Landfill methane early warning and protection distance demarcation method based on real source item

Abstract

The application discloses a landfill methane early warning and protection distance demarcation method based on real source items, and belongs to the field of environmental monitoring and industrial safety risk assessment. The method of the application synchronously acquires the methane concentration, meteorological and image data of the landfill by using a unmanned aerial vehicle, positions the methane emission hot spot through density clustering, quantitatively calculates the emission rate by using an improved mass balance method and makes source intensity effectiveness determination; inputs the high confidence source item into a three-dimensional atmospheric diffusion model, dynamically simulates in combination with real-time meteorological and terrain data, completes model posterior correction through the construction of a correction factor field by using measured data; dynamically demarcates multi-stage explosion danger zones according to the self-adaptive adjustment of the lower explosion threshold, demarcates the odor influence zone in combination with the olfactory threshold and triggers the graded early warning; and finally outputs the precise safety control instruction and the non-circular flexible sanitary protection distance scheme. The application realizes high-precision dynamic evaluation of the methane risk and improves the fine degree of the protection distance demarcation.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring and industrial safety risk assessment technology, and in particular to a method for determining methane early warning and protection distance for landfills based on real source terms. Background Technology

[0002] Sanitary landfills for municipal solid waste are important engineering facilities for disposing of municipal solid waste, but they are also significant anthropogenic sources of methane emissions. Methane's global warming potential is approximately 80 times that of carbon dioxide on a 20-year timescale, significantly impacting climate change. Furthermore, methane's explosive limit in the air is 5%–15% (by volume), and accumulation to this limit can easily lead to fires and explosions. Additionally, the volatile organic compounds, hydrogen sulfide, ammonia, and other malodorous pollutants carried by methane during its escape threaten the quality of life and health of nearby residents.

[0003] Therefore, accurately assessing the location, emission intensity, and atmospheric diffusion behavior of methane leaks at landfills is of significant practical importance for ensuring operational safety, managing environmental risks, and scientifically delineating sanitary protection distances. However, existing technical solutions for landfill methane leak risk assessment and sanitary protection distance delineation have many shortcomings: First, the source terms are distorted, and the assessment basis is unreliable. The methane source terms used in existing technologies are mostly derived from theoretical empirical formulas, historical annual average emission statistics, or the assumption of a uniform area source. These are assumed source terms and cannot reflect high-intensity point or linear unplanned methane leaks caused by engineering defects. Such abnormal leaks are the direct cause of safety accidents.

[0004] Secondly, the model is static in terms of time and space, neglecting the dynamic coupling of meteorological conditions. Existing risk assessments use fixed meteorological parameters as input to calculate the maximum static impact range. However, actual micrometeorological conditions change in real time, directly determining the diffusion characteristics of methane plumes. Static models lead to a serious disconnect between assessment results and reality.

[0005] Third, the management and control measures are crude and outdated. Fixed risk level zoning and health protection distances are insufficient for protection under high-risk weather conditions, and result in the waste of land resources during low-risk periods. Furthermore, they cannot support real-time emergency response, and managers lack dynamic decision-making basis.

[0006] In summary, existing technologies have fundamental shortcomings in terms of the accuracy of methane source term quantification, the spatiotemporal dynamics of risk assessment, and the precision of protection distance delineation. There is an urgent need for a new technical solution that is driven by real source terms, coupled with dynamic meteorological conditions, and accurately delineates risks. Summary of the Invention

[0007] The purpose of this invention is to overcome the core technical deficiencies of existing technologies, such as source term distortion, static assessment, and coarse protection distance delineation, and to provide a method for dynamic risk early warning and sanitary protection distance delineation of methane diffusion in landfills driven by real source terms. This invention, by constructing a complete technical system, achieves high-precision, real-time dynamic assessment of the internal explosion hazard zone and the external odor impact range of landfills, providing a scientific and reliable decision-making basis for on-site safety operation control and refined management of the surrounding environment of landfills.

[0008] To achieve the above objectives, the technical solution of this invention is as follows: Step S1: Synchronous acquisition of multidimensional field data and meteorological parameters A drone platform equipped with a high-precision miniature methane laser gas analyzer, ultrasonic meteorological instrument, and RTK-GPS module was used to conduct terrain-following flight inspections above the landfill, simultaneously acquiring high-resolution two-dimensional methane column integral concentration, spatial meteorological data, and high-definition visible light imagery with uniform timestamps. The gas column integral concentration refers to the total concentration value of methane emitted downwards from the drone, passing through the entire air column from the air to the ground, and summing up all the methane concentrations along that path.

[0009] Among them, the high-precision methane laser gas detector carried by the drone is based on tunable semiconductor laser absorption spectroscopy technology, with a response time of ≤0.1s and a measurement accuracy of 5ppm. The m-level; the airborne ultrasonic meteorological instrument has a sampling rate of ≥10Hz, used to synchronously collect wind speed, wind direction, temperature and atmospheric pressure; the RTK-GPS module provides centimeter-level three-dimensional coordinates and outputs a unified timing pulse signal to ensure strict alignment of data timestamps; a high-definition visible light lens assists in the identification of obvious engineering defects.

[0010] The aerial inspection adopts a ground-following flight mode. The UAV maintains a constant safe relative altitude of 30 to 50 meters above the ground surface, and the laser rangefinder is pointed vertically downwards at the surface of the pile. The flight time is preferentially selected on sunny days and during periods of weak wind (1 to 4 m / s at 10 meters above the ground). At the same time, a ground reference station is configured to monitor the background concentration synchronously.

[0011] During flight, the raw data is preprocessed, including instrument response time delay correction (0.5 to 2 seconds), time registration correction of RTK positioning data and concentration data, and removal of abnormal sampling points during UAV attitude maneuvers to ensure data quality.

[0012] Step S2: Precise location and flux quantification of the actual leakage source Based on the two-dimensional methane column integral concentration matrix obtained in step S1, after subtracting the background value, a spatial density clustering algorithm is used to identify abnormally high value areas and extract three-dimensional coordinates to accurately locate methane emission hotspots. Subsequently, an improved mass balance method is used, combined with the concentration field reconstructed by spatial interpolation and the matched prevailing wind direction, to construct multiple sets of parallel control sections for one-dimensional line integral calculation, quantitatively calculating the methane emission rate of each hotspot. Then, based on synchronous meteorological parameters, the source strength validity of the solution is automatically determined. Low-confidence source strength is replaced by the time-decay weighted average of the historical valid source strength sequence of the hotspot. Source strength refers to the pollutant emission rate of the leakage source, that is, the mass of methane or odorous pollutants leaked per unit time. In this invention, the two-dimensional methane column integral concentration matrix is ​​a coordinate-based methane concentration data table. The matrix can be used to locate the location of high concentrations and the area of ​​hotspots, thus determining the boundary of the danger zone.

[0013] Regarding the identification and precise location of methane emission hotspots: The atmospheric background methane volume concentration is obtained by combining measured values ​​from ground-based benchmark stations or upwind clean zone probes. This concentration is then multiplied by the relative altitude (h) of the UAV's terrain-following flight to obtain the baseline background gas column integral concentration. The baseline value is subtracted from the measured integrated concentration at each sampling point throughout the field to obtain the excess gas column integrated concentration. .

[0014] Set hotspot identification threshold: Exceeding the standard deviation of the baseline data Areas with methane concentrations more than three times higher than normal were identified as methane anomaly zones. Based on a two-dimensional grid methane concentration matrix obtained from UAV inspections, two-dimensional spatial density clustering algorithms such as 2D BSCAN were used to merge adjacent anomalous grid points. Adjacent anomalous grid points on the horizontal coordinate (x, y) were also merged, identifying each independent, contiguous area of ​​abnormally high methane values ​​as a methane emission hotspot. The baseline data refers to the natural atmospheric methane concentration in the upwind side of the landfill, in a clean area with no waste leakage. The concentrations from all sampling points in the baseline area were combined, and the statistical standard deviation was calculated.

[0015] The concentration-weighted centroid coordinates of each hotspot were calculated, and the corresponding surface elevation values ​​were extracted using a high-precision digital elevation model of the landfill to construct the three-dimensional spatial coordinates of the hotspots. These coordinates were then matched with physical engineering defects corresponding to visible light imagery. For each identified emission hotspot, the two-dimensional plane centroid coordinates (x, y, z) were calculated using a concentration-weighted method. s , y s The centroid coordinates are spatially matched with the landfill digital elevation model (DEM) to extract the corresponding surface elevation value (z). s ), construct the three-dimensional spatial coordinates (x) of the leakage source. s , y s ,z sThe three-dimensional coordinates are matched with the visible light image obtained in step S1 to accurately locate the corresponding physical engineering defects such as the damage to the geomembrane and the failure of the gas well cover.

[0016] Regarding the emission flux quantification based on the improved mass balance method of this invention: Using spatial interpolation algorithms such as Kriging, discrete methane column integral concentration data are mapped to a regularized two-dimensional grid, generating a high-resolution global concentration distribution map. The average wind speed and direction values, spatiotemporally matched with emission hotspots, are extracted as effective dominant wind field characteristics. Since the airborne TDLAS detection optical path has already physically completed the concentration integration of the plume along the vertical direction, the two-dimensional area integration of the vertical cross-section required in the traditional mass balance method can be simplified to a one-dimensional line integral on the horizontal plane. Based on this, the method of this invention combines spatial concentration field reconstruction with the calculation rules for averaging multiple parallel equidistant profiles, effectively reducing the random error of a single measurement caused by wind field fluctuations. Spatial layout of multiple sets of parallel control sections: Using the centroid of the target emission hotspot as a reference point, and based on the extracted effective prevailing wind direction, multiple sets of parallel integral control lines are constructed on the two-dimensional concentration distribution map: First, in the upwind and downwind directions of the hotspot area, a line segment of equal length perpendicular to the prevailing wind direction is intercepted to form a reference upwind and downwind profile. Then, using the hotspot centroid as a reference, along the wind direction axis, according to a set spatial equidistant step size (e.g., ... = 5m), and continuously extract N pairs (e.g., N=5) of parallel upwind and downwind profiles of equal length from the reference profile, either inwards towards the center of mass or outwards away from the center of mass. Wind direction lines on each pair of profiles. Downwind line The emission hotspot is spatially surrounded and enclosed, forming N independent flux calculation and control bodies.

[0017] Using the hotspot centroid as a reference point, multiple sets of parallel upwind and downwind integral control profiles are deployed according to the effective prevailing wind direction, forming multiple independent flux calculation control volumes. The methane emission rate of each control volume is calculated by the difference between the downwind one-dimensional linear integral and the upwind one-dimensional linear integral. The methane emission rate of the hotspot is obtained by averaging the calculation results of all control volumes. The calculation formula is: ; ; in, The integral concentration of methane gas column at each interpolation grid point on the profile line is expressed in ppm·m. The effective dominant wind speed component in the direction perpendicular to the profile, matched with the timestamp; The length of the infinitesimal element of the line integral is in meters (m). The formula for calculating the unit conversion and temperature / pressure compensation comprehensive coefficient is as follows: ,in The value represents the molar mass of methane (g / mol). Standard molar volume (L / mol) T and P These represent the measured temperature and air pressure, respectively. N is the number of control volumes. For the methane emission rate of the i-th control body, the set of independent emission rates calculated for N pairs of profiles { , , ..., After performing arithmetic averaging or removing outliers, the mean value is calculated to obtain the methane emission rate of the hotspot region. .

[0018] Regarding the automatic determination and confidence management of source strength validity in this invention, the source strength validity determination rule is that when the wind speed U during the measurement period is lower than a preset lower threshold... , Ideally, the wind speed should be 1.0 m / s, or the standard deviation of the wind direction. Exceeding the preset upper limit threshold At that time, the current source strength is determined to be of low confidence. 60° is preferred.

[0019] For low-confidence source strengths, the time-decayed weighted average of the historical effective source strength sequences of the hotspot is used as a substitute, calculated using the following formula: ; In the formula, The alternative methane emission rate used for low-confidence source strength is the final alternative source strength used. This represents the k-th historical effective source strength value of this hotspot. For time decay weight, Let k be the time interval between the k-th valid measurement and the current time. The preferred value is the attenuation constant. = 0.05 h -1 This translates to a half-life of approximately 14 hours. This alternative value retains the physical plausibility of the recent emission characteristics of this hotspot while giving higher weight to recent measurements through exponential decay, thus avoiding the use of overly outdated data.

[0020] When a hot spot is determined to have low confidence after three or more consecutive inspections, the system automatically marks the hot spot as a source to be supplemented for measurement. In the next UAV inspection mission planning, priority will be given to arranging multi-angle supplementary cross-sectional measurements of the hot spot under different wind conditions in order to obtain a high-confidence updated source strength as soon as possible.

[0021] Step S3: Real Source Term-Driven Dynamic Diffusion Simulation and Posterior Correction of Methane The high-confidence hotspot 3D coordinates and emission rates obtained in step S2 are used as real source terms to input the 3D atmospheric diffusion model, and the source strength of characteristic odor pollutants is equivalently converted based on the co-source and associated fugitive characteristics. Real-time meteorological data and high-precision terrain digital elevation models are connected to dynamically simulate and output the concentration matrix and contour distribution evolution map of each pollutant in multi-dimensional spatiotemporal conditions. Subsequently, in the observation area covered by the UAV, the ratio of the measured methane concentration to the model predicted concentration is extracted to construct a spatial correction factor field, and the posterior bias correction is performed on the model output concentration field.

[0022] Regarding the selection of dynamic simulation models and input parameter settings: This invention supports two three-dimensional atmospheric diffusion models, AERMOD and CALPUFF. AERMOD is suitable for scenarios with relatively flat terrain and an influence range of less than 50km, while CALPUFF is suitable for scenarios with significant terrain undulations, local wind fields, or scenarios requiring long-term dynamic simulation. The concentration matrix in multidimensional spatiotemporal dimensions refers to the set of methane or odor pollutant concentration values ​​at each time step for all grid points within the computational domain. It is a four-dimensional matrix with spatial (x, y, z) and temporal t. The matrix values ​​are generated by model calculation, specifically by combining the output of the three-dimensional diffusion model with post-hoc correction based on measured data. The contour distribution evolution map described in this application refers to a set of continuously changing pollutant concentration contour maps output based on the three-dimensional atmospheric diffusion model, combined with real-time meteorological data and dynamic simulation calculation using a high-precision digital elevation model (DEM). This diagram, centered on the landfill, uses concentration contour lines to define the spatial distribution of different pollutant concentration levels within a two-dimensional plane. It visually reflects the diffusion trajectory, concentration gradient, range of influence, and boundary drift of methane and associated odorous pollutants at different times, providing a visual representation for characterizing the spatiotemporal dynamic distribution of pollutants in the field of atmospheric diffusion simulation. Concentration contour lines connect points of equal concentration to delineate the concentration distribution range and represent diffusion boundaries.

[0023] The source parameters are the three-dimensional coordinates of the hotspot and the methane emission rate output in step S2. The emission method is set as either a ground-based point source or a vertical injection point source based on the hotspot characteristics. The source strength of the characteristic odor pollutant is calculated based on the equivalent conversion of the co-origin and co-occurrence characteristics of methane and odor pollutants, using the following formula: ; In the formula, The actual methane emission rate calculated in step S2; This is the volume concentration ratio of odorous pollutants to methane. This represents the molar mass of the corresponding malodorous pollutant; This represents the molar mass of methane. The calculated intensity of each odor pollutant source is... Using the same spatial coordinates as the methane source strength, the diffusion model is synchronously input for multi-component coupled calculations. That is, under the same meteorological field, topographic conditions and diffusion model framework, the diffusion process of methane and various odorous pollutants is calculated in parallel, sharing diffusion parameters and synchronously outputting the concentration fields of each pollutant.

[0024] The meteorological input parameters are real-time data from ground meteorological stations standardized by the AERMET or CALMET meteorological preprocessor, including wind speed at 10 meters height, wind direction, surface temperature, relative humidity, and atmospheric stability level. The terrain boundary conditions adopt a high-precision digital elevation model with a resolution of no less than 5 meters × 5 meters, covering an area with a radius of no less than 2 km centered on the landfill, and generate terrain uplift correction parameters after preprocessing.

[0025] Model Operation and Output: The model performs diffusion calculations within a 2km×2km spatial domain centered on the landfill and a 0-100m vertical stratification, with a time step of 0.5-1 hour. The output includes a three-dimensional spatiotemporal concentration matrix, contour maps of specific height layers (such as the 1.5m breathing zone and the 3m-5m equipment exhaust pipe height), and the maximum influence envelope and time series exceeding a specific concentration threshold.

[0026] Posterior bias correction involves selecting at least 20 spatially evenly distributed reference points within the UAV coverage area. The measured integral concentration of the air column from the UAV is divided by the relative flight altitude to approximate the surface methane volume concentration. C obs,j , C obs,j The average volume concentration is obtained by dividing the integral concentration of the air column by the relative flight altitude, and the simulated concentration values ​​of the model at the same coordinate and altitude layers are also extracted. C mod,j In this invention C obs,j Let be the methane concentration measured by the UAV at the j-th reference point; C mod,j The model calculates the methane concentration value at the j-th reference point and corresponding altitude layer; the local correction factor for each reference point is calculated. Remove C mod,j Points below the instrument's detection limit, when C mod,j When the value is below the instrument's detection limit, it is not included in the calibration calculation to avoid division by zero anomalies. The set of calibration factors for all valid reference points { α j By employing inverse distance weighted interpolation (IDW) or ordinary kriging interpolation, the effective local correction factors are used to generate a continuous two-dimensional spatial correction factor field. , The field of concentration coefficients is a two-dimensional continuous spatial correction coefficient field, obtained through spatial interpolation based on the reference point correction factor, covering the entire computational domain of the model. The original concentration field output by the model is multiplied grid-by-grid by the correction factor field to obtain the a posteriori-corrected concentration field. The corrected concentration field serves as the actual input for the risk zone delineation in step S4, replacing the original model output. Is the model in position? The simulated methane concentration calculated at time t. This represents the final methane spatiotemporal concentration field after a posteriori correction.

[0027] Step S4: Dynamic delineation and adaptive early warning of multi-level risk zones The lower explosive limit (LEL) of methane is 5% (volume concentration, i.e., 50,000 ppm) under standard atmospheric conditions. Traditionally, a fixed LEL percentage is used as the classification threshold, but the actual ignition risk and diffusion hazard level of landfills are influenced by a variety of environmental factors. This invention introduces an adaptive threshold adjustment mechanism to make the classification more closely reflect the real-time hazard status of the site.

[0028] Based on the percentage threshold of the lower explosive limit of methane, the concentration field, after a posteriori correction, is divided into multiple levels of explosion hazard zones in real time, with the threshold of each level adaptively adjusted according to real-time environmental conditions; at the same time, the boundary of the odor-affected zone is dynamically delineated based on the odor threshold; and the corresponding level of sound, light, communication and control early warning signals are automatically triggered according to the spatial relationship between the risk zone and sensitive targets.

[0029] The baseline classification thresholds for adaptive classification and dynamic delineation of the lower explosive limit of methane in explosion hazard zones are: Level 1 High Hazard Zone 25% LEL (12500 ppm), Level 2 Warning Zone 10% LEL (5000 ppm), and Level 3 Concern Zone 5% LEL (2500 ppm).

[0030] The adaptive adjustment rule is: (1) High temperature and low humidity ignition risk correction: When the ambient temperature Temperature above 35°C and relative humidity RH When the concentration is below 30%, the concentration threshold for each zone is multiplied by the temperature and humidity correction factor. β TH (0.8 is preferred); (2) Correction for proximity zone of non-explosion-proof equipment: When non-explosion-proof electrical equipment in operation is located in a level 3 concern zone or above, within a 50-meter safety radius centered on the equipment, the classification threshold is multiplied by the equipment proximity correction factor. β EQ = 0.7; (3) Correction for unfavorable atmospheric diffusion in strong stability: When the atmospheric stability level is E or F, a spatial expansion coefficient is applied to the boundaries of the secondary warning area and the tertiary concern area. γ stab (Option 1.3) The boundary of the Level 1 high-risk area does not expand. Spatial expansion coefficient. γ stab It is a safety amplification factor that proportionally expands the isopleth boundaries of the Level II alert zone and Level III concern zone when the atmosphere is highly stable. It is used to compensate for the risk that methane is not easy to diffuse and is easy to accumulate near the ground under highly stable conditions, thereby expanding the scope of early warning and control.

[0031] When multiple rules are triggered simultaneously, the threshold correction coefficient is multiplied, the boundary expansion coefficient is applied independently, and the threshold automatically recovers to the baseline value under standard conditions.

[0032] Temperature and humidity corrected threshold : ; Device proximity correction threshold : ; Threshold correction when multiple conditions are superimposed : ; Extension of the strongly stable atmospheric boundary: ; In the formula: This represents the temperature and humidity correction factor, when... Triggered at 35℃ and relative humidity (RH) < 30%; This indicates the proximity correction factor for non-explosion-proof equipment, effective within a 50m radius of the equipment coordinates. This represents the atmospheric boundary expansion coefficient for strongly stable atmospheres, triggered when atmospheric stability is E / F class. This represents the baseline threshold for the percentage of LEL (Leadership Level) in each partition under standard conditions. This represents the distance from the boundary of the original concentration contour lines output by the model; This represents the actual control boundary distance after applying the expansion coefficient.

[0033] Dynamic delineation of the odor impact zone: Based on the "Odor Pollutant Emission Standard," the odor thresholds of characteristic odor pollutants are determined. Isoconcentration envelopes corresponding to each pollutant reaching its odor threshold at a 1.5m breathing zone height are extracted. The envelope with the longest diffusion distance among all pollutants is taken as the outer boundary of the comprehensive odor impact zone. This boundary is updated in real-time according to meteorological conditions. The longest diffusion distance refers to the distance at which each odor pollutant reaches its odor threshold, calculated separately in the same spatial direction from the center or boundary of the landfill, and the largest value is taken as the odor impact distance in that direction. The envelope formed by connecting the farthest points in all directions is the outer boundary of the comprehensive odor impact zone. Isoconcentration envelopes are closed boundary lines connecting the same concentration values.

[0034] Automatic early warning signal issuance: The system triggers tiered early warnings based on the area of ​​the risk zone, its boundary location, and its relationship with sensitive targets. (1) Level 1 High-Risk Warning: Activate the explosion-proof sound and light alarm network, pop up a window in the central control room and send a high-frequency alarm sound to the management personnel with the highest priority blocking command; (2) Level II alert: Push risk coordinates and concentration trends to workers, remind them to wear explosion-proof equipment and prepare exhaust fans; (3) Level 3 attention and early warning: Write abnormal areas into the inspection log, generate encrypted inspection work orders and adjust the drone inspection route; (4) Odor impact warning: When the odor impact area is less than 500 meters away from the sensitive target, an early warning will be sent to the environmental protection department and the operator and odor suppression measures will be suggested; when it covers the sensitive target, it will be upgraded to a high-level warning and an early warning information with GIS vector layer will be sent to the management personnel and the environmental supervision platform.

[0035] Explosion hazard warnings and odor impact warnings operate in parallel, and control orders are implemented in overlapping areas according to the principle of choosing the higher level.

[0036] Step S5: Dynamic control strategy and output of health protection distance Based on the dynamically delineated risk zone boundaries, precise spatial safety control instructions are generated, including dynamic entry restrictions, forced equipment power outages, and precise repair work order assignments. Combining long-term meteorological probability statistics with real source terms, a non-circular flexible health protection distance vector scheme is output, which is dynamically adjusted according to the probability of exceeding the standard and seasonal meteorological characteristics.

[0037] Real-time safety control command generation: (1) Space dynamic entry prohibition command: Convert the boundary of the explosion hazard zone into a GIS vector polygon in Shapefile or GeoJSON format and send it to the on-site personnel positioning system. When personnel approach, the equipment will automatically vibrate and alarm; (2) Forced hardware intervention command: When the high-risk area covers the work surface, the power supply of non-explosion-proof electrical equipment in the area will be automatically cut off through the Internet of Things relay, and open flame work will be stopped; (3) Precise repair work order dispatch: The hot spots are sorted according to the severity of the leakage to generate repair work orders with precise three-dimensional coordinates and real scene images to guide the engineering team to carry out targeted repairs.

[0038] The dynamic delineation of flexible sanitary protection distance is based on the statutory minimum sanitary protection distance calculated under the most unfavorable meteorological conditions. The flexible sanitary protection distance is generated as follows: (1) Based on the historical hourly meteorological data of the landfill area for at least one year, combined with the real source term batch running diffusion model; with the landfill as the center, different directions are divided according to spatial orientation, and the total duration of the isopleths of the odor comprehensive impact area reaching the odor threshold in each direction is statistically analyzed to obtain the time exceedance probability distribution in different directions; (2) The impact distance corresponding to the preset exceedance probability percentile value of each direction, such as the 95th percentile value, is used as the design protection distance for that direction to form a non-circular flexible protection envelope; (3) The dominant meteorological conditions in winter and summer are distinguished, and seasonally differentiated protection distance schemes are output; (4) After the forecast of extreme meteorological events is released, the temporary extended protection distance plan is activated, pushed to the environmental protection department, and protective measures are taken in advance.

[0039] Flexible sanitary protection distances are output in GIS vector graphics formats such as Shapefile for use in landfill operation environment management.

[0040] The beneficial effects of this invention are as follows: Compared with the prior art, the present invention has the following significant advantages: 1. The source terms are real and reliable, and the accuracy of the assessment is greatly improved: Abandoning the assumed source terms estimated by traditional experience, the actual leakage location and mass flux are obtained by on-site quantification by drones. The simulation results can truly reflect the actual leakage status of the landfill, ensuring the reliability of the risk assessment from the source.

[0041] 2. Achieve dynamic spatiotemporal early warning and compensate for the deficiencies of static models: Continuously inject dynamic meteorological data into the diffusion model to realize the dynamic drift simulation of the boundary of the danger zone. When the wind direction or leakage intensity fluctuates, the boundary of the danger zone and the warning level are automatically updated to gain golden time for emergency response.

[0042] 3. Refined protection distance delineation, balancing safety and efficiency: Breaking through the static delineation model, it outputs flexible health protection distances that are adjusted according to the season or real-time weather conditions. It ensures safety during high-risk periods and releases the constraints of the protection range during low-risk periods, thereby improving the intensive use of land and operational efficiency.

[0043] 4. Forming a closed-loop engineering system to support precise remediation decisions: Simultaneously outputting the precise location and magnitude ranking of leakage hotspots, providing spatial priority suggestions for engineering remediation, and realizing closed-loop engineering management from risk identification to precise remediation.

[0044] 5. Bidirectional coupling between modules enhances system robustness: By automatically determining the effectiveness of source strength, a reverse constraint is formed on the inversion of source terms by meteorological data. Through model posterior correction, UAV measured data simultaneously drive source strength and correct concentration deviation, establishing a bidirectional information flow and overcoming the defect of unidirectional error transmission and amplification.

[0045] 6. Adaptive adjustment of risk thresholds to balance early warning accuracy and safety redundancy: Environmental factors are coupled into the threshold setting logic, the early warning range is expanded in advance during high-risk periods, and excessive early warnings are avoided during low-risk periods, taking into account both safety and operational efficiency.

[0046] 7. Source strength confidence level management to ensure continuous assessment in all weather conditions: By replacing low-confidence source strengths with historical valid source strength sequences, source strength estimates with physical basis can still be output even under unfavorable weather conditions, avoiding assessment gaps and ensuring the continuity of dynamic early warning. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the overall method of the present invention.

[0048] Figure 2 This is a schematic diagram illustrating the principle of UAV data synchronization acquisition in step S1 of the present invention.

[0049] Figure 3 This is a schematic diagram illustrating the flux calculation principle of the improved mass balance method in step S2 of the present invention.

[0050] Figure 4 This is a schematic diagram illustrating the principle of concentration field posterior correction in step S3 of the present invention, wherein... Figure 4 (a) in the figure represents the concentration field measured by the UAV. Figure 4 In the diagram, (b) represents the original output concentration field of the model. Figure 4 (c) in the equation represents the correction factor field. Figure 4 (d) in the figure represents the corrected concentration field.

[0051] Figure 5 This is a schematic diagram of the methane plume concentration distribution and dynamic evolution in step S3 of the present invention.

[0052] Figure 6 This is a schematic diagram of the dynamic delineation of multi-level risk areas in step S4 of the present invention.

[0053] Figure 7 This is a schematic diagram comparing the flexible health protection distance under different meteorological conditions in step S5 of the present invention. Detailed Implementation

[0054] 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 examples described herein are only some embodiments of this invention, not all embodiments, and are not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0055] This embodiment uses a municipal solid waste sanitary landfill as the application object, and performs a real source term-driven dynamic risk warning and sanitary protection distance determination for methane diffusion in the landfill. The flowchart of the method of this invention is as follows. Figure 1 As shown, the specific steps are as follows: Step S1: Synchronous acquisition of multidimensional field data and meteorological parameters A schematic diagram illustrating the principle of a drone equipped with multiple sensors performing terrain-following flight and synchronous data collection above a landfill is shown below. Figure 2 As shown, Figure 2 In the middle section: Item 1 represents an airborne ultrasonic meteorological instrument, used to synchronously collect real-time meteorological parameters such as wind speed, wind direction, temperature, and atmospheric pressure, with a sampling rate ≥10Hz. It provides accurate real-time meteorological input for methane emission rate calculation, source strength effectiveness determination, and atmospheric diffusion simulation, ensuring strict alignment of meteorological data with methane concentration and location data timestamps, thus improving data matching accuracy. Item 2 represents an RTK-GPS spatial positioning module, used to provide centimeter-level three-dimensional spatial coordinates to determine the location of the UAV and the leak point. It outputs a unified timing pulse to ensure millisecond-level time synchronization of concentration, meteorological, and image data. Combined with an elevation model, it constructs three-dimensional coordinates of the methane emission hotspot, achieving precise location of the leak point. Item 3 represents a high-precision methane laser gas analyzer and a high-definition visible light lens. The laser gas analyzer, based on TDLAS technology, acquires high-resolution integrated methane gas column concentration, while the visible light lens captures high-definition images of the landfill, assisting in locating physical engineering defects such as damaged geomembrane and malfunctioning manhole covers. The dotted line at number 4 represents the drone's terrain-following flight path; number 5 is the ground reference station; number 6 indicates the first methane leak point; number 7 indicates the second methane leak point; and number 8 indicates the landfill. L Indicates the laser optical path. h This indicates the relative altitude of the drone during its terrain-following flight. It utilizes an industry-grade drone platform and is equipped with a high-precision methane laser gas detector based on tunable semiconductor laser absorption spectroscopy technology (response time 0.08s, measurement accuracy 4ppm). The system includes an airborne ultrasonic weather instrument with a sampling rate of 15Hz, a centimeter-level RTK-GPS module, and a high-definition visible light lens.

[0056] The drone patrols above the landfill in a terrain-following flight mode, maintaining a constant relative altitude of 40 meters above the ground surface, with the laser rangefinder pointing vertically downwards at the surface of the landfill. The flight is scheduled for a clear day with a wind speed of 2 m / s at 10 meters above the ground. At the same time, a ground reference station is set up 500 meters upwind of the landfill to monitor the background methane concentration.

[0057] During flight, methane concentration, meteorological data, and RTK coordinates are recorded synchronously at the millisecond level using a unified timestamp. The raw data is preprocessed: the instrument response time delay is corrected by 1 second, the time registration of RTK positioning data and concentration data is completed, and abnormal sampling points during sharp turns and rapid ascents and descents of the UAV are removed.

[0058] In a specific embodiment of the present invention, the UAV conducts a terrain-following flight inspection, the laser gas meter outputs the integrated concentration of methane gas column, the meteorological instrument simultaneously collects wind speed, wind direction, temperature and air pressure, and the visible light lens simultaneously captures high-definition orthophotos. All data are stored in the airborne system with the same timestamp, and time delay correction, coordinate registration, and removal of abnormal points are performed. Finally, a two-dimensional concentration field, meteorological field and image with a unified timestamp are output.

[0059] Step S2: Precise location and flux quantification of the actual leakage source A schematic diagram of the flux calculation principle of the improved mass balance method is shown below. Figure 3 As shown, the baseline background methane volume concentration, measured at a ground-based reference station, is multiplied by a flight altitude of 40 meters to obtain the baseline background gas column integral concentration. Subtracting this baseline value from the measured integral concentrations at all sampling points yields the excess gas column integral concentration. Regions with excess gas column integral concentrations exceeding three times the standard deviation of the background data are identified as methane anomaly areas. The 2D DBSCAN algorithm is used to merge the anomaly grid points, identifying three independent methane emission hotspots.

[0060] The concentration-weighted centroid coordinates of each hotspot were calculated, and the corresponding surface elevation values ​​were extracted by combining the 1m×1m high-precision digital elevation model of the landfill. The three-dimensional spatial coordinates of the three hotspots were constructed. After matching with high-definition visible light images, it was determined that the hotspots corresponded to local damage to the geomembrane, failure of the gas well cover seal, and failure of the gas collection pipeline seal, respectively.

[0061] Kriging interpolation was used to map discrete concentration data onto a regularized two-dimensional grid, generating a high-resolution concentration distribution map. The average wind speed and direction values ​​that spatiotemporally matched each hotspot were extracted as effective dominant wind field features. Using the centroid of each hotspot as a reference point, five pairs of parallel upwind and downwind integral control profiles were deployed according to the dominant wind direction, with an adjacent profile spacing of 5m, forming five flux calculation control volumes.

[0062] The methane emission rate of each control volume was calculated using an improved mass balance method. The average of the calculation results for the five control volumes yielded methane emission rates of 0.8 kg / h, 0.5 kg / h, and 0.3 kg / h for the three hotspots.

[0063] The source strength validity is determined for the solution results. The lower limit threshold for wind speed is set to 1.0 m / s, and the upper limit threshold for wind direction standard deviation is set to 60°. In this embodiment, the wind speed during the measurement period is 2 m / s and the wind direction standard deviation is 30°. The source strength of the three hot spots is determined to be of high confidence and no replacement is required.

[0064] Step S3: Real Source Term-Driven Dynamic Diffusion Simulation and Posterior Correction of Methane A schematic diagram illustrating the principle of concentration field posterior correction is shown below. Figure 4As shown, the landfill terrain in this embodiment is relatively flat. AERMOD was selected as the three-dimensional atmospheric diffusion model, and the three-dimensional coordinates of three hotspots and the methane emission rate were used as the actual source terms input into the model. Gas composition analysis reports from recent pipeline pumping data of the landfill were collected, determining the volume concentration ratios of hydrogen sulfide, ammonia, and methane to be 0.002 and 0.01, respectively. The emission rates of hydrogen sulfide and ammonia were calculated using equivalent conversion formulas and used as the synchronous input model for the characteristic odor pollutant source strength.

[0065] The real-time meteorological data from the landfill's surface meteorological station includes wind speed of 2 m / s at a height of 10 meters, wind direction of southerly wind, surface temperature of 25°C, relative humidity of 60%, atmospheric stability of Class C, and high-precision digital elevation model data with a resolution of 5 meters × 5 meters and a coverage radius of 2 km. After preprocessing with AERMET and AERMAP, the data is input into the model.

[0066] The model performs diffusion calculations in a 2km×2km spatial domain and a 0-100m vertical stratification with a time step of 1 hour. The output includes a three-dimensional spatiotemporal concentration matrix, isopleths of methane and odor pollutants at a height of 1.5m breathing zone and 4m equipment exhaust pipe, as well as the maximum influence envelope and time series exceeding 5% LEL. A schematic diagram of the methane plume concentration distribution and dynamic evolution is shown in Figure 5.

[0067] Thirty spatially evenly distributed reference points were selected in the drone coverage area. The measured integral concentration of the air column was divided by the flight altitude of 40 meters to convert it into the surface methane volume concentration. The simulated concentration at the corresponding location of the model was extracted, and the local correction factor of each reference point was calculated. Two points with simulated concentrations lower than the instrument detection limit were removed. A two-dimensional spatial correction factor field was generated by inverse distance weighted interpolation. The original concentration field of the model was multiplied by the correction factor field grid by grid, and the a priori bias correction was performed.

[0068] Step S4: Dynamic delineation and adaptive early warning of multi-level risk zones A diagram illustrating the dynamic delineation of multi-level risk zones is shown below. Figure 6 As shown, when the concentration of associated odor pollutants at the surface breathing zone height output by the model in step S3 reaches the odor threshold (OTV) isoconcentration envelope exceeding the landfill boundary and spatially intersects with or covers sensitive targets outside the boundary, such as residential areas, schools, and hospitals, an odor nuisance risk warning instruction is immediately generated. The warning level is determined according to the type of sensitive target, population density, and duration of exceeding the standard, and is simultaneously pushed to the control platform. For areas that have triggered warnings, the system automatically extracts meteorological conditions and emission source intensity information for that period, forming a warning source tracing record to provide a basis for subsequent control and scheduling.

[0069] In this embodiment, the ambient temperature is 25℃, the relative humidity is 60%, no non-explosion-proof electrical equipment is in operation, the atmospheric stability is Class C, and the benchmark classification thresholds for the lower explosive limit of methane are adopted: Level 1 High Risk Zone 25% LEL, Level 2 Warning Zone 10% LEL, and Level 3 Concern Zone 5% LEL. Based on the a posteriori-corrected concentration field, multi-level explosion hazard zones corresponding to three hotspots are dynamically delineated. The Level 1 High Risk Zones are all located within 5 meters of the hotspots, the Level 2 Warning Zones are 5–15 meters from the hotspots, and the Level 3 Concern Zones are 15–30 meters from the hotspots.

[0070] According to the "Odor Pollutant Emission Standard", the odor threshold of hydrogen sulfide is determined to be 0.00001 ppm and the odor threshold of ammonia is 0.02 ppm. The isoconcentration envelopes of the two pollutants reaching the odor threshold at a breathing zone height of 1.5 m were extracted. Hydrogen sulfide has a longer diffusion distance and is used as the outer boundary of the comprehensive odor impact zone. The closest distance between this boundary and the landfill boundary is 200 meters. There are no sensitive targets in the impact zone.

[0071] Based on the distribution of risk areas, the system triggers a level 3 alert, writes the coordinates of the three hotspots into the inspection log, generates an encrypted inspection work order, adjusts the key routes for subsequent daily inspections of the drone, and performs encrypted monitoring of the hotspot areas.

[0072] Step S5: Dynamic control strategy and output of health protection distance A diagram comparing flexible health protection distances under different meteorological conditions is shown below. Figure 7 As shown, based on the dynamically defined risk zone boundaries, safety control instructions are generated: the boundaries of the three hotspots' Level 3 concern zones are converted into GeoJSON format GIS vector polygons, i.e., three closed danger circles are drawn on the map using coordinates, and saved as directly identifiable coordinate files for automatic entry prohibition, alarm, and location tracking, and are sent to the UWB safety helmet positioning system of on-site workers. When personnel approach the boundary, the safety helmet automatically vibrates and alarms. Since the risk zone does not cover the work surface, there is no need to cut off the power to electrical equipment. The three hotspots are sorted according to the severity of leakage, and a precise repair engineering dispatch order is generated, which includes the three-dimensional coordinates and visible light images of each hotspot, prioritizing the engineering team to carry out hot-melt repairs on the hotspots with localized damage to the geomembrane.

[0073] Historical hourly meteorological data for the landfill site over the past three years were collected. Combined with real source terms from three hotspots, the AERMOD model was run in batches to statistically analyze the temporal exceedance probability distribution of isopleths reaching the odor threshold in the east, south, west, and north directions of the comprehensive odor impact area. The design protection distance was determined using the 95th percentile distance for each direction: 500 meters to the south (prevailing wind direction), 300 meters to the east, 250 meters to the west, and 200 meters to the north, forming a non-circular, flexible protection envelope.

[0074] The meteorological characteristics of the region in winter and summer are differentiated. In winter, northerly winds prevail and atmospheric stability is relatively high, so the protection distance on the north side is adjusted to 300 meters, while the distances in other directions remain unchanged. In summer, when convection is strong, the protection distances in all directions are reduced by 10%. At the same time, an early warning plan for extreme weather events is established. When strong temperature inversion and low wind speed with stable weather are forecast, the protection distances in all directions are temporarily extended by 50%, and the early warning information is sent to the local environmental protection department.

[0075] This embodiment realizes the quantification of real source terms of methane leakage in the landfill, dynamic diffusion simulation, multi-level risk early warning, and flexible delineation of sanitary protection distances. It provides a scientific basis for the safe operation control and surrounding environmental management of the landfill, and effectively improves the precision level of methane risk control in the landfill.

[0076] The above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for landfill methane early warning and protection distance delineation based on real source terms, characterized in that, Includes the following steps: S1: Acquire high-resolution two-dimensional methane column integral concentration, space meteorological data, and high-definition visible light imagery of landfills with uniform timestamps; S2: After subtracting the background value from the two-dimensional methane column integral concentration data obtained in step S1, identify abnormally high value areas and extract three-dimensional coordinates to locate methane emission hotspots. Calculate the methane emission rate of each hotspot and then automatically determine the source strength validity of the solution based on synchronous meteorological parameters. The source strength with low confidence is replaced by the time decay weighted average of the historical valid source strength sequence of the corresponding hotspot. S3: The three-dimensional coordinates of the methane emission hotspots identified in step S2, and the high-confidence methane emission rates after source strength validity determination and processing in step S2, are used as the real source terms. The source strengths of the characteristic odor pollutants, which are converted by equivalent transformation, are input into the three-dimensional atmospheric diffusion model. Real-time meteorological data and high-precision terrain digital elevation model are connected to dynamically simulate and output pollutant concentration matrix and contour distribution evolution map. Then, the spatial correction factor field is constructed using UAV measured data to perform posterior bias correction on the model output concentration field. S4: Based on the percentage threshold of the lower explosion limit of methane, the concentration field corrected by the S3 step is divided into multi-level explosion hazard zones in real time. The boundary of the odor-affected zone is dynamically delineated based on the odor threshold. The corresponding level of early warning signal is triggered according to the spatial relationship between the risk zone and the sensitive target. S5: Based on the boundaries of the multi-level explosion hazard zones and dynamically delineated odor impact zones defined in real time in step S4, generate safety control instructions, and output protection schemes by combining long-term meteorological probability statistics and real source terms.

2. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that: The method for subtracting the background value in step S2 is to combine the measured value of the ground reference station or the detected value of the clean zone upwind to obtain the atmospheric background methane volume concentration, multiply it by the relative altitude of the UAV's terrain-following flight to obtain the reference background air column integral concentration, and subtract this reference value from the measured integral concentration of each spatial sampling point in the entire field to obtain the excess air column integral concentration.

3. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that, The criteria for identifying methane emission hotspots in step S2 are as follows: a two-dimensional spatial region where the integral concentration of the excess gas column exceeds a preset multiple of the standard deviation of the background data. An independent methane emission hotspot is obtained by merging adjacent abnormal grid points through a two-dimensional spatial density clustering algorithm, and the concentration-weighted centroid coordinates of each hotspot are calculated. The corresponding surface elevation values ​​are extracted by combining the high-precision digital elevation model of the landfill, and the three-dimensional spatial coordinates of the hotspots are constructed.

4. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that, The improved mass balance method for calculating methane emission rates in step S2 includes: S21: Spatial interpolation algorithm is used to map discrete methane column integral concentration data to a regular two-dimensional grid to generate a high-resolution global methane column integral concentration distribution map. The average wind speed and wind direction that are spatiotemporally matched with emission hotspots are extracted as effective dominant wind field features. S22: Using the centroid of the emission hotspot as the reference point, multiple sets of parallel upwind and downwind integral control profiles are set up according to the effective prevailing wind direction to form multiple independent flux calculation control bodies. S23: Calculate the methane emission rate of each control volume by the difference between the downwind one-dimensional line integral and the upwind one-dimensional line integral, and average the calculation results of all control volumes to obtain the methane emission rate of the hot spot.

5. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that, The rule for automatically determining the validity of source strength in step S2 is as follows: when the wind speed during the measurement period is lower than a preset lower threshold, or the standard deviation of the wind direction exceeds a preset upper threshold, the current source strength is determined to be of low confidence. For low-confidence source strengths, the time-decayed weighted average of the historical valid source strength sequence of the corresponding hotspot is used as a substitute. The formula for calculating the time-decayed weighted average is as follows: in The alternative methane emission rate used for low-confidence source strength. This represents the k-th historical effective source strength value of this hotspot. For time decay weight, Let k be the time interval between the k-th valid measurement and the current time. is the attenuation constant.

6. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that, The equivalent conversion formula for the source strength of characteristic odor pollutants in step S3 is as follows: In the formula, The emission rate of characteristic odorous pollutants is called the source strength. The actual methane emission rate calculated in step S2; The volume concentration ratio of characteristic odorous pollutants to methane; This represents the molar mass of the corresponding malodorous pollutant; denoted as , where is the molar mass of methane.

7. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that: In step S3, the posterior bias correction includes selecting spatially uniform reference points in the UAV coverage area, extracting the measured methane volume concentration values ​​from the UAV and the model-simulated concentration values ​​from each reference point, calculating local correction factors, generating a two-dimensional spatial correction factor field using a spatial interpolation algorithm, and multiplying the original concentration field output by the model with the correction factor field grid by grid to obtain the posteriorly corrected concentration field.

8. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that: The method for delineating the odor-affected area in step S4 includes extracting the isoconcentration envelopes of characteristic odor pollutants at the height of the surface breathing zone where they reach the odor threshold, and taking the envelope with the longest diffusion distance among all pollutants as the outer boundary of the comprehensive odor-affected area.

9. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that: The S5 step safety control instructions include: a spatial dynamic prohibition instruction to convert the boundary of the explosion hazard zone into a GIS vector polygon; a mandatory hardware intervention instruction to cut off the power supply of non-explosion-proof electrical equipment in the area; and a precise repair engineering dispatch order sorted by the severity of the leak and accompanied by accurate three-dimensional coordinates.

10. The method for landfill methane early warning and protection distance delineation based on real source terms according to claim 1, characterized in that, The process of generating the non-circular flexible sanitary protection distance in step S5 includes: based on at least one year of historical hourly meteorological data of the landfill area, combined with the batch running diffusion model of real source terms; dividing different directions according to spatial orientation with the landfill as the center, and statistically analyzing the proportion of the duration of the isopleths of the odor comprehensive impact area reaching the odor threshold in each direction to the total duration of the time, to obtain the time exceedance probability distribution; using the impact distance corresponding to the preset exceedance probability percentile value in each direction as the design protection distance for the corresponding direction, and adjusting it in combination with seasonal meteorological characteristics to obtain a seasonally differentiated flexible protection distance, and triggering a temporary extended protection distance plan when extreme weather events are forecast.

Images