A wide-area gas traceability method, device, medium and product based on multi-level sensing asynchronous cooperative triggering

Abstract

The application discloses a wide-area gas tracing method and device based on multi-stage sensing asynchronous cooperative triggering, a medium and a product, relates to the field of wide-area gas leakage tracing, and comprises the following steps: adaptively performing three-dimensional grid division on a region to be measured; detecting environmental background information; adopting a random walk strategy and introducing a mechanism for avoiding heavy areas to find a smoke plume; determining a threshold value for binary detection according to the mean value of the environmental background noise, and performing binary calculation on the sampling data; determining a near-field region according to the binary calculation result; tracking the concentration according to the near-field region, and performing real-time near-field state evaluation; adjusting the size of the near-field region according to the evaluation result; continuously judging whether to exit the concentration detection of the near-field region; when the concentration detection of the near-field region is switched to a far-field region, entering a binary detection cycle; and determining a concentration peak value according to the concentration detection data of all the near-field regions. The application can realize efficient, accurate and low-cost pollution source positioning in a complex wide-area environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wide-area gas leakage tracing, and in particular to a wide-area gas tracing method, device, medium and product based on multi-level sensing asynchronous cooperative triggering. BACKGROUND

[0002] In the field of wide-area gas leakage tracing (such as industrial area environmental monitoring, emergency response, and pollution source investigation), the existing technology mainly relies on a single type of gas sensor system for detection and positioning. Such systems usually have the following significant problems:

[0003] (1) High energy consumption and cost bottleneck In order to achieve effective tracing, the system often needs to continuously operate high-precision and high-selectivity gas sensors. Such sensors have high power consumption, which significantly shortens the endurance time of the carrying platform (such as a drone), and the overall operation and maintenance costs of the system are high.

[0004] (2) Difficulty in balancing detection accuracy and efficiency: Although the single high-precision sensor strategy has relatively high positioning accuracy, it has slow response speed and limited coverage, making it difficult to quickly locate the position of the pollution source in a large area, and is inefficient. Although the single low-precision sensor strategy has low precision (or cross-sensitivity type) sensors, it has fast response, low cost, and is easy to deploy in a large area, but it has poor selectivity and is easily disturbed by the environment background, resulting in high false positive rate, high false negative rate, and insufficient positioning accuracy. The commonly used binary region search method cannot distinguish the differences between near-field and far-field environments.

[0005] (3) Low resource utilization efficiency: Existing methods usually use a fixed grid division strategy (such as a uniform grid) for region search, which cannot dynamically adjust the detection resolution according to the size of the area to be measured and the environmental complexity, resulting in over-sampling in open areas or insufficient sampling in complex areas, causing waste of time and resources.

[0006] (4) Lack of dynamic coordination mechanism: Existing technologies fail to effectively combine the fast large-scale scanning capability of low-precision sensors with the accurate positioning capability of high-precision sensors, and lack an intelligent linkage mechanism that dynamically triggers and converts different sensor working modes according to environmental conditions (such as plume discovery and concentration gradient changes). In mobile platform (such as a drone) applications, how to achieve fast, accurate, and robust wide-area gas tracing positioning while ensuring energy efficiency has become a key technical bottleneck that needs to be broken through.

[0007] Therefore, there is an urgent need for a new gas tracing method that can effectively solve the problems of high energy consumption, low precision, resource waste, and lack of intelligent coordination mechanism, and achieve efficient, accurate, and low-cost pollution source positioning in complex wide-area environments. SUMMARY

[0008] The application aims to provide a wide-area gas tracing method, device, medium and product based on multi-level sensing asynchronous cooperative triggering, which can realize efficient, accurate and low-cost pollution source positioning in a complex wide-area environment.

[0009] To achieve the above-mentioned purpose, the application provides the following solutions.

[0010] In a first aspect, the application provides a wide-area gas tracing method based on multi-level sensing asynchronous cooperative triggering, which comprises the following steps:

[0011] Adopting a three-dimensional grid division method and an adaptive grid mechanism to perform adaptive three-dimensional grid division on the to-be-tested region to obtain a grid division result;

[0012] Detecting environmental background information according to the grid division result and the sampling data to obtain a mean value of the environmental background noise;

[0013] Regarding the to-be-tested region as a far field, adopting a random walk strategy and introducing a weight-avoiding mechanism to perform plume discovery;

[0014] Determining a threshold value for binary detection according to the mean value of the environmental background noise; and performing binary calculation on the sampling data by using the threshold value for binary detection;

[0015] Determining a near-field region based on the number of point positions of the measured data points according to the binary calculation result;

[0016] Tracking the concentration according to the near-field region, performing real-time near-field state evaluation by using an improved gradient calculation method and a relative boundary distance, and adjusting the size of the near-field region according to the evaluation result; and continuously judging whether to exit the concentration detection of the near-field region;

[0017] When the concentration detection of the near-field region is switched to the far-field region, adopting an improved Gaussian mixture model to retain historical distribution characteristics, and entering a binary detection cycle;

[0018] Determining whether to re-enter the near-field region according to real-time binary detection data;

[0019] Determining a concentration peak value according to the concentration detection data of all the near-field regions; and taking the concentration peak value as a source position.

[0020] Optionally, the adaptive three-dimensional grid division on the to-be-tested region by using the three-dimensional grid division method and the adaptive grid mechanism to obtain the grid division result specifically comprises the following steps:

[0021] Determining the volume of the to-be-tested region according to the three-dimensional size of the to-be-tested region;

[0022] Determining the size d of the cubic grid by using the formula ​

[0023] Using the cubic mesh size, an adaptive 3D mesh is generated for the area to be measured, and the mesh generation result is obtained.

[0024] Where V is the volume of the region to be measured, V threshold V is the volume threshold. max For the large spatial domain threshold, V min For the small spatial domain threshold, d max For the maximum grid size, d min α is the minimum grid size, α is the size attenuation coefficient, and k is the linear adjustment coefficient.

[0025] Optionally, the step of determining the threshold for binary detection based on the mean of the ambient background noise, and performing binary calculations on the sampled data using the threshold for binary detection, specifically includes:

[0026] The threshold T for binary detection is determined using the formula T = μ + 2σ;

[0027] Using formula Determine the output of the binary calculation;

[0028] Where μ is the mean value of the ambient background noise. x i Let be the i-th sampled data, n be the number of sampled data, and σ be the standard deviation. s j Let j be the j-th data point sampled in real time.

[0029] Optionally, based on the binary calculation results, the near-field region is determined according to the number of measured data points, specifically including:

[0030] Based on the binary calculation results, determine the set of coordinates of the data points where the gas was detected;

[0031] The spatial distribution characteristics of gas points are determined based on the set of point coordinates; the spatial distribution characteristics of gas points include: spatial range and point set density;

[0032] Based on the spatial distribution characteristics of gas points, using the formula Determine the adaptive near-field size D; where R x R y and R z Let ρ be the spatial range, ρ be the point set density, and β be the density compensation factor.

[0033] Based on the adaptive near-field size, the directional normalized range weighting method is used to adjust the gas distribution directional characteristics.

[0034] The near-field region is determined based on the gas distribution direction characteristics.

[0035] Optionally, the concentration tracking according to the near-field region adopts an improved gradient calculation method and a relative boundary distance to perform real-time near-field state evaluation; and the near-field region size is adjusted according to the evaluation result; specifically including:

[0036] The concentration slope slope is determined by using the formula i ;

[0037] The boundary distance border dist is determined by using the formula

[0038] The near-field region size is adjusted according to the concentration slope, the boundary distance, and the number of 0-concentration data points detected in the current near-field region;

[0039] Wherein, C i ,C i+1 is the concentration of data point i and data point i+1, is the three-dimensional coordinate of sampling point i, is the three-dimensional coordinate of sampling point i+1, ||·|| is the Euclidean distance, D k is the length of each axis of the near-field region, k=x, y, z, x, y, z are coordinate axes, p k is the current UAV position, c k is the center of each axis length of the near-field region.

[0040] Optionally, when the concentration detection of the near-field region is switched to the far-field region, the improved Gaussian mixture model is used to retain historical distribution characteristics, and a binary detection cycle is entered; specifically including:

[0041] The average distribution density of the last three near-field regions is obtained;

[0042] The initial size when the next near-field region is triggered is determined according to the average distribution density;

[0043] The spatial probability distribution model is determined by using the improved Gaussian mixture model;

[0044] The binary detection is performed according to the spatial probability distribution model.

[0045] Optionally, the spatial probability distribution model is determined by using the improved Gaussian mixture model, specifically including:

[0046] The confidence P(x, y, z) is determined by using the formula

[0047] Wherein, Z is a normalization coefficient, s is a spatial attenuation coefficient, w i ​​​is the time-concentration compound weight, and the concentration data is updated in real time each time the near-field region is entered, n is the total number of concentration measurement points participating in the weighted calculation, p is the coordinate vector of the current position (x, y, z) in the three-dimensional space, p i is the spatial coordinate vector of the i-th measurement point.

[0048] In a second aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the wide-area gas tracing method based on multi-level sensing asynchronous cooperative triggering.

[0049] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the wide-area gas tracing method based on multi-level sensing asynchronous cooperative triggering.

[0050] In a fourth aspect, the present application provides a computer program product comprising a computer program, wherein the computer program is executed by a processor to implement the wide-area gas tracing method based on multi-level sensing asynchronous cooperative triggering.

[0051] According to the specific embodiments provided by the present application, the present application has the following technical effects:

[0052] The present application provides a wide-area gas tracing method based on multi-level sensing asynchronous cooperative triggering, device, medium and product, which performs adaptive three-dimensional grid division on the to-be-measured region by using three-dimensional grid division method and adaptive grid mechanism, dynamically adjusts the grid size in the environment without prior information, and avoids resource waste caused by fixed grid; the present application first discovers the plume in the far-field region, determines the near-field region based on the number of measured data points according to the binary calculation result, that is, formulates the event triggering strategy, and greatly reduces the invalid working time of high-precision sensors; and the near-field-far-field dynamic conversion mechanism of the present application judges the near-field range in real time, and reduces the invalid detection area; and further can realize efficient, accurate and low-cost pollution source positioning in complex wide-area environment. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0054] Figure 1 is a flowchart of a wide-area gas tracing method based on multi-level sensing asynchronous cooperative triggering in an embodiment of the present application;

[0055] Figure 2 Flowchart for near-field region size adjustment process;

[0056] Figure 3 Flowchart for near-field region exit determination process;

[0057] Figure 4 Flowchart for overall gas traceability positioning process. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0059] The above purposes, features and advantages of the present application can be more apparent and easy to understand. The present application will be described in further detail below with reference to the drawings and specific embodiments.

[0060] In an exemplary embodiment, as shown in Figure 1 , a wide-area gas traceability method based on multi-level sensing asynchronous cooperative triggering is provided, which includes the following S101 to S109. Among them:

[0061] S101, using a three-dimensional grid division method and an adaptive grid mechanism to perform adaptive three-dimensional grid division on the to-be-measured region to obtain a grid division result;

[0062] According to the three-dimensional size (L x ,L y ,L z ) of the to-be-measured region, the volume V = L x ×L y ×L z of the to-be-measured region is determined;

[0063] The cubic grid size d is determined by the formula ; wherein d ∈ [d min ,d max ];

[0064] Using the cubic grid size, the to-be-measured region is adaptively divided into three-dimensional grids to obtain a grid division result; the grid division result is a three-dimensional cubic grid of d × d × d;

[0065] Wherein, V is the volume of the to-be-measured region, V threshold is a volume threshold value, when it is a small unmanned aerial vehicle environmental monitoring task, V threshold is 300m3 ~ 500 m 3 When tracing the source of urban / plant pollution, V threshold 1000 m 3 ~ 3000 m 3 , V max is a large airspace threshold, taking 5000 m 3 , suitable for open areas or large industrial areas, V min is a small airspace threshold, taking 200 m 3 , suitable for indoor or enclosed space, d max is the maximum grid size, d min is the minimum grid size, a is the size attenuation coefficient, and k is the linear adjustment coefficient.

[0066] S102, according to the grid division result and the sampling data, detecting the environmental background information to obtain the mean value of the environmental background noise;

[0067] As a specific example, a group of low-precision gas sensors obtains sampling data x = x1, x2,..., x n ], n is the number of sampling data (such as n = 100);

[0068] The mean value μ of the environmental background noise is determined by the formula x i is the ith sampling data; the mean value represents the baseline level of the environmental noise;

[0069] S103, regarding the area to be measured as a far field, adopting a random walk strategy and introducing a heavy avoidance mechanism to find the plume; and performing a 5-second sampling on the grid area reached;

[0070] Specifically, the one-dimensional random walk describes the process of moving in a discrete space according to a random rule as follows:

[0071]

[0072] Where, ξ i is an independent and identically distributed random variable, S0 is the starting point, and S n is the position after moving.

[0073] Further, the mathematical expression of the random walk strategy in three-dimensional space (x, y, z) is:

[0074]

[0075] Where, the independent and identically distributed random variables in three-dimensional space Each component is an independent random variable, is the starting point in three-dimensional space, The position after the movement in three-dimensional space.

[0076] Wherein, check the access state of adjacent nodes before each movement, if the target node is not accessed, allow migration; if it has been accessed or belongs to the shielding area, mark it as non-migratable and calculate the sampling point again. Each sampling point is suspended for 5 seconds for sampling to represent the final environmental data of the point.

[0077] S104, determine the threshold value of binary detection according to the mean value of the environmental background noise; and use the threshold value of binary detection to perform binary calculation on the sampling data; set to 1 if it exceeds the threshold value of binary detection, that is, detect the gas, set to 0 if it is lower than the threshold value of binary detection, that is, no gas is detected, and output binary data;

[0078] The standard deviation σ is calculated by using the sample standard deviation formula and adding the Bessel correction:

[0079]

[0080] Wherein, the denominator uses n-1 (Bessel correction) to avoid underestimating the population standard deviation;

[0081] Determine the threshold value T of binary detection by using the formula T = μ + 2σ;

[0082] Determine the binary calculation result Output by using the formula

[0083] Wherein, s j is the jth data point of real-time sampling.

[0084] S105, determine the near-field region based on the number of point positions of the measured data points according to the binary calculation result;

[0085] S105 specifically includes:

[0086] S51, determine the point position coordinate set P = {p1, p2,..., p m}(m ≥ 3) of the data points of the detected gas according to the binary calculation result.

[0087] S52, determine the gas point spatial distribution characteristics according to the point position coordinate set; the gas point spatial distribution characteristics include: spatial range and point set density.

[0088] The calculation formula of the spatial range (maximum span in each axis direction) is as follows:

[0089] R x = max(x i )-min(x i );

[0090] R y = max(y i ​) - min(y i ) ;

[0091] R z = max(z i ) - min(z i ) ;

[0092] The calculation formula of the point set density is:

[0093]

[0094] Wherein, m is the number of gas points, p is the point set density, convex_hull_volume(P) represents the convex hull volume of the point set P, the minimum convex polyhedron volume enclosing all gas points, reflecting the "point aggregation compactness, that is, the convex hull volume reflects the aggregation degree of the point set.

[0095] S53, according to the spatial distribution characteristics of the gas points, the adaptive near-field size D is determined by using the formula ; wherein, 0.8 is an empirical coefficient value, which is calibrated through a large number of experiments, balancing the "spatial range-based basic size" and the "density compensation correction amount", ensuring that the near-field size adapts to the initial range of most scenes, R x , R y and R z are the spatial range, p is the point set density, and β is the density compensation factor, which is 5; the constraint condition is determined by using the formula ; D min = 2xd grid (20m, for example), L is the initial region axial length, and d grid is the grid basic resolution.

[0096] S54, according to the adaptive near-field size, the directional normalization range weighting method is used to adjust the gas distribution direction characteristics.

[0097] The directional normalization range weighting method is:

[0098]

[0099] Wherein, γ is the direction strengthening coefficient, which is 0.3.

[0100] S55, according to the gas distribution direction characteristics, the near-field region is determined.

[0101] The near-field region is defined as a three-dimensional region with O as the geometric center:

[0102]

[0103] S106, concentration tracking is performed according to the near-field region (turn off the low-precision sensor, trigger the opening of the high-precision sensor to collect concentration information), and the improved gradient calculation method and the relative boundary distance are used to evaluate the near-field state in real time; and the size of the near-field region is adjusted according to the evaluation result; and it is continuously judged whether to exit the concentration detection of the near-field region;

[0104] Since the random walk strategy is used for sampling, the distance between two sampling points is not fixed and unknown, and the concentration gradient obtained by calculating the slope according to the time difference may have a large error for sampling points at a long distance, so the distance influence must be eliminated, and the gradient normalized by space distance is used to replace the traditional time difference slope, and the improved concentration gradient formula is:

[0105]

[0106] Wherein, C i ,C i+1 is the concentration of data point i and data point i+1, is the three-dimensional coordinate of sampling point i, is the three-dimensional coordinate of sampling point i+1, and ||·|| is the Euclidean distance,

[0107] The boundary distance border dist (the unmanned aerial vehicle and the near-field boundary) is determined by the formula border dist is [0, 1], indicating the relative distance of the current position from a certain boundary;

[0108] According to the two indexes (concentration slope slope i and boundary distance border_dist) calculated by the above near-field state and the number of 0 concentration points detected in the current near-field (set as N0), the following near-field region adjustment logic is executed, and the near-field adjustment process is as shown in Figure 2

[0109] (1) EXPAND (expand the near-field region): when the concentration shows a downward trend, and the unmanned aerial vehicle is close to the boundary, there may be gas outside the near-field region boundary, at this time, it is recommended to expand the near-field region:

[0110] If slope<-0.3and border_dist<0.2,then:state←EXPAND;

[0111] (2) SHRINK (shrink the near-field region): if the current point is far away from the boundary (located in the center area), but multiple zero concentration points are detected, indicating that the near-field region may have been expanded beyond the set range, and should be shrunk:

[0112] ​If N0≥3 and border_dist > 0.5, then: state <- SHRINK;

[0113] (3) EXIT (exit nearfield) If most points in the nearfield region are not detected to have gas (for example, 5 consecutive detections are 0), it means that there is no smoke plume in the nearfield region, and the nearfield mode should be exited immediately:

[0114] If N0≥5, then: state <- EXIT;

[0115] (4) HOLD (keep unchanged) When the above three special conditions are not met, the current nearfield region state is maintained:

[0116] Else: state <- HOLD;

[0117] On the basis of the nearfield region obtained after the dynamic adjustment in the previous step, the exit comprehensive judgment of the nearfield is performed, and whether to truly exit the nearfield and switch to the farfield mode is determined. Specifically, the nearfield region exit judgment process is as shown in Figure 3 The judgment condition (should_exit_nearfield) has the following three levels:

[0118] (1) Forced exit condition (highest priority)

[0119] Any of the following conditions is met, which is forced to exit:

[0120] 1. 5 consecutive concentration values are 0;

[0121] 2. 3 consecutive concentration decay rates > 15% (negative growth;

[0122] Judgment basis: It indicates that the gas source has disappeared or the UAV has completely left the smoke plume range;

[0123] (2) Comprehensive risk assessment (main decision mechanism)

[0124] Comprehensive risk score:

[0125] Comprehensive risk score = 0.6 x boundary risk + 0.4 x gradient stability;

[0126] Exit condition: exit score > 0.7;

[0127] Specifically, the evaluation indexes mainly include boundary risk and gradient stability:

[0128] Calculate the boundary risk value (0-1):

[0129] BoundaryRisk = min(risk x , risk y , riskz );

[0130] Specifically, for each coordinate axis (X, Y, Z) in three-dimensional space, the boundary risk calculation formula is:

[0131] p k is the current position of the UAV, is the near-field boundary coordinate; O k is the near-field region geometric center coordinate, is the left boundary distance ratio, is the right boundary distance ratio;

[0132] The final boundary risk (Boundary Risk) is the minimum value of the boundary risk values of the three axes; the proximity of the UAV to the boundary in all dimensions is comprehensively evaluated, and the minimum value principle is adopted to determine the overall risk in the most dangerous direction; decision application: (1) 0.4 safe area (suggesting to maintain or expand the near field); (2) 0.2-0.4 alert area (suggesting to maintain the current state); (3) <0.2 dangerous area (suggesting to reduce or exit the near field); the higher the value, the closer to the boundary;

[0133] Calculate the gradient stability (0-1), specifically:

[0134]

[0135] where N 梯度 is the number of gradients, N 方向变化 is the number of changes, is the average amplitude, is the gradient amplitude variance,

[0136] (3) Boundary wandering determination (auxiliary condition)

[0137] 1. Check the last 5 position records;

[0138] 2. The number of times the boundary risk value >0.6 (within 30% of the boundary area) is ≥3;

[0139] According to the boundary wandering determination, the UAV repeatedly invalidates the detection in the boundary area

[0140] S107, when the concentration detection of the exit near-field region is switched to the far-field region, the improved mixed Gaussian model is used to retain the historical distribution characteristics, and the binary detection cycle is entered;

[0141] S107 specifically includes:

[0142] S71, obtain the average distribution density of the last three times of the near field region

[0143] S72, determine the initial size D of the next trigger of the near field region according to the average distribution density init ;

[0144] S73, determine the spatial probability distribution model using the improved Gaussian mixture model;

[0145] Determine the confidence P(x, y, z) using the formula ;

[0146] where Z is a normalization coefficient (which needs to be recalculated after each update), s is a spatial attenuation coefficient, w i is a time-concentration compound weight, and real-time concentration data is used to update each time the near field region is entered, n is the total number of concentration measurement points participating in the weighted calculation, p is the coordinate vector of the current position (x, y, z) in three-dimensional space, p i is the spatial coordinate vector of the i-th measurement point. s = s0 x (1 + a0 · distribution dispersion), and real-time concentration data is used to update each time the near field region is entered.

[0147] S74, perform binary detection according to the spatial probability distribution model.

[0148] S108, determine whether to re-enter the near field region according to real-time binary detection data;

[0149] Specifically, the verification conditions include:

[0150] The specific calculation process is:

[0151] 1. For each detected gas point (x i ,y i ,z i ), calculate its confidence P(x i ,y i ,z i ) in the historical probability model;

[0152] 2. If there is at least one point that satisfies P(x i ,y i ,z i )>0.7, then pass the spatial verification;

[0153] The real-time updating rule of the probability model P(x, y, z) is: each time the near field is entered, the weight is updated with newly collected concentration data:

[0154]

[0155] According to the above conditions, it is finally determined whether to enter the near field mode again.

[0156] S109, according to the concentration detection data of all near field regions, determine the concentration peak; and the concentration peak as the source position.

[0157] As shown in Figure 4 , according to the concentration data information of the near field region obtained in the above S106 to S108, the concentration peak is locked, and when the sensor array detects a clear and significant concentration gradient in the near field region, the unmanned aerial vehicle tracks this gradient and locates to a relatively stable concentration peak point or a very small high-concentration core region, and when the position with the highest concentration no longer significantly drifts, this point / region is considered as the place where the gas leakage or release is the strongest, i.e. the source position.

[0158] The present application has the following effects:

[0159] (1) Efficiency improvement (reducing search time and resource consumption)

[0160] Through the multi-level sensing asynchronous cooperative triggering mechanism (after the low-precision sensor discovers the plume, the high-precision sensor is activated), the invalid working time of the high-precision sensor is greatly reduced. And it is shown that the high-precision sensor is activated only in 10% to 20% of the total working time, which significantly reduces the system energy consumption (the measured power consumption is reduced by 35%), prolongs the endurance time of the unmanned aerial vehicle. Through adaptive grid division and random walk strategy, in the environment without prior information, through the volume-driven grid size formula, the resolution is dynamically adjusted to avoid the waste of resources caused by fixed grid. Compared with the traditional uniform grid, the search efficiency is improved by 40% (measured data). Through the near field-far field dynamic conversion mechanism, based on the boundary risk value (Boundary Risk = min(risk_x, risk_y, risk_z)) and the gradient stability, the near field range is judged in real time, and the invalid detection area is reduced. It is shown that this mechanism reduces the near field detection area by 30% and improves the positioning speed by 50%.

[0161] (2) Positioning accuracy improvement (anti-interference and stability enhancement)

[0162] Through the spatial probability distribution model (improved Gaussian mixture model), the time-concentration composite weight and historical data fusion mechanism are introduced, which effectively suppresses the misjudgment caused by instantaneous interference. Experiments show that in the wind speed mutation scene, the positioning error is reduced by 60% (average error <1.5m) compared with the traditional gradient method. Through the direction adaptive near field adjustment, the gas distribution direction characteristics dynamically correct the near field size Furthermore, the near-field region matches the plume shape. In addition, the accuracy in the Z-axis direction is improved by 25% in actual measurement (especially suitable for complex terrain). Through the three-layer judgment of the triple near-field exit verification mechanism, i.e., forced exit (continuous zero value detection) + boundary risk assessment (multi-dimensional risk calculation) + gradient stability analysis, early exit / stay is avoided. Moreover, the false exit rate is reduced to below 5% (traditional method > 20%).

[0163] (3) Environmental adaptability improvement (complex scene robustness)

[0164] Through the background noise dynamic threshold, the automatic threshold calculation of mean + 2 times standard deviation (T = μ + 2σ) is adopted to adapt to different environmental backgrounds (such as high background noise in industrial areas). In the test, the plume recognition accuracy remains > 92% in the background noise range of 0-100 ppm. The historical data-driven near-field initialization determines the historical distribution density Further calculate the initial near-field size ( ), accelerate the repeated search process. Compared with the first round of search, the second positioning time is shortened by 70%.

[0165] (4) Resource utilization optimization (cost-benefit balance)

[0166] Through the linkage strategy of high and low precision sensors, the high precision sensor is only activated in the near-field region (duty cycle < 20%), prolonging the life of expensive sensors (document estimated life improved by 3 times), and reducing the cost of single task by 45%. The avoidance mechanism and spatial aggregation verification make it possible to shield the visited area in random walk (ξ i Generate the historical point position), and add a spatial aggregation condition (max i P(x i ,y i ,z i )>0.7), reducing redundant sampling points. The number of sampling points is reduced by 35% in actual measurement.

[0167] In an example embodiment, a computer device is provided, which can be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through network connection. The computer program is executed by the processor to implement a wide-area gas traceability method based on multi-level sensing asynchronous cooperative triggering.

[0168] In an example embodiment, a computer device is provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0169] In an example embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0170] In an example embodiment, a computer program product is provided, which includes a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0171] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0172] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc.

[0173] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0174] In the present application, all actions of obtaining signals, information or data are performed in accordance with the corresponding data protection regulations and policies of the country where the device is located, and with the authorization of the owner of the corresponding device.

[0175] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0176] The principles and implementation manners of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering, characterized in that, The wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering includes: A three-dimensional mesh generation method and an adaptive mesh mechanism are used to adaptively generate a three-dimensional mesh for the area to be measured, and the mesh generation result is obtained. Based on the grid division results and sampling data, the environmental background information is detected to obtain the mean value of the environmental background noise; The area to be measured is considered as the far field, and a random walk strategy and a weight avoidance mechanism are introduced to detect plumes. The threshold for binary detection is determined based on the mean value of the background noise; and the sampling data is then subjected to binary calculation using the threshold for binary detection. Based on the binary calculation results, the near-field region is determined according to the number of points of the measured data points; Concentration tracking is performed in the near-field region, and an improved gradient calculation method and relative boundary distance are used to evaluate the near-field status in real time; the size of the near-field region is adjusted according to the evaluation results; and concentration detection is continuously judged to determine whether to exit the near-field region. When switching from concentration detection in the near field region to the far field region, an improved Gaussian mixture model is used to preserve historical distribution characteristics, and then the binary detection loop is entered. Determine whether to re-enter the near-field region based on real-time binary detection data; Based on the concentration detection data of all near-field areas, the concentration peak was determined; and the concentration peak was used as the source location. The step of determining the threshold for binary detection based on the mean of the ambient background noise, and then using the threshold for binary detection to perform binary calculations on the sampled data, specifically includes: Using formula Determine the threshold for binary detection ; Using formula Determine the binary calculation result ; in, This represents the average value of the ambient background noise. , For the i-th sampled data, The number of sampled data, Standard deviation, , Let j be the j-th data point sampled in real time.

2. The wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering according to claim 1, characterized in that, The method employs a three-dimensional mesh generation method and an adaptive mesh mechanism to adaptively generate a three-dimensional mesh for the area to be measured, resulting in mesh generation results, specifically including: Determine the volume of the region to be measured based on its three-dimensional dimensions; Using formula Determine the cube grid size ; Using the cubic mesh size, an adaptive 3D mesh is generated for the area to be measured, and the mesh generation result is obtained. in, The volume of the region to be measured. For volume threshold, For large spatial domain threshold, For small spatial domain thresholds, For the maximum grid size, Minimum grid size, The size attenuation coefficient, This is the linear adjustment coefficient.

3. The wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering according to claim 1, characterized in that, Based on the binary calculation results, the near-field region is determined according to the number of measured data points, specifically including: Based on the binary calculation results, determine the set of coordinates of the data points where the gas was detected; The spatial distribution characteristics of gas points are determined based on the set of point coordinates; the spatial distribution characteristics of gas points include: spatial range and point set density; Based on the spatial distribution characteristics of gas points, using the formula Determine adaptive near-field size ;in, , as well as For spatial extreme difference, For point set density, Density compensation factor; Based on the adaptive near-field size, the directional normalized range weighting method is used to adjust the gas distribution directional characteristics. The near-field region is determined based on the gas distribution direction characteristics.

4. The wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering according to claim 1, characterized in that, The concentration tracking is performed based on the near-field region, and an improved gradient calculation method and relative boundary distance are used to evaluate the near-field state in real time. And adjust the near-field area size based on the evaluation results; specifically including: Using formula Determine the concentration slope ; Using formula Determine the boundary distance ; The size of the near-field region is adjusted based on the concentration slope, boundary distance, and the number of data points with 0 concentration detected in the current near-field region. in, The concentrations at data point i and data point i+1, The three-dimensional coordinates of sampling point i, The three-dimensional coordinates of sampling point i+1 For Euclidean distance For the length of each axis in the near-field region, , The current location of the drone. This is the center of each axis length in the near-field region.

5. The wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering according to claim 1, characterized in that, When switching from near-field to far-field concentration detection, an improved Gaussian mixture model is used to preserve historical distribution characteristics, and a binary detection loop is entered. Specifically, this includes: Obtain the average distribution density of the near-field region for the three most recent times; The initial size of the near-field region for the next trigger is determined based on the average distribution density. A spatial probability distribution model is determined using an improved Gaussian mixture model; Binary entry detection is performed based on a spatial probability distribution model.

6. The wide-area gas source tracing method based on multi-level sensor asynchronous collaborative triggering according to claim 5, characterized in that, The determination of the spatial probability distribution model using the improved Gaussian mixture model specifically includes: Using formula Determine the confidence level ; in, The normalization coefficient is... , Each time it enters the near-field region, it updates the data using real-time concentration data, where n is the total number of concentration measurement points participating in the weighted calculation. Let (x, y, z) be the coordinate vector of the current position (x, y, z) in three-dimensional space. Let be the spatial coordinate vector of the i-th measurement point.

7. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the wide-area gas tracing method based on multi-level sensing asynchronous collaborative triggering as described in any one of claims 1-6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the wide-area gas tracing method based on multi-level sensor asynchronous collaborative triggering as described in any one of claims 1-6.

9. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the wide-area gas tracing method based on multi-level sensor asynchronous collaborative triggering as described in any one of claims 1-6.

Images