Dynamic positioning method and system for indoor mobile pollution source

Abstract

The invention provides a dynamic positioning method and system for an indoor mobile pollution source, and relates to the technical field of indoor environment safety, and the method comprises the steps: sequentially inputting a collected pollutant concentration field distribution sequence into a trained dynamic positioning prediction model, and outputting a dynamic position coordinate of the mobile pollution source; the training method of the dynamic positioning prediction model comprises the following steps: simulating and generating a pollutant concentration field distribution sequence of a mobile pollution source in a steady state in real time by utilizing fluid dynamics; sequentially coding the pollutant concentration field distribution sequence into a grayscale image, segmenting the grayscale image into a plurality of sub-images, and screening the sub-images; performing data enhancement on each screened sub-image to generate a plurality of derivative images, and generating derivative position coordinates in the indoor space coordinates; and inputting the plurality of derivative images and the corresponding derivative position coordinates into a neural network model for training to obtain a dynamic positioning prediction model. According to the invention, the problems of strong data dependence, poor dynamic positioning precision and the like existing in pollution source positioning are solved.

Description

Technical Field

[0001] The present invention relates to the field of indoor environmental safety technology, and in particular to a dynamic positioning method and system for indoor mobile pollution sources based on a small sample neural network. Background Art

[0002] In the field of indoor gas pollution source location, traditional static positioning methods generally can only locate fixed locations and rely primarily on sensor networks. Gradient optimization methods based on fixed sensor networks are used to locate pollution sources. This method typically requires dense sensor deployment, and positioning accuracy is limited not only by the number of sensors but also by the quality of the data collected by the sensors. Furthermore, local optimal solutions can lead to error accumulation. Therefore, traditional static positioning technology has obvious limitations.

[0003] Currently, dynamic positioning is being performed using mobile robots with dynamic search capabilities. However, path planning in large-scale or complex airflow environments is inefficient, and real-time performance is difficult to guarantee. Other approaches utilize dynamic models to dynamically locate indoor mobile pollution sources, such as those based on long short-term memory (LSTM) networks. These methods not only rely on continuous time-step data but also incur high computational costs for model training and inference, making them difficult to adapt to mobile pollution sources in discrete scenarios. Furthermore, traditional machine learning methods require extensive experimental data, which is difficult to obtain. Insufficient data leads to low positioning accuracy. Furthermore, most existing methods target static pollution sources and are unable to distinguish the spatial characteristics of mobile pollution sources at different steps, leading to cumulative trajectory prediction errors. These methods are sensitive to spatial layout, have poor cross-scenario generalization capabilities, and struggle to adapt to complex indoor environments of varying sizes and shapes. Summary of the Invention

[0004] In response to the above-mentioned problems, the present invention provides a dynamic positioning method and system for indoor mobile pollution sources based on a small sample neural network, which is used to solve the current problems of pollution source positioning, such as strong dependence on data volume, poor dynamic positioning accuracy, and weak cross-scenario generalization ability.

[0005] In one aspect, the present invention provides a method for dynamically locating an indoor mobile pollution source, the method comprising: Multiple sensors are set up indoors to collect real-time data on the pollutant concentration distribution of the current mobile pollution source throughout the indoor area. The collected pollutant concentration distribution sequence is sequentially input into a trained dynamic positioning prediction model to output the dynamic position coordinates of the mobile pollution source. The training method of the dynamic positioning prediction model includes: Construct an indoor space model, establish indoor space coordinates, and use fluid dynamics real-time simulation to generate the pollutant concentration field distribution sequence of mobile pollution sources in a steady state; Encoding the pollutant concentration field distribution sequence into grayscale images in sequence, dividing the grayscale images into multiple sub-images and screening them to obtain multiple sub-images covering key feature areas of mobile pollution sources; Perform data augmentation on each filtered sub-image to generate multiple derivative images, and generate the derivative position coordinates of each derivative image in indoor space coordinates; Based on the multiple derived images and the corresponding derived position coordinates, multiple sets of training sample data sets are generated, and the multiple sets of training sample data sets are input into the neural network model for training to obtain a dynamic positioning prediction model.

[0006] Furthermore, the convection-diffusion equation simulated in real time using fluid dynamics is: ; in, express time simulated concentration at the location; Indicates the flow field where the pollution source is located; represents the concentration diffusion coefficient of the pollutant; is the source term, indicating that the pollutant Location, in Instantaneous emission at any moment.

[0007] Furthermore, when the pollutant release rate is constant and the pollution source is a point source: ; in, Dirac Function, which means that only when the coordinates are The release rate at position source of pollution.

[0008] Furthermore, a method of generating multiple derivative images is to perform rotation step processing on each sub-image.

[0009] Furthermore, the pixels of the grayscale image for: ; in, It represents the maximum value of the pollutant concentration at all locations in the simulated indoor area; It represents the minimum value of the pollutant concentration at all locations in the simulated indoor area; Indicates that the position coordinates in the simulated indoor area are concentration of pollutants.

[0010] Furthermore, the pollutant concentration field distribution sequence collected at set time intervals is input into the dynamic positioning prediction model to obtain the dynamic position coordinates of multiple mobile pollution sources in sequence; the dynamic position coordinates of the multiple mobile pollution sources obtained in sequence are combined to obtain the dynamic moving path of the mobile pollution source.

[0011] Furthermore, the performance of the dynamic positioning prediction model is improved by using the determination coefficient, the weighted accuracy based on probability density, and the accuracy of the Euclidean distance between the coordinate position predicted by the dynamic positioning prediction model and the actual coordinate position.

[0012] Furthermore, the grayscale image is segmented into a plurality of sub-images based on a grid cutting formula.

[0013] Furthermore, the grid cutting formula is: ; in, Indicates the spatial size, used to dynamically adjust the cropping ratio .

[0014] On the other hand, the present invention provides a dynamic positioning system for indoor mobile pollution sources, comprising a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of any of the above methods.

[0015] In general, the present invention provides a method and system for dynamically locating indoor mobile pollution sources. The technical solution conceived by the present invention can achieve the following beneficial effects: (1) The present invention sequentially inputs the pollutant concentration field distribution sequence collected at set time intervals into the trained dynamic positioning prediction model, outputs the dynamic position coordinates of the mobile pollution source, and models the discrete paths of the mobile pollution source moving at set time intervals to capture the spatial distribution characteristics of the pollution source at different moving stages. This not only achieves high-precision dynamic positioning of the pollution source during movement and locks the dynamic moving position of the mobile pollution source, but also accurately obtains the moving path of the mobile pollution source. This effectively solves the problems of the traditional static positioning method being unable to capture the mobile characteristics, the high computational cost of the existing model, the strong dependence on the amount of data, the poor dynamic positioning accuracy, and the weak cross-scenario generalization ability, and provides a new solution for the efficient monitoring of dynamic pollution sources.

[0016] (2) The present invention increases data samples by segmentation, screening, rotation, etc. to expand small sample data to obtain derivative images, and uses the derivative images and corresponding derivative position coordinates for training; this small sample data enhancement method significantly reduces the dependence on large-scale measured data and improves the cross-scene generalization capability.

[0017] (3) The present invention dynamically adjusts the cropping ratio by adjusting the space size, can adapt to different indoor space sizes, and flexibly adapt to various complex indoor scenes such as factory assembly lines, laboratories, hospital corridors and storage environments with mobile pollution sources. It provides an efficient and reliable solution for the fields of building environmental safety and dynamic pollution monitoring in industrial workshops, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a schematic diagram of the principle of a method and system for dynamically locating indoor mobile pollution sources provided by the present invention; Figure 2 This is a schematic diagram of the training steps of a dynamic positioning prediction model of a dynamic positioning method and system for indoor mobile pollution sources provided by the present invention; Figure 3 This is a schematic diagram of an indoor space model of a dynamic positioning method and system for indoor mobile pollution sources provided by the present invention; Figure 4 This is a schematic diagram comparing the prediction accuracy before and after optimization of a dynamic positioning method and system for indoor mobile pollution sources provided by the present invention; Figure 5 This is a schematic diagram of the verification space of a dynamic positioning method and system for indoor mobile pollution sources provided by the present invention; Figure 6 This is a schematic diagram of the relationship between Euclidean distance and accuracy of a dynamic positioning method and system for indoor mobile pollution sources provided by the present invention. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0021] It should be noted that, in the description of the embodiments of the present invention, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a method, step, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such method, step, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the method, step, or apparatus comprising the element.

[0022] This paper proposes a dynamic positioning method and system for indoor mobile pollution sources based on a small-sample neural network. By modeling the discrete paths that mobile pollution sources follow at set intervals, this method achieves high-precision dynamic positioning of pollution sources during their movement. Specifically, the method includes: installing multiple sensors indoors to collect real-time pollutant concentration distributions of the current mobile pollution source across the entire indoor area; then, inputting the collected pollutant concentration distribution sequence into a trained dynamic positioning prediction model, which outputs the dynamic position coordinates of the mobile pollution source.

[0023] like Figure 1 As shown, 16 sensors are installed indoors to collect the pollutant concentration distribution for the entire indoor area. The pollutant concentration distributions collected by each sensor are combined to obtain the pollutant concentration distribution for the entire indoor area. The data can be collected at a set interval, such as 1 minute, or at a fixed movement step, such as 0.5 meters. The pollutant concentration distribution U1 is collected at time t1, U2 at time t2, and U3 at time t3. Based on the collection times, a pollutant concentration distribution sequence [U1, U2, U3] is obtained. The pollutant concentration distributions in the sequence [U1, U2, U3] are sequentially input into the dynamic positioning prediction model, resulting in A(x1, y1, z1), B(x2, y2, z2), and C(x3, y3, z3). This indicates that the mobile pollution source is at point A at time t1, point B at time t2, and point C at time t3, thereby achieving dynamic positioning of the mobile pollution source.

[0024] Furthermore, the collected pollutant concentration field distribution sequence is input into the dynamic positioning prediction model to obtain the dynamic position coordinates of multiple mobile pollution sources in sequence; the dynamic position coordinates of the multiple mobile pollution sources obtained in sequence are combined to obtain the dynamic moving path of the mobile pollution source.

[0025] For example, if we obtain A(x1, y1, z1), B(x2, y2, z2), and C(x3, y3, z3) in sequence, then combining ABC will reveal that the mobile pollution source first moves from point A to point B, then from point B to point C, thus obtaining the dynamic movement path of the mobile pollution source. It should be noted that the shorter the set duration or the smaller the fixed step size, the more accurate the dynamic movement path of the mobile pollution source.

[0026] Among them, Figure 2 As shown, the training method of the dynamic positioning prediction model includes: Step 101: Construct an indoor space model, establish indoor space coordinates, and use fluid dynamics real-time simulation to generate a pollutant concentration field distribution sequence of a mobile pollution source in a steady state.

[0027] like Figure 3 As shown, an indoor space model of x×y×z=30m×40m×4m is constructed. Fluid dynamics simulation is used to generate the steady-state concentration field distribution of a mobile pollution source along a discrete movement path. The mobile pollution source moves along the discrete movement path; it can pause after each fixed step, such as 0.5m, or after each movement for a set duration, such as 1 minute. The fluid dynamics simulation runs until the diffusion state of the current pollution source is steady. By solving the convection-diffusion equation in the real-time fluid dynamics simulation, the steady-state pollutant concentration field distribution corresponding to each fixed step is generated. For example, a pollutant concentration field distribution is collected for each movement, and 45 pollutant concentration field distributions are obtained after 45 movements.

[0028] As an example, in order to better describe the dynamic diffusion process of mobile pollutants in a fluid, the present invention uses the convection-diffusion equation for real-time simulation of fluid dynamics as follows: ; in, express time simulated concentration at the location; Indicates the flow field where the pollution source is located; represents the concentration diffusion coefficient of the pollutant; is the source term, indicating that the pollutant Location, in Instantaneous emission at any moment.

[0029] Furthermore, when the pollutant release rate is constant and the pollution source is a point source: ; in, Dirac Function, which means that only when the coordinates are The release rate at position source of pollution.

[0030] Step 102: Encode the pollutant concentration field distribution sequence into grayscale images in sequence, divide the grayscale images into multiple sub-images and screen them to obtain multiple sub-images covering key feature areas of the mobile pollution source.

[0031] For example, after generating 45 pollutant concentration field distributions corresponding to 45 pollution source locations, the 45 pollutant concentration field distributions are encoded into grayscale images in sequence according to the generation order. The pixels of the grayscale image reflect the concentration differences through linear mapping, thereby ensuring the significant characteristics of the high-concentration area.

[0032] As an example, the pixels of a grayscale image for: ; in, It represents the maximum value of the pollutant concentration at all locations in the simulated indoor area; It represents the minimum value of the pollutant concentration at all locations in the simulated indoor area; Indicates that the position coordinates in the simulated indoor area are concentration of pollutants.

[0033] As an embodiment, the grayscale image is divided into multiple sub-images based on grid cutting. The grid cutting formula is: ; in, Indicates the spatial size, used to dynamically adjust the cropping ratio .

[0034] It should be noted that the cropping ratio is dynamically adjusted through the grid cutting formula, so that it can adapt to different space sizes and flexibly adapt to various complex scenarios.

[0035] Step 103: Perform data enhancement on each filtered sub-image to generate multiple derivative images, and generate derivative position coordinates of each derivative image in indoor space coordinates.

[0036] As another embodiment, a method of generating multiple derivative images is to perform rotation step processing on each sub-image.

[0037] In order to enhance the data, the present invention preferably divides the grayscale image into multiple sub-images based on the grid cutting formula, screens the sub-images, and rotates the screened sub-images to obtain multiple derivative images, thereby eliminating invalid data and rotating only the sub-images with concentration distribution, thereby increasing the diversity and effectiveness of the sample while avoiding information dispersion or loss.

[0038] For example, the size of the grayscale image is 40×30, based on the grid cutting strategy, such as cropping ratio α =0.75, the grayscale image is segmented into 100 4×3 sub-images. After screening, sub-images in areas where the pollution source has not spread are eliminated, resulting in 45 sub-images covering the key characteristic areas of the pollution source. Finally, by rotating with a step size of θ = 10°, each sub-image generates 36 derivative images, resulting in 45×36=1620 derivative images. This can significantly improve the adaptability of the dynamic positioning prediction model to complex spatial layouts by simulating pollutant concentration distribution sequences under different observation angles.

[0039] By increasing data samples through segmentation, screening, rotation, etc., small sample data is expanded to obtain derivative images, and the derived images and corresponding derivative position coordinates are used for training; this small sample data enhancement method significantly reduces the dependence on large-scale measured data and improves the cross-scene generalization ability.

[0040] Step 104: Generate multiple sets of training sample data sets based on the multiple derived images and the corresponding derived position coordinates, and input the multiple sets of training sample data sets into the neural network model for training to obtain a dynamic positioning prediction model.

[0041] As an embodiment, based on a nine-layer residual network (ResNet-9), an image input branch can be added, and the filtered sub-images can be sequentially input into the image input branch for rotation processing. Spatial features can be extracted through a 4×4 convolution kernel (with 64 channels), and a skip connection mechanism can be used to avoid the gradient vanishing problem. A single fully connected layer (FC=1) outputs position coordinates, thereby ensuring effective training of the deep network.

[0042] The image input branch then performs step encoding on the filtered sub-images. These are converted to corresponding multi-dimensional vectors via an embedding layer, rotated, and spatial features extracted. These features are then concatenated with the image features and fed into a single fully connected layer. For example, 45 sub-images are step-encoded to produce 1, 2, 3, ..., 45, which are then converted to 45-dimensional vectors via an embedding layer.

[0043] By introducing step coding, the present invention enables the dynamic positioning prediction model to distinguish the location characteristics of pollution sources in different movement stages while retaining the efficient feature extraction capability of the original Res Net. In addition, by optimizing network parameters (such as convolution kernel size and activation function) and simplifying the fully connected layer, feature redundancy is reduced and training efficiency is improved.

[0044] Preferably, the present invention utilizes the coefficient of determination, the weighted accuracy based on probability density, and the accuracy of the Euclidean distance between the coordinate position predicted by the dynamic positioning prediction model and the actual coordinate position to improve the performance of the dynamic positioning prediction model.

[0045] Specifically, the coefficient of determination , weighted accuracy based on probability density , and Euclidean distance The formulas are: ; ; ; According to the coefficient of determination , weighted accuracy based on probability density , and Euclidean distance The model parameters of the dynamic positioning prediction model are continuously optimized and adjusted based on the results to improve the accuracy and efficiency of the model.

[0046] Model parameters include cropping ratio , rotation step , convolution kernel size, number of fully connected layers, etc. These parameters have a significant impact on the generalization ability and positioning accuracy of the dynamic positioning prediction model; the optimal parameter combination is determined by comprehensively analyzing experimental data and model performance to improve the accuracy and efficiency of the model.

[0047] To verify the effectiveness of the present invention, experiments and simulations were conducted: in the scenario verification, training data was generated through fluid dynamics simulation, the CNN model was trained, and then verified on the test set, where the data was used for training, testing, and prediction in a ratio of 7:2:1, respectively.

[0048] like Figure 4 As shown in the figure, by comparing with the baseline model (NL=1, KS=1×1), the optimized model performs well on a 45-step linear movement path, with a weighted accuracy improvement of 17%, a single-step positioning accuracy of 88% within a 1-meter radius, and a cross-scene error of <0.3 meters, which is a 13% improvement compared to the static model. The coefficient of determination R²=0.99 indicates that the model fits the data well, and the Euclidean distance is 0.18.

[0049] To verify the generalization ability, Figure 5 and Figure 6 As shown, the present invention constructs a verification space of 10×5×3 meters, and the pollution source moves along a random jump path (8 key points). When the cropping ratio is adjusted to ,When the feature retention rate is 92.7%, the dynamic positioning prediction model can achieve an average positioning error of 0.28 meters without retraining, the positioning accuracy within a radius of 1 meter is 100%, and the Euclidean distance is reduced by 21.43%.

[0050] After training, the dynamic positioning prediction model uses a sliding window mechanism to achieve real-time tracking of dynamic pollution sources. The current steady-state concentration field distribution is collected at regular intervals and input into the dynamic positioning prediction model. This captures the spatial distribution characteristics of pollution sources at different stages of movement, outputs the location coordinates of mobile pollution sources, and optimizes trajectory coherence based on historical prediction results. The elastic weight consolidation (EWC) algorithm is used to incrementally update model parameters to prevent catastrophic forgetting and ensure high model accuracy over long-term operation. Furthermore, through adaptive adjustment of cropping ratios and rotation enhancement strategies, the model can flexibly adapt to different spatial sizes and pollution source movement patterns, significantly expanding its application scope.

[0051] In a second aspect, the present invention provides a dynamic positioning system for indoor mobile pollution sources, comprising a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of any of the aforementioned methods. The technical features of the system are consistent with those of the method and are not further detailed here.

[0052] In summary, this invention combines CFD simulation data with a small-sample-optimized residual network (ResNet) to train a dynamic positioning prediction model. This improves the model's generalization and dynamic positioning accuracy under small sample conditions, enabling efficient and accurate dynamic positioning of mobile pollution sources. Compared with traditional methods, this invention demonstrates significant advantages in positioning accuracy, computation time, and computational complexity.

[0053] It should be noted that for the aforementioned embodiments, for simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0054] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0055] In the several embodiments provided in this application, it should be understood that the disclosed methods or systems can be implemented in other ways. For example, the embodiments described above are merely illustrative, and the division of the units described is merely a logical functional division. In actual implementation, other division methods may be used, such as combining or integrating multiple units or components into another system, or ignoring or not implementing certain features.

[0056] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0057] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0058] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. This computer software product is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of this application.

[0059] Those skilled in the art will appreciate that all or part of the various circuits in the above embodiments may be implemented by instructing related hardware through a program. The program may be stored in a computer-readable memory, which may include a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0060] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure herein, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art that are not recorded in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

[0061] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A dynamic positioning method for indoor mobile pollution sources, characterized in that: The method comprises: Multiple sensors are set up indoors to collect real-time data on the pollutant concentration distribution of the current mobile pollution source throughout the indoor area. The collected pollutant concentration distribution sequence is sequentially input into a trained dynamic positioning prediction model to output the dynamic position coordinates of the mobile pollution source. The training method of the dynamic positioning prediction model includes: Construct an indoor space model, establish indoor space coordinates, and use fluid dynamics real-time simulation to generate the pollutant concentration field distribution sequence of mobile pollution sources in a steady state; Encoding the pollutant concentration field distribution sequence into grayscale images in sequence, dividing the grayscale images into multiple sub-images and screening them to obtain multiple sub-images covering key feature areas of mobile pollution sources; Perform data augmentation on each filtered sub-image to generate multiple derivative images, and generate the derivative position coordinates of each derivative image in indoor space coordinates; Based on the multiple derived images and the corresponding derived position coordinates, multiple sets of training sample data sets are generated, and the multiple sets of training sample data sets are input into the neural network model for training to obtain a dynamic positioning prediction model.

2. A method for dynamic positioning of indoor mobile pollution sources as claimed in claim 1, characterized in that: The convection-diffusion equation simulated in real time using fluid dynamics is: ； in, express time simulated concentration at the location; Indicates the flow field where the pollution source is located; represents the concentration diffusion coefficient of the pollutant; is the source term, indicating that the pollutant Location, in Instantaneous emission at any moment.

3. A method for dynamic positioning of indoor mobile pollution sources as claimed in claim 2, characterized in that: When the pollutant release rate is constant and the pollution source is a point source: ； in, Dirac Function, which means that only when the coordinates are The release rate at position source of pollution.

4. A method for dynamically locating an indoor mobile pollution source as claimed in claim 1, characterized in that: The method of generating multiple derivative images is to perform rotation step processing on each sub-image.

5. A method for dynamically locating an indoor mobile pollution source as claimed in claim 1, characterized in that: Pixels of the grayscale image for: ； in, It represents the maximum value of the pollutant concentration at all locations in the simulated indoor area; It represents the minimum value of the pollutant concentration at all locations in the simulated indoor area; Indicates that the position coordinates in the simulated indoor area are concentration of pollutants.

6. A method for dynamically locating an indoor mobile pollution source as claimed in claim 1, characterized in that: The pollutant concentration field distribution sequence collected at set time intervals is input into the dynamic positioning prediction model to obtain the dynamic position coordinates of multiple mobile pollution sources in sequence; the dynamic position coordinates of the multiple mobile pollution sources obtained in sequence are combined to obtain the dynamic moving path of the mobile pollution source.

7. A method for dynamically locating an indoor mobile pollution source as claimed in claim 1, characterized in that: The performance of the dynamic positioning prediction model is improved by using the coefficient of determination, the weighted accuracy based on probability density, and the accuracy of the Euclidean distance between the coordinate positions predicted by the dynamic positioning prediction model and the actual coordinate positions.

8. A method for dynamically locating an indoor mobile pollution source as claimed in claim 1, characterized in that: The grayscale image is divided into a plurality of sub-images based on a grid cutting formula.

9. A method for dynamically locating an indoor mobile pollution source as claimed in claim 8, characterized in that: The grid cutting formula is: ； in, Indicates the space size, used to dynamically adjust the cropping ratio .

10. A dynamic positioning system for indoor mobile pollution sources, comprising a memory, a processor, and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.

Images