Pollution source positioning method, device and equipment based on multi-source data fusion and storage medium

Abstract

The invention discloses a pollution source positioning method, device and equipment based on multi-source data fusion and a storage medium, and relates to the technical field of environment monitoring, and the method comprises the steps: obtaining environment and meteorological data through a sensor network, and obtaining filtered environment data and synchronized meteorological data through adaptive filtering and time synchronization processing. Then, pollution source information estimation is carried out by combining the data and using an improved Kriging interpolation algorithm and adaptive Kalman filtering, and an initial state vector is generated; the position of the pollution source is predicted based on the vector, a candidate coordinate set is formed, the position of the target pollution source is optimally selected through maximum likelihood estimation, the problems of data transmission delay and meteorological disturbance of a traditional method are solved, and the accuracy and reliability of real-time positioning of the VOCs pollution source in the industrial park are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and particularly to a pollution source location method, device, equipment and storage medium based on multi-source data fusion. Background Art

[0002] Volatile organic compounds (VOCs) are important atmospheric pollutants generated during industrial production processes. They can generate ozone and secondary organic aerosols in atmospheric photochemical reactions, becoming one of the main precursors of haze and photochemical pollution. The emissions of VOCs in industrial parks account for a relatively large proportion of industrial pollution sources, but traditional monitoring means are difficult to achieve rapid and accurate location of pollution sources, resulting in lagged emergency responses and increased risks of pollution diffusion.

[0003] Existing VOCs pollution source location methods mainly rely on offline laboratory analysis or a centralized calculation mode based on fixed monitoring stations. Offline laboratory analysis cannot meet the real-time requirements due to its long sampling period; while the centralized calculation mode based on fixed monitoring stations is limited by data transmission bandwidth and processing delay. Especially under complex meteorological conditions, the diffusion paths of dynamic pollution sources are time-varying and random, which makes traditional models produce significant errors.

[0004] Despite the existing progress, the existing technologies still have limitations in many aspects. First, the current technical solutions rely on the cloud for centralized processing, which requires remote transmission of a large amount of monitoring data to the server, resulting in time delays of more than a minute. Second, the layout density of fixed monitoring stations is insufficient, making it difficult to capture the dynamic gradient change characteristics of the VOCs concentration field. In addition, traditional inverse diffusion models assume steady-state meteorological conditions and do not fully consider non-ideal factors such as wind speed fluctuations and turbulence intensity, resulting in reduced location reliability in complex scenarios. Therefore, there is an urgent need for an efficient and accurate pollution source location method to meet the requirements of pollution source location in industrial parks. Summary of the Invention

[0005] The main purpose of this application is to provide a pollution source location method, device, equipment and storage medium based on multi-source data fusion, aiming to solve the technical problem of how to improve the accuracy of pollution source location in industrial parks.

[0006] To achieve the above object, this application proposes a pollution source location method based on multi-source data fusion, including: Obtain environmental data and meteorological data; Perform adaptive filtering processing on the environmental data to obtain filtered environmental data; Perform time synchronization processing on the meteorological data to obtain synchronized meteorological data; By processing the filtered environmental data and the synchronized meteorological data, pollution source information is obtained; Based on the pollution source information, an initial state vector of the pollution source is calculated; Based on the initial state vector of the pollution source, a set of candidate pollution source coordinates is calculated; The set of candidate pollution source coordinates is selected by maximum likelihood estimation to obtain the location of the target pollution source.

[0007] In one embodiment, the step of performing adaptive filtering on the environmental data to obtain the filtered environmental data includes: Obtain historical environmental data; Based on the historical environmental data, an adaptive adjustment factor is calculated. The specific formula is: ; where, is the adjustment coefficient, represents the number of the historical environmental data, represents the th concentration value of the historical environmental data, is the moving window mean, is the window standard deviation; Based on the adaptive adjustment factor, the environmental data is processed by adaptive filtering to obtain the filtered environmental data. The specific formula is: ; where, represents the concentration value of the environmental data, is the coefficient of the 3-sigma rule.

[0008] In one embodiment, the step of performing time synchronization processing on the meteorological data to obtain the synchronized meteorological data includes: According to the meteorological data, local clock information is obtained. The local clock information includes the local clock value and the clock deviation; Based on the local clock information and the theoretical synchronization time value, an edge computing correction value is calculated. The specific formula is: ; where, is the theoretical synchronization time value, is the number of sensors for obtaining the meteorological data, is the local clock value; Based on the edge computing correction value, the local clock information is corrected to obtain the corrected clock information. The specific formula is: ; Among them, is the delay clock, is the clock deviation, is the edge computing correction value; The synchronized meteorological data is obtained according to the corrected clock information, and the specific formula is: ; Among them, represents the synchronized meteorological data, including wind speed and wind direction meteorological variables, represents the said meteorological data.

[0009] In one embodiment, the step of obtaining the pollution source information by processing the filtered environmental data and the synchronized meteorological data includes: By processing the filtered environmental data using spatial interpolation method to construct the pollutant concentration field, the specific interpolation formula is: ; Among them, is the pollutant concentration at the th detection point, is the normalized weight coefficient corresponding to the th detection point, and satisfies = 1. The specific formula for obtaining the normalized weight coefficient is: ; ; ; Among them, is the spatial scale parameter, is the current position coordinate, is the wind speed at the current position, is the historical dominant wind speed, is the wind speed standard deviation, is the wind direction at the current position, is the historical dominant wind direction, is the wind direction fluctuation standard deviation; The corrected concentration value of the pollutant is obtained by coupling optimization processing of the pollutant concentration field, and the specific formula is: ; Among them, is the original concentration at the th detection point; The pollution source information is obtained by calculating according to the corrected concentration value of the pollutant.

[0010] In one embodiment, the step of calculating the initial state vector of the pollution source according to the pollution source information includes: Identifying based on the target pollutant concentration field in the pollution source information to obtain a concentration gradient value; Selecting the position where the concentration gradient value reaches a preset value as the initial position of the pollution source; Obtaining the concentration value at the initial position of the pollution source as the initial concentration; Correcting the initial position and the initial concentration of the pollution source according to the pollutant diffusion path in the pollution source information to obtain the corrected position and the corrected concentration; Combining the corrected position and the corrected concentration to obtain the initial state vector of the pollution source.

[0011] In one embodiment, the step of calculating the candidate pollution source coordinate set based on the initial state vector of the pollution source includes: According to the initial state vector and in combination with the Gaussian diffusion model, obtaining an updated state vector, and the specific formula is: ; Wherein, represents the state transition matrix, represents the process noise; Performing prediction according to the updated state vector to obtain a priori estimated state vector, and the specific formula is: ; Wherein, represents the a priori estimated state vector, represents the wind speed correction amount; Mapping the a priori estimated state vector to obtain an observation vector; Based on the observation vector and through the Kalman gain matrix, obtaining a state estimation value, and the Kalman gain matrix is expressed as: ; Wherein, is the observation vector, is the observation noise covariance matrix, is the prediction error covariance matrix; Based on the state estimation value, calculating to obtain a posterior probability, and the specific calculation formula is: ; Wherein, is the posterior probability of the pollution source position, is the state estimation value, is the a priori probability distribution of the pollution source position, and the a priori probability distribution is obtained based on historical environmental data; Generate a candidate pollution source coordinate set according to the posterior probability.

[0012] In one embodiment, the step of selecting the candidate pollution source coordinate set by maximum likelihood estimation to obtain the target pollution source position includes: Calculate according to the candidate pollution source coordinate set to obtain likelihood values corresponding to multiple pollution source coordinates; Compare multiple likelihood values to obtain the pollution source coordinates corresponding to a preset likelihood value. The specific calculation formula is: ; where is the preset distance of the sensor spacing, is the coordinate of the target pollution source, is the candidate pollution source coordinate, is the number of sensors; Convert according to the pollution source coordinates to obtain the target pollution source position.

[0013] In addition, to achieve the above object, the present application also proposes a pollution source positioning device based on multi-source data fusion. The pollution source positioning device based on multi-source data fusion includes: An acquisition module for acquiring environmental data and meteorological data; A processing module for adaptively filtering the environmental data to obtain filtered environmental data; and also for performing time synchronization processing on the meteorological data to obtain synchronized meteorological data; A fusion module for processing the filtered environmental data and the synchronized meteorological data to obtain pollution source information; A calculation module for calculating according to the pollution source information to obtain an initial state vector of the pollution source; and also for calculating based on the initial state vector of the pollution source to obtain a candidate pollution source coordinate set; A result module for selecting the candidate pollution source coordinate set by maximum likelihood estimation to obtain the target pollution source position.

[0014] In addition, to achieve the above object, the present application also proposes a storage medium. The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the pollution source positioning method based on multi-source data fusion as described above are implemented.

[0015] In addition, to achieve the above object, the present application also provides a computer program product. The computer program product includes a computer program. When the computer program is executed by a processor, the steps of the pollution source positioning method based on multi-source data fusion as described above are implemented.

[0016] This application obtains environmental data and meteorological data, performs adaptive filtering on the environmental data to obtain the filtered environmental data, performs time synchronization on the meteorological data to obtain the synchronized meteorological data, processes the filtered environmental data and the synchronized meteorological data to obtain pollution source information, calculates based on the pollution source information to obtain the initial state vector of the pollution source, calculates based on the initial state vector of the pollution source to obtain a set of candidate pollution source coordinates, and selects the set of candidate pollution source coordinates through maximum likelihood estimation to obtain the target pollution source location. By using sensors to obtain environmental and meteorological data, and through adaptive filtering and time synchronization processing, the filtered environmental data and the synchronized meteorological data are obtained respectively. Then, an improved Kriging interpolation algorithm and adaptive Kalman filtering are used in combination with these data to estimate pollution source information and generate an initial state vector. Based on this vector, the pollution source location is predicted, a set of candidate coordinates is formed, and the target pollution source location is optimized and selected through maximum likelihood estimation, realizing millisecond-level response and high-precision positioning, overcoming the problems of data transmission delay and meteorological disturbance in traditional methods, and improving the accuracy and reliability of real-time positioning of VOCs pollution sources in industrial parks. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a schematic flowchart of the first embodiment of the pollution source location method based on multi-source data fusion of the present application; Figure 2 It is a schematic flowchart of the second embodiment of the pollution source location method based on multi-source data fusion of the present application; Figure 3 It is a schematic flowchart of the third embodiment of the pollution source location method based on multi-source data fusion of the present application; Figure 4 It is a schematic module structure diagram of the pollution source location device based on multi-source data fusion in the first embodiment of the pollution source location method based on multi-source data fusion of the present application; Figure 5 It is a schematic device structure diagram of the hardware operating environment involved in the pollution source location method based on multi-source data fusion in the embodiments of the present application.

[0019] The realization of the purpose, functional features and advantages of the present application will be further described in combination with the embodiments with reference to the drawings. Detailed Embodiments

[0020] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of this application and are not used to limit this application.

[0021] To better understand the technical solutions of this application, the following will be described in detail in combination with the accompanying drawings of the specification and specific implementation manners.

[0022] Under the background of the rapid development of current industrialization, volatile organic compounds (VOCs), as important by-products in the industrial production process, have become one of the main sources of air pollution. These compounds not only directly endanger human health, but also participate in atmospheric photochemical reactions to generate ozone and secondary organic aerosols, which are the key precursors leading to haze and photochemical smog. Industrial parks, as the concentration areas of VOCs emissions, have a particularly significant impact on the air quality of the surrounding environment. However, traditional monitoring methods such as off-line laboratory analysis or centralized calculation models based on fixed monitoring stations are unable to cope when faced with complex industrial emission scenarios, especially in terms of real-time performance and accuracy. These problems are mainly reflected in aspects such as response delay, limited spatial coverage, and poor model adaptability, and cannot meet the need for quickly and accurately locating pollution sources, thus affecting the efficiency and effect of emergency response.

[0023] Therefore, in order to overcome the above problems, this application proposes a method for efficient and accurate pollution source location. The main solution of the embodiments of this application is: obtaining environmental data and meteorological data, performing adaptive filtering processing on the environmental data to obtain the filtered environmental data, performing time synchronization processing on the meteorological data to obtain the synchronized meteorological data, processing the filtered environmental data and the synchronized meteorological data to obtain pollution source information, calculating based on the pollution source information to obtain the initial state vector of the pollution source, calculating based on the initial state vector of the pollution source to obtain a set of candidate pollution source coordinates, and selecting the set of candidate pollution source coordinates through maximum likelihood estimation to obtain the target pollution source location.

[0024] Based on the above, the embodiments of this application also provide a pollution source location method based on multi-source data fusion, referring to Figure 1 , Figure 1 is a schematic flowchart of the first embodiment of the pollution source location method based on multi-source data fusion of this application.

[0025] In this embodiment, the pollution source location method based on multi-source data fusion includes steps S10 to S70: Step S10, obtaining environmental data and meteorological data.

[0026] It should be noted that environmental data is collected by deploying a high-density sensor network. These sensors are distributed at different locations in the industrial park and can monitor various environmental parameters in real time, including VOCs concentration, temperature, humidity, etc. Each sensor is equipped with advanced detection elements, ensuring high-precision and stable data collection. In addition, the sensor network uses wireless communication technology, making data transmission more convenient and efficient, reducing wiring costs and complexity. At the same time, in order to more accurately capture the impact of meteorological conditions on the diffusion of VOCs, detailed meteorological data also needs to be obtained through sensors. This includes but is not limited to key meteorological indicators such as wind speed, wind direction, and air pressure. Meteorological data not only helps to understand how pollutants migrate and disperse in the atmosphere but also provides important background information for subsequent data analysis. For example, under strong wind conditions, pollutants will quickly spread to a farther distance; while under low wind speed or stable weather conditions, pollutants will accumulate in place, forming a local high-concentration area.

[0027] Step S20, perform adaptive filtering on the environmental data to obtain the filtered environmental data.

[0028] It should be noted that in order to ensure the high precision and reliability of the environmental data obtained from the sensor network, the original environmental data needs to be effectively preprocessed. Adaptive filtering can dynamically remove noise and provide more accurate data input, which is crucial for subsequent analysis steps. The original environmental data usually contains measurement errors from the sensors themselves and noise caused by external interference. These noises may come from instantaneous interference factors such as electromagnetic pulses and mechanical vibrations.

[0029] To address these issues, a dynamic adaptive filtering method is adopted. This method is based on the Pauta criterion (3σ principle), and dynamically screens outliers by calculating the mean and standard deviation within a moving window. By default, the threshold is set to 3σ to ensure that at least more than 95% of the normal data is retained. This method can not only effectively exclude instantaneous interference but also automatically adjust the filtering threshold according to the change trend of pollutant concentration, enhancing the robustness and adaptability of the system.

[0030] Furthermore, an adaptive adjustment factor is introduced. This factor is calculated based on historical data statistics and reflects the average deviation of data over a past period. Specifically, historical environmental data is obtained and calculated according to the historical environmental data to obtain the adaptive adjustment factor. The specific formula is: ; where, is the adjustment coefficient, represents the number of historical environmental data, represents the concentration value of the th historical environmental data, is the moving window mean, is the window standard deviation.

[0031] Adaptive filtering is performed on the environmental data based on an adaptive adjustment factor to obtain the filtered environmental data. The specific formula is: ; Where, represents the concentration value of the environmental data, is the coefficient of the Pauta criterion. During the adaptive filtering process, each edge computing node will perform preliminary processing on the sensor data within its responsible area. For the data received at each time point, the moving window mean and the window standard deviation are calculated. The moving window mean refers to the average value of the data calculated within a fixed-size time window. For each newly collected data point, this time window will slide accordingly, always including the latest data sample. For example, if the window size is set to 5 minutes, then for each new data point received, the average value of all data points within the most recent 5 minutes including this data point will be calculated. The window standard deviation is calculated based on the same time window as when calculating the moving window mean. It measures the degree of dispersion of the data points within the window relative to the window mean. A higher standard deviation indicates a larger difference between the data, which may indicate the presence of outliers or significant random noise, while a lower standard deviation indicates that the data is more concentrated and more stable.

[0032] Then, the Pauta criterion is used to determine whether the current reading is an outlier. The Pauta criterion is a statistical method used to identify and eliminate outliers, where represents the standard deviation, means that any data point that deviates from the average value by more than three times the standard deviation is considered an outlier. In this embodiment, the coefficient of the Pauta criterion is set to 3, which means that when a data point deviates from the moving window mean of its window by more than 3 times the window standard deviation, this data point is considered an outlier and should be excluded or corrected. This can ensure that at least 99.7% of the normal data is retained while effectively filtering out extreme outliers.

[0033] Step S30: Perform time synchronization processing on the meteorological data to obtain the synchronized meteorological data.

[0034] It should be noted that due to factors such as the wide coverage of the sensor network, transmission delays between nodes, and local clock deviations, there are time synchronization problems in the directly collected meteorological data. For example, assume there is a sensor network consisting of multiple weather stations, and each weather station collects meteorological parameters such as wind speed and wind direction at different frequencies. Due to different geographical locations, the communication delays between each station and the central server will also vary. If time synchronization processing is not performed, then even for events occurring at the same moment, the timestamps recorded at different stations will not be consistent, thus affecting the accuracy of subsequent data analysis. To solve these problems and ensure that all meteorological data can be compared and analyzed on a unified time basis, time synchronization processing of these data is required.

[0035] First, to achieve precise time synchronization, it is necessary to calculate a correction term based on edge computing. This correction term takes into account the number of sensors and the theoretical synchronization time, reducing the cloud data transmission delay through local calculation, enabling all meteorological data to achieve a response accuracy at the millisecond level. Specifically, local clock information is obtained according to the meteorological data, and the local clock information includes the local clock value and the clock deviation. Based on the local clock information and the theoretical synchronization time value, a calculation is performed to obtain the edge computing correction value. The specific formula is as follows: ; Among them, is the theoretical synchronization time value, is the number of sensors for obtaining meteorological data, is the local clock value.

[0036] Secondly, the core goal of time synchronization processing is to eliminate the differences in data arrival times caused by different transmission paths and the deviations between local clocks. Specifically, each edge computing node will perform preliminary calibration based on the original meteorological data it receives and in combination with the local clock information. The improved time synchronization method introduces edge computing nodes for local clock calibration. By calculating the transmission delay and clock deviation compensation, the data of all sensors are mapped onto a unified time axis. Specifically, the local clock information is calibrated based on the edge computing correction value to obtain the calibrated clock information. The specific formula is as follows: ; Among them, is the delay clock, is the clock deviation, is the edge computing correction value. The delay clock represents the clock offset corresponding to the delay caused by network transmission, while the clock deviation reflects the deviation of the sensor's local clock relative to the standard time. The synchronized meteorological data is obtained according to the calibrated clock information, , where, Represents the synchronized meteorological data, including wind speed and wind direction meteorological variables, represent meteorological data. This means that regardless of where the sensors are located or when the data is collected, it can now be compared and analyzed within a unified time frame.

[0037] Step S40, by processing the filtered environmental data and the synchronized meteorological data, obtain the pollution source information.

[0038] It should be noted that the Kriging interpolation method based on flow field constraints is used to perform spatial interpolation processing on the filtered environmental data. This method not only considers the spatial characteristics of the sensor distribution but also combines the influence of key meteorological factors such as wind speed to optimize the weight allocation, ensuring that the reconstructed pollutant concentration field is unbiased and has the minimum variance.

[0039] Specifically, by analyzing the VOCs concentration at different sensor positions and introducing a wind speed weighting factor, the diffusion pattern of pollutants in the atmosphere can be more accurately simulated, compensating for the data blind spots in the sparse sensor distribution area. Then, the optimized concentration field data is combined with the synchronized meteorological data to obtain the pollution source information.

[0040] Step S50, calculate according to the pollution source information to obtain the initial state vector of the pollution source.

[0041] It should be noted that after obtaining the filtered environmental data and the time-synchronized meteorological data, the next step is to calculate the initial state vector of the pollution source based on this information. This process is very important for accurately estimating the location and release intensity of the VOCs pollution source.

[0042] Furthermore, step S50 includes: identifying based on the target pollutant concentration field in the pollution source information to obtain the concentration gradient value, and selecting the position where the concentration gradient value reaches the preset value as the initial position of the pollution source. Specifically, based on the target pollutant concentration field constructed from the filtered environmental data and the synchronized meteorological data, the concentration gradient values at different positions can be identified. The concentration gradient reflects the rate of change of pollutant concentration with space and usually reaches the maximum value near the pollution source. Select the position where the concentration gradient value reaches the maximum value as the initial position of the pollution source, and obtain the pollutant concentration value at this position as the initial concentration. Due to the existence of various interference factors in the actual environment, such as wind speed, the directly obtained concentration value deviates from the real situation.

[0043] Therefore, it is also necessary to correct the initial position and initial concentration of the pollution source according to the pollutant diffusion path in the pollution source information, obtain the corrected position and corrected concentration, and then combine the corrected position and corrected concentration to obtain the initial state vector of the pollution source. Specifically, by using advanced algorithms such as adaptive Kalman filtering and combining the latest meteorological data (such as wind speed and wind direction), the position and concentration estimation of the pollution source can be dynamically adjusted. For example, under high wind speed conditions, pollutants may spread rapidly, resulting in a deviation in the originally estimated position; while in static and stable weather, pollutants may accumulate in place, causing the concentration value to be higher than expected. By continuously updating the state transition matrix and considering the diffusion and degradation processes of pollutants in the atmosphere, the actual diffusion trajectory of pollutants can be more accurately simulated. Finally, the corrected position and concentration values are combined to form the initial state vector of the pollution source , where x and y respectively represent the coordinates of the pollution source in the two-dimensional plane represents the corrected initial pollutant concentration

[0044] Step S60: Calculate based on the initial state vector of the pollution source to obtain a set of candidate pollution source coordinates

[0045] It should be noted that the calculation based on the initial state vector combined with the Gaussian diffusion model to update the state vector and generate a set of candidate pollution source coordinates is realized through a series of complex mathematical formulas and algorithms, aiming to provide an accurate estimation of the pollution source position

[0046] Furthermore, step S60 also includes: obtaining an updated state vector according to the initial state vector combined with the Gaussian diffusion model. Specifically, according to the initial state vector , the Gaussian diffusion model is used in combination with the state transition matrix and process noise to obtain the updated state vector. The specific formula is as follows ; where represents the state transition matrix, which considers the diffusion and degradation processes of pollutants and is adjusted according to meteorological conditions (such as wind speed and wind direction). The state equation can be written as ; where the vector represents the state vector of the pollution source at time steps, , are the predicted pollution source position coordinates, is the predicted pollution source concentration value. The vector represents the state vector of the pollution source at time steps (multiplying it by the state transition matrix realizes the linear evolution of the state), , Indicates the current pollution source location coordinates, Indicates the initial concentration of the current pollution source, Indicates the pollutant concentration attenuation coefficient, which is determined by the diffusion and degradation processes, Is the degradation rate of the pollutant, which is related to the environmental conditions, Indicates the time step, Indicates the velocity, Indicates the pollution source at time The concentration change rate, Indicates the process noise, which is used to simulate the influence of uncertain factors such as environmental disturbances, Is the state vector at the next time step. Based on the updated state vector for prediction, the prior estimated state vector is obtained. In this step, a wind speed correction amount is introduced to more accurately describe the pollutant diffusion path. The specific formula is: ; Where, Indicates the prior estimated state vector, including the location coordinates of the pollution source and the pollutant concentration, Indicates the wind speed correction amount. Map the prior estimated state vector to obtain the observation vector, and then based on the observation vector through the Kalman gain matrix, obtain the state estimate value. The Kalman gain matrix is expressed as: ; Where, Is the observation vector, Is the observation noise covariance matrix, Is the prediction error covariance matrix. Based on the state estimate value for calculation, obtain the posterior probability. The specific calculation formula is: ; Where, Is the posterior probability of the pollution source location, Is the state estimate value, Is the prior probability distribution of the pollution source location. The above prior probability distribution is obtained based on historical environmental data. According to the posterior probability, a candidate pollution source coordinate set is generated. By comparing the posterior probability values at different locations, several locations with the highest probability can be selected as candidate pollution source coordinates, and screening is carried out based on historical data or regional experience.

[0047] Step S70, select from the candidate pollution source coordinate set through maximum likelihood estimation to obtain the target pollution source location.

[0048] It should be noted that after generating the candidate pollution source coordinate set, the next step is to further screen these coordinates through Maximum Likelihood Estimation (MLE) to determine the most likely location of the target pollution source. This process not only relies on statistical principles but also requires combining actual monitoring data and environmental background information to ensure the accuracy and reliability of the final positioning.

[0049] Specifically, after obtaining the candidate pollution source coordinate set, it is necessary to calculate the likelihood value of each coordinate. The likelihood value reflects the probability of the observed data occurring given that this coordinate is the location of the pollution source. Specifically, for each candidate coordinate, based on its corresponding pollutant concentration field model and actual sensor measurement values, a probability density function can be constructed. This probability density function is usually based on the Gaussian distribution assumption and, considering the existence of observation noise, describes the expected distribution of the measurement values of each sensor node at a specific pollution source location. Then, applying the maximum likelihood principle, the coordinate that maximizes the likelihood function is selected as the location of the target pollution source. This means finding a location where the likelihood of generating the current observed data is the greatest. In actual operation, since there may be multiple local maximum points, directly solving the maximum likelihood estimation may fall into a local optimal solution. Therefore, in practice, iterative optimization algorithms such as the gradient descent method or the Newton method are often used to find the global optimal solution. In addition, a multi-node collaborative verification mechanism can be combined. By setting the kernel radius to measure the spatial consistency of the positioning results of each edge node, the positioning accuracy can be further enhanced. For example, if the supports of multiple independent nodes are concentrated in a certain area, the coordinates within this area are more likely to be the true pollution source location.

[0050] Finally, the location of the target pollution source obtained through the above steps is not only the best estimate based on real-time observation data but also takes into account the influence of various factors such as historical data and meteorological conditions, thus improving the accuracy and robustness of the positioning.

[0051] In this embodiment, environmental data and meteorological data are acquired. The environmental data is processed through adaptive filtering to obtain the filtered environmental data, and the meteorological data is processed for time synchronization to obtain the synchronized meteorological data. By processing the filtered environmental data and the synchronized meteorological data, pollution source information is obtained. Based on the pollution source information, calculations are performed to obtain the initial state vector of the pollution source. Based on the initial state vector of the pollution source, calculations are performed to obtain a candidate pollution source coordinate set. The candidate pollution source coordinate set is selected through maximum likelihood estimation to obtain the target pollution source location. By using sensors to acquire environmental and meteorological data, and through adaptive filtering and time synchronization processing, the filtered environmental data and the synchronized meteorological data are obtained respectively. Then, by combining these data, an improved Kriging interpolation algorithm and adaptive Kalman filtering are used to estimate pollution source information and generate an initial state vector. Based on this vector, the location of the pollution source is predicted, a candidate coordinate set is formed, and the target pollution source location is optimized and selected through maximum likelihood estimation, achieving millisecond-level response and high-precision positioning, overcoming the problems of data transmission delay and meteorological disturbance in traditional methods, and improving the accuracy and reliability of real-time positioning of VOCs pollution sources in industrial parks.

[0052] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar content as in the above-mentioned embodiment one can be referred to the above introduction and will not be elaborated hereinafter. On this basis, please refer to Figure 2 , and step S40 of the pollution source location method based on multi-source data fusion further includes steps S201 to S203: Step S201, by processing the filtered environmental data using a spatial interpolation method, a pollutant concentration field is constructed.

[0053] It should be noted that the obtained filtered environmental data contains the VOCs concentration measurement values at each sensor node. However, due to the limited layout density of the sensor network, directly using the data of these discrete points cannot comprehensively reflect the pollutant concentration distribution in the entire area. Therefore, it is necessary to use a spatial interpolation method to expand these discrete data points into a continuous concentration field. Common interpolation methods include the inverse distance weighting method, the Kriging method, etc. In this embodiment, the Kriging interpolation method based on flow field constraint is adopted. The specific interpolation formula is: ; where, is the pollutant concentration at the th detection point, is the normalized weight coefficient corresponding to the th detection point, and satisfies = 1.

[0054] Furthermore, the Kriging interpolation method based on flow field constraints not only considers the distance relationship between sensor positions but also incorporates the influence of key meteorological parameters such as wind speed and wind direction to optimize weight allocation. Specifically, the optimal weights are determined by calculating the semivariogram, and a pollutant concentration field that is unbiased and has the minimum variance is reconstructed. This method can effectively compensate for the data blind spots in areas with sparse sensor distributions and provide a more accurate background concentration field model. For example, in the case of high wind speeds, pollutants will rapidly diffuse along a specific direction. At this time, relying solely on the distance-weighted method may lead to large errors, while introducing flow field constraints can improve the accuracy of the interpolation results. To further enhance the adaptability and robustness of the model, historical dominant wind speed and wind direction information can also be combined to adjust the interpolation weights. For example, for areas where high pollution concentrations often occur historically, higher weights are assigned to ensure that the model can better reflect the actual pollution situation. This dynamic adjustment mechanism enables high prediction accuracy to be maintained even under complex and variable meteorological conditions. Specifically: ; ; ; where is the spatial scale parameter, is the current position coordinate, is the wind speed at the current position, is the historical dominant wind speed, is the standard deviation of wind speed, is the wind direction at the current position, is the historical dominant wind direction, is the standard deviation of wind direction fluctuation.

[0055] Step S202: Through coupled optimization processing of the pollutant concentration field, the corrected concentration value of the pollutant is obtained.

[0056] It should be noted that the concentration field correction value obtained above uses the Kriging interpolation method with flow field constraints to perform coupled optimization processing on the concentration field. During the coupled optimization processing, the distribution of the concentration field will be dynamically adjusted to ensure its consistency with the actual pollutant diffusion trajectory. Specifically, for each sensor position, the corrected concentration value at that position is calculated based on the current meteorological conditions. This corrected concentration value not only considers the original concentration measured by the sensor but also incorporates the influence of meteorological parameters on pollutant diffusion. For example, under strong wind conditions, pollutants may rapidly diffuse to the downwind area, resulting in a significant decrease in the concentration value in some areas and an increase in the concentration value in other areas due to the accumulation of pollutants. The specific formula is: ; where is the The original concentration of each detection point, =1 Finally, the pollutant concentration value after coupling optimization provides a more realistic and reliable concentration distribution.

[0057] Step S203: Calculate based on the corrected concentration value of the pollutant to obtain the pollution source information.

[0058] It should be noted that after obtaining the pollutant concentration after coupling optimization, this concentration field provides the pollutant distribution in the entire area. Combining the calculated pollutant diffusion path, the trend and influence range of pollutant diffusion from the source can be observed more intuitively. For example, under high wind speed conditions, pollutants may diffuse rapidly along the dominant wind direction; while under low wind speed or stable weather conditions, pollutants may accumulate in place, forming a local high-concentration area. By superimposing these diffusion paths on the concentration field, it can clearly show how pollutants migrate over time and space. Next, using dynamic prediction methods such as adaptive Kalman filtering, combined with the optimized concentration field and diffusion path, further estimate the specific location of the pollution source and its release intensity. To further simulate the diffusion process of pollutants from the source to the sensing point, in this embodiment, a pollution diffusion path model is introduced to fuse the corrected concentration field with possible diffusion paths. By calculating the reverse diffusion probability distribution between the target point and multiple possible pollution sources, combined with the corrected concentration value of the pollutant, a joint field of pollutant concentration - diffusion path is constructed. The specific formula is: ; Wherein, represents the probability that the pollution source location is , is the corrected concentration value of the pollutant at the pollution source location, is the pollutant diffusion path, represents the data fusion function, which is used to integrate the corrected concentration value of the pollutant and the pollutant diffusion path. Finally, based on the pollution source information generated from these comprehensive analysis results, according to the probability distribution calculated from the pollutant information, determine the most likely location of the pollution source. This location not only has the highest probability value but also is highly consistent with the data collected by the sensor network and the diffusion path. This provides important decision-making support for the industrial park, enabling it to identify and respond to potential pollution events in a timely manner, take effective control measures, reduce the risk of environmental pollution, and protect the ecological environment and public health.

[0059] In this embodiment, by means of spatial interpolation and meteorological data correction, an accurate pollutant concentration field and diffusion path are constructed, and the pollution source information is obtained by combining the two, improving the positioning accuracy and enhancing the adaptability to complex meteorological conditions.

[0060] Based on the first embodiment of this application, in the third embodiment of this application, the same or similar content as in the above-mentioned Embodiment 1 can be referred to the above introduction and will not be elaborated hereinafter. On this basis, please refer to Figure 3 The pollution source location step S70 based on multi-source data fusion further includes steps S301 to S303: Step S301, calculate based on the candidate pollution source coordinate set to obtain likelihood values corresponding to multiple pollution source coordinates.

[0061] It should be noted that after generating the candidate pollution source coordinate set, the next step is to calculate the likelihood value corresponding to each coordinate to determine the most likely pollution source location. This process uses statistical methods to evaluate the possibility of different locations as potential pollution sources to ensure the accuracy and reliability of the location result.

[0062] First, for each candidate pollution source coordinate , based on its corresponding pollutant concentration field model and actual sensor measurement data, construct a probability density function , where represents the actual observation data of all sensor nodes. This probability density function is usually assumed to be a Gaussian distribution. Considering the existence of observation noise, it describes the expected distribution of the measurement values of each sensor node at a specific pollution source location. For each candidate pollution source coordinate , the likelihood value can be calculated according to the likelihood function.

[0063] Step S302, compare multiple likelihood values to obtain the pollution source coordinates corresponding to the preset likelihood value.

[0064] It should be noted that after calculating the likelihood values of multiple candidate pollution source coordinates, the next step is to compare these likelihood values to determine the pollution source coordinates corresponding to the preset likelihood value. This process ensures the accuracy and reliability of the final location result through strict statistical analysis and optimization algorithms.

[0065] Specifically, sort the likelihood values of all candidate pollution source coordinates from high to low. Each likelihood value reflects the probability of the observation data occurring when the given coordinate is used as the pollution source location. Usually, the coordinate with the highest likelihood value is the most likely true pollution source location. The specific formula is: ; where is the preset distance of the sensor spacing, is the coordinate of the target pollution source, is the candidate pollution source coordinate, is the number of sensors. However, in actual operation, to improve the robustness and accuracy of positioning, a preset likelihood threshold is usually set. Only when the likelihood value of a certain coordinate exceeds this threshold will it be recognized as a valid pollution source location. Assuming we set a preset likelihood threshold, then we need to find all candidate coordinates whose likelihood values are greater than or equal to the preset likelihood threshold. These coordinates are considered potential pollution source locations and are further verified and analyzed. For example, if the likelihood values of multiple candidate coordinates all exceed the preset threshold, a multi-node collaborative verification mechanism can be combined. By using the kernel radius (1.5 times the sensor spacing in this embodiment) to measure the spatial consistency of the positioning results of each edge node, the coordinates supported by the majority of nodes are selected as the coordinates of the final pollution source location.

[0066] Step S303, perform a conversion based on the pollution source coordinates to obtain the target pollution source location.

[0067] It should be noted that after selecting the coordinates supported by the majority of nodes as the final target pollution source coordinates based on measuring the spatial consistency of the positioning results of each edge node by setting the kernel radius, it is necessary to convert them from the local coordinate system to the global coordinate system. This step usually involves the application of geographic information system (GIS) data to ensure that all positioning information can be compared and analyzed within a unified map or spatial framework. For example, if the sensor nodes are distributed in different geographical regions, the local coordinates provided by each node need to be converted into unified global positioning system (GPS) coordinates to accurately identify the specific geographical location of the pollution source.

[0068] Finally, the target pollution source location after coordinate conversion and verification not only reflects the best estimate of real-time observations but also takes into account the influence of the environmental background and long-term trends, thereby improving the accuracy and robustness of positioning.

[0069] This embodiment realizes accurate and rapid positioning of the pollution source by calculating and comparing the likelihood values of candidate pollution source coordinates, screening the optimal coordinates by using a preset distance of the sensor spacing, and then converting to obtain the target pollution source location, enhancing the response efficiency and accuracy in dealing with environmental pollution.

[0070] Based on the first embodiment of this application, this application also provides a pollution source positioning device based on multi-source data fusion. Please refer to Figure 4 The device includes: An acquisition module 10, configured to acquire environmental data and meteorological data.

[0071] A processing module 20, configured to perform adaptive filtering processing on the environmental data to obtain the filtered environmental data.

[0072] The processing module 20 is further configured to perform time synchronization processing on the meteorological data to obtain the synchronized meteorological data.

[0073] The fusion module 30 is configured to process the filtered environmental data and the synchronized meteorological data to obtain pollution source information.

[0074] The calculation module 40 is configured to calculate based on the pollution source information to obtain an initial state vector of the pollution source.

[0075] The calculation module 40 is further configured to calculate based on the initial state vector of the pollution source to obtain a candidate pollution source coordinate set.

[0076] The result module 50 is configured to select from the candidate pollution source coordinate set through maximum likelihood estimation to obtain the target pollution source location.

[0077] The pollution source location device based on multi-source data fusion provided by this application adopts the pollution source location method based on multi-source data fusion in the above-mentioned embodiment, and can solve the technical problem of how to improve the accuracy of pollution source location in industrial parks. Compared with the prior art, the beneficial effects of the pollution source location device based on multi-source data fusion provided by this application are the same as those of the pollution source location method based on multi-source data fusion provided by the above-mentioned embodiment, and other technical features in the pollution source location device based on multi-source data fusion are the same as the features disclosed in the method of the above-mentioned embodiment, and will not be elaborated here.

[0078] In one embodiment, the processing module 20 is further configured to obtain historical environmental data; calculate based on the historical environmental data to obtain an adaptive adjustment factor; perform adaptive filtering processing on the environmental data based on the adaptive adjustment factor to obtain the filtered environmental data.

[0079] In one embodiment, the processing module 20 is further configured to obtain local clock information according to the meteorological data, where the local clock information includes a local clock value and a clock deviation; calculate based on the local clock information and a theoretical synchronization time value to obtain an edge computing correction value; correct the local clock information based on the edge computing correction value to obtain the corrected clock information; obtain the synchronized meteorological data according to the corrected clock information.

[0080] In one embodiment, the fusion module 30 is further configured to process the filtered environmental data by using a spatial interpolation method to construct a pollutant concentration field; perform coupled optimization processing on the pollutant concentration field to obtain a corrected concentration value of the pollutant; calculate based on the corrected concentration value of the pollutant to obtain pollution source information.

[0081] In one embodiment, the calculation module 40 is further configured to identify based on the target pollutant concentration field in the pollution source information to obtain a concentration gradient value; select a position where the concentration gradient value reaches a preset value as the initial position of the pollution source; obtain the concentration value at the initial position of the pollution source as the initial concentration; correct the initial position and the initial concentration of the pollution source according to the pollutant diffusion path in the pollution source information to obtain the corrected position and the corrected concentration; and combine the corrected position and the corrected concentration to obtain the initial state vector of the pollution source.

[0082] In one embodiment, the calculation module 40 is further configured to obtain an updated state vector according to the initial state vector in combination with the Gaussian diffusion model; perform prediction according to the updated state vector to obtain a priori estimated state vector; map the a priori estimated state vector to obtain an observation vector; obtain a state estimate value based on the observation vector through a Kalman gain matrix; perform calculation based on the state estimate value to obtain a posterior probability; and generate a candidate pollution source coordinate set according to the posterior probability.

[0083] In one embodiment, the result module 50 is further configured to calculate according to the candidate pollution source coordinate set to obtain likelihood values corresponding to multiple pollution source coordinates; compare the multiple likelihood values to obtain the pollution source coordinates corresponding to a preset likelihood value; and convert according to the pollution source coordinates to obtain the target pollution source position.

[0084] The present application provides a pollution source positioning device based on multi-source data fusion. The pollution source positioning device based on multi-source data fusion includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the pollution source positioning method based on multi-source data fusion in the first embodiment above.

[0085] Reference is made below Figure 5 , which shows a schematic structural diagram of a pollution source positioning device based on multi-source data fusion suitable for implementing the embodiments of the present application. The pollution source positioning device based on multi-source data fusion in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5The illustrated pollution source location device based on multi-source data fusion is merely an example and should not impose any limitations on the functions and scope of use of the embodiments of the present application.

[0086] As Figure 5 shown, the pollution source location device based on multi-source data fusion may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 into the random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the pollution source location device based on multi-source data fusion are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Generally, the following can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the pollution source location device based on multi-source data fusion to communicate with other devices wirelessly or wiredly to exchange data. Although various pollution source location devices based on multi-source data fusion are shown in the figure, it should be understood that it is not required to implement or have all the shown ones. More or fewer can be implemented or had alternatively.

[0087] Specifically, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are executed.

[0088] The pollution source positioning device based on multi-source data fusion provided by this application adopts the pollution source positioning method based on multi-source data fusion in the above-mentioned embodiments, and can solve the technical problem of how to improve the accuracy of pollution source positioning in industrial parks. Compared with the prior art, the beneficial effects of the pollution source positioning device based on multi-source data fusion provided by this application are the same as those of the pollution source positioning method based on multi-source data fusion provided by the above-mentioned embodiments, and other technical features in the pollution source positioning device based on multi-source data fusion are the same as those disclosed in the method of the previous embodiment, and will not be elaborated here.

[0089] It should be understood that each part disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0090] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0091] This application provides a computer-readable storage medium with computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the pollution source positioning method based on multi-source data fusion in the above-mentioned embodiments.

[0092] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electricity, magnetism, light, electromagnetic, infrared, or semiconductor devices, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible storage medium that contains or stores a program, and this program can be executed by an instruction, used by a device, or used in combination with it. The program code contained on the computer-readable storage medium can be transmitted using any appropriate storage medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0093] The above computer-readable storage medium can be included in the pollution source location device based on multi-source data fusion; it can also exist alone without being assembled into the pollution source location device based on multi-source data fusion.

[0094] The above computer-readable storage medium carries one or more programs. When the above one or more programs are executed by the pollution source location device based on multi-source data fusion, the pollution source location device based on multi-source data fusion can write computer program code for performing the operations of this application in one or more programming languages or combinations thereof. The above programming languages include object-oriented programming languages - such as Java, Smalltalk, C++; it also includes conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0095] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes may occur in a different order than that marked in the accompanying drawings. For example, two consecutive boxes shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, as well as the combination of boxes in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based means for performing the specified function or operation, or may be implemented by a combination of dedicated hardware and computer instructions.

[0096] The modules described in the embodiments of the present application can be implemented in software or in hardware. In some cases, the name of the module does not constitute a limitation on the unit itself.

[0097] The readable storage medium provided by the present application is a computer-readable storage medium, and the computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned pollution source location method based on multi-source data fusion, which can solve the technical problem of how to improve the accuracy of pollution source location in industrial parks. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as those of the pollution source location method based on multi-source data fusion provided by the above embodiments, and will not be elaborated here.

[0098] The present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the steps of the pollution source location method based on multi-source data fusion as described above.

[0099] The computer program product provided by the present application can solve the technical problem of how to improve the accuracy of pollution source location in industrial parks. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the pollution source location method based on multi-source data fusion provided by the above embodiments, and will not be elaborated here.

[0100] The above are only some embodiments of the present application, and do not limit the patent scope of the present application. All equivalent structural transformations made under the technical concept of the present application by using the content of the specification and drawings of the present application, or directly / indirectly applied in other related technical fields, are included in the patent protection scope of the present application.

Claims

1. A pollution source location method based on multi-source data fusion, characterized in that The method includes: Obtaining environmental data and meteorological data; Performing adaptive filtering on the environmental data to obtain filtered environmental data; Performing time synchronization on the meteorological data to obtain synchronized meteorological data; Processing the filtered environmental data and the synchronized meteorological data to obtain pollution source information; Calculating based on the pollution source information to obtain an initial state vector of the pollution source; Calculating based on the initial state vector of the pollution source to obtain a set of candidate pollution source coordinates; Selecting from the set of candidate pollution source coordinates through maximum likelihood estimation to obtain the location of the target pollution source.

2. The method according to claim 1, wherein The step of performing adaptive filtering on the environmental data to obtain filtered environmental data includes: Obtaining historical environmental data; Calculating based on the historical environmental data to obtain an adaptive adjustment factor, and the specific formula is: ； Among them, is an adjustment coefficient, represents the number of the historical environmental data, represents the th concentration value of the historical environmental data, is the moving window mean, is the window standard deviation; Performing adaptive filtering on the environmental data based on the adaptive adjustment factor to obtain filtered environmental data, and the specific formula is: ； Among them, represents the concentration value of the environmental data, is the coefficient of the Chauvenet's criterion.

3. The method according to claim 1, wherein The step of performing time synchronization on the meteorological data to obtain synchronized meteorological data includes: Obtaining local clock information according to the meteorological data, and the local clock information includes a local clock value and a clock deviation; Calculating based on the local clock information and a theoretical synchronization time value to obtain an edge computing correction value, and the specific formula is: ； Among them, is the theoretical synchronization time value, is the number of sensors for obtaining the meteorological data, is the local clock value; Correcting the local clock information based on the edge computing correction value to obtain corrected clock information, and the specific formula is: ； Among them, is the delay clock, is the clock deviation, is the edge computing correction value; Obtaining synchronized meteorological data according to the corrected clock information, and the specific formula is: ； Among them, represents the synchronized meteorological data, including wind speed and wind direction meteorological variables, represents the said meteorological data.

4. The method according to claim 1, wherein The step of processing the filtered environmental data and the synchronized meteorological data to obtain pollution source information includes: Processing the filtered environmental data by using a spatial interpolation method to construct a pollutant concentration field, and the specific interpolation formula is: ； Among them, is the pollutant concentration at the th detection point, is the normalized weight coefficient corresponding to the th detection point, and satisfies = 1. The specific formula for obtaining the normalized weight coefficient is: ； ； ； Among them, is the spatial scale parameter, is the current position coordinate, is the wind speed at the current position, is the historical dominant wind speed, is the standard deviation of wind speed, is the wind direction at the current position, is the historical dominant wind direction, is the standard deviation of wind direction fluctuation; Performing coupled optimization on the pollutant concentration field to obtain a corrected concentration value of the pollutant, and the specific formula is: ； Among them, is the original concentration of the th detection point; Calculating based on the corrected concentration value of the pollutant to obtain pollution source information.

5. The method according to claim 1, characterized in that The step of calculating based on the pollution source information to obtain an initial state vector of the pollution source includes: Identifying according to the target pollutant concentration field in the pollution source information to obtain a concentration gradient value; Selecting a position where the concentration gradient value reaches a preset value as the initial position of the pollution source; Obtaining the concentration value at the initial position of the pollution source as the initial concentration; Correcting the initial position and the initial concentration of the pollution source according to the pollutant diffusion path in the pollution source information to obtain a corrected position and a corrected concentration; Combining the corrected position and the corrected concentration to obtain an initial state vector of the pollution source.

6. The method according to claim 1, wherein The step of calculating based on the initial state vector of the pollution source to obtain a set of candidate pollution source coordinates includes: Combining the initial state vector with a Gaussian diffusion model to obtain an updated state vector, and the specific formula is: ； Among them, represents the state transition matrix, represents the process noise; Predicting according to the updated state vector to obtain a prior estimated state vector, and the specific formula is: ； Among them, represents the state vector of the prior estimate, represents the wind speed correction amount; Mapping the prior estimated state vector to obtain an observation vector; Based on the observation vector, a state estimate value is obtained through a Kalman gain matrix, and the Kalman gain matrix is expressed as: ； Among them, is the observation vector, is the observation noise covariance matrix, is the prediction error covariance matrix; Based on the state estimate value, calculations are performed to obtain a posterior probability, and the specific calculation formula is: ； Among them, is the posterior probability of the pollution source location, is the state estimate value, is the prior probability distribution of the pollution source location, and the prior probability distribution is obtained based on historical environmental data; According to the posterior probability, a candidate pollution source coordinate set is generated.

7. The method according to claim 1, characterized in that The step of selecting the target pollution source location by performing maximum likelihood estimation on the candidate pollution source coordinate set includes: Calculations are performed according to the candidate pollution source coordinate set to obtain likelihood values corresponding to multiple pollution source coordinates; The likelihood values are compared to obtain the pollution source coordinates corresponding to a preset likelihood value, and the specific calculation formula is: ； Among them, is the preset distance of the sensor spacing, is the coordinate of the target pollution source, is the coordinate of the candidate pollution source, is the number of sensors; According to the pollution source coordinates, a conversion is performed to obtain the target pollution source location.

8. A pollution source positioning device based on multi-source data fusion, characterized in that, The device includes: An acquisition module, configured to acquire environmental data and meteorological data; A processing module, configured to perform adaptive filtering processing on the environmental data to obtain filtered environmental data; and further configured to perform time synchronization processing on the meteorological data to obtain synchronized meteorological data; A fusion module, configured to process the filtered environmental data and the synchronized meteorological data to obtain pollution source information; A calculation module, configured to perform calculations according to the pollution source information to obtain an initial state vector of the pollution source; and further configured to perform calculations based on the initial state vector of the pollution source to obtain a candidate pollution source coordinate set; A result module, configured to select the target pollution source location by performing maximum likelihood estimation on the candidate pollution source coordinate set.

9. A pollution source positioning device based on multi-source data fusion, characterized in that, The device includes: a memory, a processor, and a pollution source location program based on multi-source data fusion stored on the memory and running on the processor. The pollution source location program based on multi-source data fusion is configured to implement the steps of the pollution source location method based on multi-source data fusion according to any one of claims 1-7.

10. A storage medium, characterized in that, A pollution source location program based on multi-source data fusion is stored on the storage medium. When the pollution source location program based on multi-source data fusion is executed by the processor, the steps of the pollution source location method based on multi-source data fusion according to any one of claims 1-7 are implemented.

Images