LNG tank car target positioning system and method

Abstract

The invention discloses an LNG tank car target positioning system and method, and the system comprises an image collection module which is disposed at the entrance of a loading position of an LNG loading station yard, and is used for collecting the images of tank cars at all loading positions in a visual field in real time; the image processing module is configured to identify target information of the tank car according to the tank car image, and the target information comprises loading connection object positioning information of the tank car; and the loading arm control module is configured to control the corresponding loading arm to move to the corresponding tank car gas-liquid phase interface according to the loading connection object positioning information so as to realize loading operation auxiliary control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of LNG loading operation detection, in particular to a LNG tank car target positioning system and method for an automatic loading system of an LNG loading station. BACKGROUND

[0002] Liquefied natural gas (LNG, mainly composed of methane) is a low-temperature liquid fuel cooled to its condensation point (-161.5℃) by compression. At present, based on the LNG receiving station / liquefaction plant loading station, LNG is mainly transported by road using tanker trucks. The LNG loading system mainly includes a loading pry and a loading arm, but the low-temperature land fluid loading and unloading arm (ultra-low-temperature fluid loading and unloading arm) disclosed in the prior art, including a liquid phase arm and a gas phase arm, adopts a loose flange to connect the tank car-loading arm corresponding gas phase and liquid phase flanges. During the LNG loading process, the operator manually drags the heavy hard pipe loading arm from the loading pry position to the LNG tank car tail operating box position, and simultaneously manually tightens the liquid phase arm and gas phase arm flanges, a total of 16 nuts. Therefore, during the docking process of the LNG tank car and the loading arm, the operator has a high labor intensity, and it is time-consuming and laborious to install the flanges, which leads to low LNG loading efficiency and seriously affects the operation efficiency of the LNG receiving station / liquefaction project. The adoption of LNG automatic loading technology is the key to improving the intelligence of the LNG loading station, which requires the gas phase arm and the liquid phase arm of the LNG loading arm to have the ability to automatically move to the LNG tank car operating box corresponding gas phase interface / flange and liquid phase interface / flange and quickly dock.

[0003] At present, the LNG tank car tank body is provided with multiple safety devices, including an emergency shutdown system, a safety system instrument monitoring system, etc. The LNG tank car operating box has a large number of pipelines (including liquid phase, gas phase pipelines and booster pipelines) and valves. The LNG tank cars produced by different manufacturers have great differences in the layout of the corresponding liquid phase interface, gas phase interface and booster interface in the tank car tail operating box, and the positions are different. In the traditional LNG loading process, the gas phase / liquid phase arm is connected with the LNG tank car corresponding gas phase / liquid phase interface flange by manual dragging. Moreover, when the LNG tank car is parked at the loading position before loading, the parking position is different, and the gas phase / liquid phase pipeline layout in the tank car operating box is different. Due to the differences in the positions of the tail operating box of the tank car of different types and sizes and the docking flange, in order to realize the automatic movement of the automatic loading arm joint to the position of the tank car tail operating box, it is necessary to determine the accurate positions of the liquid phase interface / flange and the gas phase interface / flange in the tail of the tank car to be filled with LNG.

[0004] In the prior art of automatic loading arm or loading system (such as patent numbers CN111637361A, CN112178451A, CN112178451B, CN213118461U, CN215489076U, CN215489076U, CN211315763U, CN214744981U), the disclosed loading system composed of loading arm (loading crane pipe) does not have the function of automatically identifying the gas phase and liquid phase interface positions of different LNG tank cars, and cannot realize the automatic displacement, docking and loading of the loading arm of the LNG loading system. In this case, only manual control of the loading arm can be used to move to the corresponding tank car operation box gas phase / liquid phase interface position. For this purpose, manual control of the loading arm is still required before LNG loading, which requires multiple people to operate together, wastes manpower, and is low in loading efficiency, and cannot truly realize LNG loading automation.

[0005] In addition, in the prior art of patent number CN115818552A, a one-key full-automatic intelligent loading crane pipe and its control system are disclosed. This scheme uses an industrial camera lens installed on a stand to only capture the vehicle parking position; a depth camera installed on the outer wall of the gas phase loading arm automatic vertical pipe and the liquid phase automatic vertical pipe determines the flange position, which has the problems that the depth camera installation position is easy to be damaged due to collision, the loading arm is frozen and fogged due to low-temperature LNG transportation during the loading process, the frozen fogging seriously blocks the depth camera lens, and affects the accuracy of automatic positioning of the LNG receiving station / liquefied plant continuous loading process tank car flange.

[0006] In summary, the prior art needs to provide an LNG tank car target positioning scheme for an LNG loading station automatic loading system to accurately control the displacement of the gas phase arm and the liquid phase arm, realize the automation of LNG loading operation, and improve the efficiency and safety of LNG loading operation. SUMMARY

[0007] The purpose of the present application is to provide an LNG tank car target positioning scheme for an LNG loading station automatic loading system to accurately control the displacement of the gas phase arm and the liquid phase arm, realize the automation of LNG loading operation, and improve the efficiency and safety of LNG loading operation.

[0008] In order to solve the above technical problems, the embodiment of the present application provides a LNG tank car target positioning system, comprising: an image acquisition module arranged at the entrance of a loading position of a LNG loading station, configured to acquire tank car images on all loading positions in a field of view in real time; an image processing module configured to identify target information of the tank car according to the tank car images, the target information comprising loading connection object positioning information of the tank car; and a loading arm control module configured to control corresponding loading arms to move to corresponding tank car gas-liquid phase interfaces according to the loading connection object positioning information, so as to realize auxiliary control of loading operation.

[0009] Preferably, the image acquisition module comprises two high-definition cameras and matching light supplementing equipment arranged on the two side columns of the canopy at the entrance of the loading position, wherein the LNG tank car target positioning system further comprises: a data switch assembly located between the image acquisition module and the image processing module, configured to forward the tank car images to the image processing module; and an alarm prompt module connected with the image processing module, configured to respond to alarm prompt information sent by the image processing module.

[0010] Preferably, the target information further comprises tank car position, vehicle type and identity information, wherein the image processing module comprises: an edge computing unit assembly configured to identify the position of the tank car to be loaded in the image according to the tank car images, and identify the vehicle type and identity information of the current tank car when detecting that the tank car is parked at a designated loading position, and then locate the loading connection object positioning information of the tank car, the loading connection object positioning information comprising gas phase interface positioning information, gas phase flange positioning information, liquid phase interface positioning information and liquid phase flange positioning information.

[0011] Preferably, the edge computing unit assembly is further configured to send an alarm instruction to the alarm prompt module when detecting that the tank car is not parked at the designated loading position.

[0012] Preferably, the vehicle type and identity information is obtained through license plate recognition.

[0013] Preferably, the edge computing unit assembly is further configured to detect whether the gas-liquid phase loading arm moves to the right position according to the tank car images acquired during the movement of the loading arm to the tank car gas-liquid phase interface, wherein when detecting that the movement is to the right position, the alarm prompt module is sent a prompt information about allowing to continue loading; and when detecting that the movement is not to the right position, the alarm prompt module is sent a prompt information about that the loading arm movement is not to the right position.

[0014] Preferably, the edge computing unit assembly is further configured to identify the loading connection object positioning information by the following steps: collecting different angle tank car images of the same time scene from the first camera and the second camera after double target positioning, wherein the internal parameter, the external parameter and the distortion parameter of the two cameras are obtained after double target positioning, and the distortion parameter includes radial distortion parameter and tangential distortion parameter; after the current two tank car images are sequentially subjected to stereo correction and parallax calculation, a corresponding re-projection matrix is obtained; according to the coordinate position of the pixel of the loading connection object to be positioned in the two-dimensional image, the three-dimensional space coordinates of the current loading connection object are obtained by using the re-projection matrix, which is recorded as the loading connection object positioning information.

[0015] Preferably, the re-projection matrix is represented by the following expression:

[0016]

[0017] Wherein, Q represents the re-projection matrix, u o , v o respectively represent the horizontal coordinate and the vertical coordinate of the optical center of the main view camera during parallax correction, f represents the focal length, T x represents the translation vector in the camera parameter, and the three-dimensional space coordinates of the loading connection object are obtained by the following expression:

[0018]

[0019] Wherein, X W , Y W and Z W respectively represent the three-dimensional space coordinates of the loading connection object, b represents the baseline parameter of the camera, and d represents the parallax.

[0020] Preferably, the edge computing unit assembly is further configured to obtain the calculated value of the straight-line distance between each connection object and the main view camera according to the three-dimensional space coordinates of the loading connection object, so as to judge the accuracy of the current binocular positioning measurement by analyzing the error between the calculated value and the measured value by using the calculated value of the straight-line distance and combining the measured value of the straight-line distance.

[0021] Preferably, during double target positioning, the LNG loading station loading arm fixed column or the fixed parking mark line of the LNG tank car is taken as the reference zero baseline.

[0022] Preferably, the LNG tank car target positioning system further comprises a data storage and monitoring module configured to store the target information of the tank car after completing the loading operation.

[0023] In another aspect, the LNG tank car target positioning method is realized by the LNG tank car target positioning system as described above, and the LNG tank car target positioning method comprises the following steps: real-time image acquisition of tank cars on all loading positions in the field of view is performed by the image acquisition module arranged at the entrance of the loading position of the LNG loading station; target information of the tank car is identified from the tank car image by the image processing module, and the target information comprises loading connection object positioning information of the tank car; and the corresponding loading arm is moved to the corresponding gas-liquid interface of the tank car under the action of the loading arm control module according to the loading connection object positioning information, so as to realize auxiliary control of the loading operation.

[0024] Compared with the prior art, one or more embodiments in the above scheme can have the following advantages or beneficial effects:

[0025] The application discloses an LNG tank car target positioning system and method for an automatic loading system of an LNG loading station. The image recognition and position positioning system of the LNG tank car liquid / gas interface constructed by the system and the method can detect the parking position of the tank car, guide the accurate parking of the LNG tank car, quickly identify the type of the tank car, and accurately position the position coordinates of the liquid / gas interface / flange of the tank car, thereby providing feedback parameters for the movement control of the loading arm, and being beneficial to the unmanned, digitization and intelligent construction of a large LNG receiving station.

[0026] Thus, the application solves the problems of manual and manual operation of the existing LNG tank car loading, such as many steps, complicated valve opening and closing operation, time-consuming and laborious tank car docking process of the loading arm, large working strength of the operator, and the like, and improves the LNG high turnover efficiency and safety level of the LNG receiving station and the LNG factory.

[0027] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the application. The objects and other advantages of the application will be realized and attained by the structure particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0028] The accompanying drawings are included to provide a further understanding of the application, and constitute a part of the specification, illustrate embodiments of the application, and together with the description serve to explain the application. In the drawings:

[0029] Figure 1 A structure schematic diagram of the LNG tank car target positioning system of the embodiment of the present application.

[0030] Figure 2 An application layout schematic diagram of the LNG tank car target positioning system of the embodiment of the present application.

[0031] Figure 3 A target positioning principle schematic diagram of the LNG tank car target positioning system of the embodiment of the present application.

[0032] Figure 4 A flowchart for realizing target positioning in the LNG tank car target positioning system of the embodiment of the present application.

[0033] Figure 5 An implementation step flowchart of the LNG tank car target positioning method of the embodiment of the present application. DETAILED DESCRIPTION

[0034] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and embodiments, by which how the present application applies technical means to solve technical problems and achieves technical effects can be fully understood and implemented. It should be noted that, as long as there is no conflict, each embodiment in the present application and each feature in each embodiment can be combined with each other, and the technical solutions formed thereby are all within the protection scope of the present application.

[0035] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Moreover, although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown.

[0036] The terms used herein are merely used to describe specific embodiments and are not intended to limit the exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" used herein are also intended to include the plural. It should also be understood that the terms "comprise" and / or "include" as used herein specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or combinations thereof.

[0037] Liquefied natural gas (LNG, mainly composed of methane) is a low-temperature liquid fuel compressed and cooled to its condensation point (-161.5℃). At present, based on the LNG receiving station / liquefaction plant loading station, LNG is mainly transported by land using tank trucks. The LNG loading system mainly includes a loading pry and a loading arm, but the low-temperature land fluid loading and unloading arm (ultra-low-temperature fluid loading and unloading arm) disclosed in the prior art, including a liquid-phase arm and a gas-phase arm, adopts a loose flange to connect the tank truck-gas-phase and liquid-phase arm corresponding flanges. During the LNG loading process, the operator needs to manually pull the heavy hard pipe loading arm from the loading pry position to the LNG tank truck tail operating box position, and manually tighten the liquid-phase arm and gas-phase arm flanges, a total of 16 nuts. Therefore, during the docking process of the LNG tank truck and the loading arm, the operator has a high labor intensity, and it is time-consuming and laborious to install the flanges, which leads to low LNG loading efficiency and seriously affects the operating efficiency of the LNG receiving station / liquefaction project. The adoption of LNG automatic loading technology is the key to improving the intelligence of the LNG loading station, which requires the gas-phase arm and the liquid-phase arm of the LNG loading arm to have the ability to automatically move to the LNG tank truck operating box corresponding gas-phase interface / flange and liquid-phase interface / flange and quickly dock.

[0038] At present, the LNG tank truck tank is provided with multiple safety devices, including an emergency shutdown system, a safety system instrument monitoring system, etc. The LNG tank truck operating box has a large number of pipelines (including liquid-phase, gas-phase pipelines and booster pipelines) and valves. The LNG tank trucks produced by different manufacturers have great differences in the layout of the corresponding liquid-phase interface, gas-phase interface and booster interface in the tank truck tail operating box, and the positions are different. In the traditional LNG loading process, the gas-phase / liquid-phase arm is connected with the LNG tank truck corresponding gas-phase / liquid-phase interface flange by manual pulling. Moreover, when the LNG tank truck is parked at the loading position before loading, the parking position is different, and the gas-phase / liquid-phase pipeline layout in the tank truck operating box is different. Due to the differences in the positions of the tail operating box of the tank truck of different types and sizes and the docking flange, in order to realize the automatic movement of the automatic loading arm joint to the position of the tank truck tail operating box, it is necessary to determine the accurate positions of the liquid-phase interface / flange and the gas-phase interface / flange in the tail of the LNG tank truck to be filled.

[0039] In the prior art of automatic loading arm or loading system (such as patent numbers CN111637361A, CN112178451A, CN112178451B, CN213118461U, CN215489076U, CN215489076U, CN211315763U, CN214744981U), the disclosed loading system composed of loading arm (loading crane) does not have the function of automatically identifying the gas phase and liquid phase interface positions of different LNG tank cars, and cannot realize the automatic displacement, docking and loading of the loading arm of the LNG loading system. In this case, only manual control of the loading arm can be used to move to the corresponding tank car operation box gas phase / liquid phase interface position. For this purpose, manual control of the loading arm is still required before LNG loading, which requires multiple people to operate together, wastes manpower, and also cannot truly realize LNG loading automation.

[0040] In addition, in the prior art of patent number CN115818552A, a one-key full-automatic intelligent loading crane and its control system are disclosed. This scheme uses an industrial camera lens installed on a stand to only capture the vehicle parking position; a depth camera installed on the outer wall of the gas phase loading arm automatic vertical pipe and the liquid phase automatic vertical pipe determines the flange position, which has the problems that the installation position of the depth camera is easy to be damaged due to collision, the loading arm is frozen and fogged due to low-temperature LNG transportation during the loading process, and the icing and atomization seriously block the depth camera lens, affecting the accuracy of automatic positioning of the LNG receiving station / liquefied plant continuous loading process tank car flange.

[0041] In summary, the prior art needs to provide a LNG tank car target positioning scheme for an automatic loading system of an LNG loading station, to accurately control the displacement of the gas phase arm and the liquid phase arm, realize the automation of LNG loading operation, and improve the efficiency and safety of LNG loading operation.

[0042] To solve one or more of the above technical problems, the present application provides a LNG tank car target positioning system and method for an automatic loading system of an LNG loading station. The system and method identify the gas phase and liquid phase interface / flange position and accurate coordinates in the operation box of the parked LNG tank car to be loaded through the photographed tank car image, to accurately control the displacement of the gas phase arm and the liquid phase arm, so that the LNG receiving station / liquefied plant uses a power-assisted loading arm, realizes the automation of LNG loading operation, and improves the efficiency and safety of LNG loading operation.

[0043] Example One

[0044] Figure 1 The structure diagram of the LNG tank car target positioning system of the embodiment of the present application is shown in FIG. 1. The following refers to the structure diagram of the LNG tank car target positioning system of the embodiment of the present application. Figure 1The specific structure of the LNG tank truck target positioning system according to the embodiment of the present application is described.

[0045] As shown in Figure 1 The LNG tank truck target positioning system comprises an image acquisition module 1, an image processing module (not numbered) and a loading arm control module 5.

[0046] The image acquisition module 1 is arranged at the entrance of the loading position of the LNG loading station. The image acquisition module 1 is configured to acquire images of all tank trucks on the loading position in the field of view in real time. The tank truck images contain information of the docking process of the loading arm of the LNG loading pry and the tank truck.

[0047] The image processing module is configured to identify the target information of the tank truck according to the tank truck images acquired by the image acquisition module 1. The target information includes the positioning information of the loading connection object of the tank truck. In the embodiment of the present application, the loading connection object includes a gas phase interface, a gas phase flange, a liquid phase interface and a liquid phase flange. Further, the loading connection object positioning information includes gas phase interface positioning information, gas phase flange positioning information, liquid phase interface positioning information and liquid phase flange positioning information.

[0048] In one embodiment, the image processing module comprises an edge computing module assembly 3.

[0049] The loading arm control module 5 is configured to control the corresponding loading arm to move to the corresponding tank truck gas phase and liquid phase interface according to the loading connection object positioning information, so as to realize auxiliary control of the loading operation.

[0050] In one embodiment, the image acquisition module 1 comprises two high-definition cameras 1-1, 1-2 and matching light supplementing equipment arranged on the two side columns (9-1, 9-2) of the canopy at the entrance of the loading position. In addition, the LNG tank truck target positioning system further comprises a data switch assembly 2. The data switch assembly 2 is located between the image acquisition module 1 and the image processing module.

[0051] The data switch assembly 2 is configured to forward the tank truck images transmitted from the image acquisition module 1 to the image processing module.

[0052] In addition, the LNG tank truck target positioning system further comprises an alarm prompt module 7. The alarm prompt module 7 is connected with the image processing module. The alarm prompt module 7 is configured to respond to the alarm prompt information issued by the image processing module.

[0053] Figure 2 The application layout diagram of the LNG tank truck target positioning system according to the embodiment of the present application. Figure 3 The target positioning principle diagram of the LNG tank truck target positioning system according to the embodiment of the present application.

[0054] AsFigure 2 and Figure 3 As shown, the two cameras 1-1 and 1-2 in the image acquisition system simultaneously capture images of the tank truck, including the gas phase interface 13-1 and liquid phase interface 13-2 of the tank truck to be loaded in the loading area on both sides of the loading skid. The image processing module then generates coordinate information for the gas phase interface 13-1 and the liquid phase interface 13-2. In subsequent steps, the loading arm control module 5 controls the automatic loading arm drive device to automatically move the gas phase arm 12-1 and the liquid phase arm 12-2 to the loading station loading position during the loading arm movement and docking operation. Specifically, the gas phase arm 12-1 is moved to the gas phase interface position 13-1 of the tank truck to be loaded, and the liquid phase arm 12-2 is moved to the liquid phase interface position 13-2 of the tank truck to be loaded. After the gas and liquid phase arms are connected to the corresponding interface flanges of the tank truck to be loaded 13, loading operations on the tank truck to be loaded 13 are initiated by controlling the valves on the loading skid 8 and the loading arm 12.

[0055] Column 10 is used to install and secure the gas phase arm and liquid phase arm at the LNG loading station. A universal joint is secured to column 10 via a support frame. The gas phase arm 12-1 and liquid phase arm 12-2 are connected to the fixed LNG infusion pipe section on column 10 via the universal joint, enabling rotation of gas phase arm 12-1 and liquid phase arm 12-2. The LNG infusion pipe section is secured to column 10, which is then connected to the LNG loading skid via a flange. This allows LNG to be transferred to the LNG tanker during the loading process, via the pipelines, valves, and instrumentation between the LNG storage tank, loading skid, and loading arm.

[0056] The LNG loading station tank truck parking area line 11 is used as the LNG parking area and location division, and can also be used as the reference baseline for identifying the spatial coordinate position of the gas and liquid phase interfaces (different LNG tank truck parking locations and gas / liquid phase pipeline layouts result in different gas-liquid phase interface positions of the tank trucks, and 11 is used as the on-site position reference baseline). A fixed point reference coordinate zero point can be set to calculate the position coordinates of the gas-liquid phase interface of the tank truck.

[0057] In addition, in the embodiment of the present invention, the target information includes not only the tank truck's loading connection object positioning information, but also the tank truck's position, vehicle type, and identity information.

[0058] like Figure 4 As shown, the edge computing unit assembly 2 in the image processing module is configured to identify the position of the tank truck to be loaded in the image based on the tank truck image, and when the tank truck is detected to be parked at the designated loading position 11, identify the vehicle type and identity information of the current tank truck, and then locate the loading connection object positioning information of the tank truck.

[0059] In one embodiment, the edge computing unit assembly 2 is further configured to issue an alarm instruction and send the alarm instruction to the alarm prompt module 7 when it is detected that the tank truck is not parked at the designated loading position 11, so that the alarm prompt module 7 responds to the current alarm instruction about the tank truck not being parked well and the parking indication.

[0060] The two high-definition cameras in the image acquisition module 1 constitute a binocular machine vision positioning system. The binocular machine vision positioning system can accurately position the LNG tank truck in the LNG loading station loading operation video. If the LNG tank truck is not parked at the designated position, the alarm prompt module 7 reminds the driver through the sound and light alarm mode, plays the specified voice, flashes the specified color, such as "vehicle not parked well, please continue to drive", "your vehicle has been parked well", etc., to guide the driver to drive, pass the marker line and park; during the process of the driver correcting the parking position, the tank truck parking position is continuously detected; the driver corrects the parking position, guides the driver to park, gets off and prompts to touch the human body static electricity eliminator for 5s.

[0061] In one embodiment, the edge computing unit assembly 2 is further configured to obtain the vehicle type and identity information by recognizing the license plate.

[0062] In addition, with reference to Figure 1 , the LNG tank truck target positioning system described in the embodiments of the present application further comprises a data storage and monitoring module 1. The data storage and monitoring module 1 is configured to store the target information of the tank truck that has completed the loading operation.

[0063] The image acquisition module 1 is composed of two sets of high-definition cameras (cameras) 1-1 and 1-2 and a matching light sensing and light supplementing system. The high-definition cameras (cameras) 1-1 and 1-2 are respectively fixed on the left and right stand columns 9-1 and 9-2 of the shed at the entrance of the LNG loading station loading position by supports. The image acquisition module 1 uploads the images of the movement of the LNG tank truck 13 and the management layout of the operation box to the edge computing module assembly 3 through the data exchange system assembly 2; the edge computing module assembly 3 calculates and recognizes the coordinates of the gas phase interface and the liquid phase interface / flange of the LNG tank truck to be loaded in the LNG loading station, and transmits the related interface / flange position coordinates to the PLC control system 5, and transmits the related information to the data storage and data monitoring system 4. The edge computing module assembly 3 and the data exchange system assembly 2 are installed in the explosion-proof box in the LNG loading station. The explosion-proof box is installed at the lower part of the loading pry or the shed stand column (such as 9-1) of the loading position. The data exchange system assembly 2 is connected with the gas sensing system 6 and the alarm prompt module 7 installed near the LNG loading position through signal lines and power lines. The loading arm PLC control system 5 is installed in the explosion-proof control cabinet of the LNG loading station or directly installed in the instrument cabinet of the LNG receiving station / liquefied plant.

[0064] The image recognition and data processing system (including an image acquisition system 1, a data exchange system assembly 2, and an edge computing module assembly 3) is composed of software and hardware, wherein the hardware mainly includes a camera, a light sensing and light supplementing system, an edge computing box, and hardware parameters of each device thereof. The software mainly includes video recognition algorithms, positioning algorithms, and requirements of each algorithm. The system is composed of an edge computing mainboard unit as the core, and real-time calculation is performed on the data fed back on site through a pre-set program and a trained model. The latest ARM architecture of NVIDIA is selected, the CPU is not less than 6 cores and is an ARM 64-bit architecture, the GPU is equipped with an architecture of not less than 384 NVIDIA CUDA cores and 48 Tensor Cores, the AI computing power is not less than 14 / 21 TOPS (INT8), the video memory is not less than 8 GB 128-bit LPDDR4, the storage capacity is not less than 128 G, the storage mode can include not less than 16 GBeMMC 5.1 storage mode, the basic requirements of deep learning AI calculation for the position and motion measurement of the tank car are basically met, multiple video streams can be processed at the same time, and parallel tasks can be executed. The high-definition camera group is composed of a high-definition pan-tilt camera, the infrared night vision is 100 meters, the pan-tilt control can horizontally rotate 360° and vertically rotate -15-90°. The light source of the audible and visual alarm uses LED stroboscopic light, and has red and green warning light colors, the speaker decibel is not less than 100 dB, the alarm voice can be customized, and the audible and visual alarm meets the basic requirements of product waterproofing, explosion-proofing, and dust-proofing. The remote model training and testing operation host needs to support two or more GPUs (NVIDIA Tesla series, GTX series, RTX series, etc.) to run, the number of NVIDIACUDA cores of a single GPU is not less than 8704, the acceleration frequency is not less than 1.67 GHz, the basic frequency is not less than 1.26 GHz, the server memory is not less than 16 G, and the CPU is not less than 10 cores and 20 threads.

[0065] The PLC control system 5 controls the automatic loading arm driving device to automatically move the gas phase arm 12-1 and the liquid phase arm 12-2 to the loading station and the loading position, for example, the gas phase arm 12-1 is moved to the LNG tank truck gas phase interface position 13-1, and the liquid phase arm 12-2 is moved to the LNG tank truck liquid phase interface 13-2. The gas and liquid phase arms are connected to the corresponding interfaces and flanges of the LNG tank truck, and the valve on the loading pry 8 and the loading arm 12 is controlled to start the loading operation to the tank truck 13.

[0066] The gas sensing system 6 is a combustible gas and toxic gas detection sensor installed on site at the LNG loading station, which is used to detect the methane volatilized from the leaked LNG into a gaseous state. When the gaseous methane volatilizes to the sensor of the gas sensing system 6, an audible and visual alarm signal is sent through the alarm prompt module 7.

[0067] In one embodiment, the edge computing unit assembly 2 is also configured to identify the loading connection object positioning information of the tank car to be loaded by the following steps, which include:

[0068] Different angle tank car images at the same time from the first camera (for example: left camera 1-1) and the second camera (for example: right camera 1-2) after double target setting. Wherein, after double target setting, the internal parameters, external parameters and distortion parameters of the two cameras are obtained, and the distortion parameters include radial distortion parameters and tangential distortion parameters;

[0069] After the current two tank car images are sequentially subjected to stereo correction and disparity calculation, the corresponding re-projection matrix is obtained;

[0070] According to the coordinate position of the pixel of the loading connection object to be positioned in the two-dimensional image, the three-dimensional space coordinates of the current loading connection object are obtained by using the above re-projection matrix, which is recorded as the loading connection object positioning information.

[0071] The image acquisition and data processing system uses the image acquisition module 1 of the binocular camera (camera) to position the flange interface 13-1, 13-2 in space, so as to control the loading arm to move to the position of the flange interface according to the position information.

[0072] As shown in Figure 3 The ranging of the binocular camera is based on the triangulation technology, which uses two cameras to shoot different angle images of the same object at the same time.

[0073] The flow of binocular ranging is double target setting, stereo correction (including distortion elimination), stereo correction, disparity calculation, three-dimensional coordinate calculation and output.

[0074] In the double target setting, the LNG loading station loading arm fixed column or the fixed parking mark line of the LNG tank car is taken as the reference zero point baseline.

[0075] (1) Double target setting:

[0076] When the double target setting is used, the binocular camera is installed at a height of 5m above the two side columns 9-1 and 9-2 of the LNG loading station. When the reflected light of the LNG tank car interface passes through the camera optical system, distortion is generated. In order to eliminate the distortion and obtain the true LNG tank car interface position information, the distortion is modeled:

[0077] One kind of distortion caused by lens shape is called radial distortion, which can be modeled as:

[0078]

[0079] Another is that tangential distortion can be introduced during the assembly of the camera due to the inability to make the lens and imaging surface strictly parallel, where the tangential distortion can be modeled as:

[0080]

[0081] where r 2 = x 2 +y 2 , x, y are normalized plane coordinates, and r represents the distance from the normalized plane coordinate point (x, y) to the optical center. r is the radial distance in the polar coordinate system, which represents the distance from the point (x, y) to the origin, so r is used to calculate the distortion parameters.

[0082] According to the above modeling method, a total of 5 distortion parameters are required, which are:

[0083] Distortion_coefficients = (k1, k2, p1, p2, k3) (3)

[0084] The goal of the dual target calibration is to obtain the intrinsic parameters, extrinsic parameters and distortion coefficients of the left and right cameras. Among them, the intrinsic parameters include fx, fy, cx, cy of the left and right cameras, the extrinsic parameters include the rotation matrix and translation vector of the left camera relative to the right camera, and the distortion coefficients include the radial distortion coefficients (k1, k2, k3) and the tangential distortion coefficients (p1, p2). The intrinsic and extrinsic parameters are usually provided with preset values by the camera manufacturer, but in actual application, due to factors such as lens quality and installation position, small deviations can occur, so these parameters need to be accurately determined through calibration. The calibrated intrinsic parameters can be used to correct the distortion in the image, so that the objects in the image are more consistent with the actual geometric shape. The extrinsic parameters are used to convert the coordinate system of one camera to the coordinate system of another camera, which can provide accurate correspondence between the left and right cameras.

[0085] (2) Stereoscopic correction:

[0086] The image recognition and data processing system performs mathematical projection transformation on the left and right views of the same LNG tank scene, so that the two image planes are coplanar and parallel to the baseline, which is simply called coplanar line alignment. After this correction process, the epipolar lines in the two images will be completely horizontal, resulting in the same point in space having the same pixel position in the left and right images.

[0087] (3) Disparity calculation:

[0088] The binocular vision camera simultaneously captures left and right images, and each pixel point in the left image finds its corresponding point in the right image, so that the disparity d can be calculated:

[0089] d = u l -u r(4)

[0090] wherein, u l , u r respectively represent the horizontal coordinate of the same pixel in the tank image captured by the first camera and the horizontal coordinate of the same pixel in the tank image captured by the second camera.

[0091] The main purpose of calculating the disparity is to estimate the distance and depth information of the object in space through the pixel offset between the corresponding points in the same scene in the left and right images. After obtaining the disparity, the disparity information can be combined with the intrinsic and extrinsic parameters of the camera to calculate the three-dimensional coordinates.

[0092] (4) Three-dimensional coordinate calculation

[0093] After a series of calibrations, the calculated re-projection matrix Q is shown in formula 5. According to the re-projection matrix, the parameters: u o , v o , f, b can be obtained. Wherein, f represents the focal length of the camera, that is, the ratio of the distance from the object to the imaging plane to the imaging size on the camera imaging plane. b represents the baseline of the camera, that is, the distance between the two cameras in the binocular camera.

[0094] According to the formula, the three-dimensional coordinates of each pixel point in the camera coordinate system can be obtained. The (u, v) coordinate system in pixels, the (x, y) coordinate system in physical size (mm) (camera coordinate system), and the (x, y) coordinate system take the intersection of the camera principal axis and the image plane as the origin. The camera coordinate system is a very key coordinate system in binocular vision, which describes the coordinates of the image in three-dimensional space through the camera's angle of view.

[0095] In one embodiment, the re-projection matrix is represented by the following expression:

[0096]

[0097] wherein, Q represents the re-projection matrix, u o , v o respectively represent the horizontal coordinate and the vertical coordinate of the principal camera (the principal camera is the first camera in the above formula (4)) during the disparity correction, f represents the focal length, and T x represents the translation vector in the camera parameter. Wherein, the horizontal and vertical coordinates of the principal camera, the focal length and the translation vector can be obtained in the binocular calibration process.

[0098] The three-dimensional space coordinates of the loading connection object are obtained by the following expression:

[0099]

[0100] wherein, X W , YW and Z W respectively represent the three-dimensional space coordinates of the loading connection object, b represents the baseline parameter of the camera, and d represents the parallax.

[0101] First, the positions of the interfaces, flanges and other parts are identified according to the tank truck image captured by the front-view camera, and then the two-dimensional pixel point coordinates of these specific parts in the two-dimensional tank truck image can be obtained. According to the above expression (5), the three-dimensional coordinates of each specific part (of each loading connection object) in the left camera coordinate system can be calculated.

[0102] In addition, the edge computing unit assembly 2 is also configured to obtain the calculated value of the straight-line distance between each connection object and the front-view camera according to the three-dimensional space coordinates of the loading connection object, so as to use the calculated value of the straight-line distance to judge the accuracy of the current binocular positioning measurement by analyzing the error between the calculated value and the measured value in combination with the measured value of the straight-line distance.

[0103] The straight-line distance l of each loading connection object from the front-view camera can be obtained according to the three-dimensional space coordinates of the loading connection object:

[0104] In addition, with reference to Figure 4 , the edge computing unit assembly 2 is also configured to detect whether the gas-liquid phase loading arm moves to the right position according to the tank truck images collected during the movement of the loading arm to the tank truck gas-liquid phase interface.

[0105] When it is detected that the movement is to the right position, the edge computing unit assembly 2 sends a prompt information about allowing to continue loading to the alarm prompt module 7.

[0106] When it is detected that the movement is not to the right position, the edge computing unit assembly 2 sends a prompt information about the loading arm movement not being to the right position to the alarm prompt module 7.

[0107] Example Two

[0108] Based on the above LNG tank truck target positioning system, the embodiment of the present application also provides a verification method and error test of the LNG tank truck pre-loading gas phase and liquid phase structure positioning identification system (i.e. LNG tank truck target positioning system), the principle of which is mainly based on the positioning of the characteristic straight line, and the main process and method are as follows:

[0109] (1) Set the camera parameters

[0110] A binocular stereo camera is used, and the camera parameters (here, the camera parameters are the internal and external parameters obtained during the binocular target positioning process) have been calibrated. According to the actual working conditions of LNG, the two cameras are fixedly installed on the shed stand, and the angle is adjusted to obtain a proper field of view.

[0111] (2) Camera calibration using calibration board and corresponding calibration algorithm to obtain the intrinsic and extrinsic parameters of the camera

[0112] The purpose of calibration is to accurately determine the geometric characteristics and relative position relationship of the camera, obtain the intrinsic, extrinsic and distortion coefficients of the left and right cameras, wherein the intrinsic parameters include fx, fy, cx and cy of the left and right cameras, the extrinsic parameters include the rotation matrix and translation vector of the left camera relative to the right camera, and the distortion coefficients include the radial distortion coefficients (k1, k2, k3) and the tangential distortion coefficients (p1, p2), which provide an accurate basis for subsequent disparity calculation and ranging.

[0113] (3) Data acquisition

[0114] After fixing the camera installation position, two ways are used to select the characteristic straight line of LNG loading site: ① Draw a fixed parking line 11 at the parking position of the tank truck and mark it yellow as the camera calibration characteristic straight line; ② Take the fixed column 10 of the loading arm of the LNG tank truck loading station as a specific characteristic straight line. The above parking line or column is used as a reference baseline, and left and right images are collected using binocular cameras to ensure that the images have sufficient contrast, clarity and texture characteristics.

[0115] (4) Stereo matching

[0116] The left and right images are matched by algorithm to calculate the disparity value of each pixel. The main purpose of calculating the disparity is to estimate the distance and depth information of the object in space through the pixel offset between the corresponding points in the left and right images of the same scene.

[0117] (5) Disparity conversion

[0118] According to the calibration parameters of the camera, the disparity value is converted into the actual physical distance (the distance from the camera to the parking line or column). According to the baseline distance and the intrinsic parameters of the camera, the conversion relationship of disparity to distance is calculated to obtain the accurate object distance value. After calibration, the re-projection matrix Q can be obtained u o , v o , f, b, according to the formula, the three-dimensional coordinates of each pixel point in the camera coordinate system can be obtained.

[0119] (6) Accuracy evaluation

[0120] The distance between the target reference baseline and the camera is measured by a laser range finder, and the distance measured by the binocular vision positioning camera is compared with the true distance of the reference baseline (parking line or column) (see Table 1 for comparison), and the accuracy of binocular vision ranging is evaluated. The error index such as average error can be calculated to quantify the accuracy of the ranging result, see Table 2. According to the error test result, the on-site camera measurement parameters are fine-tuned to obtain the best tank truck interface positioning effect.

[0121] Table 1 target baseline reference test error examples

[0122] / m Actual Coordinates Measured Coordinates Actual Coordinates Measured Coordinates Actual Coordinates Measured Coordinates X 3.2 3.16 4.8 4.65 3.9 3.94 Y 1.8 1.77 2.3 2.26 1.5 1.53 Z 4.5 4.46 3.7 3.66 5.2 5.15 X 3.5 3.48 4.0 3.96 4.2 4.18 Y 2.0 1.97 2.5 2.47 1.2 1.18 Z 4.0 3.98 3.8 3.28 3.3 3.28

[0123] Table 2 error analysis calculations

[0124]

[0125] Example Three

[0126] Based on the LNG tank car target positioning system described in Embodiment One and Embodiment Two, the LNG tank car target positioning method is also provided. The LNG tank car target positioning method uses the LNG tank car target positioning system described in Embodiment One and Embodiment Two to achieve the LNG tank car target positioning method.

[0127] Figure 5 The implementation steps of the LNG tank car target positioning method of the embodiments of the present application are shown in the flowchart. As shown in the flowchart, the LNG tank car target positioning method includes the following steps: Figure 5

[0128] Step S501, through the image acquisition module arranged at the entrance of the loading position of the LNG loading station, real-time acquisition of the tank car image on all loading positions in the field of view is realized;

[0129] Step S502, through the image processing module, the target information of the tank car is identified according to the tank car image, and the target information includes the loading connection object positioning information of the tank car;

[0130] Step S503, under the action of the loading arm control module, the corresponding loading arm is controlled to move to the corresponding tank car gas-liquid interface according to the loading connection object positioning information, so as to realize the auxiliary control of the loading operation.

[0131] The present application discloses a kind of LNG tank car target positioning system and method for LNG loading station automatic loading system.The LNG tank car liquid / gas interface image recognition and position positioning system constructed by the system and method is as LNG automatic loading system component, both can detect tank car parking position, guide LNG tank car accurate parking, can also quickly identify tank car type and accurately position tank car liquid, gas interface / flange position coordinates, provide feedback parameter for loading arm movement control, be conducive to the unmanned, digitization and intelligent construction of large-scale LNG receiving station.In addition, image acquisition system is installed on the canopy column of LNG loading station, there is no image acquisition camera and other collision damage risk, simultaneously, image acquisition point is located at high place, is not affected by LNG loading arm loading process icing atomization, will not block light, image acquisition is fast and accurate, does not affect the continuous loading operation efficiency of loading station.

[0132] ​Thus, the application solves the problems of the existing LNG tank car loading, such as manual operation, many steps, complicated valve opening and closing operation, time-consuming and laborious docking process of the loading arm and the tank car, and large working strength of the operator, and improves the LNG high turnover efficiency and safety level of the LNG receiving station and LNG factory.

[0133] The above merely describes preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0134] In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more; the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail" and the like is based on the orientation or positional relationship shown in the drawings, and is merely for the convenience of describing the present application and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third" and the like are only for descriptive purposes, and cannot be understood as indicating or implying relative importance.

[0135] In the description of the present application, it should be noted that, unless otherwise specified and limited, the terms "connected", "connected" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0136] It should be understood that the embodiments disclosed in the present application are not limited to the specific structure, processing steps or materials disclosed herein, but should extend to equivalent alternatives of these features understood by those skilled in the relevant art. It should also be understood that the terms used herein are only for the purpose of describing specific embodiments and do not mean limitation.

[0137] The phrase "one embodiment" or "an embodiment" appearing in the specification means that the specific feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Therefore, the phrase "one embodiment" or "an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment.

[0138] Although the present application has been described with reference to the above embodiments, the contents described are only the embodiments adopted for facilitating the understanding of the present application, and are not intended to limit the present application. Any modification and change in the form and details of the present application can be made by any person skilled in the art without departing from the spirit and scope of the present application, and the patent protection scope of the present application shall be subject to the scope defined by the appended claims.

Claims

1. An LNG tanker target positioning system, characterized in that: include: An image acquisition module installed at the entrance of the loading bay of the LNG loading station is configured to collect images of tank trucks at all loading bays within the field of view in real time; An image processing module configured to identify target information of the tank truck based on the tank truck image, wherein the target information includes location information of a loading connection object of the tank truck; The loading arm control module is configured to control the corresponding loading arm to move to the corresponding tank truck gas-liquid interface according to the positioning information of the loading connection object, so as to realize auxiliary control of the loading operation.

2. The LNG tanker target positioning system according to claim 1, characterized in that: The image acquisition module includes two high-definition cameras and supporting supplementary lighting equipment respectively installed on the pillars on both sides of the canopy at the entrance of the loading bay. The LNG tanker target positioning system also includes: a data switch assembly located between the image acquisition module and the image processing module, configured to forward the tank truck image to the image processing module; An alarm prompt module is connected to the image processing module and is configured to respond to the alarm prompt information issued by the image processing module.

3. The LNG tanker target positioning system according to claim 2, characterized in that: The target information also includes the tank truck location, vehicle type and identity information, wherein the image processing module has: The edge computing unit assembly is configured to identify the position of the tank truck to be loaded in the image based on the tank truck image, and when the tank truck is detected to be parked at the designated loading position, identify the vehicle type and identity information of the current tank truck, and then locate the loading connection object positioning information of the tank truck, wherein the loading connection object positioning information includes gas phase interface positioning information, gas phase flange positioning information, liquid phase interface positioning information and liquid phase flange positioning information.

4. The LNG tanker target positioning system according to claim 3, characterized in that: The edge computing unit assembly is further configured to issue an alarm instruction and send the alarm instruction to the alarm prompt module when it detects that the tank truck is not parked at the designated loading space.

5. The LNG tanker target positioning system according to claim 3 or 4, characterized in that: The vehicle type and identity information are obtained through license plate recognition.

6. The LNG tanker target positioning system according to any one of claims 3 to 5, characterized in that: The edge computing unit assembly is further configured to detect whether the gas-liquid phase loading arm has moved into place based on the tank truck image collected during the movement of the loading arm toward the gas-liquid phase interface of the tank truck, wherein: When the vehicle is detected to be in position, a prompt message is sent to the alarm prompt module to allow the vehicle to continue loading; When it is detected that the loading arm is not moved into place, a prompt message about the loading arm not being moved into place is sent to the alarm prompt module.

7. The LNG tanker target positioning system according to any one of claims 2 to 6, characterized in that: The edge computing unit assembly is further configured to identify the location information of the loading connection object through the following steps, including: Images of the tank truck at different angles in the same scene at the same moment are captured by a first camera and a second camera through binocular calibration, wherein intrinsic parameters, extrinsic parameters, and distortion parameters of the two cameras are obtained after binocular calibration, wherein the distortion parameters include radial distortion parameters and tangential distortion parameters; After stereo rectification and parallax calculation of the two current tank car images, the corresponding reprojection matrix is ​​obtained; According to the coordinate position of the pixel of the loading connection object to be located in the two-dimensional image, the three-dimensional space coordinates of the current loading connection object are obtained using the reprojection matrix, which is recorded as the loading connection object positioning information.

8. The LNG tanker target positioning system according to claim 7, characterized in that: The reprojection matrix is ​​expressed as follows: Among them, Q represents the reprojection matrix, u o 、v o They represent the horizontal and vertical coordinates of the optical center of the main camera during parallax correction, f represents the focal length, T x represents the translation vector in the camera parameter, and the three-dimensional space coordinates of the loading connection object are obtained using the following expression: Among them, X W 、Y W and Z W They represent the three-dimensional spatial coordinates of the loading connection object, b represents the baseline parameter of the camera, and d represents the parallax.

9. The LNG tanker target positioning system according to claim 7 or 8, characterized in that: The edge computing unit assembly is further configured to obtain a calculated value of the straight-line distance between each connected object and the main view camera based on the three-dimensional spatial coordinates of the loaded connected object, and to use the calculated value of the straight-line distance in combination with the measured value of the straight-line distance to judge the accuracy of the current binocular positioning measurement by analyzing the error between the calculated value and the measured value.

10. The LNG tanker target positioning system according to any one of claims 7 to 9, characterized in that: In dual-target timing, the fixed column of the loading arm at the LNG loading station or the fixed parking mark line of the LNG tank truck is used as the reference zero baseline.

11. The LNG tanker target positioning system according to any one of claims 1 to 10, characterized in that: The LNG tanker target positioning system further includes: The data storage and monitoring module is configured to store target information of tank trucks that have completed loading operations.

12. A method for LNG tanker target positioning, characterized in that: The LNG tank truck target positioning method is implemented by the LNG tank truck target positioning system according to any one of claims 1 to 11, wherein the LNG tank truck target positioning method includes: The image acquisition module installed at the entrance of the loading bay of the LNG loading station can collect images of tank trucks in all loading bays within the field of view in real time. Identifying target information of the tank truck based on the tank truck image through an image processing module, wherein the target information includes location information of a loading connection object of the tank truck; Under the action of the loading arm control module, the corresponding loading arm is controlled to move to the corresponding tank truck gas-liquid interface according to the positioning information of the loading connection object, so as to realize auxiliary control of the loading operation.

Images