A natural gas leakage cloud edge collaborative monitoring system for large-scale open environment

Abstract

The present application relates to the technical field of cloud edge collaboration, in particular to a natural gas leakage cloud edge collaborative monitoring system for large-scale open environment, the system comprises: a point planning unit generates sampling points and scanning ranging coordinates, a scanning acquisition unit controls the pan-tilt to cruise and collect methane concentration, a concentration mapping unit generates spatial concentration grid by coordinate sorting and interpolation, a source position determination unit determines clustering radius according to the distance between high concentration points and outputs source coordinates and average concentration, an alarm output unit generates position and concentration alarm information, and the system guides orderly cruise through the spatial correlation of risk position and scanning coordinates, presents leakage cloud distribution in continuous concentration grid, adjusts clustering scale according to the density change of high concentration points, reduces monitoring blind area and repeated inspection, reduces the interference of discrete readings and fixed threshold on source position judgment, improves leakage source positioning stability and alarm information checkability, and shortens the on-site review and disposal link.

Description

Technical Field

[0001] This invention relates to the field of cloud-edge collaborative technology, and in particular to a cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments. Background Technology

[0002] The field of cloud-edge collaborative technology involves a computer implementation method where data collected by field terminals is aggregated by edge nodes and transmitted to cloud storage resources, whereby the server classifies, manages, and retrieves the monitoring data. Traditional cloud-edge collaborative monitoring of natural gas leaks in large-scale open environments involves multiple monitoring terminals periodically collecting natural gas concentrations and environmental parameters. Edge nodes convert and encapsulate the collected data before sending it to a cloud server. The cloud server writes the monitoring data for each monitoring area into storage resources, and the client reads the corresponding historical monitoring data based on the area identifier, device identifier, and collection time.

[0003] Existing natural gas leak monitoring in large-scale open areas relies heavily on the dispersed deployment of fixed probes or manual point-by-point inspections. There is a lack of unified spatial correlation between monitoring points and risky locations such as pipelines, valves, and flanges. The coverage is limited and the inspection paths are repetitive. Concentration data is usually presented as isolated readings, which makes it difficult to continuously reflect the spatial distribution of the leak cloud. Anomaly point identification is easily affected by wind field disturbances and distance measurement deviations. Fixed threshold clustering is difficult to adapt to changes in the density of points. The stability of leak source location is insufficient, and the alarm information does not correspond clearly with the actual coordinates, which increases the workload of verification and delays on-site handling. Summary of the Invention

[0004] This invention aims to provide a cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments. By collaboratively configuring sampling point planning, pan-tilt-zoom (PTZ) scanning, synchronous methane concentration acquisition, spatial concentration mapping, adaptive leak source determination, and alarm output, the system can generate monitoring data with spatial correspondences around the locations of natural gas pipelines, valves, and flange connections. Based on the spatial concentration distribution, the system can determine the location of the leak source, thus providing a technical solution with a clear structure, well-defined data correlation, and suitability for location output for natural gas leak monitoring in large-scale open environments.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments includes: The point planning unit is used to establish a spatial coordinate system in the monitoring area, generate sampling points based on the connection positions of natural gas pipelines, valves and flanges, and output the pan-tilt scanning angle and ranging coordinates corresponding to the sampling points. The scanning and acquisition unit is connected to the point planning unit and is used to control the gimbal equipped with a laser methane telemetry instrument and a laser range sensor to cruise according to the gimbal scanning angle and collect methane concentration data at the range coordinates. The concentration mapping unit, connected to the scanning acquisition unit, is used to receive the methane concentration data, construct a cubic spline interpolation relationship according to the spatial sorting of the ranging coordinates, and generate a spatial concentration grid. The source location determination unit, connected to the concentration mapping unit, is used to filter high-concentration measurement points according to the spatial concentration grid, determine the clustering radius based on the inter-point distance distribution of the high-concentration measurement points, and perform adaptive clustering of the high-concentration measurement points based on the clustering radius, and output the leakage source coordinates and average concentration. An alarm output unit, connected to the source location determination unit, is used to receive the coordinates of the leakage source and the average concentration, and generate alarm information that records the location marker of the leakage source and the concentration display.

[0006] Specifically, the point planning unit includes a coordinate establishment submodule, a component extraction submodule, a point generation submodule, and an angular distance output submodule; The coordinate establishment submodule is used to obtain the monitoring area boundary, the gimbal installation reference point and the height reference point, establish a spatial coordinate system based on a unified length unit, and generate the regional coordinate reference. The component extraction submodule is used to extract the coordinates of candidate components based on the projection relationship of the natural gas pipeline, valve and flange connection positions in the regional coordinate reference. The sampling point generation submodule is used to generate sampling points according to the coordinates of the candidate components; The angular distance output submodule is used to acquire the gimbal mounting reference point, the gimbal horizontal zero position direction, the gimbal pitch zero position direction, and the sampling point. The spatial line connecting the sampling point and the gimbal mounting reference point is determined as the scanning pointing line. Based on the deflection relationship of the scanning pointing line relative to the gimbal horizontal zero position direction and the gimbal pitch zero position direction, the horizontal scanning angle and the pitch scanning angle are generated. The ranging distance is determined based on the spatial length of the scanning pointing line, and the horizontal scanning angle, the pitch scanning angle, and the ranging distance are associated and written into the same sampling point record, and the pan-tilt scanning angle and the ranging coordinates are output.

[0007] Specifically, the process of generating the sampling points is as follows: Obtain the component type corresponding to the coordinates of the candidate components, set the pipeline sampling interval according to the centerline of the natural gas pipeline, set the dense sampling interval according to the valve boundary and flange connection end face, and merge the sampling positions between adjacent candidate component coordinates that are less than the point merging distance threshold. According to the spatial order of each sampling position, write the component type identifier and point order identifier for each sampling position to generate the sampling point.

[0008] Specifically, the scanning and acquisition unit includes an angle analysis submodule, a cruise control unit, a synchronous sampling submodule, and a data binding unit; The angle analysis submodule is used to obtain the gimbal scanning angle and ranging coordinates, convert the gimbal scanning angle into horizontal rotation and pitch rotation commands, and write the ranging coordinates into the sampling point queue. The cruise control unit is used to drive the gimbal to the corresponding attitude in sequence according to the queue of sampling points; The synchronous sampling submodule is used to trigger the laser methane telemetry instrument and the laser range sensor to collect synchronous data after the gimbal attitude is stabilized, and generate synchronous sampling data. The data binding unit is used to associate the synchronous sampling data with the ranging coordinates and output the methane concentration data at the ranging coordinates.

[0009] Specifically, the process of generating the synchronous sampling data includes: The gimbal attitude feedback angle and angle stability threshold are obtained. The gimbal attitude feedback angle is compared with the horizontal rotation command and the pitch rotation command respectively. When the horizontal deviation and the pitch deviation do not exceed the angle stability threshold, a sampling trigger signal is generated. Based on the sampling trigger signal, the concentration reading of the laser methane telemetry instrument and the distance reading of the laser rangefinder are read simultaneously to generate the synchronous sampling data.

[0010] Specifically, the concentration mapping unit includes a data receiving submodule, a coordinate sorting submodule, a spline construction submodule, and a grid generation submodule; The data receiving submodule is used to receive the methane concentration data and extract the concentration reading, the ranging coordinates and the acquisition time identifier, and generate a concentration coordinate record; The coordinate sorting submodule is used to spatially sort the concentration coordinate records according to the horizontal and vertical positions of the ranging coordinates in the spatial coordinate system, thereby generating an ordered concentration sequence. The spline construction submodule is used to construct a cubic spline interpolation relationship based on the coordinate interval of adjacent measurement points in the ordered concentration sequence. The mesh generation submodule is used to assign concentration values ​​to the preset mesh nodes according to the cubic spline interpolation relationship to generate the spatial concentration mesh.

[0011] Specifically, the process of constructing the cubic spline interpolation relationship is as follows: The ranging coordinates, concentration readings, and coordinate intervals of adjacent measurement points in the ordered concentration sequence are obtained. The coordinate intervals are determined as the interpolation segment boundaries. Segmented interpolation constraints are established according to the conditions that the concentrations of adjacent segments are continuous at the connection points and the concentration change trends are continuous. The interpolation curves of each segment are constrained according to the segmented interpolation constraints to generate the cubic spline interpolation relationship.

[0012] Specifically, the source location determination unit includes a high value filtering submodule, a distance analysis submodule, a radius determination submodule, and a source location output submodule; The high-value screening submodule is used to obtain the grid concentration value, background concentration benchmark value, and concentration increment threshold in the spatial concentration grid, compare the difference between the grid concentration value and the background concentration benchmark value, and determine the measurement point whose difference is not lower than the concentration increment threshold as a candidate high-value point; when there are no less than two consecutive points of the candidate high-value point in adjacent grid nodes, the candidate high-value point is determined as the high-concentration measurement point, and a high-value point sequence is generated. The distance analysis submodule is used to calculate the distance distribution between adjacent points in the high-value point sequence and generate a distance distribution record. The radius determination submodule is used to determine the cluster radius based on the distance distribution record; The source location output submodule is used to obtain the point clusters obtained by adaptive clustering, the distance coordinates and concentration readings of each high concentration measurement point within the point cluster, perform weighted summation on each distance coordinate based on the concentration readings, determine the center coordinates of the point cluster, the center coordinates of the point cluster are determined as the coordinates of the leakage source, and calculate the mean of each concentration reading within the point cluster, outputting the average concentration associated with the coordinates of the leakage source.

[0013] Specifically, the process of determining the cluster radius includes: Obtain the distance values ​​between points and the corresponding number of points in the distance distribution record, generate a distance sorting sequence in ascending order of the distance values ​​between points, and determine the distance boundary position in the distance sorting sequence where the increase in adjacent distance exceeds the radius mutation threshold; The maximum distance between points before the distance boundary position is determined as the candidate cluster radius, and the candidate cluster radius is expanded and corrected according to the upper limit of the gimbal ranging error to generate the cluster radius.

[0014] Specifically, the alarm output unit includes a coordinate mapping submodule, a concentration labeling submodule, and an information output submodule; The coordinate mapping submodule is used to receive the coordinates of the leakage source and convert the coordinates of the leakage source into the coordinates of the monitoring area plane marker, thereby generating location marker data; The concentration labeling submodule is used to receive the average concentration and truncate the average concentration according to the concentration display accuracy to generate concentration display data. The record generation submodule is used to associate the location marker data, the concentration display data, and the alarm time identifier to generate an alarm record, and to issue alarm information based on the alarm record.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, spatial coordinate associations are established around risky locations such as pipeline valves and flanges. Sampling points are converted into pan-tilt-zoom (PTZ) scanning angles and ranging coordinates, enabling patrol data collection to cover key areas in spatial order, reducing redundant inspections and monitoring blind spots. Based on the coordinate order, cubic spline interpolation is performed on discrete methane concentrations to form a continuous spatial concentration grid, enhancing the ability to represent the distribution of leakage clouds. In high-concentration measurement points, the clustering scale is determined by the distance distribution between points, allowing the clustering range to be adjusted according to the density of points, reducing the interference of fixed thresholds on the location results. Furthermore, the coordinates of the leakage source are associated with the average concentration to generate location markers and concentration displays, improving the stability of source location identification and the verifiability of alarm information, and shortening the on-site verification and handling process. Attached Figure Description

[0016] Figure 1 This is a flowchart of the cloud-edge collaborative monitoring process for natural gas leaks according to the present invention; Figure 2 This is a schematic diagram of the deployment of the cloud-edge collaborative monitoring system of the present invention; Figure 3 This is a schematic diagram illustrating the sampling point planning and cruise data collection effect of the present invention; Figure 4 This is a schematic diagram illustrating the determination of spatial concentration grid and leakage source coordinates according to the present invention; Figure 5 This is a schematic diagram illustrating the relationship between alarm information generation and tracing in this invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] Example Please see Figures 1 to 5This embodiment provides a cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments. In practical applications, such as in open monitoring areas where natural gas stations, pipeline corridors, and valve chambers of long-distance pipelines operate continuously, the edge-side monitoring pole carries a pan-tilt unit, a laser methane telemetry instrument, and a laser ranging sensor. The cloud-side processing node receives the sampling point records, synchronous sampling data, and alarm records uploaded from the edge side. The system forms a closed loop around the natural gas pipeline, valve, and flange connection locations, from point planning, cruise data acquisition, concentration mapping, source location determination to alarm output. It includes a point planning unit, a scanning acquisition unit, a concentration mapping unit, a source location determination unit, and an alarm output unit.

[0019] Point planning unit: This is a structural unit used to form cruise targets in the cloud-edge collaborative monitoring system. In this embodiment, the cloud-edge collaborative monitoring system refers to a monitoring system formed by connecting edge-side acquisition nodes, cloud-side processing nodes, and alarm display nodes according to a data transmission link. The edge-side acquisition nodes are responsible for attitude control and sensor reading, the cloud-side processing nodes are responsible for spatial point organization, concentration grid generation, and leakage source coordinate determination, and the alarm display nodes are responsible for receiving location markers and concentration display fields. The point planning unit receives the monitoring area boundary, natural gas pipeline direction, valve shape and position, flange connection end face position, pan-tilt unit installation reference point, and height reference point to form a sampling point record. The sampling point record is a data object corresponding to the same sampling location, carrying component type identifier, point sequence identifier, spatial coordinate field, horizontal scan angle field, pitch scan angle field, and distance measurement field, and serves as the cruise input for the scanning acquisition unit.

[0020] During edge-side initialization, the point planning unit reads the PTZ installation reference point and PTZ zero-position direction from the local configuration, and reads the registered pipe, valve, and flange connection positions in the monitoring area layer on the cloud side. When the monitoring area boundary is missing, the point planning unit writes the area coordinate reference status flag to the pending completion status, the sampling point record is not entered into the patrol queue, and a boundary completion request is returned to the cloud-side processing node. When duplicate records exist for pipe, valve, or flange connection positions, deduplication is performed according to spatial coordinate consistency and component type identifier, retaining records that can match the area coordinate reference. The source of the duplication is written to the planning log for subsequent traceability.

[0021] The coordinate establishment submodule is used to create a regional coordinate reference. This reference is a spatial reference data object shared by the positions of various components, the gimbal, and sampling locations within the monitoring area. Its content includes a planar position reference, an altitude position reference, a gimbal installation reference point, a boundary closure status indicator, and a coordinate validity status indicator. The coordinate establishment submodule receives the monitoring area boundary, the gimbal installation reference point, and the altitude reference point. It unifies regional layers, field survey records, and installation registration records from different sources into a spatial coordinate system with the same length and caliber, and writes the coordinate reference into the cloud-side planning cache and the edge-side cruise configuration cache.

[0022] The coordinate establishment submodule performs integrity checks on the input data. When the monitoring area boundary cannot form a closed region, the boundary closure status is marked as unclosed, and the subsequent component extraction submodule only receives component projection records that have fallen within the valid boundary segments. When the gimbal installation reference point and the height reference point cannot establish a spatial relationship, the coordinate validity status is marked as reference abnormal, and the angular distance output submodule stops generating scanning angles. The verified area coordinate reference is sent to the component extraction submodule as the sole coordinate source for candidate component coordinate extraction.

[0023] The component extraction submodule extracts candidate component coordinates from the regional coordinate reference. Candidate component coordinates are positional data objects of the natural gas pipeline centerline, valve boundaries, and flange connection faces within the regional coordinate reference, carrying component type identifiers, spatial projection positions, geometric boundary sources, and record validity status identifiers. The component extraction submodule reads the projection relationships of pipeline, valve, and flange connection positions within the regional coordinate reference, projects the pipeline centerline as a continuous linear object with deployable sampling positions, projects the valve boundaries and flange connection faces as local objects requiring dense scanning, and sends the candidate component coordinates to the point generation submodule.

[0024] The component extraction submodule writes an "out-of-bounds status" flag to the coordinates of components that cannot fall within the monitoring area boundary, and writes a "type pending confirmation" flag to the coordinates that lack a component type flag. Out-of-bounds coordinates are not included in the sampling point generation, while coordinates with a "type pending confirmation" flag enter the cloud-based manual review queue but do not block the processing of other valid candidate component coordinates. When there is an overlap in the projection relationship, an overlap processing record is formed according to the component type priority. The overlapping part of the valve boundary and flange connection end face covering the pipeline centerline is processed as a local object in the point generation submodule.

[0025] The sampling point generation submodule is used to generate sampling points based on the coordinates of candidate components. Sampling points are spatial location data objects that the PTZ (pan-tilt-zoom) system needs to point to during pan-tilt-zoom (PTZ) navigation, carrying spatial location, component type identifier, point sequence identifier, and merging status identifier. The sampling point generation submodule reads the component type corresponding to the candidate component coordinates, generates pipeline sampling positions along the natural gas pipeline centerline according to the pipeline sampling interval, and generates local sampling positions around valve boundaries and flange connection faces according to the dense sampling interval. The pipeline sampling interval, dense sampling interval, and point merging distance threshold are all derived from the cloud-side planning rule table. The rule table is maintained by the monitoring object type, on-site inspection strategy, and equipment installation diameter. Fields are written to the planning configuration storage area and refreshed when the regional coordinate benchmark is updated or the inspection strategy is updated.

[0026] The sampling point generation submodule merges sampling locations whose coordinates fall within the sampling point merging distance threshold boundary between adjacent candidate components. The sampling point merging distance threshold is a configurable field, representing the spatial boundary of sampling locations that can be covered by the same pan-tilt unit. It originates from equipment installation calibration records and patrol coverage rules and is subsequently used to reduce duplicate sampling locations and maintain the correspondence between sampling points and component types. The merged sampling locations retain the source component type identifier. When there are multiple associated pipe, valve, and flange connection faces, a composite component type identifier is written into the sampling point record, allowing the alarm output unit to display the leak source associated component in the location marker.

[0027] The sampling point generation submodule writes a sampling point sequence identifier according to the order of each sampling location in the spatial coordinate system. The sampling point sequence identifier is a cruise queue sorting field, derived from the spatial arrangement of sampling locations along the pipeline direction, valve boundaries, and flange connection faces, and is used by the scanning and acquisition unit to generate a queue of sampling points. If the sampling point coordinates are missing, the height field is missing, or it is inconsistent with the area coordinate reference, the merge status identifier is written invalidly, the record is entered into the planning anomaly log, and it is not entered into the angular distance output submodule.

[0028] The angular distance output submodule is used to output the gimbal scanning angle and ranging coordinates. The gimbal scanning angle is attitude control data composed of the horizontal scanning angle field and the pitch scanning angle field. The ranging coordinates are position data composed of the spatial coordinate field corresponding to the sampling point in the regional coordinate reference and the distance field of the scanning pointing line. The angular distance output submodule reads the gimbal installation reference point, the gimbal horizontal zero position direction, the gimbal pitch zero position direction, and the sampling point. It determines the spatial line connecting the sampling point and the gimbal installation reference point as the scanning pointing line, and generates the horizontal scanning angle and pitch scanning angle based on the deflection relationship of the scanning pointing line relative to the horizontal zero position direction and the pitch zero position direction.

[0029] The angular distance output submodule simultaneously determines the ranging distance based on the spatial length of the scanning pointing line, and associates the horizontal scanning angle, pitch scanning angle, and ranging distance into the same sampling point record. The gimbal zero-position direction comes from the edge-side installation calibration record, and the gimbal installation reference point comes from the regional coordinate reference generated by the coordinate establishment submodule. The scanning angle field and ranging distance field are written together into the cruise configuration cache. If the scanning pointing line exceeds the gimbal attitude coverage boundary, the sampling point record is marked as unreachable, does not enter the sampling point queue, and retains the corresponding spatial coordinate field and component type identifier in the planning log for maintenance to adjust the gimbal installation position or sampling point rules.

[0030] Scanning and Acquisition Unit: Connected to the point planning unit, the scanning and acquisition unit converts sampling point records into gimbal cruise control and synchronous sensor reads. It receives the gimbal scanning angle and ranging coordinates, splitting the same sampling point record into control fields and binding fields. The control fields include horizontal rotation and pitch rotation commands, while the binding fields include ranging coordinates, point sequence identifier, component type identifier, and acquisition time identifier. The scanning and acquisition unit operates at the edge, and after sampling, outputs the methane concentration data at the ranging coordinates to the concentration mapping unit.

[0031] The scanning and acquisition unit manages the sampling point records output by the point planning unit in a queue. Records in the queue are primarily sorted by point sequence identifier, with unreachable, invalid, and sampled status identifiers serving as state boundaries. When transmission between the edge side and the cloud side is interrupted, the queue of points to be sampled that have already been written into the edge side's cruise configuration cache remains, and the sampling results are entered into the edge side's pending upload cache. After transmission resumes, the data binding unit re-transmits the sampling records according to the acquisition time identifier and point sequence identifier, preventing the spatial concentration grid from using concentration readings lacking coordinate binding.

[0032] The angle resolution submodule is used to resolve the pan-tilt-zoom (PTZ) scanning angle into rotation commands that the PTZ controller can receive. This submodule reads the horizontal scanning angle field, the pitch scanning angle field, and the ranging coordinate field. It converts the horizontal scanning angle into a horizontal rotation command, the pitch scanning angle into a pitch rotation command, and writes the ranging coordinates into the sampling point queue. The horizontal rotation and pitch rotation commands are attitude control data in the edge-side control bus, and the ranging coordinates are the spatial indexes for subsequent concentration readings.

[0033] The angle parsing submodule performs range and format checks on the angle field. If the angle field is missing, parsing fails, or it is inconsistent with the gimbal attitude coverage boundary, the corresponding sampling point is written with an angle parsing anomaly status flag, and the cruise control unit does not drive the gimbal to point to that record. If the format is consistent but there are duplicate point sequence flags, the angle parsing submodule retains the currently valid record according to the generation time flag of the sampling point record and writes the duplicate records to the edge-side control log.

[0034] Cruise Control Unit: Used to drive the gimbal to the corresponding attitude sequentially according to the queue of points to be sampled. The queue of points to be sampled is the cruise control sequence output by the angle analysis submodule, carrying the point sequence identifier, horizontal rotation command, pitch rotation command, ranging coordinates, and sampling status identifier. The cruise control unit reads the valid record at the head of the queue, sends the horizontal rotation command and pitch rotation command to the gimbal controller, and receives the gimbal attitude feedback angle. The gimbal attitude feedback angle is the current attitude data returned by the gimbal encoder or attitude feedback interface, which is used by the synchronous sampling submodule to determine the sampling trigger condition.

[0035] When the cruise control unit receives an unconfirmed instruction from the gimbal controller, it writes a sampling status flag as "attitude unconfirmed" and retains the sampling point in the queue for retry scheduling. If the gimbal controller returns a motion-limited status, the cruise control unit writes this record to the "cannot cruise" status flag and uploads the attitude-limited record to the cloud-side processing node. Sampling points that have achieved attitude confirmation enter the synchronous sampling submodule; sampling points that have not achieved confirmation do not trigger readings from the laser methane telemetry device and the laser rangefinder sensor.

[0036] Synchronous Sampling Submodule: This module triggers the laser methane telemetry instrument and laser rangefinder to synchronously acquire data after the gimbal attitude stabilizes, generating synchronous sampling data. The synchronous sampling data consists of a data object composed of concentration readings, distance readings, gimbal attitude feedback angles, acquisition time markers, and sampling point status markers read under the same sampling trigger signal. The synchronous sampling submodule receives the gimbal attitude feedback angle and angle stability threshold fields output by the cruise control unit, and compares the feedback angle with the horizontal and pitch angle commands respectively. When both the horizontal and pitch deviation states enter the stable category, a sampling trigger signal is generated.

[0037] The angle stability threshold field, derived from the gimbal installation calibration record and the controller attitude stability rule table, represents the attitude deviation boundary that allows synchronous sampling to be triggered. It is stored in the edge-side control configuration cache and updated after gimbal maintenance calibration. The synchronous sampling submodule simultaneously reads the concentration reading from the laser methane telemetry instrument and the distance reading from the laser rangefinder sensor based on the sampling trigger signal. The concentration reading characterizes the methane concentration response along the gas path traversed by the scanning pointing line, while the distance reading verifies the consistency between the measured distance and the actual echo position. If either sensor fails to return a reading, the synchronous sampling data is written to an incomplete sampling status flag. The data binding unit only stores this record as an abnormal sampling record and does not output valid methane concentration data to the concentration mapping unit.

[0038] The data binding unit associates synchronous sampling data with ranging coordinates and outputs the methane concentration data at the ranging coordinates. The methane concentration data is a data object consisting of concentration reading, distance reading, ranging coordinates, point sequence identifier, component type identifier, acquisition time identifier, and data validity status identifier. The data binding unit reads the binding fields from the sampling point queue and the synchronous sampling data generated by the synchronous sampling submodule, establishes a correspondence according to the point sequence identifier and acquisition time identifier, and binds the concentration reading to the ranging coordinates.

[0039] The data binding unit performs consistency verification between the distance reading and the measured distance field, and writes the verification result into the data validity status identifier. When the distance reading cannot be associated with the measured coordinates, the acquisition time identifier is missing, or the point sequence identifier is duplicated, the record is written into the binding anomaly status identifier and enters the edge-side anomaly cache. Records with a valid data validity status identifier are sent to the concentration mapping unit; binding anomaly records retain the original concentration reading, distance reading, and sampling trigger status for maintenance to trace the sensor reading link, and do not participate in the spatial concentration grid generation.

[0040] Concentration mapping unit: Connected to the scanning acquisition unit, it receives methane concentration data, constructs a cubic spline interpolation relationship according to the spatial order of the ranging coordinates, and generates a spatial concentration grid. The spatial concentration grid is a data object formed by associating preset grid nodes and concentration assignment results within the monitoring area, carrying grid node coordinates, grid concentration values, acquisition time identifiers, interpolation source status identifiers, and data validity status identifiers. The concentration mapping unit runs on the cloud-side processing node, receives valid methane concentration data uploaded from the edge side, and sends the generated spatial concentration grid to the source location determination unit.

[0041] Before generating the grid, the concentration mapping unit performs temporal aggregation and spatial consistency verification on the methane concentration data. If the acquisition time identifier is missing, the record is placed in a time anomaly cache and not included in the current grid assignment. If the ranging coordinates are inconsistent with the regional coordinate baseline, the record is placed in a coordinate anomaly cache. If duplicate valid concentration readings exist for the same ranging coordinates, the concentration mapping unit selects the record that represents the current cruise round based on the acquisition time identifier and sampling status identifier, and retains a traceability index for duplicate records. The verified concentration coordinate records are then entered into the coordinate sorting submodule.

[0042] The data receiving submodule receives methane concentration data and extracts concentration readings, ranging coordinates, and acquisition time identifiers to generate concentration coordinate records. These records serve as grid assignment input objects used within the concentration mapping unit, carrying concentration readings, ranging coordinates, acquisition time identifiers, point sequence identifiers, and data validity status identifiers. The data receiving submodule receives methane concentration data from the edge side, checks the data validity status identifiers, extracts fields suitable for interpolation processing, and writes them to the cloud-side concentration cache.

[0043] The data receiving submodule writes a reception anomaly status flag for records lacking concentration readings, ranging coordinates, or acquisition time identifiers. Concentration readings come from the laser methane telemetry instrument; ranging coordinates come from spatial coordinate fields generated by the point planning unit and confirmed by the data binding unit; and acquisition time identifiers come from the sampling trigger record of the synchronous sampling submodule. When the field sources are inconsistent, the concentration coordinate record is not entered into the coordinate sorting submodule, but the edge-side device identifier, point sequence identifier, and anomaly field name are retained in the cloud-side log for verification by the acquisition link.

[0044] The coordinate sorting submodule spatially sorts concentration coordinate records based on their horizontal and vertical positions in the spatial coordinate system, generating an ordered concentration sequence. The ordered concentration sequence is a set of concentration coordinate records organized according to the sampling path and height distribution within the monitoring area, carrying the sorting position identifier, distance coordinates, concentration readings, collection time identifiers, and adjacency identifiers. The coordinate sorting submodule first sorts the records horizontally according to the pipeline direction and local component boundaries, then organizes the records near the same horizontal position according to their height, creating a continuous interpolated input sequence of concentration readings in space.

[0045] The coordinate sorting submodule writes sorting status identifiers for records with duplicate spatial locations, broken spatial relationships, and missing height fields. Records with duplicate spatial locations are merged and sorted according to the valid status retained by the data receiving submodule; records with broken spatial relationships have breakpoint identifiers written into the ordered concentration sequence, and the spline construction submodule stops connecting across breakpoints at those breakpoints; records with missing height fields do not participate in height sorting but are retained in the concentration cache for tracking collection anomalies. The sorted ordered concentration sequence is then sent to the spline construction submodule.

[0046] The spline construction submodule is used to construct cubic spline interpolation relationships based on the coordinate intervals of adjacent measurement points in an ordered concentration sequence. The cubic spline interpolation relationship is a segmented concentration assignment rule object used within the concentration mapping unit, carrying segment boundaries, coordinates of adjacent measurement points, concentration readings, continuity status indicators at connection points, and trend continuity status indicators. The spline construction submodule reads the distance coordinates, concentration readings, and coordinate intervals of adjacent measurement points in the ordered concentration sequence, determines the coordinate intervals as the interpolation segment boundaries, and maintains concentration continuity and concentration change trend continuity at the connection points of adjacent segments.

[0047] The spline construction submodule does not cross the breakpoint markers written by the coordinate sorting submodule during processing. When there is a lack of effective concentration readings between adjacent measurement points, the segment is marked as non-interpolable, and the grid generation submodule retains the pending sampling status for the corresponding grid node. The continuous status markers for connection positions and trends are derived from the adjacency relationships and concentration reading change directions of the ordered concentration sequence. After formation, they are written into the cubic spline interpolation relationship and read by the grid generation submodule. The interpolation segment boundaries, status markers, and source measurement point indexes are stored in the cloud-side concentration mapping log for alarm record backtracking.

[0048] The grid generation submodule assigns concentration values ​​to preset grid nodes according to cubic spline interpolation relationships, generating a spatial concentration grid. Preset grid nodes are spatial node data objects within the monitoring area that are consistent with the regional coordinate baseline, carrying node coordinates, the region to which the node belongs, the interpolation source identifier, and the grid concentration value. The grid generation submodule reads the cubic spline interpolation relationships and the preset grid nodes, assigns each grid node to its corresponding interpolation segment, generates a grid concentration value based on the concentration continuity rules and trend continuity rules of that segment, and writes it into the spatial concentration grid.

[0049] The grid generation submodule writes an unassigned status flag to nodes that cannot be included in a valid interpolation segment, and writes a pending sampling status flag to nodes that fall into an uninterpolation segment. Unassigned nodes and nodes pending sampling do not participate in the high-value screening of the source location determination unit. After the spatial concentration grid is completed, the grid node coordinates, grid concentration values, interpolation source flags, and data validity status flags are sent to the source location determination unit together, so that subsequent high-concentration measurement point screening can be traced back to the original sampling point records and synchronous sampling data.

[0050] Source location determination unit: Connected to the concentration mapping unit, it filters high-concentration measurement points based on the spatial concentration grid, determines the cluster radius based on the distance distribution between high-concentration measurement points, and performs adaptive clustering of high-concentration measurement points based on the cluster radius, outputting the leakage source coordinates and average concentration. High-concentration measurement points are those whose grid concentration values ​​enter the high-value category relative to the background concentration benchmark and form a continuous high-value chain in adjacent grid nodes. The leakage source coordinates are the coordinates of the center of the point cluster, and the average concentration is the concentration display source field after the concentration readings within the point cluster are processed according to a unified standard.

[0051] After receiving the spatial concentration grid, the source location determination unit reads the background concentration baseline value, concentration increment threshold, radius mutation threshold, and PTZ ranging error upper limit. The background concentration baseline value is the environmental baseline field in the cloud-side rule table, derived from historical cruise records and operational management configurations under non-alarm conditions within the monitoring area, serving as the benchmark for comparing grid concentration value differences. The concentration increment threshold is the high-value judgment boundary field, derived from alarm rule configurations and monitored object risk level configurations. The radius mutation threshold is the rule field for identifying distance boundaries in point-to-point distance sorting, derived from the clustering rule table. The PTZ ranging error upper limit is the error boundary field in the equipment calibration record, used by the radius determination submodule to correct candidate cluster radii. These fields are stored in the cloud-side monitoring rule base and refreshed when equipment calibration, alarm rules, or monitored object files are updated.

[0052] The high-value screening submodule is used to obtain the grid concentration values, background concentration benchmark values, and concentration increment thresholds in the spatial concentration grid. It compares the grid concentration values ​​with the background concentration benchmark values ​​and identifies measurement points whose differences are not lower than the concentration increment threshold as candidate high-value points. Candidate high-value points are grid nodes that initially meet the concentration increment rules, and they carry grid node coordinates, grid concentration values, background concentration benchmark source identifiers, concentration increment status identifiers, and adjacent node relationship identifiers.

[0053] When candidate high-value points form a continuous high-value chain with adjacent grid nodes, the high-value screening submodule identifies these points as high-concentration measurement points and generates a high-value point sequence. The determination of a continuous high-value chain is based on the spatial adjacency relationship of grid nodes and the concentration increment status identifier. Isolated candidate high-value points are written into a single-point high-value status identifier and enter the verification cache, but are not entered into the distance analysis submodule. When the background concentration baseline value is missing, the high-value screening submodule writes the spatial concentration grid into a baseline missing status identifier. The source location determination unit does not output the leakage source coordinates and sends a rule missing prompt field to the alarm output unit.

[0054] The distance analysis submodule calculates the inter-point distance distribution between adjacent points in the high-value point sequence and generates a distance distribution record. The inter-point distance distribution is a data object that serves as the sorting basis for the spatial intervals between adjacent high-concentration measurement points, carrying adjacent point identifiers, inter-point distance values, corresponding point quantity fields, and distance source status identifiers. The distance analysis submodule reads the distance coordinates of adjacent high-concentration measurement points according to the spatial order in the high-value point sequence, forms the inter-point distance values, and writes the inter-point distance values ​​and their corresponding point relationships into the distance distribution record.

[0055] The distance analysis submodule writes anomaly status identifiers for high-concentration measurement points with missing distance coordinates, broken adjacency relationships, or abnormal source status. Points with distance analysis anomalies are not included in the cluster radius determination but are retained in the high-value point sequence as verification objects. After the distance distribution record is completed, it is sent to the radius determination submodule as input for cluster radius generation. The distance distribution record also retains the source grid node coordinates and spatial concentration grid index for the source location output submodule to trace when generating leakage source coordinates.

[0056] The radius determination submodule is used to determine the cluster radius based on the distance distribution record. The cluster radius is the spatial boundary field used in adaptive clustering to determine whether high-concentration measurement points belong to the same cluster. It is derived from the point distance sorting results in the distance distribution record and the upper limit of the gimbal ranging error. The radius determination submodule reads the point distance values ​​and the corresponding number of points, generates a distance sorting sequence according to the point distance values ​​in ascending order, and determines the distance boundary position in the distance sorting sequence where the increase in adjacent distances enters the mutation category.

[0057] The radius determination submodule determines the maximum inter-point distance before the distance boundary position as the candidate cluster radius, and expands and corrects the candidate cluster radius according to the upper limit of the gimbal ranging error to generate the final cluster radius. The upper limit of the gimbal ranging error comes from the equipment calibration record and means the error boundary that the laser ranging sensor is allowed to use for spatial clustering correction under field installation conditions. It is stored in the edge-side equipment file and synchronized to the cloud-side monitoring rule base. When the distance sorting sequence does not form a valid distance boundary position, the radius determination submodule generates a cluster radius field according to the conservative boundary state in the clustering rule table and writes a radius insufficient boundary status flag, so that the source output submodule can distinguish between clustering results with clear rule boundaries and those with insufficient rule boundaries.

[0058] The source location output submodule is used to acquire the point clusters obtained by adaptive clustering, the distance coordinates and concentration readings of each high-concentration measurement point within the point cluster, and output the leakage source coordinates and average concentration. Adaptive clustering refers to the point grouping rules within the source location determination unit, a mechanism that, under the constraint of the cluster radius, groups spatially connected high-concentration measurement points into the same point cluster. A point cluster is a set of high-concentration measurement point data objects, carrying a cluster identifier, point coordinates within the cluster, concentration readings within the cluster, cluster radius source identifier, and cluster status identifier.

[0059] The source location output submodule groups high-value point sequences based on cluster radius, reads the ranging coordinates and concentration readings of each high-concentration measurement point within the point cluster, uses the concentration readings as the source field for coordinate contribution, and performs weighted processing on the ranging coordinates within the cluster to form the center coordinates of the point cluster, which are then identified as the leakage source coordinates. The closer the concentration reading is to the high-value center state within the point cluster, the higher its contribution to the center coordinates. Points with abnormal reading states or abnormal source identifiers do not participate in the coordinate contribution processing but are retained in the cluster traceability record. The source location output submodule also performs mean processing on the concentration readings within the point cluster and outputs the average concentration associated with the leakage source coordinates. The average concentration field, carrying the cluster identifier, concentration source status identifier, and acquisition time identifier, is sent to the alarm output unit.

[0060] Alarm Output Unit: Connected to the source location determination unit, this unit receives the leak source coordinates and average concentration, and generates alarm information that records the leak source location marker and concentration display. The alarm information is a shared data object used by the alarm display node and cloud-side maintenance records, carrying the leak source location marker, concentration display data, alarm time identifier, associated component type identifier, source location status identifier, and traceability index. The alarm output unit receives the leak source coordinates, average concentration, and clustering status identifier output by the source location determination unit, and completes coordinate mapping, concentration labeling, and alarm record generation.

[0061] The alarm output unit generates an alarm message when the source location status identifier is valid, and generates a review prompt record when the source location status identifier indicates missing rules, insufficient radius boundaries, or insufficient data. The alarm message is sent to the alarm display node via the cloud-side alarm interface and simultaneously written to the alarm record storage area. The alarm record storage area saves location marker data, concentration display data, alarm time identifiers, sampling point record indexes, spatial concentration grid indexes, and point cluster indexes, enabling maintenance personnel to trace back to the original patrol point and synchronous sampling data from the alarm message.

[0062] The coordinate mapping submodule receives the coordinates of the leak source and converts them into plane marker coordinates for the monitoring area, generating location marker data. This location marker data is the planar display object used by the alarm display node, carrying the plane marker coordinates, associated component type identifiers, monitoring area identifiers, and source location status identifiers. The coordinate mapping submodule reads the area coordinate reference and the leak source coordinates, projects the spatial coordinates onto the monitoring area plane representation, and associates them with the natural gas pipeline, valve, or flange connection end face based on the component type identifiers in the sampling point records.

[0063] When the coordinate mapping submodule detects a leak source coordinate falling outside the monitoring area boundary, it writes a location out-of-bounds status flag and converts the alarm information into a verification prompt record. When the leak source coordinates are inconsistent with the area coordinate reference, the coordinate mapping submodule reads the point cluster index retained by the source location determination unit and calls the spatial concentration grid index to verify the coordinate source. If the verification fails, it does not generate planar marker coordinates and outputs a coordinate mapping exception field to the record generation submodule.

[0064] The concentration labeling submodule receives the average concentration and truncates it according to the concentration display precision to generate concentration display data. The concentration display precision is the display caliber field in the alarm display rules table, derived from the alarm interface configuration and maintenance record specifications, stored in the cloud-based alarm configuration library, and refreshed when the display rules are updated. The concentration display data consists of the concentration field in the alarm information used for display and recording, carrying the average concentration source identifier, display caliber identifier, and concentration status identifier.

[0065] The concentration labeling submodule reads the average concentration and concentration source status identifier. For average concentrations with valid sources, it truncates the data according to the concentration display precision and retains the original average concentration index for traceability. If the average concentration is missing, the source status is abnormal, or it cannot be associated with the point cluster identifier, the concentration labeling submodule writes a concentration labeling abnormal status identifier and forwards the alarm information to the review prompt path. After the concentration display data is completed, it is sent to the record generation submodule, where it forms an alarm record together with the location marker data and the alarm time identifier.

[0066] The record generation submodule is used to associate location marker data, concentration display data, and alarm time identifiers to generate alarm records and issue alarm information based on these records. The alarm time identifier comes from the time record generated when the cloud-side processing node receives the source location determination result, and is used to associate the alarm record with the same patrol acquisition link. The alarm record is a traceable data object in the cloud-side alarm storage area, containing location marker data, concentration display data, alarm time identifier, source location coordinate index, spatial concentration grid index, sampling point record index, and status identifier.

[0067] The record generation submodule issues an alarm message and writes the alarm record to the cloud-side alarm storage area when both the location marker data and concentration display data are valid. If the alarm interface returns an error, a pending-send status flag is written to the alarm record, and the alarm output unit resends it according to the alarm interface's recovery status. If log writing fails, the record generation submodule retains the edge-side sampling link index and the cloud-side source location index, and outputs the storage error field to the operations and maintenance interface. The alarm records, after writing and sending are completed, are used for subsequent patrol verification, alarm display, and tracing of historical leak source locations.

[0068] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments, characterized in that: include: The point planning unit is used to establish a spatial coordinate system in the monitoring area, generate sampling points based on the connection positions of natural gas pipelines, valves and flanges, and output the pan-tilt scanning angle and ranging coordinates corresponding to the sampling points. The scanning and acquisition unit is connected to the point planning unit and is used to control the gimbal equipped with a laser methane telemetry instrument and a laser range sensor to cruise according to the gimbal scanning angle and collect methane concentration data at the range coordinates. The concentration mapping unit, connected to the scanning acquisition unit, is used to receive the methane concentration data, construct a cubic spline interpolation relationship according to the spatial sorting of the ranging coordinates, and generate a spatial concentration grid. The source location determination unit, connected to the concentration mapping unit, is used to filter high-concentration measurement points according to the spatial concentration grid, determine the clustering radius based on the inter-point distance distribution of the high-concentration measurement points, and perform adaptive clustering of the high-concentration measurement points based on the clustering radius, and output the leakage source coordinates and average concentration. An alarm output unit, connected to the source location determination unit, is used to receive the coordinates of the leakage source and the average concentration, and generate alarm information that records the location marker of the leakage source and the concentration display.

2. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 1, characterized in that, The point planning unit includes a coordinate establishment submodule, a component extraction submodule, a point generation submodule, and an angular distance output submodule. The coordinate establishment submodule is used to obtain the monitoring area boundary, the gimbal installation reference point and the height reference point, establish a spatial coordinate system based on a unified length unit, and generate the regional coordinate reference. The component extraction submodule is used to extract the coordinates of candidate components based on the projection relationship of the natural gas pipeline, valve and flange connection positions in the regional coordinate reference. The sampling point generation submodule is used to generate sampling points according to the coordinates of the candidate components; The angular distance output submodule is used to acquire the gimbal mounting reference point, the gimbal horizontal zero position direction, the gimbal pitch zero position direction, and the sampling point. The spatial line connecting the sampling point and the gimbal mounting reference point is determined as the scanning pointing line. Based on the deflection relationship of the scanning pointing line relative to the gimbal horizontal zero position direction and the gimbal pitch zero position direction, the horizontal scanning angle and the pitch scanning angle are generated. The ranging distance is determined based on the spatial length of the scanning pointing line, and the horizontal scanning angle, the pitch scanning angle, and the ranging distance are associated and written into the same sampling point record, and the pan-tilt scanning angle and the ranging coordinates are output.

3. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 2, characterized in that, The process of generating the sampling points is as follows: Obtain the component type corresponding to the coordinates of the candidate components, set the pipeline sampling interval according to the centerline of the natural gas pipeline, set the dense sampling interval according to the valve boundary and flange connection end face, and merge the sampling positions between adjacent candidate component coordinates that are less than the point merging distance threshold. According to the spatial order of each sampling position, write the component type identifier and point order identifier for each sampling position to generate the sampling point.

4. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 1, characterized in that, The scanning and acquisition unit includes an angle analysis submodule, a cruise control unit, a synchronous sampling submodule, and a data binding unit; The angle analysis submodule is used to obtain the gimbal scanning angle and ranging coordinates, convert the gimbal scanning angle into horizontal rotation and pitch rotation commands, and write the ranging coordinates into the sampling point queue. The cruise control unit is used to drive the gimbal to the corresponding attitude in sequence according to the queue of sampling points; The synchronous sampling submodule is used to trigger the laser methane telemetry instrument and the laser range sensor to collect synchronous data after the gimbal attitude is stabilized, and generate synchronous sampling data. The data binding unit is used to associate the synchronous sampling data with the ranging coordinates and output the methane concentration data at the ranging coordinates.

5. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 4, characterized in that, The process of generating the synchronous sampling data specifically includes: The gimbal attitude feedback angle and angle stability threshold are obtained. The gimbal attitude feedback angle is compared with the horizontal rotation command and the pitch rotation command respectively. When the horizontal deviation and the pitch deviation do not exceed the angle stability threshold, a sampling trigger signal is generated. Based on the sampling trigger signal, the concentration reading of the laser methane telemetry instrument and the distance reading of the laser rangefinder are read simultaneously to generate the synchronous sampling data.

6. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 1, characterized in that, The concentration mapping unit includes a data receiving submodule, a coordinate sorting submodule, a spline construction submodule, and a grid generation submodule; The data receiving submodule is used to receive the methane concentration data and extract the concentration reading, the ranging coordinates and the acquisition time identifier, and generate a concentration coordinate record; The coordinate sorting submodule is used to spatially sort the concentration coordinate records according to the horizontal and vertical positions of the ranging coordinates in the spatial coordinate system, thereby generating an ordered concentration sequence. The spline construction submodule is used to construct a cubic spline interpolation relationship based on the coordinate interval of adjacent measurement points in the ordered concentration sequence. The mesh generation submodule is used to assign concentration values ​​to the preset mesh nodes according to the cubic spline interpolation relationship to generate the spatial concentration mesh.

7. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 6, characterized in that, The process of constructing the cubic spline interpolation relationship is as follows: The ranging coordinates, concentration readings, and coordinate intervals of adjacent measurement points in the ordered concentration sequence are obtained. The coordinate intervals are determined as the interpolation segment boundaries. Segmented interpolation constraints are established according to the conditions that the concentrations of adjacent segments are continuous at the connection points and the concentration change trends are continuous. The interpolation curves of each segment are constrained according to the segmented interpolation constraints to generate the cubic spline interpolation relationship.

8. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 1, characterized in that, The source location determination unit includes a high value screening submodule, a distance analysis submodule, a radius determination submodule, and a source location output submodule; The high-value screening submodule is used to obtain the grid concentration value, background concentration benchmark value, and concentration increment threshold in the spatial concentration grid, compare the difference between the grid concentration value and the background concentration benchmark value, and determine the measurement point whose difference is not lower than the concentration increment threshold as a candidate high-value point; when there are no less than two consecutive points of the candidate high-value point in adjacent grid nodes, the candidate high-value point is determined as the high-concentration measurement point, and a high-value point sequence is generated. The distance analysis submodule is used to calculate the distance distribution between adjacent points in the high-value point sequence and generate a distance distribution record. The radius determination submodule is used to determine the cluster radius based on the distance distribution record; The source location output submodule is used to obtain the point clusters obtained by adaptive clustering, the distance coordinates and concentration readings of each high concentration measurement point within the point cluster, perform weighted summation on each distance coordinate based on the concentration readings, determine the center coordinates of the point cluster, the center coordinates of the point cluster are determined as the coordinates of the leakage source, and calculate the mean of each concentration reading within the point cluster, outputting the average concentration associated with the coordinates of the leakage source.

9. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 8, characterized in that, The process of determining the cluster radius specifically includes: Obtain the distance values ​​between points and the corresponding number of points in the distance distribution record, generate a distance sorting sequence in ascending order of the distance values ​​between points, and determine the distance boundary position in the distance sorting sequence where the increase in adjacent distance exceeds the radius mutation threshold; The maximum distance between points before the distance boundary position is determined as the candidate cluster radius, and the candidate cluster radius is expanded and corrected according to the upper limit of the gimbal ranging error to generate the cluster radius.

10. The cloud-edge collaborative monitoring system for natural gas leaks in large-scale open environments according to claim 1, characterized in that, The alarm output unit includes a coordinate mapping submodule, a concentration labeling submodule, and an information output submodule; The coordinate mapping submodule is used to receive the coordinates of the leakage source and convert the coordinates of the leakage source into the coordinates of the monitoring area plane marker, thereby generating location marker data; The concentration labeling submodule is used to receive the average concentration and truncate the average concentration according to the concentration display accuracy to generate concentration display data. The record generation submodule is used to associate the location marker data, the concentration display data, and the alarm time identifier to generate an alarm record, and to issue alarm information based on the alarm record.

Images