Three-dimensional tomographic imaging method and system for detecting spatial distribution of gas concentration based on unmanned aerial vehicle

Abstract

The application belongs to the technical field of gas concentration detection, and discloses a three-dimensional tomographic imaging method and system for gas concentration spatial distribution detection based on a UAV, which comprises the following steps: step 1, a space rectangular coordinate system is established, and the longitude and latitude coordinates are converted into XYZ coordinates; step 2, a field of view grid and a grid number are divided; step 3, the intersection grid number and the intersection line segment length of a light beam and the field of view are calculated; step 4, a weight coefficient matrix and an integral concentration vector are calculated; and step 5, a SART algorithm is used to reconstruct a three-dimensional gas concentration distribution. The application combines a UAV and telemetry technology, realizes gas cloud scanning and three-dimensional gas cloud concentration distribution reconstruction based on the UAV, and has the characteristics of non-contact measurement, fast imaging speed, high precision, autonomous detection, strong environmental adaptability and the like, can realize real-time monitoring and prediction of gas diffusion trends, and provides a new solution for realizing accurate tracing of methane gas leakage and reconstruction of leakage spatial distribution in a complex scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gas concentration detection, in particular to a three-dimensional tomographic imaging method and system for detecting spatial distribution of gas concentration based on a UAV. BACKGROUND

[0002] With the rapid development of industrialization and urbanization, air pollution and air pollution control problems are becoming increasingly serious, and how to monitor the emission of pollution gas and locate the pollution source is of great significance for controlling air pollution. Infrared monitoring technology has the characteristics of high selectivity, non-contact, long service life and easy maintenance, and is widely used in gas detection field. According to the different working modes, the infrared gas detection technology can be divided into active and passive. Among them, the passive detection method has the advantages of simple structure, large detection range and strong real-time performance, but has the disadvantages of low sensitivity and short detection distance; the active detection method has high sensitivity, strong anti-interference ability and long working distance, but compared with the passive detection method, it has high cost and few types of detected gas.

[0003] Tunable diode laser absorption spectroscopy (TDLAS) is a commonly used active gas detection method, which changes the wavelength of the laser output to make the laser sweep through the absorption peak of the target gas, thereby realizing the detection of gas concentration. However, this method can only measure the path integral concentration of the gas, and cannot obtain the internal concentration distribution of the gas, so it does not have distance resolution capability. Tomographic imaging (CL) is a method of obtaining the internal structure of a target object using multi-angle scanning and image reconstruction, which can be used in gas detection to obtain the two-dimensional or three-dimensional distribution of the internal concentration of the gas cloud, thereby realizing the positioning of the gas leakage source, the visualization of the gas diffusion path, and the quantitative analysis of the concentration distribution, etc. SUMMARY

[0004] The present application relates to the technical field of gas concentration detection, in particular to a three-dimensional tomographic imaging method and system for detecting spatial distribution of gas concentration based on a UAV.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] A three-dimensional tomographic imaging method for detecting spatial distribution of gas concentration based on a UAV, comprising:

[0007] Step 1, establishing a spatial rectangular coordinate system to convert the latitude and longitude coordinates into XYZ coordinates;

[0008] Step 2, dividing the field of view grid and grid number;

[0009] Step 3, calculating the intersection grid number and intersection line segment length of the light beam and the field of view;

[0010] Step 4, calculating the weight coefficient matrix and integral concentration vector;

[0011] Step 5, reconstruct the three-dimensional concentration distribution of the gas using the SART algorithm.

[0012] Further, the step 1 specifically comprises:

[0013] Step 1.1, selecting a point on the ground near the gas cloud as the origin, wherein, and are the latitude and longitude of the point, respectively, and the position of the UAV is denoted as , wherein, and are the latitude and longitude of the UAV, respectively, is the height above ground measured by the ranging module;

[0014] Step 1.2, establishing a space rectangular coordinate system with the longitude and latitude directions as the X and Y axes, respectively, and the vertical direction upward from the ground as the Z axis;

[0015] Step 1.3, calculating the coordinates of the UAV in the space rectangular coordinate system:

[0016] the latitude difference between the UAV and the origin , the longitude difference , using the local tangent plane projection, obtaining the coordinate x A and coordinate y A of the UAV:

[0017]

[0018]

[0019] In the formula, R is the radius of the earth;

[0020] According to the return value of the ranging module, the z coordinate z A of the UAV is obtained:

[0021] .

[0022] Further, the step 2 specifically comprises:

[0023] Step 2.1, selecting the field center F according to the location of the gas cloud, the F coordinate is , the lengths of the field in the XYZ three directions are W, L and H, respectively, the number of grids divided in the XYZ three directions is N, M and K, respectively, and the total number of grids in the field is NMK;

[0024] ​The spatial resolution of the field of view in the XYZ three directions is respectively Δx, Δy, Δz, according to the field of view size and the grid division quantity, the following is obtained: , , ;

[0025] Step 2.2, grid number, the numbering method is as follows: starting from the position S point closest to the coordinate origin, according to the Z axis coordinate, the field of view is divided into 1-K layers, from low layer to high layer, in each layer, along the Y axis direction, it is divided into 1-M rows, from low row number to high row number, in each row, along the X axis direction, it is divided into 1-N columns, from low column number to high column number;

[0026] Step 2.3, conversion of coordinates and grid number, according to the field of view center F point coordinates and the field of view size, the S point coordinates are obtained as follows:

[0027] , ,

[0028] In the formula, x S is the S point x coordinate, y S is the S point y coordinate, and z S is the S point z coordinate;

[0029] For any point P in the field of view , the coordinates relative to the S point are as follows:

[0030] , ,

[0031] In the formula, , and are respectively the XYZ coordinates of the P point relative to the S point;

[0032] The grid number of the P point is obtained as follows The calculation method is as follows:

[0033]

[0034] Wherein, “[]” represents rounding down, .

[0035] Further, the step 3 specifically includes:

[0036] Step 3.1, according to the light beam equation and the field of view plane equation, the intersection point coordinates and the intersection point matrix are calculated; the intersection point coordinates are obtained through the straight line equation and the plane equation, and the plane equation is as follows:

[0037] The plane perpendicular to the X axis: ;

[0038] The plane perpendicular to the Y axis: ;

[0039] The plane perpendicular to the Z axis: ;

[0040] According to the pitch angle and heading angle of the UAV and the angle data returned by the galvanometer, the azimuth angle and the elevation angle of the light beam are obtained, , where T represents the number of effective light beams intersecting the field of view; the position of the UAV is , and the linear parameter equation of the light beam is obtained as:

[0041]

[0042] The plane equation is brought into the linear parameter equation to obtain the parameter t, and then the intersection point coordinates are obtained;

[0043] All intersection point coordinates are stored in the intersection point matrix ; remove the repeated intersection points in the matrix , and then sort all intersection points according to the XYZ coordinate order to obtain the intersection point coordinate matrix , where is the total number of intersection points of the straight line and all planes;

[0044] Step 3.2, calculate the length of the grid line segment intersected by the light beam and the field of view. For intersection points, there are line segments, and the length d u of the th line segment is calculated as follows:

[0045]

[0046] Step 3.3, calculate the grid number intersected by the light beam and the field of view. The midpoint V coordinates (V x , V y , V z ) of the th line segment are:

[0047]

[0048]

[0049]

[0050] Step 3.4, according to the grid number where the P point is located coordinates, to obtain the first The grid number of the line segment .

[0051] Furthermore, step 4 specifically includes:

[0052] Calculate the weighting coefficient matrix and the integral concentration vector, and store the lengths of the intersecting line segments into the weighting coefficient matrix according to their indices. The corresponding position, that is The gas concentration of all grids in the field of view is represented by a concentration vector of length NMK. Indicates that the beam The corresponding path integral concentration is Then the following relationship exists:

[0053]

[0054] In the formula, It is the length of the line segment of beam i in the j-th grid. It is the average gas concentration within the j-th grid;

[0055] Will Transforming it into matrix multiplication form, it can be expressed as:

[0056]

[0057] Use symbols respectively , and If we represent the matrix and vector in the above expression, then the expression can be written as:

[0058]

[0059] in, It is the integral concentration vector. It is a weight coefficient matrix. It is a concentration vector.

[0060] Furthermore, step 5 specifically includes:

[0061] Step 5.1, first, the concentration vector Initialize to a zero vector;

[0062] Step 5.2, adjust the concentration vector according to the weighting coefficient matrix and the integral concentration vector. Iterative calculations are performed, and the calculation method is as follows:

[0063]

[0064] Where p is the number of iterations. It is a relaxation factor used to control the iteration speed;

[0065] Step 5.3, determining whether to stop iteration according to iteration error, selecting a threshold As a stopping iteration condition, when the root mean square error of the concentration distribution of the adjacent two iterations is less than the threshold, the iteration is stopped, that is:

[0066]

[0067] Step 5.4, reconstructing the three-dimensional concentration distribution of the gas, according to the numbering mode, the obtained concentration vector is restored to the three-dimensional space concentration distribution, stored in the three-dimensional array with a size of , and the actual gas cloud three-dimensional distribution is reconstructed according to the field resolution; the host computer displays the internal concentration distribution of the gas cloud through three-dimensional point cloud, horizontal section and vertical section.

[0068] The application also provides a three-dimensional tomography system for detecting spatial distribution of gas concentration based on a UAV, which is used for realizing the three-dimensional tomography method for detecting spatial distribution of gas concentration based on a UAV as described above, and comprises a GNSS positioning module, a ranging module and a gas cloud scanning system.

[0069] Further, the gas cloud scanning system comprises a main control board, a laser, a detector, a signal generator, an acquisition card, a receiving optical system and a galvanometer.

[0070] The signal generator is used for generating a modulation signal to drive the laser to emit scanning laser.

[0071] The laser is used for emitting laser for scanning the gas cloud.

[0072] The acquisition card is used for sampling and demodulating the output signal of the detector.

[0073] The detector is used for receiving the echo signal converged by the receiving optical system, and converting the optical signal into an electrical signal output to the acquisition card.

[0074] The galvanometer is used for changing the direction of the laser emission to realize scanning of the gas cloud.

[0075] The optical system is used for receiving a return signal of the scanning laser and focusing the reflected laser to a detector.

[0076] Compared with the prior art, the present application has the beneficial effects that: the present application combines unmanned aerial vehicles and telemetry technology, realizes gas cloud scanning and three-dimensional gas cloud concentration distribution reconstruction based on unmanned aerial vehicles, has the characteristics of non-contact measurement, fast imaging speed, high precision, autonomous detection, strong environmental adaptability, etc., can monitor and predict the gas diffusion trend in real time, and provides a new solution for realizing precise tracing of methane gas leakage and reconstruction of leakage space distribution in complex scenes. The present application can be applied to the fields of industrial safety and environmental protection monitoring, public safety, air pollution control and urban management, etc., and can monitor flammable and explosive, toxic gas leakage, quickly locate the leakage source and evaluate the pollution diffusion range. BRIEF DESCRIPTION OF DRAWINGS

[0077] Figure 1 is a three-dimensional tomographic imaging method flow chart of the present application for detecting gas concentration spatial distribution based on unmanned aerial vehicles.

[0078] Figure 2 is a multi-view scanning schematic diagram of the three-dimensional tomographic imaging method for detecting gas concentration spatial distribution based on unmanned aerial vehicles of the present application.

[0079] Figure 3 is a field of view gridding and laser path integral concentration calculation principle schematic diagram in the three-dimensional tomographic imaging method for detecting gas concentration spatial distribution based on unmanned aerial vehicles of the present application.

[0080] Figure 4 is a structure block diagram of the three-dimensional tomographic imaging system for detecting gas concentration spatial distribution based on unmanned aerial vehicles of the present application, in which the black solid arrows represent electrical signals, and the arrows with blank interiors represent optical signals. DETAILED DESCRIPTION

[0081] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0082] Please refer to Figures 1-3 A three-dimensional tomographic imaging method for detecting gas concentration spatial distribution based on unmanned aerial vehicles, comprising the following steps:

[0083] Step 1, a spatial rectangular coordinate system is established, and the latitude and longitude coordinates are converted into XYZ coordinates. Specifically, it includes:

[0084] Step 1.1, selecting a point on the ground near the gas cloud As the origin, among which, and These are the latitude and longitude of that point. The drone scans and images the clouds from different perspectives, such as... Figure 2 As shown. The location of the drone is denoted as... ,in, and It refers to the latitude and longitude of the drone. It is the altitude above the ground measured by the ranging module.

[0085] Step 1.2: Establish a spatial rectangular coordinate system with longitude and latitude as the X-axis and Y-axis respectively, and the vertical upward direction as the Z-axis.

[0086] Step 1.3, the coordinates of the UAV in this spatial rectangular coordinate system are calculated as follows:

[0087] Latitude difference between the drone and point O Longitude difference By using local tangent plane projection, the image of the UAV can be obtained. coordinate x A and coordinates y A :

[0088] (1)

[0089] (2)

[0090] The z-coordinate of the UAV is obtained from the return value of the ranging module. A :

[0091]

[0092] Where R is the Earth's radius, which is taken as 6,371,000 meters here.

[0093] Step 2, divide the field of view into grids and assign grid numbers, such as... Figure 3 As shown, the specific steps are as follows:

[0094] Step 2.1: Based on the location of the gas cloud, select the center of the field of view F, where the coordinates of F are... The lengths of the field of view in the XYZ directions are W, L, and H, respectively, and the number of grids in the field of view in the XYZ directions are N, M, and K, respectively. Therefore, the total number of grids in the field of view is NMK.

[0095] The spatial resolutions of the field of view in the X, Y, and Z directions are Δx, Δy, and Δz, respectively. Based on the field of view size and the number of grid divisions, the spatial resolutions of the field of view in the X, Y, and Z directions are as follows: , , The field of view spatial resolution represents the minimum spatial size that can distinguish the concentration of the gas cloud. The smaller the resolution, the more detailed information of the spatial distribution of the concentration of the gas cloud is contained, but higher requirements are also put forward for the system positioning longitude, detection data and gas detection longitude.

[0096] Step 2.2, grid numbering. The numbering method is as follows: starting from the position S closest to the coordinate origin, as shown in FIG. 2.2. According to the Z-axis coordinate, the field of view is divided into 1-K layers, and the numbering is performed from low layer to high layer. In each layer, it is divided into 1-M rows along the Y-axis direction, and the numbering is performed from low row number to high row number. In each row, it is divided into 1-N columns along the X-axis direction, and the numbering is performed from low column number to high column number. Figure 2

[0097] First, the first row of the first layer is numbered from 1-N, and then a row is added along the Y-axis direction, numbered from N+1 to 2N. In this way, the last grid of the first layer is numbered as NM. Then a layer is added along the Z-axis direction, i.e. the second layer, numbered from NM+1, and the last grid of the second layer is numbered as 2NM. Then the 3-K layer is numbered according to the above method, and the last grid of the K-th layer is numbered as NMK.

[0098] Step 2.3, conversion of coordinates and grid numbers. According to the coordinates of the field of view center F and the size of the field of view, the coordinates of the S point can be obtained as follows:

[0099] , ,

[0100] In the formula, x S is the x coordinate of the S point, y S is the y coordinate of the S point, and z S is the z coordinate of the S point.

[0101] For any point P in the field of view , the coordinates of P relative to the S point are:

[0102] , ,

[0103] The grid number of the P point can be obtained as follows: The calculation method is as follows:

[0104] (3)

[0105] Where, “[]” represents rounding down, .

[0106] ​Step 3, calculate the grid number and the length of the intersection line segment of the light beam and the field of view, the specific steps are as follows:

[0107] Step 3.1, calculate the intersection coordinates and intersection matrix according to the light beam equation and the field of view plane equation. The intersection coordinates can be obtained by the straight line equation and the plane equation, according to the grid division of step 2, including three types of planes, namely the plane perpendicular to the X axis, the plane perpendicular to the Y axis and the plane perpendicular to the Z axis, and the plane equation is as follows:

[0108] The plane perpendicular to the X axis: ;

[0109] The plane perpendicular to the Y axis:

[0110] The plane perpendicular to the Z axis:

[0111] Here n represents an integer between 0 and N, without specific physical meaning, and m and k are the same. According to the pitch angle and heading angle of the unmanned aerial vehicle and the angle data returned by the galvanometer, the azimuth angle and the elevation angle of the light beam can be obtained, where T represents the number of effective light beams intersecting the field of view. The position of the unmanned aerial vehicle is , and the straight line parameter equation of the light beam can be obtained as follows:

[0112] (4)

[0113] Substitute the plane equation into the straight line parameter equation to obtain the parameter t, and then obtain the intersection coordinates. The calculation method of the intersection of the straight line and the plane perpendicular to the X axis is as follows:

[0114] The x-coordinate of the intersection is:

[0115]

[0116] Substitute into equation to obtain the parameter as follows:

[0117] (5)

[0118] Further, substitute into the straight line parameter equation to obtain the y-coordinate and z-coordinate, respectively as follows:

[0119] (6)

[0120] (7)

[0121] The intersection point of the straight line and the plane perpendicular to the Y-axis is calculated as follows:

[0122] The intersection point y coordinate is:

[0123]

[0124] Substitute into the equation , and the parameter is obtained as:

[0125]

[0126] Further, substitute into the straight line parameter equation, and the x coordinate and z coordinate are obtained as follows, respectively:

[0127]

[0128]

[0129] The intersection point of the straight line and the plane perpendicular to the Z-axis is calculated as follows:

[0130] The intersection point z coordinate is:

[0131]

[0132] Substitute into the equation , and the parameter is obtained as:

[0133]

[0134] Further, substitute into the straight line parameter equation, and the x coordinate and y coordinate are obtained as follows, respectively:

[0135]

[0136]

[0137] Store all intersection point coordinates into the intersection point matrix . Since the straight line may intersect with the intersection line of the planes, repeated intersection points may occur, so it is necessary to remove the repeated intersection points in the matrix , and then sort all intersection points according to the XYZ coordinate order to obtain the intersection point coordinate matrix , wherein is the total number of intersection points of the straight line and all planes.

[0138] Step 3.2, calculate the light beam and the field of view intersection grid line segment length. For There are The midpoint of the i-th line segment The length of the i-th line segment d u The calculation method is as follows:

[0139] (8)

[0140] Step 3.3, calculate the intersection grid number of the light beam and the field of view. The midpoint V coordinate (V x , V y , V z ) of the i-th line segment is:

[0141]

[0142]

[0143]

[0144] Step 3.4, according to formula (3) and the midpoint coordinate, the grid number where the i-th line segment is obtained. .

[0145] Step 4, calculate the weight coefficient matrix and the integral concentration vector. Specifically, it includes:

[0146] According to the number, store the intersection line segment length to the corresponding position of the weight coefficient matrix , that is . The gas concentration of all grids in the field of view is represented by the concentration vector with a length of NMK, and the corresponding path integral concentration of the light beam is , then there is the following relationship:

[0147] (9)

[0148] Where, is the line segment length of the i-th light beam in the j-th grid, is the average gas concentration in the j-th grid.

[0149] Write formula (9) in the form of matrix multiplication, which can be expressed as:

[0150] (10)

[0151] Use symbols , and to represent the matrix and vector of formula (10) respectively, then formula (10) can be written as:

[0152] (11)

[0153] wherein, is the integral concentration vector, is the weight coefficient matrix, is the concentration vector.

[0154] Step 5, reconstruct the three-dimensional concentration distribution of the gas using the SART algorithm. The specific method is as follows:

[0155] Step 5.1, first initialize the concentration vector to a zero vector.

[0156] Step 5.2, according to the weight coefficient matrix and the integral concentration vector, iteratively calculate the concentration vector , the calculation method is as follows:

[0157] (12)

[0158] wherein, k is the iteration number. is the relaxation factor, used to control the iteration speed.

[0159] Step 5.3, determine whether to stop iteration according to the iteration error. Select a threshold as the stopping iteration judgment condition, when the root mean square error of the concentration distribution of the adjacent two iterations is less than the threshold, stop iteration, that is:

[0160] (13)

[0161] Step 5.4, reconstruct the three-dimensional concentration distribution of the gas. According to the numbering method, restore the obtained concentration vector to the three-dimensional space concentration distribution, store it in the three-dimensional array with the size of , and reconstruct the actual gas cloud three-dimensional distribution according to the field resolution. The host computer displays the internal concentration distribution of the gas cloud through three-dimensional point cloud, horizontal section and vertical section. Further, according to the reconstructed three-dimensional gas cloud distribution, the precise source tracing, leakage rate estimation and diffusion area detection of the leakage source can be carried out.

[0162] Please refer to Figure 4 , the application also provides a three-dimensional tomography system for detecting spatial distribution of gas concentration based on unmanned aerial vehicle, which is carried on the unmanned aerial vehicle 2, and is used for realizing the three-dimensional tomography method for detecting spatial distribution of gas concentration based on unmanned aerial vehicle as described above. The system comprises a GNSS positioning module 3, a ranging module 10 and a gas cloud scanning system. The unmanned aerial vehicle 2 is also provided with an unmanned aerial vehicle attitude control module 1.

[0163] The unmanned aerial vehicle attitude control module 1 is used to control the heading angle, pitch angle and roll angle of the unmanned aerial vehicle.

[0164] Specifically, the attitude control module controls the flight attitude of the unmanned aerial vehicle, adjusts the center of the scanning field of view to the position of the gas cloud center, and the heading angle, the pitch angle and the roll angle are used to calculate the actual angle of the emitted laser.

[0165] The GNSS positioning module 3 measures the longitude and latitude of the position of the unmanned aerial vehicle, and is used to obtain the position information of the unmanned aerial vehicle. The measurement data of the GNSS positioning module 3 is used for establishing the spatial rectangular coordinate system in the above-mentioned three-dimensional tomographic imaging method.

[0166] Preferably, the system adopts a high-precision GNSS positioning module. This module can support multiple satellite systems such as GPS, BDS, GLONASS, etc. The single-point GNSS horizontal positioning error is less than 1.5 meters. Combined with the RTK (Real-time kinematic) technology, the positioning error can be controlled within centimeters. The module sends the positioning information to the unmanned aerial vehicle 2 through the serial port.

[0167] It should be noted that the coordinates directly returned by the GNSS positioning module 3 are longitude and latitude. In order to facilitate subsequent data processing, it is necessary to convert them into coordinates in the spatial rectangular coordinate system using projection transformation.

[0168] The spatial rectangular coordinate system is established as follows: select a point on the ground near the target gas cloud as the coordinate origin, measure its longitude and latitude, and take the extension direction of the meridian and the latitude as the X-axis and the Y-axis respectively, and the vertical ground direction as the Z-axis. A new spatial rectangular coordinate system is established.

[0169] When the unmanned aerial vehicle 1 is flying, the GNSS positioning module 3 is used to measure the longitude and latitude coordinates, the relative longitude and latitude of the above-mentioned origin are calculated, and then the X-coordinate and Y-coordinate in the spatial rectangular coordinate system are calculated according to the obtained relative longitude and latitude.

[0170] The ranging module measures the height information of the unmanned aerial vehicle 1, and the measurement data of the ranging module is used for establishing the spatial rectangular coordinate system in the above-mentioned three-dimensional tomographic imaging method.

[0171] Preferably, the ranging module 10 uses laser ranging to accurately measure distance information.

[0172] Specifically, the ranging module 10 measures the distance vertically to the ground to obtain the height of the unmanned aerial vehicle 1 relative to the ground. The height data and the plane coordinates obtained by the GNSS positioning module 3 form the spatial coordinates of the unmanned aerial vehicle, which are used for positioning the unmanned aerial vehicle and positioning the laser light path.

[0173] The gas cloud scanning system emits scanning infrared laser and receives echo signal after gas absorption, measures gas cloud integral concentration image, and the gas cloud integral concentration image is used for calculation of grid number, intersection line length, weight coefficient matrix and integral concentration vector of light beam and field of view intersection in the three-dimensional tomography method.

[0174] The gas cloud scanning system comprises a main control board 5, a laser 7, a detector 8, a signal generator 4, a collection card 6, a receiving optical system 11 and a galvanometer 9. The main control board 5 is electrically connected with the unmanned aerial vehicle 1, the signal generator 4, the collection card 6, a ranging module 10 and the galvanometer 9 respectively. The signal generator 4 is electrically connected with the laser 7. The collection card 6 is electrically connected with the detector 8.

[0175] The signal generator 4 is used for generating a modulation signal to drive the laser 7 to emit scanning laser.

[0176] Specifically, according to the wavelength modulation spectroscopy technology principle, the signal generator 4 generates a low-frequency sawtooth wave and a high-frequency sine wave superposition signal as a driving signal of the laser 7 to control the laser 7 to output scanning laser. In addition, the signal generator 4 outputs a trigger signal of the same frequency as the low-frequency sawtooth wave as a sampling clock of the collection card 6.

[0177] Preferably, the signal generator 4 uses a high signal-to-noise ratio and 16-bit high-precision DAC, and the main control board 5 communicates with the DAC through the SPI protocol to control the DAC to output the low-frequency sawtooth wave and the high-frequency sine wave superposition signal.

[0178] According to the wavelength modulation spectroscopy technology principle and the requirements of imaging time and image spatial resolution, the frequencies of the low-frequency sawtooth wave and the high-frequency sine wave can be selected as needed. The typical waveform frequency can be selected as: low-frequency sawtooth wave 50Hz and high-frequency sine wave 10kHz.

[0179] The laser 7 is used for emitting laser for scanning gas cloud.

[0180] Specifically, the laser 7 is a distributed feedback laser and can emit near-infrared continuous wave laser. The central wavelength of the laser is 1653nm, and the output laser wavelength range is 1652nm-1656nm. The laser uses the modulation signal output by the signal generator as the driving signal, and the output wavelength and the driving current are approximately linearly related.

[0181] The strong absorption peak of methane gas in the near-infrared band is 1653.7nm, and the tuning band of the laser 7 contains the absorption peak. Therefore, when the gas cloud to be measured is methane, the above-mentioned laser 7 can be used.

[0182] The collection card 6 is used for sampling and demodulating the signal output by the detector 8.

[0183] Preferably, the acquisition card selects the FPGA chip of EP4CE10F17C8 as the control core, uses the 14-bit high-precision ADC to collect signals, and uses the SPI communication protocol and the FPGA chip to transmit the sampling data. The acquisition card has a DDR4 memory with a storage space of 2 GB, which is used to store the waveforms and demodulated signals collected by the ADC.

[0184] Specifically, the acquisition card 6 receives the trigger signal of the signal generator 4. When the rising edge of the trigger signal is detected, the acquisition card 6 starts the ADC to sample the output signal of the detector 8. The sampling period is the same as the period of the low-frequency sawtooth wave, and the typical sampling rate is 100 kHz. The waveform data collected by the ADC is stored in the DDR4 memory.

[0185] Further, in the FPGA chip, a digital quadrature phase lock amplifier is used to generate software to demodulate the collected waveforms to obtain a first harmonic 1f signal and a second harmonic 2f signal, respectively, and send the first harmonic and the second harmonic signals to the main control board 5 through the serial port.

[0186] The detector 8 is used to receive the echo signal converged by the optical system (11) and convert the optical signal into an electrical signal output to the acquisition card 6.

[0187] Preferably, the detector 8 adopts an indium gallium arsenide single-point photodetector. The response wavelength range of the detector 8 is 800 nm-1700 nm, the peak response wavelength is 1550 nm, which meets the requirements of methane gas detection. The photosensitive area is 2 mm in diameter, and the equivalent noise power is 22 , which meets the demand of gas cloud remote sensing. The working temperature range is 10℃-50℃, and it can be used at room temperature.

[0188] The galvanometer 9 is used to change the direction of the laser output to realize scanning of the gas cloud.

[0189] Specifically, a two-axis galvanometer system is selected, and through two mirror reflections, laser scanning can be realized in the X and Y axes.

[0190] Preferably, the galvanometer 9 is a known technology, which is composed of a motor, a lens, and a motor driving board. The galvanometer 9 rotates the motor through current control, thereby driving the lens to deflect the laser beam. The motor resolution is 15μRad, the maximum scanning angle is ±12.5°, and the angle maximum zero drift is not more than 10μRad / ℃. The lens can receive a maximum beam diameter of 5 mm, and the mirror surface uses gold coating material, which can reflect laser in the wavelength range of 800 nm~20um, meeting the demand of scanning wavelength. The motor driving board is connected with the main control board through the serial port, and can output motor driving current and transmit galvanometer scanning angle data.

[0191] The optical system 11 is used to receive the echo signal of the scanned laser and focus the reflected laser to the detector 8.

[0192] Specifically, the receiving optical system 11 selects a Fresnel condenser lens. Compared with a general lens, the Fresnel condenser lens has a larger numerical aperture, which can improve the receiving efficiency of the echo signal. At the same time, the large numerical aperture makes the Fresnel lens have a larger field of view, which improves the cloud scanning field of view range.

[0193] Preferably, the Fresnel lens is made of acrylic material with a focal length of 76 mm, which ensures the miniaturization and light weight of the system, and is convenient for carrying on the unmanned aerial vehicle 2.

[0194] The main control board 5 is used to control the unmanned aerial vehicle and each sub-module, receive the data returned by each module, and complete the subsequent data analysis and processing task.

[0195] Preferably, the main control board 5 selects a Raspberry Pi 5 development board. The development board CPU is a quad-core ARM Cortex-A76 with 8GB of memory and a 32GB SD storage card. It has a large number of communication interfaces, including four USB interfaces, an HDMI video display output interface, an MIPI CSI camera interface, a gigabit Ethernet port, Bluetooth, a serial port, and a general IO port. Raspberry Pi 5 uses 5V power supply, and the size is 85mmx56mm, which can realize the miniaturization of the equipment and meet the unmanned aerial vehicle detection requirements.

[0196] Raspberry Pi 5 uses a dedicated Linux operating system, Raspberry Pi OS system. The system and the on-board hardware resources have good compatibility, and the calling of serial ports, network ports and other interfaces can be realized through the system's own function library, which is convenient for developing host computer programs to control the unmanned aerial vehicle platform and other sub-modules.

[0197] Specifically, the host computer program of the main control board 5 is used to control the unmanned aerial vehicle and the modules of the cloud imaging system, and analyze and process the received data. The host computer controls the unmanned aerial vehicle to fly according to the preset trajectory, and receives the attitude angle data and GNSS positioning data from the unmanned aerial vehicle. The host computer calls the ranging module 10 to measure the height of the unmanned aerial vehicle to the ground. The host computer sends the sawtooth wave and sine wave frequency to the signal generator according to the set parameters, which is used to generate the laser driver signal. The host computer initializes the galvanometer and sends the scanning angle range and scanning speed to the galvanometer 9. After the galvanometer 9 starts scanning, when the rising edge of the trigger signal is detected, the angle data is returned to the main control board through the network port. The host computer opens the acquisition card, and the acquisition card 6 starts working when the rising edge of the trigger signal is detected. The main control board 5 receives the first harmonic 1f and the second harmonic 2f data returned by the acquisition card 6.

[0198] Further, the host computer receives the data sent by each module and processes it. The 2f signal is normalized using the 1f signal to obtain the normalized 2f curve. According to the peak value of the 2f curve and the laboratory calibration result, the actual gas path integral concentration is obtained by inversion. The integral concentration is one-to-one corresponding to the X-axis and Y-axis angle data returned by the galvanometer, and a two-dimensional integral concentration image is reconstructed. The image is median filtered to remove noise.

[0199] While embodiments of the present application have been shown and described with reference to particular embodiments thereof, it will be understood by those skilled in the art that various changes in form and details can be made therein without departing from the spirit and scope of the application. The scope of the application is defined by the appended claims and their equivalents.

Claims

1. A three-dimensional tomographic imaging method for detecting the spatial distribution of gas concentration based on unmanned aerial vehicles (UAVs), characterized in that, include: Step 1: Establish a spatial rectangular coordinate system and convert latitude and longitude coordinates into XYZ coordinates; Step 2: Divide the field of view into grids and assign grid numbers; Step 3, calculate the grid number and intersection line length of the beam and the field of view, specifically including: Step 3.1: Calculate the intersection point coordinates and intersection point matrix based on the beam equation and the field-of-view plane equation. The intersection point coordinates are obtained through the line equation and the plane equation, which are as follows: A plane perpendicular to the X-axis: ; A plane perpendicular to the Y-axis: ; A plane perpendicular to the Z-axis: ; The field of view is divided into N, M, and K grids in the X, Y, and Z directions, respectively, with a total grid size of NMK. The spatial resolutions of the field of view in the X, Y, and Z directions are Δx, Δy, and Δz, respectively. S Let S be the x-coordinate and y-coordinate. S Let S be the y-coordinate and z-coordinate. S Let S be the z-coordinate of point S, where S represents the position closest to the origin. Based on the UAV's pitch and yaw angles and the angle data returned by the galvanometer, the beam is obtained. azimuth and elevation angle , Where T represents the number of effective beams that intersect with the field of view; the location of the UAV is... x A Represents the x-coordinate and y-coordinate of the drone. A Represents the y-coordinate of the drone, z-coordinate A The z-coordinate of the UAV is used to obtain the beam. The parametric equation of the line is: Substituting the plane equation into the parametric equation of the line, we obtain the parameter t, and then obtain the coordinates of the intersection point; Store the coordinates of all intersection points into an intersection matrix. Remove matrix Repeated intersection points are identified, and then all intersection points are sorted according to their XYZ coordinates to obtain the intersection point coordinate matrix. ,in, It is the total number of intersections between the line and all planes; Step 3.2, calculate the length of the grid line segment where the beam and the field of view intersect. There are intersection points, and there exist. line segment, number Line segment length d u The calculation method is as follows: Step 3.3, calculate the grid number where the beam intersects with the field of view, the first... The coordinates of the midpoint V of the line segment (V x V y V z )for: Step 3.4, based on the grid number where point P is located. Calculation method and midpoint coordinates, to obtain the first The grid number of the line segment ; Step 4: Calculate the weighting coefficient matrix and the integral concentration vector; Step 5: Reconstruct the three-dimensional gas concentration distribution using the SART algorithm.

2. The three-dimensional tomographic imaging method for detecting the spatial distribution of gas concentration based on an unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step 1 specifically includes: Step 1.1: Select a point on the ground near the gas cloud. As the origin, among which, and These are the latitude and longitude of the point, respectively, and the location of the drone is denoted as... ,in, and These are the latitude and longitude of the drone. It is the altitude above the ground measured by the ranging module; Step 1.2: Establish a spatial rectangular coordinate system with longitude and latitude as the X-axis and Y-axis respectively, and the direction perpendicular to the ground upward as the Z-axis; Step 1.3, calculate the coordinates of the UAV in this spatial rectangular coordinate system: Latitude difference between the drone and the origin Longitude difference By using local tangent plane projection, the UAV's coordinate x A and coordinate y A : In the formula, R is the Earth's radius; The z-coordinate of the UAV is obtained from the return value of the ranging module. A : 。 3. The three-dimensional tomographic imaging method for detecting the spatial distribution of gas concentration based on an unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step 2 specifically includes: Step 2.1: Based on the location of the gas cloud, select the center of the field of view F, where the coordinates of F are... The lengths of the field of view in the XYZ directions are W, L, and H, respectively. Based on the field of view size and the number of grid divisions, we obtain: , , ; Step 2.2, grid numbering, the numbering method is as follows: start numbering from the position S closest to the coordinate origin, divide the field of view into layers 1-K according to the Z-axis coordinate, and number from the lower layer to the higher layer. In each layer, divide into rows 1-M along the Y-axis direction, and number from the lower row number to the higher row number. In each row, divide into columns 1-N along the X-axis direction, and number from the lower column number to the higher column number. Step 2.3, Coordinate and Grid Number Conversion: Based on the coordinates of the field of view center F and the field of view size, the coordinates of point S are obtained as follows: ， ， For any point within the field of view Its coordinates relative to point S are: ， ， In the formula, , and These are the XYZ coordinates of point P relative to point S; Obtain the grid number of point P. The calculation method is as follows: Where "[]" indicates rounding down, .

4. The three-dimensional tomographic imaging method for detecting the spatial distribution of gas concentration based on an unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step 4 specifically includes: Calculate the weighting coefficient matrix and the integral concentration vector, and store the lengths of the intersecting line segments into the weighting coefficient matrix according to their indices. The corresponding position, that is The gas concentration of all grids in the field of view is represented by a concentration vector of length NMK. Indicates that the beam The corresponding path integral concentration is Then the following relationship exists: In the formula, It is the length of the line segment of beam i in the j-th grid. It is the average gas concentration within the j-th grid; Will Transforming it into matrix multiplication form, it can be expressed as: Use symbols respectively , and If we represent the matrix and vector in the above expression, then the expression can be written as: in, It is the integral concentration vector. It is a weight coefficient matrix. It is a concentration vector.

5. A three-dimensional tomographic imaging method for detecting the spatial distribution of gas concentration based on an unmanned aerial vehicle (UAV) according to claim 1, characterized in that, Step 5 specifically includes: Step 5.1, first, the concentration vector Initialize to a zero vector; Step 5.2, adjust the concentration vector according to the weighting coefficient matrix and the integral concentration vector. Iterative calculations are performed, and the calculation method is as follows: Where p is the number of iterations. It is a relaxation factor used to control the iteration speed; Step 5.3: Determine whether to stop iteration based on the iteration error, and select a threshold. As a criterion for stopping iterations, iteration stops when the root mean square error of the concentration distribution between two consecutive iterations is less than this threshold. Step 5.4: Reconstruct the three-dimensional gas concentration distribution. According to the numbering method, the obtained concentration vector is... Restore the concentration distribution in three-dimensional space and store it in a file of size [size missing]. The actual three-dimensional distribution of the gas cloud is reconstructed in the three-dimensional array and the actual gas cloud three-dimensional distribution is reconstructed according to the field resolution; the host computer displays the internal concentration distribution of the gas cloud through three-dimensional point cloud, horizontal section and vertical section.

6. A three-dimensional tomographic imaging system for detecting the spatial distribution of gas concentration based on an unmanned aerial vehicle (UAV), used to implement the three-dimensional tomographic imaging method for detecting the spatial distribution of gas concentration based on an UAV as described in any one of claims 1-5, characterized in that, The system includes a GNSS positioning module, a ranging module, and an air cloud scanning system. The GNSS positioning module measures the latitude and longitude coordinates of the UAV's location, and the ranging module measures the UAV's altitude above the ground. The measurement data from the GNSS positioning module and the ranging module are used to establish a spatial rectangular coordinate system. The air cloud scanning system measures the integrated density image of the air cloud, and the integrated density image is used to calculate the grid number of the intersection between the beam and the field of view, the length of the intersection line segment, the weighting coefficient matrix, and the integrated density vector.

7. A three-dimensional tomographic imaging system for detecting the spatial distribution of gas concentration based on an unmanned aerial vehicle (UAV) according to claim 6, characterized in that, The gas cloud scanning system includes a main control board, a laser, a detector, a signal generator, a data acquisition card, a receiving optical system, and a galvanometer. The main control board is electrically connected to the signal generator, the data acquisition card, the ranging module, and the galvanometer. The signal generator is electrically connected to the laser, and the data acquisition card is electrically connected to the detector. The signal generator is used to generate a modulation signal to drive the laser to emit a scanning laser. The laser is used to emit laser light to scan the gas cloud; The acquisition card is used to sample and demodulate the detector output signal; The detector is used to receive the echo signal converged by the optical system and convert the optical signal into an electrical signal to be output to the acquisition card; The galvanometer is used to change the laser emission direction to scan the gas cloud; The optical system is used to receive the echo signal of the scanning laser and focus the reflected laser onto the detector.

Images