement the high-efficiency water pollution source tracing method described in any one of the above.

[0016] The beneficial effects of the present invention are: Quick initial orientation: Multiple stations sample synchronously, instantaneously obtain the spatial concentration distribution, without single-ship exploration, and shorten the response time.

[0017] High search efficiency: multiple subsystems operate in parallel, covering a wide range, and coordinated movement avoids duplication.

[0018] Highly adaptable: It dynamically adjusts its movement direction and sampling formation to adapt to different pollution diffusion conditions.

[0019] Strong anti-interference capability: Multi-point data fusion eliminates single-point noise and water flow vectors, resulting in more accurate gradient estimation. The collaborative strategy can cope with complex flow fields. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the control principle of the water pollution source tracing system of the present invention; Figure 2 This is a traceability roadmap for Embodiment 1 of the present invention; Figure 3 This is a source tracing roadmap for Embodiment 2 of the present invention. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. All directional indicators (such as up, down, left, right, front, back, etc.) in the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indicator will also change accordingly.

[0022] like Figure 1 As shown, the efficient water pollution source tracing system in this embodiment includes a main system and multiple subsystems. The main system is fixedly deployed at a key section in the middle reaches of the river (such as a bridge pier, sluice gate, or other fixed structure). Its lateral position is not limited to the center of the river width and can be determined according to the actual installation conditions. In non-source tracing situations, all subsystems are docked at the main system. The subsystems have the function of moving in the river. Both the subsystems and the main system have the function of monitoring the water quality parameters and water flow velocity vector at their respective locations.

[0023] The main system is used for: communicating with a cloud platform that has official pollution monitoring data; issuing warnings based on preset warning conditions; collecting and analyzing monitoring data; managing all subsystems, releasing and recovering subsystems; and controlling subsystems to trace sources according to the source tracing path. All subsystems and the main system can communicate with each other.

[0024] This system will utilize official pollution monitoring data and communicate with a cloud platform through the main system. The cloud platform has official water pollution monitoring data from multiple monitoring stations that are regularly and permanently set up in the river.

[0025] The subsystem samples at a sampling frequency of one second and uploads the sampling data in real time. The scanning mode sampling means that the main system calculates using all the sampling data of the subsystem; at other times, the main system calculates using the sampling data of the specified points of the subsystem.

[0026] Taking a river width of 50 m as an example, the specific source tracing methods of the embodiments of the present invention include the following steps: (1) After the main system is triggered with a pollution warning, the main system initially infers the possible pollution source area based on the water flow velocity and official monitoring data, and determines the initial sampling points of the subsystem. (1.1) The main system monitors the water quality parameters at its location in real time. When the concentration of a certain pollutant exceeds the preset warning line, a warning is triggered, and the type of pollutant with the most exceeding standards is recorded as the concerned pollutant. All subsequent samplings are for detecting the concentration of the concerned pollutant.

[0027] (1.2) The main system obtains the current water flow velocity vector and obtains the official monitoring site data from the cloud platform.

[0028] (1.3) Based on the water flow velocity and official data, initially infer the possible pollution source area and determine the initial sampling point layout strategy: a If the water flow velocity ≥ threshold X, the main system releases N subsystems (N≥2), which are arranged at equal intervals d along the reverse water flow direction. For example, they are respectively at d, 2d,... upstream of the main system (the longitudinal interval d between adjacent subsystems takes an empirical value of 20 - 50 m), and one subsystem is arranged on each side of the main system adjacent to the river bank to expand the coverage range and avoid omission.

[0029] b If the water flow velocity < threshold X and the official data does not indicate the pollution source. Then adopt a radial layout: with the main system as the center, evenly distribute N subsystems on a circle with a radius of R (such as R = 20 m, angular interval 360° / N). 1 / 3W < R < 1 / 2W, where W is the river width.

[0030] c If the water flow velocity < threshold X and the official data indicates that there is an abnormal site in a certain direction, the main system releases N subsystems (N≥2), which are arranged at equal intervals d along the abnormal direction. For example, they are respectively at d, 2d,... in this direction from the main system (the longitudinal interval d between adjacent subsystems takes an empirical value of 20 - 50 m), and one subsystem is arranged on each side of the main system adjacent to the river bank to expand the coverage range and avoid omission.

[0031] The threshold X is taken as 0.1 m / s, which is an empirical value for pollution investigation. When the water flow velocity ≥ 0.1 m / s, the water body is in an obvious flowing state (non-static water), and the pollutant migration effect with the water flow is significant, and it can be reasonably inferred that the pollution source mainly comes from upstream; when the water flow velocity < 0.1 m / s, the water body is approximately static, and the pollutant may come from all directions.

[0032] (2) The main system calculates the local concentration gradient vector of the region where the subsystem is located based on the initial sampling data of the subsystem. ; (2.1) After all subsystems are in place, all subsystems simultaneously collect the concentration of pollutants of interest at the current location. and coordinates And transmit it to the main system in real time via communication; (2.2) After the main system collects data from all subsystems, it calculates the local concentration gradient vector of the region where the subsystem is located. Multiple points of information can be utilized by using interpolation or weighted averaging methods, which are existing technologies.

[0033] (3) Correct the local concentration gradient vector based on the water flow velocity of the main system. Determine the direction of the pollution source ; (3.1) When the water flow velocity is greater than or equal to the threshold X, the water flow will cause the pollutant diffusion to deviate, so the observation gradient needs to be corrected to obtain a direction closer to the true direction of the pollution source. , ; No correction is needed when the water flow velocity is less than the threshold X. ; in This is an empirical coefficient; : Local concentration gradient vector; : Water flow velocity vector.

[0034] In flowing water bodies (rivers, canals), pollutant transport is primarily driven by advection (carried by the water flow), with diffusion playing a secondary role. Pollution plumes are "elongated" downstream, forming a plume pattern: high concentrations near the source, gradually decreasing downstream, but the concentration decreases more slowly along the direction of the flow, while decreasing more rapidly laterally (perpendicular to the flow direction). If directly along... Moving the search upstream too much can cause it to miss the true source (especially in cases of shoreline discharges or lateral inflows). This is particularly problematic when the pollution source is located on the shoreline; directly following the flow... This can easily cause the search path to skew towards the upstream center of the river, rather than directly towards the source on the bank. The essence of flow correction is to subtract the "dragging effect" of the flow from the gradient. Correction can subtract the "false upstream component" caused by the flow from the observed gradient, making the direction more accurate. For slow or still water bodies, no correction is necessary.

[0035] (4) Subsystem dynamic tracing: (4.1) Navigation Iteration: Select the subsystem with the highest concentration as the navigation subsystem. As the system moves in the direction of flow, other subsystems rejoin the lead subsystem according to proximity. Following the lead subsystem, the other subsystems are evenly distributed in a fan shape centered on it, with the flow direction of the lead subsystem as the reference point. The fan shape has a radius of 10-30m. The main system calculates the local concentration gradient vector of the region where the iteration ends using the sampling data from the subsystem at the iteration's end point. The local concentration gradient vector is corrected based on the water flow velocity of the navigation subsystem. The correction method is the same as the correction method in step (3.1), and the direction of the pollution source is obtained again. L is greater than 10m.

[0036] This navigation search strategy allows the navigation subsystem to focus on high-concentration areas while also following the subsystem to maintain peripheral searches, thus better adjusting the tracing direction and preventing misjudgments.

[0037] (4.2) After the navigation iteration ends, determine whether the local fine search conditions are met. If not, repeat step (4.1) until a certain subsystem meets the local fine search conditions. Repeat step (4.1) no more than 10 times to avoid invalid tracing. If it meets the conditions, execute step (5). The local fine-grained search criteria are: At the end of each round of pilot iteration, the concentration change rate Y of a certain subsystem is 0≤Y≤0.1, where Y is calculated by the main system based on the real-time sampling data of the subsystem.

[0038] b. The concentration of this subsystem at the end of the current navigation iteration is more than 20% higher than the concentration measured by all other subsystems; or the concentration at this point is the highest value among all current subsystem sampling points and the difference between it and the second highest value exceeds the set threshold Z (ammonia nitrogen concentration Z≥0.5mg / L, the threshold is different for different monitoring parameters).

[0039] (5) Conduct a fine search around the subsystem that meets the local fine search conditions to determine the location of the pollution source; (5.1) The subsystems that meet the local fine search conditions keep their positions unchanged and set as the center. Other subsystems are arranged at equal intervals along the same radial direction and sampled in a circular scanning mode. Local fine sampling is carried out. After each complete scan, the scanning radius is reduced by 1m towards the center and circular fine sampling is continued until the distance between the subsystem closest to the center and the center is less than or equal to 1m. When sampling stops, no more sampling is performed.

[0040] (5.2) The point with the highest concentration obtained from the scan is selected as the candidate point for the pollution source location.

[0041] (5.3) Verify candidate points. Candidate points that meet the following two conditions are determined as pollution source locations; Supplementary sampling was conducted in multiple directions (such as east, south, west, and north) around the candidate point within a range of less than or equal to 1 / 5 of the river width, and the concentration at each point was lower than that point. b. Multiple samplings of candidate points show pollutant concentration fluctuations below a set threshold; If candidate points are sampled at least three times, and the concentration fluctuation in each sample is less than a set threshold (e.g., fluctuation range < 10% of the mean), the interference of instantaneous water quality fluctuations can be eliminated to confirm the stability and reliability of the concentration at that point.

[0042] (5.4) For candidate points that do not meet the verification conditions, repeat steps (5.1) to (5.4) with the candidate point as the center.

[0043] During the source tracing process, if any subsystem is ≤1m away from the riverbank, it will automatically adjust to travel along the riverbank. If the set route does not meet this condition, it will restore the set route to avoid colliding with the river.

[0044] The main system records information such as the coordinates, concentration, and time of pollution sources, generates a source tracing report, and recycles all subsystems.

[0045] The following are the key processes and data of two specific embodiments executed according to the above tracing method.

[0046] Example 1 like Figure 2 As shown, a river is approximately 50 meters wide, and the main system is deployed at a cross-section in the middle reaches of the river. On a certain day, the main system detected a COD concentration of 45 mg / L, exceeding the warning threshold and triggering an alert. The water flow velocity was 0.02 m / s (<0.1 m / s), flowing from west to east. Data from upstream and downstream official stations were normal, with no clear indication of an abnormal direction.

[0047] Initial sampling point deployment strategy: The main system releases four subsystems (numbered A, B, C, and D), which move 20 meters to their positions in the four directions of due west, due north, due south, and due east, respectively, and simultaneously collect COD concentration and coordinates at their current locations. Sampling results are as follows: Subsystem A (20m due west of the main system): COD concentration 45mg / L; Subsystem B (20m due north of the main system): COD concentration 42 mg / L; Subsystem C (20m south of the main system): COD concentration 41 mg / L; Subsystem D (20m due east of the main system): COD concentration 38mg / L; Gradient estimation and flow correction: The main system uses interpolation to fit the concentration field based on the concentration and coordinates at four points, calculating the local concentration gradient direction as approximately 8° west of north. Due to the extremely low flow velocity, the correction amount is negligible, and the direction remains essentially unchanged, requiring no further correction. The source direction is... It is still about 8° west of north.

[0048] Subsystem dynamic tracing: The preset movement distance for the first iteration is 50m, and the movement distance for the second and subsequent iterations is 30m.

[0049] First iteration: The main system uses 8° west of north as the initial direction of movement for the navigation subsystem. The subsystem with the highest initial concentration is subsystem A (westernmost side), designated as the initial navigation subsystem. Other subsystems are arranged in a 90° sector with the navigation subsystem as the center. After the navigation subsystem moves to 50m, the sampling data is as follows: Subsystem A (Navigation Subsystem): COD concentration 55 mg / L; Subsystem B (currently located to the right rear of the navigation subsystem): COD concentration 52 mg / L; Subsystem C (originally due south, now located to the left and rear of the navigation subsystem): COD concentration 48 mg / L; Subsystem D (originally due east, now located at the rear): COD concentration 46 mg / L; The main system recalculates the local concentration gradient vector, and the direction of the local concentration gradient is 24° west of north. A is still the highest concentration and is retained as the navigation subsystem, entering the second round.

[0050] Second iteration: The pilot station remains subsystem A. After moving 30 meters along a direction of 24° west-northwest, the concentrations at each station are as follows: Subsystem A (Navigation Station): COD concentration 70 mg / L; Subsystem B: COD concentration 55 mg / L; Subsystem C: COD concentration 53 mg / L; Subsystem D: COD concentration 50 mg / L; At this point, subsystem A's own concentration change rate Y is less than 1%, and its concentration is more than 20% higher than that of other subsystems. The main system determines that the local fine-grained search conditions are met and triggers the local fine-grained search mode. In this iteration, when subsystem B is ≤1m from the riverbank, it automatically adjusts to travel along the riverbank.

[0051] Localized fine-grained search: Taking subsystem A as the center, subsystems B, C, and D are distributed radially along the same radial line, with initial radial distances from A of 3m, 6m, and 9m respectively. All three subsystems are equally distributed along the same radial line. Localized fine-grained sampling is conducted using a circular sampling scanning mode for all three subsystems. Then, the scanning radius is reduced by 1m towards A, and circular fine-grained sampling continues. This is stopped after three cycles, at which point the distances of B, C, and D from A are 1m, 5m, and 7m respectively. Point P with a COD concentration of 80mg / L is found 1m from the north bank of the river at a direction 11° north of east of A. Supplementary sampling is performed at points 1m east, south, and west of point P. To protect the subsystems, sampling is not performed in directions less than 1m from the riverbank. Point P is sampled three times repeatedly, all meeting the verification criteria, thus identifying point P as the pollution source location. When subsystems C and D are ≤1m from the riverbank, they automatically adjust to travel along the riverbank; if this condition is not met, they revert to the set route. This forms a... Figure 2 The sampling line is in the middle ring shape.

[0052] Example 2 like Figure 3 As shown, a river is approximately 50 meters wide, with its main monitoring system deployed at a fixed cross-section in the middle reaches. On a certain day, the main system detected an ammonia nitrogen concentration of 3.5 mg / L, exceeding the preset warning threshold and triggering an alert. At this time, the water flow velocity was relatively high, at 1.2 m / s, flowing from west to east (due east, 90°). Data from official monitoring stations upstream and downstream showed no significant anomalies, leading to the preliminary assessment that the pollution originated from the upstream (western) direction.

[0053] Initial sampling point deployment strategy: Since the water flow velocity is greater than the threshold of 0.1 m / s, the main system releases 4 subsystems (numbered A, B, C, and D).

[0054] Two subsystems are equally spaced along the counter-current direction (i.e., from east to west), located 40m and 80m upstream of the main system, respectively. Subsystem A is located 40m west of the main system upstream, and subsystem B is located 80m west of the main system upstream. A subsystem is also established on each side of the main system adjacent to the riverbank. Subsystem C: 20m due north of the main system; Subsystem D: 20m due south of the main system; Simultaneous sampling of ammonia nitrogen concentration and coordinates yielded the following results: Subsystem A: Ammonia nitrogen concentration 4.0 mg / L; Subsystem B: Ammonia nitrogen concentration 4.7 mg / L; Subsystem C: Ammonia nitrogen concentration 4.1 mg / L; Subsystem D: Ammonia nitrogen concentration 3.4 mg / L; Gradient estimation and flow correction: Based on the subsystem concentrations and coordinates, the system uses interpolation to fit the concentration field, calculating the initial local concentration gradient direction as 6° west of north. The flow velocity vector is known. The direction is due east (90°), and the speed is 1.2 m / s. The empirical coefficient α is taken as 0.7. According to the correction formula, the corrected direction is 3° west of north.

[0055] Subsystem dynamic tracing: The initial preset movement distance is set to 50m, and the movement distance decreases by 10m each round, until it reaches 10m and is no longer adjusted.

[0056] First iteration: The main system uses the corrected direction (3° west of north) as the initial movement direction of the navigation subsystem. The subsystem with the highest initial concentration is subsystem B (the westernmost part), designated as the initial navigation subsystem. Other subsystems are arranged in a 90° sector with the navigation subsystem as the center. After the navigation subsystem moves to 50m, the concentrations at each point are as follows: Subsystem A: Ammonia nitrogen concentration 4.8 mg / L; Subsystem B: Ammonia nitrogen concentration 5.4 mg / L; Subsystem C: Ammonia nitrogen concentration 4.4 mg / L; Subsystem D: Ammonia nitrogen concentration 3.9 mg / L; The main system recalculates the local concentration gradient vector, and the direction of the local concentration gradient is 2.8° west of north; B is still the highest, so it is retained as the navigation subsystem and enters the second round.

[0057] Second iteration: After the navigation subsystem B moves 40 meters along a direction of 2.8° west of north, the concentrations at each station are as follows: Subsystem A: Ammonia nitrogen concentration 5.3 mg / L; Subsystem B: Ammonia nitrogen concentration 6.2 mg / L; Subsystem C: Ammonia nitrogen concentration 5.1 mg / L; Subsystem D: Ammonia nitrogen concentration 4.3 mg / L; The main system recalculates the local concentration gradient vector, and the direction of the local concentration gradient is 2.5° west of north; B is still the highest, and it is retained as the navigation subsystem, entering the third round.

[0058] Third iteration: After the navigation subsystem B moves 30 meters in a direction of 2.5° west of north, the concentrations at each station are as follows: Subsystem A: Ammonia nitrogen concentration 5.8 mg / L; Subsystem B: Ammonia nitrogen concentration 7.3 mg / L; Subsystem C: Ammonia nitrogen concentration 5.5 mg / L; Subsystem D: Ammonia nitrogen concentration 4.7 mg / L; If the concentration change rate Y of subsystem B is less than 1% and its concentration is more than 20% higher than that of other subsystems, the main system determines that it has entered the vicinity of the pollution source and triggers the local fine search mode.

[0059] Local fine-grained search: Taking subsystem B as the center, subsystems A, C, and D are distributed radially along the same radial line, with initial radial distances from B of 3m, 6m, and 9m respectively. All three subsystems are equally distributed along the same radial line. Local fine-grained sampling is performed using a circular sampling scanning mode for all three subsystems. Then, the scanning radius is reduced by 1m towards B, and circular fine-grained sampling continues. This process is stopped after three cycles, at which point the distances of A, C, and D from B are 1m, 5m, and 7m respectively, forming a pattern as shown below. Figure 3 A circular sampling route was used. Point P with a concentration of 7.9 mg / L was found 4 m away from point B at a direction 5° west of north. Supplementary sampling was conducted at distances of 1 m in the east, south, west, and north directions from point P. Three repeated samplings of point P were performed, and all met the verification conditions, thus identifying point P as the pollution source location.

[0060] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A highly efficient method for tracing the source of water pollution, comprising a main system and a subsystem for mobile source tracing along a river, wherein the main system is fixedly deployed in the middle reaches of the river, and the subsystem docks with the main system when not tracing the source, characterized in that: There are multiple such subsystems, and all subsystems can communicate with the main system; The tracing method includes the following steps: (1) After the main system is triggered with a pollution warning, the main system initially infers a possible pollution source area based on the water flow velocity and official monitoring data, and determines the initial sampling points of the subsystems; (2) The main system releases at least four subsystems, and calculates the local concentration gradient vector of the region where the subsystem is located based on the initial sampling data of the subsystems. ; (3) Correct the local concentration gradient vector based on the water flow velocity of the main system. Determine the direction of the pollution source ; (4) Subsystem Dynamic Source Tracing: (4.1) Navigation Iteration: Select the subsystem with the highest concentration as the navigation subsystem. The main system moves in a direction, with other subsystems following behind the navigation subsystem, arranged in a fan shape centered on the navigation subsystem. The fan shape is symmetrical about the radial direction of the water flow. After the navigation subsystem moves a distance L, the main system calculates the local concentration gradient vector of the region where the iteration end point subsystem is located using the sampling data of the iteration end point subsystem. The local concentration gradient vector is corrected based on the water flow velocity of the navigation subsystem. Once again, the direction of the pollution source was determined. (4.2) Determine whether the local fine search conditions are met. If not, repeat step (4.1) until a certain subsystem meets the local fine search conditions. (5) Conduct a fine search around the subsystem that meets the local fine search conditions to determine the location of the pollution source.

2. The efficient water pollution source tracing method according to claim 1, characterized in that: The step (5) includes: (5.1) The subsystems that meet the local fine search condition remain in place and are set as the center of the circle. Other subsystems are arranged at equal distances along the same radial direction and perform circular scanning mode sampling to carry out local refined sampling. After each complete scan, they shrink a certain scanning radius towards the center of the circle and continue to perform circular refined sampling until the distance between the subsystem closest to the center of the circle and the center of the circle is less than or equal to the safe distance for subsystem sampling, and then stop sampling; (5.2) Take the point with the highest concentration obtained by scanning as the candidate point for the pollution source location; (5.3) Verify the candidate point, and the candidate point that meets the verification condition is determined as the pollution source location; (5.4) For the candidate points that do not meet the verification condition, take the candidate point as the center of the circle and re-execute steps (5.1) to (5.4).

3. The efficient water pollution source tracing method according to claim 1, characterized in that: The local fine search condition is: a. The concentration change rate Y of a certain subsystem itself at the end of each round of leader iteration, 0 ≤ Y ≤ 0.1; b. The concentration of this subsystem at the end point of this round of leader iteration is more than 20% higher than the concentrations measured by all other subsystems; or the concentration at this point is the highest among the sampling points of all current subsystems and the difference from the second highest value exceeds the set threshold Z.

4. The efficient water pollution source tracing method according to claim 1, characterized in that: The method for determining the initial sampling points of the subsystems is as follows: If the water flow velocity ≥ threshold X, the main system releases N subsystems, which are arranged at equal intervals d along the reverse water flow direction, and one subsystem is arranged on each side of the main system adjacent to the river bank; If the water flow velocity < threshold X and the official data does not indicate the pollution source, then a radial layout is adopted: with the main system as the center, N subsystems are evenly distributed on a circle with a radius of R, 1 / 3W < R < 1 / 2W, where W is the river width; If the water flow velocity < threshold X and the official data indicates that there is an abnormal site in a certain direction, the main system releases N subsystems, which are arranged at equal intervals d along the abnormal direction, and one subsystem is arranged on each side of the main system adjacent to the river bank; N ≥ 2, and the arranged position of d is the initial sampling point of the subsystem.

5. The efficient water pollution source tracing method according to claim 4, characterized in that: X = 0.1m / s, 1 / 5W < d < 1 / 2W.

6. The efficient water pollution source tracing method according to claim 1, characterized in that: When the water flow velocity is greater than or equal to the threshold X, the following... The calculation method is as follows: ； When the water flow velocity is less than the threshold X ; in This is an empirical coefficient; : Local concentration gradient vector; : Water flow velocity vector.

7. The efficient water pollution source tracing method according to claim 1, characterized in that: 1 / 5W < L ≤ W, and the moving distance L of the leader subsystem in each round of iteration is not greater than the moving distance L of the leader subsystem in the previous round of iteration.

8. The efficient water pollution source tracing method according to claim 1, characterized in that: During the tracing process, when any subsystem is ≤ 1m away from the river bank, it automatically adjusts to travel along the river bank and resumes the set route when the set route does not meet this condition.

9. The efficient water pollution source tracing method according to claim 2, characterized in that: The verification condition in the step (5.3) is: a. Supplementary sampling is carried out in multiple directions within a range less than or equal to 1 / 5 of the river width around the candidate point, and the concentration at each point is lower than that at this point; b. The fluctuation of the pollutant concentration during multiple samplings of the candidate point is less than the set threshold.

10. A highly efficient water pollution source tracing system, characterized in that: For implementing the efficient water pollution tracing method described in any one of claims 1 - 9.

Images