Laser methane detector

Abstract

The invention belongs to the technical field of laser methane detection, and particularly relates to a laser methane detector which comprises a laser emitting and detecting unit. The gyroscope positioning module synchronously works with the laser emission and detection unit and outputs a data stream containing a timestamp, a triaxial angular velocity and an attitude parameter; the signal processing module is electrically connected with the laser emission and detection unit and the gyroscope positioning module respectively and is electrically connected with the communication module, and the methane concentration is calculated according to the reflected signal received by the laser emission and detection unit by adopting a differential absorption spectrometry; and the communication module is electrically connected with the signal processing module. According to the laser methane detection system and method, dynamic positioning and spatial data fusion technologies can be combined, and methane concentration three-dimensional distribution diagram generation and leakage source accurate inversion can be achieved.

Description

Technical Field

[0001] This invention belongs to the field of laser methane detection technology, and specifically relates to a laser methane detector. Background Technology

[0002] Laser methane detectors primarily rely on laser absorption spectroscopy technology to detect methane gas. When a laser beam passes through a medium containing methane gas, the methane molecules absorb laser energy of a specific wavelength.

[0003] Problems with existing technology: Currently, some advanced laser methane detection technologies have been applied to remote sensing of methane leaks. For example, differential absorption spectroscopy (DAS) calculates methane concentration by emitting a laser beam of a specific wavelength and measuring the attenuation of the laser after passing through methane gas in the air. However, existing laser methane detectors typically require scanning at a fixed location or rely on manually marked paths for mobile detection. This static or semi-static detection method makes it difficult to quickly generate methane concentration distribution maps over large areas, especially in large or complex environments, resulting in low detection efficiency. Furthermore, due to the lack of precise spatial positioning and attitude information, existing technologies struggle to effectively analyze methane concentration data collected from different locations or scanning angles. The lack of effective spatial correlation and fusion makes it impossible to accurately construct a three-dimensional concentration distribution map, and even more difficult to accurately invert the location of the leak source through concentration gradient analysis. For example, some existing telemetry systems may provide two-dimensional concentration scanning results, but ignore the impact of the movement and attitude changes of the detector in three-dimensional space on the detection data, thus affecting the accuracy of the concentration distribution map. At the same time, for mobile detection scenarios, existing technologies often cannot correct detection errors caused by factors such as changes in detection distance and laser incident angle in real time, resulting in deviations in the generated concentration data and affecting the accuracy of subsequent leak source location. Therefore, existing technologies are insufficient in terms of rapid, large-scale detection, accurate spatial location, and leak source inversion of methane leaks. Summary of the Invention

[0004] The purpose of this invention is to provide a laser methane detector that combines dynamic positioning and spatial data fusion technology to generate a three-dimensional distribution map of methane concentration and accurately invert the leakage source.

[0005] The specific technical solution adopted by this invention is as follows: A laser methane detector, comprising: Laser emission and detection unit; The gyroscope positioning module works synchronously with the laser emission and detection unit, outputting a data stream containing timestamps, three-axis angular velocities, and attitude parameters. The signal processing module is electrically connected to the laser emission and detection unit and the gyroscope positioning module, and is also electrically connected to the communication module. It uses differential absorption spectroscopy to calculate the methane concentration based on the reflected signal received by the laser emission and detection unit. And a communication module, which is electrically connected to the signal processing module; The process involves calculating methane concentration based on the absorption intensity of the reflected signal, then associating and storing the methane concentration data with the dynamic spatial coordinate data obtained from the data stream output by the gyroscope positioning module in a time-series manner. This constructs a dynamic dataset containing spatial coordinates and corresponding methane concentration values, representing a three-dimensional methane concentration distribution. Multiple scans are performed by the laser emission and detection unit through scanning control commands, including the scan range and scan step size. A registration algorithm based on feature point matching is used to align the local dataset containing spatial coordinates and corresponding methane concentration values ​​obtained from each scan with the spatial coordinates calculated from the data stream output by the gyroscope positioning module. Multiple local datasets are then stitched together using a weighted average or maximum value method to construct a global leakage distribution dataset covering the target area. Finally, a concentration gradient is calculated on the global leakage distribution dataset to fit the location of the leakage source.

[0006] In one possible implementation, the signal processing module is further configured to perform spatial position correction on the methane concentration data using a geometric correction model based on the detection range and detection distance of the laser emission and detection unit and the attitude parameters output by the gyroscope positioning module.

[0007] In one possible implementation, the signal processing module is configured to preprocess the data using an interpolation algorithm and / or combine it with a leakage diffusion model for auxiliary analysis when calculating the concentration gradient of the global leakage distribution dataset and fitting the location of the leakage source.

[0008] In one possible implementation, the signal processing module is further configured to compensate for the methane concentration data using a signal correction algorithm based on a laser propagation model and / or a laser attenuation model.

[0009] In one possible implementation, the output data of the gyroscope and the output data of the accelerometer are input to the attitude calculation module, which uses a Kalman filter algorithm to fuse the two types of data to calculate the attitude.

[0010] In one possible implementation, the steps of aligning multiple local datasets using a registration algorithm based on feature point matching and stitching together multiple local datasets using a weighted average method or a maximum value method include using a geometric correction algorithm based on the flip angle of the detector and the laser pointing relationship to correct the overlapping areas of data in different scanning directions.

[0011] In one possible implementation, the device further includes an explosion-proof housing at the end of the detector and an optical resonant cavity assembly for mounting the laser emission and detection unit. An air bladder is filled between the laser emission and detection unit and the explosion-proof housing, and a graphene heat-conducting strip is attached to the outside of the air bladder. The outer sides of the explosion-proof housing and the optical resonant cavity assembly are made of heat-conducting material.

[0012] In one possible implementation, the graphene heat-conducting strip surrounding the outside of the airbag may be arranged in a manner including, but not limited to, a parallel ring shape or a spiral shape.

[0013] In one possible implementation, a detection method for a laser methane detector includes the following steps: The output data of the gyroscope is acquired, and the initial detection position parameters and attitude parameters of the laser detector are calculated. When moving the laser detector, the output signal of the laser emission and detection unit and the output data of the gyroscope are read simultaneously. Methane concentration data is associated with spatial coordinate data and stored together to construct a local dataset containing spatial coordinates and corresponding methane concentration values. Repeat the steps of simultaneously reading, associating and storing data, and building a local dataset until the target area is covered; A registration algorithm based on feature point matching is used to align multiple local datasets; Multiple local datasets are spliced ​​together to construct a global leak distribution dataset covering the target area; It also calculates the concentration gradient on the global leak distribution dataset and fits the location of the leak source.

[0014] In one possible implementation, the step of associating and storing the methane concentration data with the spatial coordinate data calculated from the movement trajectory and attitude change data includes: A geometric correction model based on the detection range and distance of the laser emission and detection unit, as well as the attitude parameters output by the gyroscope, is used to correct the spatial position of the methane concentration data.

[0015] The technical effects achieved by this invention are as follows: This invention constructs a dynamic dataset containing spatial coordinates and corresponding methane concentration values ​​by associating and storing calculated methane concentration data with corresponding dynamic spatial coordinate data in a time-series manner. This dataset intuitively represents the three-dimensional methane concentration distribution. Multiple scans are performed by a laser emission and detection unit via scanning control commands. Each scan yields a local dataset containing spatial coordinates and corresponding methane concentration values, achieving more comprehensive regional coverage. A registration algorithm based on feature point matching is used to align the local datasets obtained from each scan with the spatial coordinates calculated from gyroscope data, integrating the local datasets into a unified dataset. A complete global view is obtained, and after alignment, multiple local datasets are stitched together using a weighted average method or a maximum value method to construct a global leakage distribution dataset covering the target area. Concentration gradients are calculated on this global dataset, and optimization algorithms such as gradient ascent or least squares are used to fit the location of the leakage source. By combining the methane concentration information obtained from laser detection with the precise spatial positioning information provided by the gyroscope, and using data processing and fusion algorithms, the precise location of the methane leakage source and the construction of a three-dimensional concentration distribution are achieved. This enables the localization of the leakage source in three-dimensional space and precise positioning, further realizing high-precision, high-efficiency, and three-dimensional visualization detection.

[0016] This invention achieves the dual functions of heat dissipation and shock absorption through graphene heat-conducting strips. The spiral arrangement of the graphene heat-conducting strips combined with the elastic deformation of the airbags achieves a tight fit of the structure. The synergistic effect of the airbags and graphene heat-conducting strips not only improves heat dissipation efficiency but also achieves dynamic shock absorption. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system structure of the detector of the present invention; Figure 2 This is a schematic diagram of the mechanical structure of the detector in this invention; Figure 3 This is a structural disassembly diagram of the detector in this invention; Figure 4 This is a schematic diagram of the airbag structure in this invention; Figure 5 This is a flowchart of the detection method of the detector in this invention.

[0018] The attached diagram lists the components represented by each number as follows: 1. Detector; 2. Explosion-proof housing; 3. Optical resonant cavity assembly; 4. Airbag; 5. Graphene heat-conducting strip. Detailed Implementation

[0019] To make the objectives and advantages of this invention clearer, the invention will be specifically described below with reference to embodiments. It should be understood that the following text is merely used to describe one or more specific embodiments of the invention and does not strictly limit the scope of protection specifically claimed by the invention.

[0020] like Figure 1 As shown, a laser methane detector includes a laser emitting and detection unit; a gyroscope positioning module that works synchronously with the laser emitting and detection unit and outputs a data stream containing timestamps, three-axis angular velocities, and attitude parameters; a signal processing module that is electrically connected to the laser emitting and detection unit and the gyroscope positioning module, and is also electrically connected to the communication module, which calculates the methane concentration based on the reflected signal received by the laser emitting and detection unit using differential absorption spectroscopy; and a communication module that is electrically connected to the signal processing module.

[0021] Furthermore, the reflected signal is received, the absorption intensity at a specific wavelength is extracted, and the methane concentration is calculated according to the Lambert-Beer law. The methane concentration data is then associated and stored with the dynamic spatial coordinate data calculated from the data stream output by the gyroscope positioning module in a time series, constructing a dynamic dataset containing spatial coordinates and corresponding methane concentration values. This dataset represents the three-dimensional methane concentration distribution. Multiple scans are controlled by sending scan control commands, including the scan range and scan step size. A registration algorithm based on feature point matching is used to align the local dataset containing spatial coordinates and corresponding methane concentration values ​​obtained from each scan with the spatial coordinates calculated from the data stream output by the gyroscope positioning module. Multiple local datasets are then stitched together using a weighted average method or a maximum value method to construct a global leakage distribution dataset covering the target area. The concentration gradient of the global leakage distribution dataset is then calculated, and the location of the leakage source is fitted using a gradient ascent method or a least squares method.

[0022] Furthermore, to address the issues of dynamically locating leak sources and rapidly generating large-scale, high-precision methane concentration distribution maps, the principle is as follows: This embodiment measures methane concentration based on differential absorption spectroscopy (DAS) and the Lambert-Beer law. Simultaneously, a gyroscope positioning module is used to acquire the spatial position and attitude information of the detection system in real time. By dynamically correlating laser detection data with spatial coordinate data provided by the gyroscope, a dynamic dataset representing the three-dimensional methane concentration distribution can be constructed. By controlling the laser emission and detection unit to perform multiple scans covering the target area, the local dataset obtained from each scan is aligned with the spatial coordinates calculated from the gyroscope data. These coordinates are then integrated into a global leak distribution dataset through registration and stitching techniques. Finally, concentration gradient analysis is performed on the global dataset, and an optimization algorithm is used to fit the precise location of the leak source. Based on the above, the synergy of laser detection, dynamic positioning, spatial mapping, data fusion, and leak source inversion is achieved.

[0023] Based on the above, by integrating a gyroscope positioning module and a laser emission and detection unit, and using a signal processing module for data fusion and processing, dynamic, rapid, and high-precision detection and location of methane leaks are achieved. Compared with traditional static or manual detection methods, this embodiment can quickly generate a large-scale, high-precision three-dimensional methane concentration distribution map and accurately invert the location of the leak source. By combining concentration gradient analysis with optimization algorithms to invert the leak source, the positioning accuracy can reach the ppb level. Dynamic correlation and multi-round scanning stitching overcome the limitations of single-point or fixed-area detection and provide comprehensive leak information.

[0024] In the above embodiments, the laser emission and detection unit can use a tunable semiconductor laser as the laser emitter, with the operating wavelength locked near the strong absorption line of methane (e.g., 1653.7 nm), and the laser receiver can use an InGaAs detector; the gyroscope positioning module can use a high-precision MEMS gyroscope, supporting a data output rate of not less than 100 The signal processing module can employ a high-performance embedded processor or DSP chip, possessing sufficient computing power for real-time data processing and complex algorithm calculations. The communication module supports wireless communication methods such as Wi-Fi, Bluetooth, Starflash, or cellular networks to transmit data to mobile terminals or cloud platforms. The scanning range of the scanning control command can be dynamically adjusted according to the size and shape of the target area, and the scanning step size can be set according to the required spatial resolution. The feature point matching and registration algorithm can employ algorithms such as SIFT, SURF, or ORB, combined with RANSAC for robust matching. When using the weighted average method or the maximum value method for stitching, the weights can be determined based on data quality, signal-to-noise ratio, or distance information. The concentration gradient calculation can employ the finite difference method or a surface fitting-based method. The gradient ascent method or the least squares method can be selected or combined according to the actual situation. The dynamic dataset and the global leakage distribution dataset can be stored and represented using data structures such as point clouds, voxel grids, or raster maps.

[0025] As an optional embodiment, the signal processing module is also configured to perform spatial position correction on the methane concentration data using a geometric correction model based on the detection range and detection distance of the laser emission and detection unit and the attitude parameters output by the gyroscope positioning module.

[0026] Furthermore, when a laser beam propagates in space, its pointing direction and detection distance are affected by the attitude of the detection system itself (such as pitch angle, roll angle, and heading angle). Traditional simple correlation methods may ignore these geometric factors, resulting in a deviation between the correspondence between methane concentration data and actual spatial location. This deviation will affect the accuracy of the final concentration distribution map, especially when performing large-scale scanning or rapid movement.

[0027] This embodiment introduces a geometric correction model to correct the spatial position associated with methane concentration data based on the detection range and distance of the laser emission and detection unit and the attitude parameters output by the gyroscope positioning module, ensuring the accuracy of spatial coordinate data and thus improving the accuracy of the constructed dynamic dataset and global leakage distribution dataset.

[0028] Furthermore, in addition to performing data fusion and processing functions, the signal processing module is also configured to use a geometric correction model based on the data stream of the detection range and detection distance of the laser emission and detection unit and the attitude parameters output by the gyroscope positioning module to perform spatial position correction on the methane concentration data. It should be further explained that before or during the process of constructing a dynamic dataset by associating and storing the methane concentration data with the dynamic spatial coordinate data calculated from the data stream output by the gyroscope positioning module in a time series, the signal processing module uses a geometric correction model to calculate the precise position of the spatial point actually pointed to by the laser beam in the global coordinate system based on the current laser emission direction, detection distance, and the real-time attitude of the detection system (provided by the gyroscope positioning module), and uses this precise position as the spatial coordinates corresponding to the methane concentration data.

[0029] Based on the above, the geometric correction model can effectively correct spatial position errors caused by changes in the attitude of the detection system, the divergence angle of the laser beam, and the detection distance, making the correlation between methane concentration data and spatial coordinate data more accurate. This significantly improves the spatial accuracy of the subsequently constructed dynamic dataset and global leakage distribution dataset, laying a solid foundation for accurately inverting the location of the leakage source. Through this correction, the uniformity error of the concentration distribution map can be reduced.

[0030] Furthermore, the geometric correction model can be a mathematical model that takes into account the emission angle of the laser emitter, the field of view of the laser receiver, the detection distance, and the real-time attitude (including position and orientation) of the detection system in three-dimensional space provided by the gyroscope positioning module. The correction process can be achieved through matrix transformation or vector calculation, transforming the position of the measurement point in the sensor coordinate system to the global coordinate system, and fine-tuning it by taking into account the effects of laser beam divergence and receiving field of view. Alternatively, the geometric correction model can be obtained through pre-calibration, for example, by collecting data in a known environment and establishing a mapping relationship between the sensor's original output and the precise spatial position. Attitude parameters may include Euler angles, quaternions, or rotation matrices. Detection range and detection distance information can be obtained from the parameters of the laser emission and detection unit or through signal processing calculations.

[0031] As an optional embodiment, the signal processing module is configured to preprocess the data using an interpolation algorithm and / or combine it with a leakage diffusion model for auxiliary analysis when calculating the concentration gradient of the global leakage distribution dataset and fitting the location of the leakage source.

[0032] In this embodiment, the global leakage distribution dataset obtained through multiple rounds of scanning may not be completely continuous, and there may be gaps between data points. This will affect the accuracy of concentration gradient calculation. At the same time, the process of fitting the leakage source is an inverse problem, which may have multiple solutions or be sensitive to noise. Combining with a physical model can provide more reliable constraints and auxiliary information. This embodiment enhances the accuracy and robustness of concentration gradient calculation and leakage source fitting by introducing interpolation algorithms and / or leakage diffusion models into the signal processing module.

[0033] Furthermore, before calculating the concentration gradient, the signal processing module can use interpolation algorithms (such as Kriging interpolation, inverse distance weighted interpolation, or radial basis function interpolation) to interpolate the discrete concentration data points in the global leakage distribution dataset, generating a denser and more continuous concentration field or concentration surface. This helps to calculate the concentration gradient at each point in space more accurately. When fitting the location of the leakage source, the signal processing module combines a leakage diffusion model (such as a Gaussian plume model, a box model, or a more complex CFD model) for auxiliary analysis. The leakage diffusion model predicts the concentration distribution formed by the diffusion of methane from the potential leakage source based on environmental parameters (such as wind speed, wind direction, and atmospheric stability level). The signal processing module can compare or fuse the actual measured concentration distribution dataset with the prediction results of the leakage diffusion model. For example, the model prediction can be introduced into the fitting algorithm as prior information, or the model results can be used to verify the rationality of the fitted leakage source location.

[0034] Furthermore, using interpolation algorithms for preprocessing can overcome the impact of data sparsity on gradient calculation and improve the accuracy of gradient information. Combining this with a leakage diffusion model for auxiliary analysis introduces physical constraints, making the leakage source fit more in line with the actual diffusion law, thus improving the reliability and accuracy of the location results. In particular, when environmental parameters are known, the accuracy of the location can be significantly improved, with the location accuracy reaching the ppb level.

[0035] In the above implementation, the selection of the interpolation algorithm can be optimized according to the characteristics of the data distribution and computing resources. Kriging interpolation is suitable for data with spatial correlation. The parameters of the leakage diffusion model (such as wind speed, wind direction, turbulence parameters) can be provided by external environmental sensors or set according to the terrain, building distribution and other information of the detection area. The auxiliary analysis method can be to use the location of the leakage source to be fitted as the model input, calculate its predicted concentration distribution, and perform error analysis with the measured distribution. The fitting result can be optimized by minimizing the error.

[0036] As an optional embodiment, the signal processing module is also configured to compensate for the methane concentration data using a signal correction algorithm based on a laser propagation model and / or a laser attenuation model.

[0037] It should be noted that during the propagation of the laser beam in the air, in addition to being absorbed by methane, it is also affected by absorption, scattering, diffraction, and other factors from other atmospheric components (such as water vapor and carbon dioxide), leading to signal attenuation. Especially at longer detection distances or under complex atmospheric conditions, this non-methane-induced signal attenuation can affect the accuracy of methane concentration calculation by differential absorption spectroscopy. In this embodiment, a signal correction algorithm based on the laser propagation model and / or laser attenuation model is used in the signal processing module to compensate for the methane concentration data, thereby eliminating the influence of these interference factors and improving the accuracy of concentration measurement.

[0038] In the above embodiments, the signal processing module is further configured to compensate for the methane concentration data using a signal correction algorithm based on a laser propagation model and / or a laser attenuation model. This signal correction algorithm can be applied after the signal processing module receives the signal output by the laser emission and detection unit, before or during the calculation of the methane concentration. The laser propagation model describes the propagation characteristics of the laser beam in free space and the atmosphere, including divergence and diffraction. The laser attenuation model describes the change of laser signal intensity with propagation distance and atmospheric conditions. For example, the extended Lambert-Beer law can be used to describe attenuation other than target gas absorption (such as background gas absorption and aerosol scattering). The signal correction algorithm uses these models, combined with the current detection distance (which can be obtained from the signal of the laser emission and detection unit or measured by other means), and possible environmental parameters (such as temperature, humidity, and atmospheric pressure), to calculate the expected non-methane-induced signal attenuation and compensate for the received reflected signal intensity or the preliminarily calculated methane concentration data to obtain a more accurate methane concentration value.

[0039] By employing a signal correction algorithm to compensate for methane concentration data, the influence of the background environment on the laser signal can be effectively eliminated, improving the accuracy and reliability of concentration measurement. This is crucial for accurate measurement under different environmental conditions and detection distances, ensuring that the constructed concentration distribution dataset reflects the true methane concentration, thereby improving the accuracy of leak source location. Through compensation, the concentration data deviation caused by changes in detection distance can be corrected, reducing the uniformity error of the concentration distribution map.

[0040] It should be further explained that the laser propagation model and laser attenuation model can be constructed based on physical principles or obtained through experimental calibration. The specific implementation of the signal correction algorithm can be to calculate a correction factor based on the model and multiply the original concentration data by the factor, or to directly correct the received signal intensity and then calculate the concentration. Environmental parameters can be obtained through other sensors integrated into the detection system or input by the user.

[0041] As an optional embodiment, the output data of the gyroscope and the output data of the accelerometer are input to the attitude calculation module, which uses the Kalman filter algorithm to fuse the two types of data to calculate a more accurate attitude.

[0042] In the above implementation, the angular velocity data output by the gyroscope will drift over time when integrating to calculate the attitude, while the acceleration data output by the accelerometer (usually integrated with the gyroscope in the IMU) can provide a gravity vector when there is no external acceleration interference, which can be used to correct the attitude (especially the pitch and roll angles). In order to obtain more accurate and stable instrument attitude information, an attitude calculation module is set in the signal processor to fuse the data from the gyroscope and the accelerometer.

[0043] Furthermore, the attitude calculation module uses high-frequency angular velocity data from the gyroscope as the prediction input and predicts the attitude at the next moment based on the motion model. At the same time, it uses the gravity vector measured by the accelerometer as the observation input and corrects the predicted attitude through the Kalman filter algorithm. The Kalman filter algorithm can comprehensively consider the noise characteristics of the gyroscope and accelerometer data and fuse them in the optimal way to output a more accurate and smoother attitude estimate. The Kalman filter algorithm describes the attitude change through a state-space model and iteratively performs prediction and update steps.

[0044] By inputting data from the gyroscope and accelerometer into the attitude calculation module and fusing them using the Kalman filter algorithm, this detector can obtain more accurate and stable real-time attitude parameters than using only the gyroscope. Accurate attitude information is crucial for calculating precise spatial coordinate data, which directly affects the spatial accuracy of the dynamic dataset and the global leak distribution dataset, thereby improving the accuracy of the leak source location fitting.

[0045] In the above implementation, the attitude calculation module can be a software module or hardware acceleration unit in the signal processor. The Kalman filtering algorithm can be standard Kalman filtering, extended Kalman filtering (EKF), or unscented Kalman filtering (UKF). The specific choice depends on the nonlinearity of the system and the computing resources. In addition to gyroscopes and accelerometers, magnetometer data can be further fused to correct the heading angle, or GPS data can be fused to provide an absolute position reference to further improve the accuracy of positioning and attitude estimation.

[0046] As an optional embodiment, please refer to Figure 2-4 The detector 1 also includes an explosion-proof housing 2 at the end of the detector 1 and an optical resonant cavity assembly 3 for mounting the laser emission and detection unit. An air bladder 4 is filled between the laser emission and detection unit and the explosion-proof housing 2. A graphene heat-conducting strip 5 is attached to the outside of the air bladder 4. The explosion-proof housing 2 and the outside of the optical resonant cavity assembly are made of heat-conducting materials, including but not limited to hard plastics with high thermal conductivity and heat-conducting metal materials.

[0047] Furthermore, the graphene heat-conducting strips 5 surrounding the outer side of the airbag 4 are arranged in a manner including but not limited to parallel rings (see reference). Figure 3 ), spiral-shaped surrounding arrangement (refer to) Figure 4 ).

[0048] Based on the above, the graphene heat-conducting strip 5 achieves the dual functions of heat dissipation and shock absorption. The spiral arrangement of the graphene heat-conducting strip 5 and the elastic deformation of the airbag 4 combine to achieve a tight fit of the structure. The synergistic effect of the airbag 4 and the graphene heat-conducting strip 5 not only improves heat dissipation efficiency, but also achieves dynamic shock absorption, avoiding the impact on the detection accuracy of the detector 1 during use and transportation.

[0049] Please refer to Figure 5 As an optional embodiment, a laser methane detection method includes the following steps: S1. Read the output data of the gyroscope and calculate the initial detection position parameters and attitude parameters of the laser detector; S2. When moving the laser detector, the output signal of the laser emission and detection unit and the output data of the gyroscope are read simultaneously through the data interface. S3. Associate and store the methane concentration data with the spatial coordinate data calculated based on the movement trajectory and attitude change data to construct a local dataset containing spatial coordinates and corresponding methane concentration values. S4. Repeat the steps of simultaneously reading data through the data interface, associating and storing data, and building a local dataset until the target area is covered; S5. A registration algorithm based on feature point matching is used to align multiple local datasets; S6. Use a weighted average method or a maximum value method to stitch together multiple local datasets to construct a global leakage distribution dataset covering the target area; S7. Calculate the concentration gradient of the global leakage distribution dataset and fit the location of the leakage source using the gradient ascent method or the least squares method.

[0050] To enable methane detection methods to quickly acquire information on the distribution of methane leaks over a wide area and accurately locate the leak source, this embodiment provides a laser methane detection method. By using a dynamically moving detector and combining sensor data fusion technology, it achieves rapid scanning, three-dimensional distribution construction, and leak source localization of regional methane leaks.

[0051] Furthermore, by simultaneously acquiring laser detection data and gyroscope positioning data, concentration information is correlated with spatial location information in real time to construct a local dataset. Through repeated scanning and data acquisition, the target area is covered, and the obtained multiple local datasets are aligned and stitched together to form a global leakage distribution dataset. Then, using concentration gradient analysis and optimization algorithms, the location of the leakage source is inverted from the global dataset.

[0052] A laser methane detection method combined with a laser methane detector specifically includes the following steps: S101. Read the output data of the gyroscope and calculate the initial detection position parameters and attitude parameters of the laser detector, which provides an initial reference for subsequent dynamic positioning. S201. When moving the laser detector, the output signal (i.e., reflection signal) of the laser emission and detection unit and the output data of the gyroscope are read simultaneously through the data interface. The synchronous acquisition ensures the time correspondence between the concentration data and the position / attitude data. S301. The methane concentration data calculated based on the reflected signal is associated and stored with the spatial coordinate data calculated based on the movement trajectory and attitude change data (calculated from gyroscope data), and a local dataset containing spatial coordinates and corresponding methane concentration values ​​is constructed, which can realize the initial mapping of concentration information in space. S401. Repeat the steps of simultaneously reading data through the data interface, associating and storing the data, and constructing a local dataset until the target area is covered. By controlling the movement of the detector and the scanning range / step size, the concentration data of the target area can be systematically collected. S501. After covering the target area, a registration algorithm based on feature point matching is used to align multiple local datasets. The registration algorithm uses feature points in the data to transform local datasets acquired at different times or from different scanning angles into the same spatial reference system. S601. Use the weighted average method or the maximum value method to stitch together multiple aligned local datasets to construct a global leakage distribution dataset covering the target area. The stitching method can be selected according to the requirements. For example, the weighted average method can smooth the data in the overlapping areas, while the maximum value method can highlight the area with the highest concentration. S701. Calculate the concentration gradient of the global leakage distribution dataset, and fit the leakage source location using the gradient ascent method or the least squares method. By analyzing the rate of change and direction (gradient) of concentration in space, the diffusion trend of methane can be inferred, and the possible leakage source location can be traced back. The gradient ascent method or the least squares method is an optimized algorithm for solving this inverse problem.

[0053] Based on the above methods, rapid and accurate detection and location of methane leaks over a large area were achieved through dynamic data acquisition, spatial correlation, data fusion, and gradient analysis.

[0054] In the above embodiments, reading the gyroscope output data can adopt a standard sensor interface protocol. Simultaneously reading the output signal of the laser emission and detection unit and the gyroscope output data can be achieved through hardware synchronization triggering or software timestamp alignment. Associative storage can use a spatial database or a specific data structure (such as a kd-tree) to manage spatial concentration data. Repeated scanning can adopt a preset scanning path or can be adaptively adjusted according to real-time data feedback. The feature point matching and registration algorithm can be combined with technologies such as RANSAC to improve robustness. When using the weighted average method or the maximum value method for splicing, the weights can be determined based on factors such as the freshness of the data points, the signal-to-noise ratio, or the distance from the sensor. Concentration gradient calculation can use numerical methods (such as finite difference) or analytical methods based on the fitted surface. When using the gradient ascent method or the least squares method to fit the location of the leakage source, parameters such as the number of iterations and the convergence threshold can be set.

[0055] As an optional embodiment, the step of associating and storing methane concentration data with spatial coordinate data calculated based on movement trajectory and attitude change data includes using a geometric correction model based on the detection range and detection distance of the laser emission and detection unit and the attitude parameters output by the gyroscope to correct the spatial position of the methane concentration data.

[0056] In this embodiment, for each methane concentration measurement, the precise three-dimensional spatial coordinates corresponding to the measurement value are calculated using the current detection distance (acquired by the laser emission and detection unit) and the real-time attitude information (data stream) of the laser detector provided by the gyroscope through a geometric correction model. The precise spatial coordinates take into account factors such as the propagation direction and divergence angle of the laser beam, as well as the tilt and rotation of the detector, and correct the original spatial coordinate calculation error caused by these geometric factors. The corrected spatial coordinates are then associated and stored with the corresponding methane concentration data to form a more accurate local dataset.

[0057] Furthermore, by integrating geometric correction during the data association stage, it is ensured that each concentration value is accurately mapped to its corresponding spatial location, which improves the quality of the local dataset. This is crucial for subsequent local dataset alignment, global dataset stitching, and final leak source fitting, effectively reducing errors caused by inaccurate spatial location and improving the overall accuracy of the method.

[0058] In the above embodiments, the geometric correction model can be a mathematical model. This model calculates the precise intersection position of the laser beam in the global coordinate system based on the geometric characteristics of the laser emission and detection unit (such as laser emission angle, receiving field of view angle, detection range parameters), detection distance, and attitude parameters output by the gyroscope (such as pitch angle, roll angle, and heading angle). The correction process can be implemented in software by transforming the measurement point position in the sensor coordinate system to the global coordinate system through a coordinate transformation matrix. The detection distance can be obtained through the laser ranging function or calculated from the flight time of the reflected signal.

[0059] As an optional embodiment, the steps of calculating the concentration gradient and fitting the location of the leakage source on the global leakage distribution dataset include preprocessing the data using an interpolation algorithm and / or combining it with a leakage diffusion model for auxiliary analysis.

[0060] In the above implementation, before calculating the concentration gradient, an interpolation algorithm (such as Kriging interpolation, natural neighbor interpolation, or spline interpolation) can be used to interpolate the discrete and possibly non-uniformly distributed concentration data points in the global leakage distribution dataset to generate a continuous or high-density concentration field. This makes it convenient to calculate the concentration value and its gradient at any point, improving the accuracy and smoothness of the gradient calculation. When fitting the location of the leakage source, the leakage diffusion model can be used for auxiliary analysis. For example, based on known environmental parameters (such as wind speed, wind direction, temperature, and atmospheric pressure) and topographic information, a leakage diffusion model can be used to predict the methane concentration distribution that diffuses from a hypothetical leak source location. Then, the predicted distribution is compared with the measured global leak distribution dataset. By optimizing the hypothetical leak source location, the difference between the model prediction and the measured data is minimized, thereby more accurately determining the leak source location. Auxiliary analysis can also incorporate the leakage diffusion model as a constraint or prior information into fitting algorithms such as gradient ascent or least squares.

[0061] Furthermore, by introducing interpolation algorithms and / or leakage diffusion models, the accuracy and reliability of inverting the location of leakage sources from the global leakage distribution dataset are enhanced. Interpolation algorithms solve the problem of directly calculating gradients for discrete data points, while leakage diffusion models provide physical constraints and guidance, making the fitting results more consistent with the actual situation and improving the accuracy and robustness of leakage source localization.

[0062] In the above implementation, the selection of interpolation algorithm and parameter settings can be optimized according to the data density and spatial distribution characteristics of the global dataset. The selection of leakage diffusion model can be based on the leakage scenario (such as point source, line source), environmental conditions (such as open space, indoor environment) and required accuracy. Environmental parameters can be obtained through external sensors or meteorological data services. The specific methods of auxiliary analysis can be based on model residual minimization, Bayesian inference or data assimilation.

[0063] As an optional embodiment, the steps of aligning multiple local datasets using a registration algorithm based on feature point matching and stitching together multiple local datasets using a weighted average method or a maximum value method include using a geometric correction algorithm based on the flip angle of the detector and the laser pointing relationship to correct the overlapping areas of data in different scanning directions.

[0064] In the above implementation, the problem is solved that when the laser detector performs multiple scans to cover the target area, there may be overlapping areas between the local datasets obtained from different scan cycles or different scan directions (such as horizontal scan, vertical scan or tilt scan). Due to the changes in the detector's posture (flip angle) and the changes in the laser pointing direction, the data in these overlapping areas may have inconsistencies or deviations. Direct alignment and splicing may introduce errors and affect the accuracy of the global leakage distribution dataset.

[0065] Furthermore, after aligning multiple local datasets using a registration algorithm based on feature point matching, or before or during stitching using a weighted average or maximum value method, overlapping areas between different local datasets will be identified. For data points in overlapping areas, a geometric correction algorithm is used to calculate and correct the spatial position or concentration value of these data points based on the flip angle of the detector when acquiring these data points (provided by gyroscope data) and the laser pointing relationship (determined by the scanning control command of the laser emission and detection unit and its own attitude), so as to eliminate measurement deviations or spatial misalignments caused by differences in scanning angles. For example, the effective optical path can be calculated or the laser beam landing point can be corrected based on the flip angle and the laser pointing relationship, thereby correcting the concentration value or its corresponding spatial coordinates in the overlapping area. The corrected overlapping area data is then fused (e.g., by weighted average or maximum value) to construct a more accurate and consistent global leakage distribution dataset.

[0066] In this embodiment, by performing fine geometric correction on the overlapping areas, the accuracy and consistency of the alignment and stitching of multiple local datasets are improved. This effectively avoids "ghosting" or data conflicts in the overlapping areas and ensures that the constructed global leakage distribution dataset accurately reflects the real methane concentration distribution, which is crucial for subsequent concentration gradient calculation and leakage source fitting.

[0067] In the above embodiments, the geometric correction algorithm can be a mathematical model or a set of correction rules. The model takes into account the flip angle of the detector around different axes and the pointing direction of the laser beam relative to the detector body. The correction can be to fine-tune the spatial position of the data points in the overlapping area or to compensate for the concentration value so that the data obtained by different scans are consistent in the overlapping area. The flip angle can be directly obtained from the attitude parameters of the gyroscope. The laser pointing relationship can be determined by the scan control command and the internal structure of the laser emission and detection unit. The identification of the overlapping area can be achieved by comparing the spatial range or data point distribution of different local datasets.

[0068] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A laser methane detector, characterized in that, include: Laser emission and detection unit; The gyroscope positioning module works synchronously with the laser emission and detection unit, outputting a data stream containing timestamps, three-axis angular velocities, and attitude parameters. The signal processing module is electrically connected to the laser emission and detection unit and the gyroscope positioning module, and is also electrically connected to the communication module. It uses differential absorption spectroscopy to calculate the methane concentration based on the reflected signal received by the laser emission and detection unit. And a communication module, which is electrically connected to the signal processing module; Among them, the methane concentration is calculated based on the absorption intensity of the reflected signal, and the dynamic spatial coordinate data obtained by calculating the methane concentration data with the data stream output by the gyroscope positioning module is stored in a time series. Construct a dynamic dataset containing spatial coordinates and corresponding methane concentration values, representing a three-dimensional methane concentration distribution; The laser emission and detection unit is controlled to perform multiple scans by sending scan control commands. The control commands include the scan range and scan step size. A registration algorithm based on feature point matching is used to align the local dataset containing spatial coordinates and corresponding methane concentration values ​​obtained in each scan with the spatial coordinates calculated from the data stream output by the gyroscope positioning module. Multiple local datasets are stitched together using a weighted average method or a maximum value method to construct a global leakage distribution dataset covering the target area. Concentration gradients are then calculated on the global leakage distribution dataset to fit the location of the leakage source.

2. The laser methane detector according to claim 1, characterized in that: The signal processing module is also configured to perform spatial position correction on the methane concentration data using a geometric correction model based on the detection range and detection distance of the laser emission and detection unit and the attitude parameters output by the gyroscope positioning module.

3. The laser methane detector according to claim 1, characterized in that: The signal processing module is configured to preprocess the data using an interpolation algorithm and / or combine it with a leakage diffusion model for auxiliary analysis when calculating the concentration gradient and fitting the location of the leakage source on the global leakage distribution dataset.

4. The laser methane detector according to claim 1, characterized in that: The signal processing module is also configured to compensate for the methane concentration data using a signal correction algorithm based on a laser propagation model and / or a laser attenuation model.

5. A laser methane detector according to claim 1, characterized in that: The output data from the gyroscope and the output data from the accelerometer are input to the attitude calculation module, which uses a Kalman filter algorithm to fuse the two types of data to calculate the attitude.

6. A laser methane detector according to claim 1, characterized in that, The steps of aligning multiple local datasets using a registration algorithm based on feature point matching and stitching together multiple local datasets using a weighted average method or a maximum value method include: using a geometric correction algorithm based on the flip angle of the detector and the laser pointing relationship to correct the overlapping areas of data in different scanning directions.

7. A laser methane detector according to claim 1, characterized in that, It also includes an explosion-proof housing at the end of the detector and an optical resonant cavity assembly for mounting the laser emission and detection unit. An air bladder is filled between the laser emission and detection unit and the explosion-proof housing. A graphene heat-conducting strip is attached to the outside of the air bladder. The explosion-proof housing and the outside of the optical resonant cavity assembly are made of heat-conducting material.

8. A laser methane detector according to claim 7, characterized in that: The graphene heat-conducting strips surrounding the outside of the airbag can be arranged in a manner including, but not limited to, parallel rings or spirals.

9. A laser methane detector according to claims 1-8, characterized in that, The following detection methods are included: The output data of the gyroscope is acquired, and the initial detection position parameters and attitude parameters of the laser detector are calculated. When moving the laser detector, the output signal of the laser emission and detection unit and the output data of the gyroscope are read simultaneously; Methane concentration data is associated with spatial coordinate data and stored together to construct a local dataset containing spatial coordinates and corresponding methane concentration values. Repeat the steps of simultaneously reading, associating and storing the data, and building a local dataset until the target area is covered; A registration algorithm based on feature point matching is used to align multiple local datasets; Multiple local datasets are spliced ​​together to construct a global leak distribution dataset covering the target area; It also calculates the concentration gradient on the global leak distribution dataset and fits the location of the leak source.

10. A laser methane detector according to claim 9, characterized in that, The steps for associating and storing methane concentration data with spatial coordinate data calculated from movement trajectory and attitude change data include: A geometric correction model based on the detection range and distance of the laser emission and detection unit, as well as the attitude parameters output by the gyroscope, is used to correct the spatial position of the methane concentration data.

Images