Abnormal odor online monitoring system, method, device and storage medium

Abstract

The present application relates to the technical field of odor monitoring, and particularly relates to an abnormal odor online monitoring system, method, device and storage medium, the steps of the method comprising: acquiring abnormal odor concentration distribution data, personnel walking data and region space data of a monitoring area through a sensor array; analyzing the personnel walking action amplitude and the human body scanning volume proportion relationship through the personnel walking data and the region space data, and determining the airflow disturbance range caused by personnel walking; combining the abnormal odor concentration distribution data to analyze the odor concentration fluctuation degree in the airflow disturbance range and perform odor concentration calibration; and using the calibrated abnormal odor concentration distribution data to perform abnormal odor tracing. The present application dynamically analyzes the personnel walking action amplitude and the scanning volume proportion, accurately identifies the airflow disturbance range, and then calibrates the odor concentration distribution data, thereby effectively improving the accuracy and reliability of abnormal odor tracing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of odor monitoring, and in particular to an abnormal odor online monitoring system, method, device and storage medium. BACKGROUND

[0002] Abnormal odor refers to an odor that brings people stimulating, unpleasant or annoying olfactory perception. The abnormal odor perception in an indoor environment not only affects the living experience but also harms human health. Traditional odor detection methods rely on manual sampling and laboratory analysis, which have defects such as long detection period, limited spatial coverage and untimely response, and are difficult to meet the needs of high spatiotemporal resolution and continuous online monitoring.

[0003] With the development of sensor technology, abnormal odor online monitoring methods based on sensor arrays have gradually become a research hotspot. By deploying multi-point gas sensors in the monitoring area, the odor concentration distribution data at different spatial positions can be collected in real time, thereby realizing abnormal odor tracing. However, in actual application, personnel activities are frequent in the monitoring area, and personnel movement will disturb the local airflow, thereby causing short-term fluctuations or even distortion of the odor concentration data collected by the sensors.

[0004] Existing abnormal odor online monitoring systems ignore the airflow disturbance effect caused by personnel movement, and do not effectively calibrate the sensor data in the disturbed area, resulting in concentration misjudgment and tracing misdirection, which is more obvious in closed or semi-closed spaces.

[0005] To solve the above problems, the present application provides a solution. SUMMARY

[0006] To overcome the above-mentioned defects of the prior art, embodiments of the present application provide an abnormal odor online monitoring system, method, device and storage medium, which accurately identifies the airflow disturbance range by dynamically analyzing the personnel movement action amplitude and the body scanning volume proportion, and calibrates the odor concentration distribution data, thereby effectively eliminating the interference of personnel activities on abnormal odor monitoring and improving the accuracy and reliability of abnormal odor tracing.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] In a first aspect, the present application provides an abnormal odor online monitoring method, comprising the following steps:

[0009] acquiring abnormal odor concentration distribution data, personnel movement data and regional space data of a monitoring area through a sensor array;

[0010] analyzing the personnel movement action amplitude and the relationship between the body scanning volume proportion through the personnel movement data and the regional space data, and determining the airflow disturbance range caused by personnel movement;

[0011] analyze the odor concentration fluctuation degree in the airflow disturbance range combined with the abnormal odor concentration distribution data and perform odor concentration calibration;

[0012] trace the abnormal odor source using the calibrated abnormal odor concentration distribution data.

[0013] Further, the personnel walking action amplitude is analyzed, including:

[0014] acquire the body limb activity data in the personnel walking data and extract the body skeleton key points using the human body posture estimation algorithm;

[0015] quantify the personnel action amplitude by the Euclidean distance change between the extracted skeleton key points and the reference skeleton key points in the static standing state;

[0016] normalize the personnel action amplitude by taking the personnel height as the normalization factor and take the normalized personnel action amplitude mean value as the average action amplitude index of the personnel walking, which is used to quantify the personnel walking action amplitude.

[0017] Further, the human body sweeping volume proportion relationship is analyzed, including:

[0018] calculate the minimum envelope volume of the human body contour at each monitoring moment by the extracted skeleton key points combined with the convex hull algorithm, and perform the set operation on the minimum envelope volumes corresponding to all monitoring moments to obtain the human body sweeping volume;

[0019] take the ratio of the human body sweeping volume to the region space volume of the monitoring region as the human body sweeping volume proportion index, which is used to quantify the relationship between the human body sweeping volume and the monitoring region space volume.

[0020] Further, the airflow disturbance range caused by the personnel walking is determined, including:

[0021] acquire the average action amplitude index and the human body sweeping volume proportion index of the personnel walking;

[0022] perform model parameter mapping by the average action amplitude index and the human body sweeping volume proportion index, the model parameters including the grid update frequency, the grid maximum deformation amount and the dynamic grid region radius;

[0023] construct a dynamic grid model based on the model parameters to perform airflow disturbance analysis and extract the airflow disturbance range.

[0024] Further, the model parameter mapping includes:

[0025] perform model parameter mapping of the grid update frequency and the grid maximum deformation amount by the average action amplitude index, wherein the mapped grid update frequency is f grid=f0×(1+λ1×AI norm ), wherein f0 represents an initial grid update frequency, λ1 represents a grid update frequency mapping weight, f grid represents a mapped grid update frequency, and a maximum deformation variable of the mapped grid is Δ max =Δ0×(1+λ2×AI norm ), wherein Δ0 represents an initial maximum grid deformation variable, λ2 represents a maximum grid deformation variable mapping weight, and Δ max represents a mapped maximum grid deformation variable.

[0026] The human body scanning volume proportion index is used to map model parameters of the dynamic grid area radius, wherein the mapped dynamic grid area radius is R=R0×(1+λ3×V swept ), wherein R0 represents an initial dynamic grid area radius, λ3 represents a dynamic grid area radius mapping weight, and R represents the mapped dynamic grid area radius.

[0027] Further, the analysis of the odor concentration fluctuation degree in the airflow disturbance range and the odor concentration calibration comprise:

[0028] Obtaining abnormal odor concentration distribution data and extracting first local abnormal odor concentration distribution data in the airflow disturbance range and second local abnormal odor concentration distribution data in the non-airflow disturbance range;

[0029] Calculating the abnormal odor concentration coefficient of variation in the airflow disturbance range by using the first local abnormal odor concentration distribution data, which is used to evaluate the odor concentration fluctuation degree in the airflow disturbance range;

[0030] When the abnormal odor concentration coefficient of variation in the airflow disturbance range is greater than or equal to a preset abnormal odor concentration coefficient of variation threshold, the odor concentration distribution data in the airflow disturbance range is calibrated by using the second local abnormal odor concentration distribution data in combination with a spatial interpolation algorithm, and the spatial interpolation algorithm comprises an inverse distance weighted method, a spline interpolation method, a radial basis function method, and a Kriging interpolation method.

[0031] Further, the use of the calibrated abnormal odor concentration distribution data for abnormal odor source tracing comprises:

[0032] Constructing an abnormal odor source tracing model, wherein the abnormal odor source tracing model is any one of a Gaussian inversion model, a reverse Lagrange model, and a computational fluid dynamics inversion model;

[0033] Inputting the calibrated abnormal odor concentration distribution data into the abnormal odor source tracing model to perform abnormal odor source tracing, and determining an abnormal odor release source.

[0034] In a second aspect, the present application provides an abnormal odor online monitoring system, comprising:

[0035] a data acquisition module configured to acquire abnormal odor concentration distribution data, personnel movement data and area space data of a monitoring area through a sensor array;

[0036] a disturbance range determination module configured to analyze the amplitude of personnel movement and the proportion of human body scanning volume based on the personnel movement data and the area space data, and determine the airflow disturbance range caused by personnel movement;

[0037] an odor concentration calibration module configured to analyze the odor concentration fluctuation degree in the airflow disturbance range based on the abnormal odor concentration distribution data, and perform odor concentration calibration;

[0038] an abnormal odor tracing module configured to perform abnormal odor tracing using the calibrated abnormal odor concentration distribution data;

[0039] a control module configured to control the operation of the data acquisition module, the disturbance range determination module, the odor concentration calibration module and the abnormal odor tracing module.

[0040] In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes the abnormal odor online monitoring method by calling the computer program stored in the memory.

[0041] In a fourth aspect, the present application provides a computer readable storage medium storing instructions, which, when executed on a computer, cause the computer to execute the abnormal odor online monitoring method.

[0042] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0043] The present application can effectively identify and quantify the airflow disturbance caused by personnel movement through multi-source sensor information fusion and personnel behavior modeling, and by introducing the average movement amplitude index and the human body scanning volume proportion index of personnel movement, the grid model parameters are dynamically mapped, the extraction accuracy of the airflow disturbance range is improved, the accuracy and reliability of abnormal odor tracing are improved by quantifying the odor concentration fluctuation degree in the airflow disturbance range and performing odor concentration calibration, so that real-time and accurate tracing of abnormal odor in a dynamic environment is realized. BRIEF DESCRIPTION OF DRAWINGS

[0044] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings:

[0045] Figure 1 is the overall flow schematic diagram of the abnormal odor online monitoring method provided by the embodiments of the present application;

[0046] Figure 2 This is a schematic diagram of the structure of the online abnormal odor monitoring system provided in an embodiment of the present invention;

[0047] Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0048] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0049] Please see Figure 1 , Figure 1 This is a schematic diagram of the overall process of the online abnormal odor monitoring method provided in this embodiment of the invention, which specifically includes the following steps:

[0050] Step S110: Acquire abnormal odor concentration distribution data, personnel movement data, and regional spatial data of the monitoring area through a sensor array.

[0051] Step S120: Analyze the amplitude of personnel movement and the ratio of human body scan volume by analyzing personnel movement data and regional spatial data to determine the range of airflow disturbance caused by personnel movement;

[0052] The amplitude of human movement directly affects the intensity and range of airflow disturbance. Larger amplitude movements, such as rapid walking or large arm swings, result in stronger mechanical energy input to the surrounding air, leading to more significant airflow vortices and turbulence. This results in greater air mixing and odor diffusion disturbance over a wider area. Quantifying the amplitude of human movement allows for precise assessment of its dynamic impact on the local airflow field, providing key parameters for subsequent airflow disturbance range analysis. It effectively distinguishes between odor fluctuations caused by natural diffusion and anthropogenic disturbances. The analysis of human movement amplitude includes:

[0053] The limb movement data in the personnel walking data is acquired, and a human posture estimation algorithm is used to extract the skeletal key points of the human body, the human posture estimation algorithm mainly includes a two-dimensional posture estimation algorithm and a three-dimensional posture estimation algorithm, wherein the two-dimensional posture estimation algorithm includes OpenPose, AlphaPose and HRNet, the key skeletal points (such as shoulders, elbows, knees and ankles) of the human body in an image are detected and positioned through a deep neural network, and the two-dimensional posture estimation algorithm has high precision and robustness; the three-dimensional posture estimation algorithm includes VideoPose3D, PoseFormer and VIBE, and further introduces depth estimation or time sequence modeling on the basis of the two-dimensional key points, and the dynamic posture of the human body in a three-dimensional space is reconstructed, thereby providing a reliable data basis for extracting the human body swept volume and the action amplitude and the like parameters;

[0054] The action amplitude of the personnel is quantified by the Euclidean distance between the extracted skeletal key points and the reference skeletal key points when standing still;

[0055] The personnel height is taken as a normalization factor to normalize the personnel action amplitude, and the average value of the normalized personnel action amplitude is taken as an average action amplitude index of the personnel walking, which is used to quantify the personnel walking action amplitude, in an embodiment of the present application, the calculation formula of the average action amplitude index can be:

[0056]

[0057] In the formula, P k (t) represents the spatial position of the skeletal key point k at the tth monitoring moment, P k (0) represents the spatial position of the reference skeletal key point k when standing still, ||P k (t)-P k (0)|| represents the Euclidean distance, T represents the number of monitoring moments, that is, the monitoring duration, K represents the number of skeletal key points, S represents the personnel height as a normalization factor, and AI norm represents the average action amplitude index of the personnel walking.

[0058] The human body swept volume proportion is a proportion of a volume swept by the human body in space to a total volume of a monitoring area, and the index reflects the coverage degree of the human body to the local space gas disturbance, the greater the proportion is, the wider the space range involved in the personnel movement is, and the stronger the gas distribution interference is, so that a larger range of airflow disturbance is easily caused, especially in a closed or semi-closed environment, the high swept volume proportion can destroy the original concentration gradient and airflow structure, and the human body swept volume proportion is analyzed, including:

[0059] The minimum envelope volume of the human body contour at each monitoring time is calculated by the extracted skeleton key points combined with the convex hull algorithm, the minimum envelope volume corresponding to all monitoring times is set operation to obtain the human body swept volume, the convex hull algorithm is a geometric algorithm for calculating the minimum convex hull of a point set, the target is to find a minimum convex polygon or convex polyhedron that just encloses all points in two-dimensional or three-dimensional space, the minimum envelope volume refers to the space envelope of the skeleton key points obtained by human posture estimation in three-dimensional space, forming the minimum convex geometry covering the human body contour, which is used to represent the volume distribution of the human body in space at this time, in an embodiment of the present application, the calculation formula of the human body swept volume can be:

[0060]

[0061] In the formula, P(t) represents the skeleton key point set at the tth monitoring time, The set operation is performed on all monitoring times to obtain the human body swept volume. swept The set operation is performed on all monitoring times to obtain the human body swept volume.

[0062] The ratio of the human body swept volume to the region space volume of the monitoring region is taken as the human body swept volume proportion index, which is used to quantify the relationship between the human body swept volume and the monitoring region space volume.

[0063] The dynamic mesh model is a computational fluid dynamics (CFD) method, which can simulate the dynamic change of flow field structure during the movement of the object, by introducing dynamic mesh that can change in real time with the movement of the object in the simulation area, the dynamic mesh model can accurately capture the influence of non-stationary disturbance behavior such as personnel walking on the surrounding airflow, in the technical scheme of the present application, by obtaining the average action amplitude index of personnel walking and the human body swept volume proportion index, and mapping them as key parameters in the dynamic mesh model, the quantitative correlation between personnel behavior characteristics and airflow disturbance modeling can be realized, making the airflow disturbance analysis more realistic, determining the airflow disturbance range caused by personnel walking, including:

[0064] Obtaining the average action amplitude index of personnel walking and the human body swept volume proportion index;

[0065] Mapping the model parameters by the average action amplitude index and the human body swept volume proportion index, the model parameters include grid update frequency, maximum grid deformation and dynamic mesh region radius;

[0066] Constructing the dynamic mesh model based on the model parameters to analyze the airflow disturbance and extract the airflow disturbance range;

[0067] wherein the model parameter mapping comprises: model parameter mapping of the grid update frequency and the grid maximum deformation variable by the average action amplitude index, wherein the mapped grid update frequency is f grid =f0×(1+λ1×AI norm ), wherein f0 represents an initial grid update frequency, the initial grid update frequency corresponds to an action frequency when a human body walks slowly (such as standing breathing or small-step walking), and the initial grid update frequency can be obtained by counting the step frequency or arm swing frequency by an action capture device, λ1 represents a grid update frequency mapping weight, f grid represents the mapped grid update frequency, and the mapped grid maximum deformation variable is Δ max =Δ0×(1+λ2×AI norm ), wherein Δ0 represents an initial grid maximum deformation variable, the initial grid maximum deformation variable can be selected as a grid deformation variable corresponding to a maximum gas concentration gradient in the monitoring area, λ2 represents a grid maximum deformation variable mapping weight, Δ max represents the mapped grid maximum deformation variable, in an embodiment of the present application, f0=1 Hz, λ1=2, Δ0=0.01 m, and λ2=2;

[0068] model parameter mapping of the dynamic grid region radius by the human body scanning volume proportion index, wherein the mapped dynamic grid region radius is R=R0×(1+λ3×V swept ), wherein R0 represents an initial dynamic grid region radius, the initial dynamic grid region radius can be determined by a gas flow disturbance radius when the human body is static, λ3 represents a dynamic grid region radius mapping weight, and R represents the mapped dynamic grid region radius, in an embodiment of the present application, R0=0.3 m and λ3=0.5.

[0069] Step S130: analyzing the odor concentration fluctuation degree in the gas flow disturbance range and performing odor concentration calibration in combination with the abnormal odor concentration distribution data;

[0070] By dividing the odor concentration data in the gas flow disturbance area and the non-disturbance area, and introducing the variation coefficient to quantify the disturbance intensity, the abnormal concentration fluctuation caused by personnel activities can be effectively identified, when the disturbance exceeds the threshold, the concentration data in the surrounding stable area is used as a reference to perform calibration by using a spatial interpolation algorithm, which has the following advantages: (1) the variation coefficient objectively reflects the dispersion degree of the odor concentration in the disturbance area, and provides a quantitative basis for whether calibration is needed; (2) the spatial interpolation method is selected, which conforms to the spatial continuity principle of gas diffusion, and ensures that the calibrated concentration field is physically self-consistent; (3) the dynamic zoning processing takes into account the reliability and integrity of the monitoring data, and lays a foundation for subsequent accurate tracing; analyzing the odor concentration fluctuation degree in the gas flow disturbance range and performing odor concentration calibration, comprising:

[0071] Obtain abnormal odor concentration distribution data and extract first local abnormal odor concentration distribution data in the airflow disturbance range and second local abnormal odor concentration distribution data in the non-airflow disturbance range;

[0072] Calculate the abnormal odor concentration variation coefficient in the airflow disturbance range through the first local abnormal odor concentration distribution data, for evaluating the odor concentration fluctuation degree in the airflow disturbance range;

[0073] When the abnormal odor concentration variation coefficient in the airflow disturbance range is greater than or equal to a preset abnormal odor concentration variation threshold, calibrate the odor concentration distribution data in the airflow disturbance range by using the second local abnormal odor concentration distribution data combined with a spatial interpolation algorithm, the spatial interpolation algorithm includes inverse distance weighting method, spline interpolation method, radial basis function method and Kriging interpolation method, in an embodiment of the present application, the preset abnormal odor concentration variation threshold is obtained by: obtaining abnormal odor concentration distribution data, personnel walking data and regional space data to construct a data set, substituting into the calculation of the abnormal odor concentration variation coefficient, and obtaining the judgment result of the expert on the odor concentration fluctuation degree, inputting the abnormal odor concentration variation coefficient and the judgment result into the fitting software, and outputting the preset abnormal odor concentration variation threshold that meets the maximum judgment accuracy;

[0074] When the abnormal odor concentration variation coefficient in the airflow disturbance range is less than the preset abnormal odor concentration variation threshold, the first local abnormal odor concentration distribution data is retained.

[0075] Step S140: abnormal odor tracing is performed by using the calibrated abnormal odor concentration distribution data;

[0076] By introducing a mature abnormal odor tracing model, on the basis of the odor concentration disturbance calibration, the traceability analysis is performed, the accuracy and interpretability of the traceability result are ensured, the abnormal odor tracing is performed by using the calibrated abnormal odor concentration distribution data, including:

[0077] An abnormal odor tracing model is constructed, the abnormal odor tracing model is any one of a Gaussian inversion model, a reverse Lagrange model and a computational fluid dynamics inversion model;

[0078] The calibrated abnormal odor concentration distribution data is input into the abnormal odor tracing model to perform abnormal odor tracing, and the abnormal odor release source is determined.

[0079] Please refer to Figure 2 , Figure 2 is a structural schematic diagram of an abnormal odor online monitoring system provided by an embodiment of the present application, the present embodiment provides an abnormal odor online monitoring system, which comprises:

[0080] The data acquisition module 210 is configured to acquire abnormal odor concentration distribution data, personnel movement data and area space data of the monitoring area through the sensor array.

[0081] The disturbance range determination module 220 is configured to analyze the action amplitude of personnel movement and the volume proportion relationship of human body sweeping through the personnel movement data and the area space data, and determine the airflow disturbance range caused by personnel movement.

[0082] The odor concentration calibration module 230 is configured to analyze the odor concentration fluctuation degree in the airflow disturbance range in combination with the abnormal odor concentration distribution data and perform odor concentration calibration.

[0083] The abnormal odor tracing module 240 is configured to perform abnormal odor tracing by using the calibrated abnormal odor concentration distribution data.

[0084] The control module 250 is configured to control the operation of the data acquisition module, the disturbance range determination module, the odor concentration calibration module and the abnormal odor tracing module.

[0085] In an embodiment of the present application, the disturbance range determination module 220 is configured to analyze the action amplitude of personnel movement and the volume proportion relationship of human body sweeping through the personnel movement data and the area space data, and determine the airflow disturbance range caused by personnel movement.

[0086] The limb activity data in the personnel movement data is acquired and the human body skeleton key points are extracted by using the human body posture estimation algorithm;

[0087] The action amplitude of personnel is quantified by the Euclidean distance change between the extracted skeleton key points and the reference skeleton key points when standing still;

[0088] The action amplitude of personnel is normalized by taking the height of personnel as a normalization factor, and the average action amplitude index of personnel movement after normalization is taken as the average action amplitude index of personnel movement, which is used to quantify the action amplitude of personnel movement;

[0089] The volume proportion relationship of human body sweeping is analyzed, including:

[0090] The minimum envelope volume of human body contour at each monitoring time is calculated by using the extracted skeleton key points and the convex hull algorithm, and the union operation is performed on the minimum envelope volumes corresponding to all monitoring times to obtain the human body sweeping volume;

[0091] The ratio of the human body sweeping volume to the area space volume of the monitoring area is taken as the human body sweeping volume proportion index, which is used to quantify the relationship between the human body sweeping volume and the monitoring area space volume;

[0092] The airflow disturbance range caused by personnel movement is determined, including:

[0093] An average motion amplitude index and a human body scanning volume proportion index of the moving personnel are acquired;

[0094] Model parameters are mapped through the average motion amplitude index and the human body scanning volume proportion index, and the model parameters include a grid update frequency, a grid maximum deformation variable and a dynamic grid region radius;

[0095] A dynamic grid model is constructed based on the model parameters to analyze the air flow disturbance and extract an air flow disturbance range;

[0096] The model parameters are mapped, including:

[0097] The grid update frequency and the grid maximum deformation variable are mapped through the average motion amplitude index, wherein the mapped grid update frequency is f grid =f0×(1+λ1×AI norm ), wherein f0 represents an initial grid update frequency, λ1 represents a grid update frequency mapping weight, f grid represents the mapped grid update frequency, and the mapped grid maximum deformation variable is Δ max =Δ0×(1+λ2×AI norm ), wherein Δ0 represents an initial grid maximum deformation variable, λ2 represents a grid maximum deformation variable mapping weight, and Δ max represents the mapped grid maximum deformation variable;

[0098] The dynamic grid region radius is mapped through the human body scanning volume proportion index, wherein the mapped dynamic grid region radius is R=R0×(1+λ3×V swept ), wherein R0 represents an initial dynamic grid region radius, λ3 represents a dynamic grid region radius mapping weight, and R represents the mapped dynamic grid region radius.

[0099] In an embodiment of the present application, the smell concentration calibration module 230 is used to analyze the smell concentration fluctuation degree in the air flow disturbance range in combination with the abnormal smell concentration distribution data and perform smell concentration calibration, and the analysis of the smell concentration fluctuation degree in the air flow disturbance range and the smell concentration calibration include:

[0100] Abnormal smell concentration distribution data are acquired, and first local abnormal smell concentration distribution data in the air flow disturbance range and second local abnormal smell concentration distribution data in a non-air flow disturbance range are extracted;

[0101] An abnormal smell concentration variation coefficient in the air flow disturbance range is calculated through the first local abnormal smell concentration distribution data, and is used to evaluate the smell concentration fluctuation degree in the air flow disturbance range;

[0102] When the abnormal odor concentration variation coefficient in the air flow disturbance range is greater than or equal to the preset abnormal odor concentration variation threshold, the odor concentration distribution data in the air flow disturbance range is calibrated by using the second local abnormal odor concentration distribution data in combination with a spatial interpolation algorithm, and the spatial interpolation algorithm includes an inverse distance weighting method, a spline interpolation method, a radial basis function method and a Kriging interpolation method.

[0103] In an embodiment of the present application, the abnormal odor source tracing module 240 is configured to trace the abnormal odor source by using the calibrated abnormal odor concentration distribution data, and the tracing of the abnormal odor source by using the calibrated abnormal odor concentration distribution data includes:

[0104] constructing an abnormal odor source tracing model, the abnormal odor source tracing model being any one of a Gaussian inversion model, a reverse Lagrange model and a computational fluid dynamics inversion model;

[0105] inputting the calibrated abnormal odor concentration distribution data into the abnormal odor source tracing model to trace the abnormal odor source, and determining the abnormal odor release source.

[0106] The above steps about the implementation of the respective functions of the parameters and the unit modules in the abnormal odor online monitoring system of the present application can refer to the parameters and steps in the embodiments of the abnormal odor online monitoring method, and will not be repeated here.

[0107] Please refer to Figure 3 The embodiment of the present application also provides an electronic device 300, which includes a memory 310, a processor 320 and a communication bus 330; the memory 310 and the processor 320 are connected through the communication bus 330. The memory 310 stores instructions that can be loaded and executed by the processor 320 to implement the abnormal odor online monitoring method provided in the above embodiments.

[0108] The memory 310 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 310 can include a storage program area and a storage data area, wherein the storage program area can store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the abnormal odor online monitoring method provided in the above embodiments, etc.; the storage data area can store data involved in the abnormal odor online monitoring method provided in the above embodiments, etc.

[0109] The processor 320 can include one or more processing cores. The processor 320 performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 310, calling data stored in the memory 310. The processor 320 can be at least one of an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a Central Processing Unit (CPU), a controller, a microcontroller, and a microprocessor. It can be understood that, for different devices, the electronic device for implementing the functions of the processor 320 described above can also be other, and the embodiments of the present application are not specifically limited.

[0110] The communication bus 330 can include a path for transmitting information between the above components. The communication bus 330 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 330 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 3 Only one double-headed arrow is used in the middle, but it does not mean that there is only one bus or one type of bus.

[0111] The embodiments of the present application provide a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to perform the abnormal smell online monitoring method provided by the above embodiments.

[0112] In the embodiments of the present application, the computer readable storage medium can be a tangible device that keeps and stores instructions for use by an instruction execution device. The computer readable storage medium can be, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the above. Specifically, the computer readable storage medium can be a portable computer diskette, a hard disk, a U disk, a random access memory (RAM), a read only memory (ROM), an erasable programmable read only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, and any combination of the above.

[0113] The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0114] The above description is merely preferred embodiments of the present application and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above application concept. For example, the above features can be replaced with technical features with similar functions applied in the present application (but not limited to) to form technical solutions.

Claims

1. An online method for monitoring abnormal odors, characterized in that, Includes the following steps: Data on abnormal odor concentration distribution, personnel movement, and spatial data of the monitored area are acquired through a sensor array. By analyzing personnel movement data and regional spatial data, the range of airflow disturbances caused by personnel movement can be determined based on the relationship between the amplitude of personnel movement and the proportion of human body scan volume. The degree of odor concentration fluctuation within the airflow disturbance range was analyzed by combining abnormal odor concentration distribution data, and odor concentration calibration was performed. The source of abnormal odors is traced using calibrated abnormal odor concentration distribution data. The analysis of odor concentration fluctuations within the airflow disturbance range and the calibration of odor concentration include: Acquire abnormal odor concentration distribution data and extract the first local abnormal odor concentration distribution data within the airflow disturbance range and the second local abnormal odor concentration distribution data outside the airflow disturbance range; The coefficient of variation of abnormal odor concentration within the airflow disturbance range is calculated using the first local abnormal odor concentration distribution data to assess the degree of odor concentration fluctuation within the airflow disturbance range. When the coefficient of variation of abnormal odor concentration within the airflow disturbance range is greater than or equal to the preset abnormal odor concentration variation threshold, the odor concentration distribution data within the airflow disturbance range is calibrated using the second local abnormal odor concentration distribution data combined with a spatial interpolation algorithm. The spatial interpolation algorithm includes inverse distance weighting, spline interpolation, radial basis function, and kriging interpolation.

2. The online monitoring method for abnormal odors according to claim 1, characterized in that, The analysis of the range of motion of the personnel includes: Acquire limb movement data from personnel walking data and extract human skeletal key points using human pose estimation algorithms; The range of human movement is quantified by the change in Euclidean distance between the extracted skeletal key points and the reference skeletal key points when standing still. Personnel height is used as a normalization factor to normalize the range of motion of personnel, and the mean of the normalized range of motion is used as the average range of motion index of personnel walking, which is used to quantify the range of motion of personnel walking.

3. The online monitoring method for abnormal odors according to claim 1, characterized in that, The analysis of the human body scan volume ratio includes: The minimum envelope volume of the human body contour at each monitoring time is calculated by combining the extracted skeletal key points with the convex hull algorithm. The human body sweep volume is obtained by performing a union operation on the minimum envelope volumes corresponding to all monitoring times. The ratio of the human body scan volume to the spatial volume of the monitoring area is used as the human body scan volume ratio index to quantify the relationship between the human body scan volume and the spatial volume of the monitoring area.

4. The online monitoring method for abnormal odors according to claim 1, characterized in that, The determination of the range of airflow disturbance caused by personnel movement includes: Obtain the average movement amplitude index and the human body scan volume percentage index of personnel movement; Model parameters are mapped using the average motion amplitude index and the human body scanning volume percentage index. The model parameters include the mesh update frequency, the maximum mesh deformation, and the dynamic mesh region radius. A dynamic mesh model is constructed based on the model parameters to analyze airflow disturbances and extract the range of airflow disturbances.

5. The online monitoring method for abnormal odors according to claim 4, characterized in that, The process of mapping model parameters includes: The model parameters are mapped to the mesh update frequency and the maximum mesh deformation using the average action amplitude exponent, where the mapped mesh update frequency is f. grid =f0×(1+λ1×AI) norm In the formula, f0 represents the initial grid update frequency, λ1 represents the grid update frequency mapping weight, and f grid This represents the update frequency of the mapped mesh, and the maximum deformation of the mapped mesh is Δ. max =Δ0×(1+λ2×AI) norm In the formula, Δ0 represents the maximum deformation of the initial mesh, λ2 represents the mapping weight of the maximum deformation of the mesh, and Δ max This represents the maximum deformation of the mapped mesh; The model parameters of the dynamic mesh region radius are mapped using the human body scan volume percentage index, where the mapped dynamic mesh region radius is R = R0 × (1 + λ3 × V) swept In the formula, R0 represents the initial dynamic grid region radius, λ3 represents the dynamic grid region radius mapping weight, and R represents the mapped dynamic grid region radius.

6. The online monitoring method for abnormal odors according to claim 1, characterized in that, The method of tracing the source of abnormal odors using calibrated abnormal odor concentration distribution data includes: Construct an abnormal odor source tracing model, which can be any one of the Gaussian inversion model, the inverse Lagrange model, and the computational fluid dynamics inversion model; The calibrated abnormal odor concentration distribution data is input into the abnormal odor tracing model to trace the source of the abnormal odor and determine the source of the abnormal odor release.

7. An online monitoring system for abnormal odors, applied to the online monitoring method for abnormal odors according to any one of claims 1-6, characterized in that, The system includes: The data acquisition module is used to acquire data on the distribution of abnormal odor concentrations, personnel movement data, and regional spatial data in the monitoring area through a sensor array; The disturbance range determination module is used to determine the range of airflow disturbances caused by personnel movement by analyzing the amplitude of personnel movement and the ratio of human body scan volume through personnel movement data and regional spatial data. The odor concentration calibration module is used to analyze the degree of odor concentration fluctuation within the airflow disturbance range by combining abnormal odor concentration distribution data and to perform odor concentration calibration. The abnormal odor tracing module is used to trace the source of abnormal odors using calibrated abnormal odor concentration distribution data. The control module is used to control the operation of the data acquisition module, the disturbance range determination module, the odor concentration calibration module, and the abnormal odor tracing module; The analysis of odor concentration fluctuations within the airflow disturbance range and the calibration of odor concentration include: Acquire abnormal odor concentration distribution data and extract the first local abnormal odor concentration distribution data within the airflow disturbance range and the second local abnormal odor concentration distribution data outside the airflow disturbance range; The coefficient of variation of abnormal odor concentration within the airflow disturbance range is calculated using the first local abnormal odor concentration distribution data to assess the degree of odor concentration fluctuation within the airflow disturbance range. When the coefficient of variation of abnormal odor concentration within the airflow disturbance range is greater than or equal to the preset abnormal odor concentration variation threshold, the odor concentration distribution data within the airflow disturbance range is calibrated using the second local abnormal odor concentration distribution data combined with a spatial interpolation algorithm. The spatial interpolation algorithm includes inverse distance weighting, spline interpolation, radial basis function, and kriging interpolation.

8. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes the online abnormal odor monitoring method as described in any one of claims 1-6 by calling the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that, The device stores instructions that, when executed on a computer, cause the computer to perform the online abnormal odor monitoring method as described in any one of claims 1-6.

Images