Industrial park hidden gas leakage-oriented pollution source dynamic tracing method and equipment

Abstract

The invention discloses a pollution source dynamic tracing method and equipment for industrial park hidden gas leakage. Comprising the following steps: acquiring meteorological monitoring data at each observation point through fixed monitoring equipment and / or mobile monitoring equipment; correcting the coordinates of the observation points; predicting a first gas pollutant predicted concentration corresponding to each observation point after coordinate correction according to a Gaussian plume diffusion model; inputting the first gas pollutant predicted concentration and meteorological monitoring data into a CatBoost model to obtain a second gas pollutant predicted concentration; taking the difference between the actual gas pollutant concentration at each observation point and the second gas pollutant predicted concentration as a target function, and solving the target function to obtain the pollutant source intensity and the source position; completing one iteration; and setting the observation point near the current predicted pollutant source position, and so on, and obtaining the final pollutant source intensity and source position after n iterations.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental monitoring, and particularly relates to a method and device for dynamically tracing the source of pollution of hidden gas leakage in industrial parks. Background Art

[0002] In industrial parks, the emission of gaseous pollutants poses a serious threat to the ecological environment and human health. Traditional means of identifying pollution sources mainly rely on monitoring stations at fixed locations, which are difficult to comprehensively and accurately capture the complex diffusion patterns of pollutants and cannot meet the need for quickly and accurately tracing the sources of hidden gas leakage pollution sources in industrial parks.

[0003] With the rapid development of industry, gas leakage accidents often occur in industrial parks. These leaks may come from hidden sources, and their source strengths and locations are difficult to determine. Therefore, there is an urgent need for a tracing method to solve these problems. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a method and device for dynamically tracing the source of pollution of hidden gas leakage in industrial parks.

[0005] In a first aspect, an embodiment of the present invention provides a method for dynamically tracing the source of pollution of hidden gas leakage in industrial parks, the method comprising:

[0006] Obtaining meteorological monitoring data at each observation point through fixed monitoring devices and / or mobile monitoring devices;

[0007] Correcting the coordinates of the observation point; predicting the predicted concentration of the first gaseous pollutant corresponding to each observation point after correcting the coordinates according to the Gaussian plume diffusion model;

[0008] Inputting the predicted concentration of the first gaseous pollutant and the meteorological monitoring data into the CatBoost model to obtain the predicted concentration of the second gaseous pollutant;

[0009] Taking the difference between the actual concentration of the gaseous pollutant and the predicted concentration of the second gaseous pollutant at each observation point as the objective function, solving the objective function to obtain the source strength and source location of the pollutant; completing the first iteration;

[0010] Setting the observation point near the currently predicted pollutant source location through a mobile monitoring device, and so on. After n iterations, the final source strength and source location of the pollutant are obtained.

[0011] In a second aspect, an embodiment of the present invention provides an electronic device, comprising a memory and a processor, the memory being coupled to the processor; wherein, the memory is used for storing program data, and the processor is used for executing the program data to implement the above-mentioned method for dynamically tracing the source of pollution of hidden gas leakage in industrial parks.

[0012] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned dynamic source tracing method for hidden gas leakage in an industrial park is implemented.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the above-mentioned dynamic source tracing method for hidden gas leakage in an industrial park is implemented.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0015] The present invention provides a dynamic source tracing method for hidden gas leakage in an industrial park. By using fixed monitoring devices and mobile monitoring devices, more comprehensive and accurate meteorological monitoring data are obtained, providing a reliable basis for source tracing. By correcting the coordinates of the observation points, the corrected coordinates of the observation points are input into the improved Gaussian plume diffusion model to predict the preliminary predicted concentration of gaseous pollutants. Then, combined with the meteorological monitoring data and the CatBoost model, the diffusion of pollutants in the complex environment of the industrial park is simulated, further improving the accuracy of source tracing. Taking the difference between the actual concentration of gaseous pollutants and the second predicted concentration of gaseous pollutants at each observation point as the objective function, and combining the wind direction factor, GA algorithm, and IPSO algorithm to solve the objective function, effectively avoiding the problem that the traditional algorithm is prone to falling into local optimum, and improving the accuracy and efficiency of estimating the source location and source strength. At the same time, the present invention can also adjust the position of the mobile monitoring device. In each iteration, according to the source location information obtained from the previous inversion calculation, the position of the observation point is dynamically adjusted, improving the accuracy of estimating the source location and source strength. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings without creative efforts.

[0017] Figure 1 It is a flowchart of the dynamic source tracing method for hidden gas leakage in an industrial park provided by an embodiment of the present invention;

[0018] Figure 2 It is a schematic diagram of the dynamic source tracing method for hidden gas leakage in an industrial park provided by an embodiment of the present invention;

[0019] Figure 3 It is a schematic diagram of coordinate transformation provided by an embodiment of the present invention;

[0020] Figure 4 Schematic diagram of the equivalent height of the observation point provided by the embodiment of the present invention;

[0021] Figure 5 Schematic diagram of the improved Gaussian-CatBoost diffusion model (IG-CatBoost) provided by the embodiment of the present invention;

[0022] Figure 6 Flow chart of the hybrid wind direction factor-GA-IPSO algorithm provided by the embodiment of the present invention;

[0023] Figure 7 Schematic diagram of an electronic device provided by the embodiment of the present invention. Detailed implementation manners

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with 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 the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0025] It should be noted that, without conflict, the features in the following embodiments and implementation manners can be combined with each other.

[0026] As Figure 1 and Figure 2 shown, the embodiment of the present invention provides a dynamic source tracing method for hidden gas leakage in industrial parks, and the method includes the following steps:

[0027] Step S1, obtaining meteorological monitoring data at each observation point through fixed monitoring devices and / or mobile monitoring devices.

[0028] Specifically, in this example, ground fixed monitoring stations are arranged in the industrial park to be monitored, and the basic meteorological monitoring data is obtained by using their characteristics of long-term stable monitoring; at the same time, mobile monitoring vehicles and unmanned aerial vehicle monitoring devices are equipped, and the monitoring positions are flexibly adjusted according to the actual situation in the industrial park, and encrypted monitoring is carried out on key areas or suspected pollution areas.

[0029] Step S2, correcting the coordinates of the observation point; predicting the predicted concentration of the first gas pollutant corresponding to each observation point with the corrected coordinates according to the Gaussian plume diffusion model.

[0030] Specifically, as Figure 3 shown, the process of correcting the coordinates of the observation point includes:

[0031] Perform coordinate transformation on the observation point (x, y) in the geographic coordinate system to the observation point (x ′ , y ′ ) in the coordinate system with the wind direction as the abscissa; the expression is as follows:

[0032]

[0033] In the formula, α is the angle between the wind direction and the due north direction, (x 0 , y 0 ) represents the pollution source point, and (x, y) represents the observation point;

[0034] As Figure 4 shown, according to the terrain height of the pollution source point, the terrain height of the observation point, and the height of the gas sensor set at the observation point from the ground, obtain the equivalent height at the observation point; the expression is as follows:

[0035] z ′ [x, y] = z 0 [x 0 , y 0 - z[x, y] + h z

[0036] In the formula, z ′ [x, y] represents the equivalent height of the observation point (x, y), z 0 [x 0 , y 0 represents the terrain height of the pollution source point (x 0 , y 0 ), z[x, y] represents the terrain height of the observation point (x, y), and h z represents the height of the gas sensor from the ground.

[0037] It should be noted that the improved Gaussian plume model significantly enhances the accuracy of pollutant diffusion simulation through coordinate transformation and equivalent height calculation: on the one hand, the coordinates of the observation points in the geographical coordinate system are transformed into a new coordinate system with the wind direction as the abscissa, making the pollutant diffusion process under the action of the wind field more intuitive in the mathematical model, facilitating parameter calculation and physical law analysis; on the other hand, the equivalent height comprehensively considers the terrain height difference between the pollution source and the observation point and the height of the sensor from the ground, quantifying the complex terrain factors into a unified parameter and incorporating it into the model, effectively correcting the influence of terrain undulation on the vertical diffusion path of pollutants. These two improvements jointly optimize the applicability of the Gaussian plume model, enabling it to more realistically reflect the three-dimensional diffusion law of pollutants in the actual environment and providing a more accurate concentration prediction basis for subsequent pollution source positioning and source strength inversion. In terms of the improvement of the wind direction coordinate axis, a geographical coordinate system is constructed with the gas release source as the origin, and the points in the geographical coordinate system are transformed to the coordinate system with the wind direction as the abscissa through a rotation matrix. The coordinates of point A(x, y) in the geographical coordinate system are transformed to a point A ′ (x ′ ,y ′ ) in the new coordinate system along the wind direction, making the calculation and simulation of the wind diffusion process more intuitive and accurate.

[0038] As Figure 5 shown, the coordinates of the corrected observation points are input into the improved Gaussian plume diffusion model to predict the predicted concentration of the first gaseous pollutant corresponding to each observation point after coordinate correction; among them, the expression of the improved Gaussian plume diffusion model is as follows:

[0039]

[0040] In the formula, C(x, y, z) represents the concentration at (x, y, z), σ′ y represents the diffusion coefficient in the horizontal direction, σ′ z represents the diffusion coefficient in the vertical direction, H is the vertical height of the release source point from the ground, q represents the pollution source strength, and u represents the wind speed.

[0041] Step S3, input the predicted concentration of the first gaseous pollutant and the meteorological monitoring data into the CatBoost model to obtain the predicted concentration of the second gaseous pollutant.

[0042] Furthermore, in the CatBoost model part, its core is based on the gradient boosting algorithm. Let the training data set be where x i represents the feature vector of the i-th sample, including meteorological parameters (such as wind speed, wind direction, temperature, atmospheric stability, etc.), terrain parameters, and parameters related to the pollution source (such as release height, release rate, etc.), y iis the corresponding target value (i.e., gas concentration), and N is the total number of samples. The initial predicted value of the CatBoost model is usually set to a constant (such as the mean). In the t-th iteration, the predicted value of the CatBoost model is the predicted value of the CatBoost model in the previous t - 1 iterations and the predicted value of the t-th decision tree h t (x i ). The expression is as follows:

[0043]

[0044] where α t is the weight of the t-th decision tree, which is determined by minimizing the loss function L. Commonly used loss functions such as mean squared error (MSE) In each iteration, according to the prediction error (negative gradient) of the current CatBoost model fit a new decision tree on the residuals to gradually approximate the true model and improve the prediction accuracy.

[0045] In the gas diffusion problem, there are many categorical features, such as atmospheric stability level, wind direction, wind speed, temperature, etc. For the j-th category of the categorical feature c, the calculation method for converting it into a numerical feature is as follows:

[0046]

[0047] where is the indicator function, which is 1 when is equal to j and 0 otherwise, is the gradient value of the k-th sample in the t-th iteration, and Prior is the prior value (which can be set according to the data distribution).

[0048] At the same time, CatBoost adopts a series of advanced techniques to prevent overfitting. In terms of regularization, penalize the complexity of the decision tree. For example, by controlling parameters such as the depth d of the decision tree and the number of leaf nodes m, add a regularization term to the loss function as follows:

[0049]

[0050] where γ and λ are regularization coefficients, and ω j is the weight of leaf node j.

[0051] It should be noted that in the IG-CatBoost model, the improved Gaussian plume model provides preliminary gas concentration predictions based on physical mechanisms for the CatBoost model, and the CatBoost model then performs non-linear fitting and optimization on these preliminary prediction results. Through continuous iterative training, using its ability to handle complex data relationships and prevent overfitting, it outputs more accurate predicted gas concentrations.

[0052] Step S4: Using the difference between the actual gas pollutant concentration and the second predicted gas pollutant concentration at each observation point as the objective function, solve the objective function to obtain the pollutant source strength and source location; complete 1 iteration.

[0053] In this example, the pollutant tracing problem is transformed into an unconstrained extreme value problem to construct the objective function, and the expression is as follows:

[0054]

[0055] In the formula, Q 0 represents the source release rate of the pollutant, (x 0 , y 0 , z 0 ) represents the location of the pollutant, and n is the number of gas sensors; represents the gas pollutant concentration detected by the i-th gas sensor; is the second predicted gas pollutant concentration corresponding to the i-th gas sensor point output based on the CatBoost model.

[0056] Furthermore, as Figure 6 shown, the process of solving the objective function to obtain the pollutant source strength and source location includes:

[0057] (1) Initialize the population and set the initial positions and velocities of the particles;

[0058] (2) Calculate the fitness of each particle;

[0059] (3) Based on the GA algorithm, select the particles with lower fitness to enter the crossover and mutation stages, and generate new particles through crossover and mutation operations to increase the diversity of the population;

[0060] (4) Update the inertia weight, particle velocity, and particle position; including:

[0061] The update formula for the inertia weight ω is:

[0062]

[0063] Among them, ω max and ω min are the initial and minimum inertia weights respectively, and k maxis the maximum number of iterations, and k is the current iteration number;

[0064] The update formula for the particle velocity is as follows:

[0065]

[0066] In the formula, is the velocity of the i-th particle, ω is the inertia weight, c 1 , c 2 is the acceleration constant, r 1 , r 2 is a random number, pbest i is the best position in the history of the i-th particle, gbest is the best position of all particles, r 3 is a random number, ω v is the wind speed correction weight, is the wind speed.

[0067] (5) Re-evaluate the fitness of each particle, update the historical best position and the global best position of the particle, and output the global optimal solution, that is, the estimated pollution source strength and pollution source location.

[0068] It should be noted that in this example, the difference between the actual gas pollutant concentration and the predicted concentration of the second gas pollutant at each observation point is used as the objective function, and the wind direction factor, GA algorithm, and IPSO algorithm are combined to solve the objective function, effectively avoiding the problem that the traditional algorithm is easily trapped in the local optimum, and improving the accuracy and efficiency of the pollution source location and source strength estimation. Because after considering the wind direction factor, the movement of particles in the search space is more in line with the actual pollutant diffusion path, and can explore the search space more comprehensively, rather than being limited to the local area. This greatly improves the accuracy of the pollution source location and source strength estimation, making the estimated pollution source strength and source location closer to the actual situation. At the same time, due to more reasonably simulating the pollutant propagation and search process, the solving efficiency is also improved, and accurate results can be obtained faster.

[0069] Step S5, set the observation point near the currently predicted pollutant source location, and so on. After n iterations, the final pollutant source strength and source location are obtained.

[0070] It should be noted that the mobile monitoring platform (such as unmanned aerial vehicles, mobile vehicle platforms, etc.), as the carrier for deploying gas sensors, dynamically adjusts the position of the observation point according to the pollution source location information obtained from the previous inversion calculation in each iteration. For example, if it is initially estimated that there is a pollution source in a small area but the data is not accurate enough, the mobile monitoring platform can be controlled to maneuver towards this area in the next iteration to collect more accurate data.

[0071] Correspondingly, the present application further provides an electronic device, including: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the source dynamic tracing method for hidden gas leakage in industrial parks as described above. As Figure 7 shown, it is a hardware structure diagram of any device with data processing capabilities where the source dynamic tracing method for hidden gas leakage in industrial parks provided by the embodiment of the present invention is located. In addition to Figure 7 the processors, memory, and network interfaces shown, any device with data processing capabilities where the device in the embodiment is located usually may further include other hardware according to the actual functions of the any device with data processing capabilities, which will not be elaborated here.

[0072] Correspondingly, the present application further provides a computer-readable storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the source dynamic tracing method for hidden gas leakage in industrial parks as described above is implemented. The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device, such as a plug-in hard disk, a Smart Media Card (SMC), an SD card, a Flash Card, etc. equipped on the device. Further, the computer-readable storage medium may also include both an internal storage unit of any device with data processing capabilities and an external storage device. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store the data that has been output or will be output.

[0073] The above embodiments are only used to illustrate the design concept and features of the present invention, and the purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes or modifications made according to the principles and design ideas disclosed by the present invention are within the protection scope of the present invention.

Claims

1. A method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks, characterized in that: The method comprises: Obtain meteorological monitoring data at each observation point through fixed monitoring equipment and / or mobile monitoring equipment; Correct the coordinates of the observation points; predict the predicted concentration of the first gas pollutant corresponding to each observation point after the corrected coordinates according to the Gaussian plume diffusion model; Inputting the predicted concentration of the first gas pollutant and the meteorological monitoring data into the CatBoost model to obtain the predicted concentration of the second gas pollutant; The difference between the actual gas pollutant concentration at each observation point and the predicted concentration of the second gas pollutant is used as the objective function, and the objective function is solved to obtain the pollutant source intensity and source location; completing the first iteration; The observation point is set near the currently predicted pollutant source location through mobile monitoring equipment, and so on. After n iterations, the final pollutant source intensity and source location are obtained.

2. According to claim 1, a method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks is characterized in that: The process of correcting the coordinates of the observation points includes: The coordinates of the observation point (x, y) in the geographic coordinate system are converted to the observation point (x ′ ,y ′ ); the expression is as follows: In the formula, α is the angle between the wind direction and the north direction, (x0, y0) represents the pollution source point, and (x, y) represents the observation point; According to the terrain height of the pollution source, the terrain height of the observation point, and the height of the gas sensor set at the observation point from the ground, the equivalent height at the observation point is obtained; the expression is as follows: z ′ [x,y]=z0[x0,y0]-z[x,y]+h z In the formula, z ′ [x,y] represents the equivalent height of the observation point (x,y), z0[x0,y0] represents the terrain height of the pollution source point (x0,y0), z[x,y] represents the terrain height of the observation point (x,y), h z Indicates the height of the gas sensor from the ground.

3. According to claim 1, a method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks is characterized in that: The meteorological monitoring data includes wind direction, wind speed, temperature, and atmospheric stability.

4. According to claim 1, a method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks is characterized in that: The expression of the objective function is as follows: Where Q0 represents the source release rate of the pollutant, (x0, y0, z0) represents the location of the pollutant, and n is the number of gas sensors; represents the concentration of gas pollutants detected by the i-th gas sensor; is the predicted concentration of the second gas pollutant corresponding to the i-th gas sensor point based on the CatBoost model output.

5. According to claim 1, a method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks is characterized in that: The process of solving the objective function to obtain the pollutant source intensity and source location includes: Initialize the population and set the initial position and velocity of the particles; Calculate the fitness of each particle; Select particles with lower fitness to enter the crossover and mutation phase, and generate new particles through crossover and mutation operations; Update inertia weight, particle velocity, and particle position; Re-evaluate the fitness of each particle, update the particle's historical best position and global best position, and output the global optimal solution, that is, the estimated pollution source strength and pollution source location.

6. A method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks according to claim 5, characterized in that: The update formula of inertia weight ω is: Among them, ω max and ω min are the initial and minimum inertia weights, k max is the maximum number of iterations, and k is the current number of iterations.

7. A method for dynamic tracing of pollution sources for hidden gas leakage in industrial parks according to claim 5, characterized in that: The update formula of particle velocity is: In the formula, is the velocity of the ith particle, ω is the inertia weight, c1, c2 are acceleration constants, r1, r2 are random numbers, pbest i is the best position of the i-th particle in history, gbest is the best position of all particles, r3 is a random number, ω v is the wind speed correction weight, It's the wind speed.

8. An electronic device, comprising a memory and a processor, characterized in that: The memory is coupled to the processor; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the dynamic source tracing method for hidden gas leakage in industrial parks as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, a dynamic source tracing method for hidden gas leakage in an industrial park as described in any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the method for dynamically tracing pollution sources for hidden gas leaks in industrial parks as described in any one of claims 1-7 is implemented.

Images