A tank bottom plate detection device based on an oil detection robot

Abstract

The present application discloses a tank bottom plate detection device based on an in-oil detection robot, which relates to the technical field of tank detection. The device comprises: an in-oil detection robot, a navigation module and a host computer; the navigation module is used to collect the positioning points of the in-oil detection robot at each moment, and according to the collected positioning points and reference paths, a path tracking algorithm based on linear regression is used to track the path of the in-oil detection robot; the oil detection robot comprises a magnetic flux leakage detection module and an ultrasonic and acoustic emission detection module connected to the host computer, and the host computer is used to perform defect detection on the tank bottom plate according to the magnetic field signal collected by the magnetic flux leakage detection module, the ultrasonic echo signal and the acoustic reflection signal collected by the ultrasonic and acoustic emission detection module. The present application can reduce positioning errors and improve the reliability of path control.

Description

Technical Field

[0001] The present application relates to the technical field of storage tank detection, and in particular to a storage tank bottom plate detection device based on an oil detection robot. Background Art

[0002] In the field of petrochemicals, large vertical storage tanks are metal storage tanks for storing crude oil and refined oil, and their diameters are generally between a dozen meters and several dozen meters. The safety of oil and refined oil storage tanks is an important part of public safety. Once a leak occurs, it often causes catastrophic accidents such as fire, explosion, and pollution. The existing safety assurance model of regular tank cleaning and inspection has prominent problems such as high cost, low efficiency, high pollution, and dangerous operations. There is an urgent need for online inspection without opening the tank.

[0003] With the development of technology, robot detection technology is being continuously applied to tank bottom plate inspection. Among them, the tank bottom plate in-oil detection robot refers to a robot that enters the storage medium through the manhole on the top of the tank to detect defects in the tank bottom plate. However, the oil in the crude oil storage tank is invisible and there are many obstacles, so the robot must have the function of obstacle detection; the crude oil storage tank is huge, which will cause the acoustic positioning signal to attenuate significantly and reduce the positioning accuracy; the manhole on the top of the tank is narrow, which limits the overall size and structural design of the robot; there is a mixture of oil, water and sediment at the bottom of the tank, which puts higher requirements on the comprehensive performance of the detection equipment. In summary, although the in-oil detection robot is the best solution to achieve non-opening tank detection, it is still a global problem.

[0004] The existing tank bottom oil detection robots have the following two problems. First, a single detection module cannot adapt to the complex working conditions of the tank bottom. Due to the presence of obstacles and sediment layers on the tank bottom, the robot cannot detect the area around the obstacles. At the same time, a single detection module is difficult to have both strong penetration and high precision requirements. Therefore, the robot's detection module needs to be completed by multiple devices. Second, the navigation software ignores the impact of positioning point errors on the control system. Since the accuracy of the acoustic positioning method is limited by many factors, it is far lower than the positioning accuracy of traditional sensors under visible and air medium conditions, which may cause positioning errors of several centimeters or even tens of centimeters, which will increase the possibility of misjudgment by the navigation control system. Summary of the invention

[0005] The purpose of this application is to provide a tank bottom plate inspection device based on an oil inspection robot. The composite inspection module carried by the oil inspection robot can adapt to the complex working conditions of the tank bottom plate and realize full coverage inspection of the tank bottom plate. At the same time, the navigation module can reduce the positioning error and improve the reliability of path control.

[0006] To achieve the above-mentioned purpose, the present application provides the following scheme: a tank bottom plate inspection device based on an in-oil inspection robot comprises: an in-oil inspection robot, a navigation module and a host computer; the navigation module is used to collect the positioning points of the in-oil inspection robot at each moment, and according to the collected positioning points and the reference path, a path tracking algorithm based on linear regression is used to track the path of the in-oil inspection robot; the oil inspection robot comprises a magnetic flux leakage detection module and an ultrasonic and acoustic emission detection module connected to the host computer, and the host computer is used to perform defect detection on the tank bottom plate according to the magnetic field signal collected by the magnetic flux leakage detection module, and the ultrasonic echo signal and acoustic reflection signal collected by the ultrasonic and acoustic emission detection module.

[0007] According to the specific embodiments provided in the present application, the present application discloses the following technical effects: the present application provides a tank bottom plate detection device based on an in-oil detection robot, which uses a variety of detection modules for switching according to the magnetic flux leakage detection module, the ultrasonic and acoustic emission detection module, and can detect the corrosion and thinning of the bottom plate below the sediment layer or around obstacles, thereby improving the detection coverage and achieving full coverage detection of the tank bottom plate; in addition, according to the collected positioning points and reference paths, a path tracking algorithm based on linear regression is used to track the path of the in-oil detection robot, which can estimate the position information of the in-oil detection robot with the smallest error, reduce the positioning error, and improve the reliability of path control. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0009] Figure 1 A schematic diagram of a working scenario provided for an embodiment of the present application.

[0010] Figure 2 A working framework diagram of an in-oil detection robot provided in one embodiment of the present application.

[0011] Figure 3 A top view of an in-oil detection robot provided in one embodiment of the present application.

[0012] Figure 4 A side view of an in-oil detection robot provided in one embodiment of the present application.

[0013] Figure 5 A bottom view of an in-oil detection robot provided in one embodiment of the present application.

[0014] Figure 6A front view of an in-oil detection robot provided in one embodiment of the present application.

[0015] Figure 7 This is a first oblique view of an in-oil detection robot provided in one embodiment of the present application.

[0016] Figure 8 A second oblique view of the in-oil detection robot provided in one embodiment of the present application.

[0017] Fig. 9 A schematic diagram of the internal structure of a fuselage provided in an embodiment of the present application.

[0018] Fig.10 A schematic diagram of the inner sealed cavity structure between the body and the top cover provided in one embodiment of the present application.

[0019] Fig.11 A framework diagram of the electromechanical system of an in-oil detection robot provided in one embodiment of the present application.

[0020] Fig.12 A framework diagram of a positioning unit provided in one embodiment of the present application.

[0021] Fig.13 A schematic diagram of a positioning unit provided in an embodiment of the present application.

[0022] Fig.14 This is a schematic diagram of the structures of a magnetic flux leakage detection module and an ultrasonic and acoustic emission detection module provided in one embodiment of the present application.

[0023] Fig.15 This is a framework diagram of a magnetic flux leakage detection module and an ultrasonic and acoustic emission detection module provided in one embodiment of the present application.

[0024] Fig.16 A schematic diagram of an ultrasonic signal processing algorithm provided in one embodiment of the present application.

[0025] Fig.17 A schematic diagram of the collaborative operation of multiple detection modules provided in one embodiment of the present application.

[0026] Fig.18 This is a flow chart of the overall control algorithm of a tank bottom oil detection robot provided in one embodiment of the present application.

[0027] Fig.19 A schematic diagram of the inspection path of a tank bottom oil inspection robot provided in one embodiment of the present application.

[0028] Fig. 20 A schematic diagram of a path tracking algorithm based on piecewise linear regression provided in one embodiment of the present application.

[0029] Fig.21 A schematic diagram of a path tracking algorithm flow chart provided in one embodiment of the present application.

[0030] Figure numerals: 1-in-oil detection robot, 2-positioning signal receiver, 3-positioning signal wireless transceiver, 4-positioning signal wireless transceiver host, 5-host computer, 6-winch, 7-composite cable, 8-guide frame, 9-tripod, -first positioning sonar rod, 1012-second positioning sonar rod, 102-sound velocity calibration sonar, 103-cable connector, 104-dredging wheel, 105-leakage magnetic flux detection module, 106-ultrasonic and sound velocity emission detection module, 107-track motion mechanism, 108-obstacle avoidance sensor, 109-leakage magnetic flux lifting mechanism, 110-top cover, 111-fuselage body, 201-dredging wheel drive motor, 202-track wheel drive motor, 301-hydraulic pump, 302-oil pipe, 303-pulling shaft, 304-lifting beam , 305-hydraulic cylinder, 401-motor driver, 402-sonar excitation circuit, 403-leakage magnetic flux detection circuit, 404-network port switch, 405-switch state monitor, 406-ultrasonic and acoustic emission detection circuit, 407-main control board, 408-electromagnetic relay switch, 409-optical fiber network port converter, 410-power management module, 501-permanent magnet magnetizer, 502-magnetic focusing sheet, 503-magnetic field probe array, 504-lifting guide shaft, 505-ultrasonic probe, 506-acoustic emission probe, 10-manhole, 11-reference path, 12-preset obstacle, 13-initial path point, 14-obstacle avoidance path point, 15-new obstacle, 16-obstacle avoidance path, 17-ultrasonic detection area, 18-leakage magnetic flux detection area. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0032] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0033] The present application provides a tank bottom plate detection device based on an in-oil detection robot, comprising: an in-oil detection robot 1, a navigation module and a host computer 5; the navigation module is used to collect the positioning points of the in-oil detection robot 1 at each moment, and according to the collected positioning points and a reference path 11, a path tracking algorithm based on linear regression is used to track the path of the in-oil detection robot 1; the oil detection robot comprises a magnetic flux leakage detection module 105 and an ultrasonic and acoustic emission detection module 106 connected to the host computer 5, and the host computer 5 is used to perform defect detection on the tank bottom plate according to the magnetic field signal collected by the magnetic flux leakage detection module 105, and the ultrasonic echo signal and acoustic reflection signal collected by the ultrasonic and acoustic emission detection module 106.

[0034] The present application adopts a path tracking algorithm based on linear regression to track the path of the in-oil detection robot 1, which can estimate the position information of the in-oil detection robot 1 with the smallest error, reduce the positioning error, and improve the reliability of path control.

[0035] In an exemplary embodiment, the storage tank is a large floating roof tank. The tank bottom plate detection device based on the oil detection robot also includes a winch 6, a composite cable 7, a guide frame 8, a tripod 9, a positioning signal receiver 2, a positioning signal wireless transceiver 3 and a positioning signal wireless transceiver host 4. Figure 1 As shown, the oil detection robot 1 is connected to the composite cable 7, and the composite cable 7 is connected to the winch 6; the positioning signal receiver 2 is connected to the positioning signal wireless transceiver 3, and the positioning signal wireless transceiver 3 is connected to the positioning signal wireless transceiver host 4 through wireless signal transmission; the winch 6 and the positioning signal wireless transceiver host 4 are connected to the host computer 5. A positioning signal transmitter is arranged on the top of the oil detection robot 1, and the positioning signal transmitter includes a first positioning sonar and a second positioning sonar. The function of the positioning signal receiver 2 is to receive the positioning signal transmitter located on the top of the oil detection robot 1. The positioning signal transmitter excites ultrasonic waves, so as to determine the position of the oil detection robot 1 according to the propagation time of ultrasonic waves. The positioning signal transmitter has a built-in piezoelectric ceramic chip, and the operating frequency is about 10kHz to 35kHz, which can convert acoustic vibration into electrical signals. The positioning signal receiver 2 is installed around the outer wall of the measured storage tank, and the installation height is at the same level as the first positioning sonar and the height of the first positioning sonar after erection. Multiple positioning signal receivers 2 are arranged at equal intervals along the outer wall of the tank, with as many as possible to reduce the error caused by a certain receiver. Generally, the number should be no less than 8, and the angle interval is 45°. Fig.13 shown.

[0036] like Figure 2As shown, the function of the positioning signal wireless transceiver 3 is to collect the electrical signal captured by the positioning signal receiver 2, and convert the voltage signal exceeding the threshold into a digital signal, and finally send it to the positioning signal wireless transceiver host 4 in the form of a digital signal through a wireless signal and attach a timestamp; the timestamp refers to the local time when the voltage signal exceeding the threshold is received. This application adopts the form of wireless signal transmission to transmit the positioning sound signal in order to avoid the trouble caused by laying cables, especially for large crude oil storage tanks with a diameter of tens of meters.

[0037] The function of the positioning signal wireless transceiver host 4 is to receive data transmitted by multiple positioning signal wireless transceivers 3, including the positioning signal wireless transceiver number, timestamp, and voltage value.

[0038] The function of the host computer 5 is to send control commands to the oil detection robot 1 and receive the data sent back; at the same time, the host computer 5 is connected to the positioning signal wireless transceiver host 4 to receive the positioning signal sent by the oil detection robot 1, which is used to calculate the position of the oil detection robot 1. The signals received by the host computer 5 include sensor signals, detection signals and positioning signals sent back by the positioning signal transceiver host carried on the oil detection robot 1. The sensor signals include the internal temperature of each detection module and the oil detection robot 1, the obstacle distance measured by the obstacle avoidance sensor 108, the posture of each actuator, the posture of the oil detection robot, etc.; the detection signals include magnetic leakage detection data, ultrasonic detection data and acoustic emission detection data.

[0039] The function of the winch 6 is to coil the composite cable 7, and transmit the optical signal and electrical signal in the composite cable 7 to the host computer 5 through the optical slip ring on the side of the winch 6 through the network cable. The cable retraction and release of the winch 6 can be controlled by the host computer 5, and the cable can also be retracted and released manually. The function of the composite cable 7 includes transmitting signals between the oil detection robot 1 and the winch 6, including 2 optical fibers for transmitting digital signals and 2 power supply cables. The guide frame 8 is installed near the manhole 10 on the top of the storage tank, and is located in the same plane as the manhole 10 and the winch 6. The function of the guide frame 8 is to guide the composite cable 7 to enter the inside of the storage tank through the manhole 10 on the top of the storage tank, and bear the pressure of the composite cable 7 during the movement and retraction of the oil detection robot 1. The tripod 9 is installed directly above the manhole 10 on the top of the storage tank, and its function is to be vertically suspended on the oil detection robot 1 when the oil detection robot 1 enters or leaves the manhole 10.

[0040] The oil detection robot 1 as a whole has a narrow and long rectangular structure. The purpose is that the overall size of the oil detection robot 1 is limited by the manhole 10 on the top of the tank, and its diameter is about 65 cm. The present application reduces the diameter of the oil detection robot 1 by integrating various modules, so as to facilitate flexible entry and exit from the manhole 10; the manhole 10 is a circular opening on the top of the tank, and a vertical ladder is provided at the opening for personnel to enter and exit the tank.

[0041] The tank bottom plate detection device based on the oil detection robot also includes a host computer 5, such as Figure 3-Figure 8 As shown, the oil detection robot 1 includes a body 111, an obstacle avoidance sensor 108 and a magnetic flux leakage lifting mechanism 109; the magnetic flux leakage detection module 105, the ultrasonic and acoustic emission detection module 106, the obstacle avoidance sensor 108 and the magnetic flux leakage lifting mechanism 109 are all connected to the host computer 5. The magnetic flux leakage detection module 105 and the magnetic flux leakage lifting mechanism 109 are installed in the first groove at the bottom of the body 111, the magnetic flux leakage lifting mechanism 109 is used to adjust the distance between the magnetic flux leakage detection module 105 and the bottom plate of the storage tank, and the ultrasonic and acoustic emission detection module 106 is located in the second groove at the bottom of the body 111; the obstacle avoidance sensor 108 is located in front of the body 111, and the obstacle avoidance sensor 108 is used to detect whether there is an obstacle in front of the oil detection robot 1 and the distance to the obstacle.

[0042] The oil detection robot 1 also includes a top cover 110, a first positioning sonar, a second positioning sonar, a sound speed calibration sonar 102, a cable connector 103, a dredging wheel 104 and a crawler motion mechanism 107; the first positioning sonar, the second positioning sonar and the sound speed calibration sonar 102 are all connected to the host computer 5.

[0043] The dredging wheel 104 is located in front of the fuselage 111 and below the obstacle avoidance sensor 108. The crawler motion mechanism 107 is located on both sides of the fuselage 111. The top cover 110 is located above the fuselage 111. The first positioning sonar is located in the third groove of the top cover 110. The second positioning sonar is located in the fourth groove of the top cover 110. The third groove and the fourth groove are at different distances from the front end of the top cover 110. The third groove is located in the front section above the top cover 110, and the fourth groove is located in the middle section above the top cover 110. The sound velocity calibration sonar 102 is located The rear end of the top cover 110; the first positioning sonar and the second positioning sonar are used to detect the steering angle of the oil detection robot 1. Specifically, the first positioning sonar is used to determine the first positioning coordinates, and the second positioning sonar is used to determine the second positioning coordinates. The host computer 5 determines the steering angle of the oil detection robot 1 according to the first positioning coordinates and the second positioning coordinates; the sound velocity calibration sonar 102 is used to collect the sound velocity of the medium in the storage tank at the current position; the cable connector 103 is fixed at the rear end of the fuselage 111 and is connected to the rear of the fuselage 111 through a rotating shaft. The composite cable 7 enters the interior of the fuselage 111 from below the rotating shaft. The cable connector 103 is used to connect the leakage magnetic detection module 105, the ultrasonic and acoustic emission detection module 106, the obstacle avoidance sensor 108, the leakage magnetic lifting mechanism 109, the first positioning sonar, the second positioning sonar and the sound velocity calibration sonar 102 to the host computer 5 through the composite cable 7.

[0044] The tank bottom plate detection equipment based on the oil detection robot includes a first positioning sonar rod 1011, a second positioning sonar rod 1012, a first positioning sonar rod 1011 driving motor and a second positioning sonar rod 1012 driving motor. One end of the first positioning sonar rod 1011 is movably connected to the third groove, and the other end is provided with a first positioning sonar. One end of the second positioning sonar rod 1012 is movably connected to the fourth groove, and the other end is provided with a second positioning sonar. The first positioning sonar rod 1011 driving motor is used to adjust the posture of the first positioning sonar rod 1011, and the second positioning sonar rod 1012 driving motor is used to adjust the posture of the second positioning sonar rod 1012.

[0045] When the first positioning sonar and the second positioning sonar are working, the first positioning sonar rod 1011 rotates from a horizontal posture to a vertical posture in the third groove, and the second positioning sonar rod 1012 rotates from a horizontal posture to a vertical posture in the fourth groove. The horizontal posture is a posture parallel to the third groove or the fourth groove, and the vertical posture is a posture perpendicular to the plane where the top cover 110 is located.

[0046] The present application uses a first positioning sonar and a second positioning sonar that are fired one after the other. The direction of the front end of the oil detection robot 1, that is, the steering angle, can be calculated based on the two positioning coordinates and the orientation of the transmitting source. The transmitting source is the first positioning sonar and the second positioning sonar. The first positioning sonar and the second positioning sonar are oblique and have an angle with the robot's forward direction. The direction of the front end of the oil detection robot 1 can be determined based on the angle and the positioning coordinates of the first positioning sonar and the second positioning sonar.

[0047] The sound velocity calibration sonar 102 is located at the rear end of the top cover 110, and its axial direction is slightly tilted upward. When the second positioning sonar rod 1012 located in the middle of the oil detection robot 1 is raised, the axis of the sound velocity calibration sonar 102 is directly opposite to the second positioning sonar rod 1012 located in the middle of the top cover 110. The front end of the sound velocity calibration sonar 102 is an ultrasonic transducer, which can excite ultrasonic waves propagating along the axial direction. The ultrasonic waves can be received by the piezoelectric ceramic sheet on the top of the second positioning sonar, so as to calculate the sound velocity of the medium in the tank at the current position.

[0048] The dredging wheel 104 is located in front of the oil detection robot 1 and is connected to the front end of the body 111. The function of the dredging wheel 104 is to stir the storage medium through its spiral structure so as to remove the soft silt deposited on the bottom plate of the tank from the bottom plate of the tank, thereby reducing the interference of the silt on the detection module signal. The dredging wheel 104 is driven by the dredging wheel driving motor 201 located at the front end of the body 111.

[0049] The magnetic flux leakage detection module 105 is used to detect the material loss caused by corrosion of the bottom plate. It is stored in the first groove at the bottom of the fuselage 111. The first groove and the second groove are both external grooves. Its mechanical structure is connected to the fuselage 111 through a set of magnetic flux leakage lifting mechanisms 109, and its signal line is connected to other electronic components inside the fuselage 111 through explosion-proof cables. The central part of the magnetic flux leakage detection module 105 is a magnetic field probe array 503 arranged in a transverse direction. The transverse direction is a direction perpendicular to the forward direction of the oil detection robot 1. A permanent magnet magnetizer 501 is respectively arranged on both sides of the transverse magnetic field probe array 503. The permanent magnet magnetizer 501 is used to generate the magnetic field required for magnetic flux leakage detection. The function of the magnetic flux leakage detection module 105 is to magnetize the steel bottom plate of the storage tank through the permanent magnet magnetizer 501, and at the same time use the magnetic field probe to detect the sudden change of the magnetic field caused by the defect. The magnetic flux leakage detection module 105 requires the magnetic flux leakage lifting mechanism 109 to complete the lifting and lowering of the module, so as to adjust the distance between the magnetic field sensor array and the bottom plate of the storage tank, that is, the lifting height.

[0050] The ultrasonic and acoustic emission detection module 106 is located in front of the magnetic flux leakage detection module 105. The bottom end surface of the ultrasonic and acoustic emission detection module 106 is 30 mm from the ground, and its lead is connected to other electronic components inside the fuselage 111 through an explosion-proof cable. The ultrasonic and acoustic emission detection module 106 includes a plurality of ultrasonic probes 505 and an acoustic emission probe 506, wherein the former is used to send ultrasonic excitation signals through the ultrasonic probe 505 and detect ultrasonic echoes to determine the thickness of the bottom plate, and the latter is used to monitor the acoustic signals caused by bottom plate corrosion and cracking around the oil detection robot 1 through the acoustic emission probe 506 to determine whether there is an area with severe bottom plate corrosion in the surrounding environment.

[0051] The crawler motion mechanism 107 located on both sides of the fuselage 111 is in contact with the bottom plate of the tank. The crawler motion mechanism 107 is the moving mechanism of the oil detection robot 1. The crawler motion mechanism 107 on each side is composed of a driving wheel located at the rear, a tensioning wheel located at the front, four rollers located at the bottom, two support wheels located at the top, and crawlers wrapped around the outside of each wheel, and the rollers are located between the driving wheels and the tensioning wheels. The driving wheel is driven by a crawler wheel driving motor 202 located at the rear of the internal cavity of the fuselage 111.

[0052] The obstacle avoidance sensor 108 is mounted under the pull rod in front of the body 111, and its function is to detect obstacles in front of the oil detection robot 1. The obstacle avoidance sensor 108 consists of two ultrasonic obstacle avoidance sensors 108 located on the outermost layer and four ultrasonic obstacle avoidance sensors 108 located in the middle ( Fig.11The ultrasonic obstacle avoidance sensors 108A to 108F have a detection distance of about 50 cm. The outermost ultrasonic obstacle avoidance sensors 108 face forward to detect obstacles with a high height inside the tank. The ultrasonic obstacle avoidance sensors 108 in the middle are grouped in pairs and adopt a one-transmit-one-receive mode to improve the detection accuracy. The probes face the ground to detect obstacles protruding from the ground that can be crossed.

[0053] The magnetic flux leakage lifting mechanism 109 is located above the magnetic flux leakage detection module 105. The oil detection robot 1 also includes four lifting guide shafts 504, which are respectively connected to the four corners of the magnetic flux leakage detection module 105, one end of each lifting guide shaft 504 is connected to the fuselage 111, and the other end is connected to the magnetic flux leakage detection module 105. The four lifting guide shafts 504 are used to lift or lower the magnetic flux leakage detection module 105 to reduce the obstruction effect of the permanent magnet magnetizer 501 on the movement.

[0054] The magnetic flux leakage lifting mechanism 109 is located inside the fuselage 111. Fig. 9 As shown, the magnetic flux leakage lifting mechanism 109 includes a hydraulic pump 301, an oil pipe 302, a drawing shaft 303, a lifting beam 304 and two hydraulic cylinders 305; the two hydraulic cylinders 305 are connected to the hydraulic pump 301 through the oil pipe 302, and the two ends of the lifting beam 304 are respectively connected to a hydraulic cylinder 305, and the two hydraulic cylinders 305 are used to lift the hydraulic cylinders 305, the drawing shaft 303 is located on the lifting beam 304, the drawing shaft 303 is connected to the magnetic flux leakage detection module 105, and the lifting beam 304 is used to transfer the lifting force of the hydraulic cylinder 305 to the drawing shaft 303. The hydraulic pump 301 is used to provide high-pressure hydraulic oil; the oil pipe 302 is used to deliver high-pressure hydraulic oil to the two hydraulic cylinders 305. In addition, in order to monitor whether the lifting beam 304 is lifted to the highest point, a toggle switch or displacement sensor is provided on the top of the lifting beam 304.

[0055] There is a sealed cavity between the top cover 110 and the body 111, which is used to accommodate components inside the oil detection robot 1, specifically various types of drive motors, circuit boards, power supplies and other electromechanical equipment. There is another cavity at the bottom for accommodating the leakage magnetic detection module 105 and the ultrasonic and acoustic emission detection module 106.

[0056] like Fig.10 As shown, the sealed cavity includes a motor driver 401, a sonar excitation circuit 402, a magnetic leakage detection circuit 403, a network port switch 404, a switch status monitor 405, an ultrasonic and acoustic emission detection circuit 406, a main control board 407, an electromagnetic relay switch 408, a fiber optic network port converter 409 and a power management module 410.

[0057] The motor driver 401 is used to set the direction and speed of each motor. The motor driver 401 is also used to collect the power and actual speed of the motor. The motor driver 401 is connected to the main control board 407 through a bus. The power supply of the motor driver 401 is 24V. Each motor corresponds to a motor driver 401. Fig.11 The first motor to the fifth motor are respectively the silting wheel driving motor 201, the first positioning sonar rod 1011 driving motor, the second positioning sonar rod 1012 driving motor, the left track wheel driving motor 202 and the right track wheel driving motor 202. Each motor corresponds to a motor driver 401.

[0058] The switch status monitor 405 is used to monitor the position switch of the magnetic flux leakage lifting mechanism 109 and the position switch of the first positioning sonar rod 1011 ( Fig.11 The first position switch in the middle) and the position switch of the second positioning sonar rod 1012 ( Fig.11 (second position switch in the middle).

[0059] The sonar excitation circuit 402 is used to send pulse signals with specific time intervals, amplitudes and frequencies to the first positioning sonar, the second positioning sonar and the sound speed calibration sonar 102 respectively. The sonar excitation circuit 402 is connected to the main control board 407 via a serial port line, and the sonar excitation circuit 402 is powered by 24V.

[0060] The magnetic field probe array 503 includes 1 row of 21 columns of magnetic field probes, each of which is a Hall element. The detection width of the magnetic flux leakage detection module 105 is the width of the 21 columns of magnetic field probes, and the detection width is the path width of the reference path 11.

[0061] The detection module of the present application includes a magnetic flux leakage detection module 105 and an ultrasonic and acoustic emission detection module 106. The mechanical design of the detection module is as follows: Fig.14 As shown, the signal transmission framework of the detection module is as follows Fig.15 shown.

[0062] The magnetic flux leakage detection module 105 includes a magnetic field probe array 503, an ultrasonic detection unit includes a plurality of ultrasonic probes 505, and an acoustic emission detection unit includes an acoustic emission probe 506. The magnetic flux leakage detection module 105 also includes a magnetic field collecting plate 502, which is two long iron plates located on both sides of the magnetic field probe array 503. The magnetic field probe array 503 detects the surrounding magnetic field strength. The magnetic field collecting plate 502 is used to increase the magnetic field strength. Two permanent magnet magnetizers 501 are located on both sides of the magnetic field collecting plate 502. When the oil detection robot 1 moves forward, the detection width each time is equal to the probe array width W. mflThe bottom plate corrosion state. The lifting guide shaft 504 is used to guide the movement direction of the leakage magnetic detection module 105 to remain perpendicular to the bottom surface of the fuselage 111 during the lifting process. The leakage magnetic detection circuit 403 reads the voltage signal of the 21-channel Hall element through the analog-to-digital converter, and converts the measured voltage signal into a digital signal and transmits it to the host computer 5. The single packet leakage magnetic data is based on 1 channel N mfl sampling points, 2 channels N mfl sampling points, ..., 21 channels N mfl The order of the sampling points is uploaded in sequence, and then unpacked by the host computer 5. mfl The host computer 5 can process the magnetic flux leakage detection data in the following steps 101 to 104.

[0063] Step 101: Data unpacking, specifically, the original leakage magnetic digital signal is converted into a 1-dimensional array, i.e., 1×21N mfl Vector, disassembled into a 21-dimensional array, that is, 21×N mfl matrix.

[0064] Step 102: wavelet transform, specifically, the wavelet basis function uses the Mexican Hat function. According to the commonly used wavelet transform method, the correlation between the single-channel magnetic flux leakage digital signal and the wavelet basis function in the window can be calculated. If the wavelet basis function in the window function is more similar to the magnetic flux leakage signal, the higher the correlation, the higher the wavelet function value will be. If the base value fluctuation of the single-channel magnetic flux leakage digital signal is within the preset fluctuation range, this step can be closed, that is, if the magnetization is sufficient, the base value is stable, and wavelet transform is not required, thereby improving the online detection speed, or after the detection, the magnetic flux leakage digital signal in the database is processed offline.

[0065] In order to improve the signal-to-noise ratio of the magnetic leakage digital signal, the present application introduces wavelet transform to identify the magnetic leakage signal with obvious characteristics.

[0066] Step 103: Amplitude compensation, specifically including: before defect detection, wavelet transform is performed on standard parts with defects of different sizes, and the ratio between the signal output value and the real defect size is calibrated. The signal output value is the single-channel magnetic leakage digital signal after wavelet transform. After wavelet transform, the result is scaled according to the relationship between the real defect size and the signal output value, so as to restore the real size of the detected defect.

[0067] Step 104: Defect calibration, specifically marking the defect results on the two-dimensional map according to the coordinates of the positioning points of the signals where the defects appear in the database.

[0068] The ultrasound and acoustic emission detection module 106 includes an ultrasound detection unit and an acoustic emission detection unit.

[0069] In an exemplary embodiment, the number of ultrasonic probes 505 in the ultrasonic detection unit is 8, and the ultrasonic probes 505 are arranged in a staggered array of 2 rows and 4 columns to improve the detection resolution, and the detection width is W ut .

[0070] The magnetic flux leakage detection circuit 403 is used to collect the voltage signal of each magnetic field probe in the magnetic field probe array 503, and convert the voltage signal of each magnetic field probe into a digital signal and upload it to the host computer 5 through the Ethernet in the composite cable 7. The magnetic flux leakage detection circuit 403 contains multiple analog-to-digital converters to convert the 5V voltage into 16-bit data. The magnetic flux leakage detection circuit 403 is connected to the network port switch 404 through an Ethernet cable, and the power supply is 5V, such as Fig.11 shown.

[0071] The function of the network port switch 404 is to connect a single Ethernet line output from the optical fiber converter to multiple network devices including the magnetic flux leakage detection circuit 403 , the ultrasonic and acoustic emission detection circuit 406 , and the main control board 407 .

[0072] The function of the switch status monitor is to monitor the current position of each actuator in the oil detection robot 1. The switch status monitor is used to monitor the status information of the travel switch of the lifting beam 304 and the positioning sonar rod; if a displacement sensor is installed on the top of the lifting beam 304, the switch status monitor is not connected to the lifting beam 304. The switch status monitor is connected to the main control board 407 through a serial port line, and the power supply is 5V.

[0073] The ultrasonic and acoustic emission detection circuit 406 is used to collect the voltage signal of each ultrasonic probe 505 and the voltage signal of the acoustic emission probe 506, and upload the voltage signal of each ultrasonic probe 505 and the voltage signal of the acoustic emission probe 506 to the host computer 5 through the Ethernet in the composite cable 7. The ultrasonic and acoustic emission detection circuit 406 is connected to the network port switch 404 through an Ethernet line, and the power supply is 24V.

[0074] The ultrasonic and acoustic emission detection circuit 406 includes an ultrasonic detection circuit and an acoustic emission detection circuit. The ultrasonic detection circuit is composed of a digital signal processor that can generate low-voltage pulses with adjustable width, a transmitting circuit that can amplify low-voltage pulses, and an amplifying and filtering circuit that receives ultrasonic echoes; the acoustic emission nondestructive detection circuit is composed of a signal amplifier that amplifies low-level acoustic emission signals to a moderate level, an analog filter that filters and denoises the amplified signals, and an analog-to-digital converter that records and analyzes the filtered signals. In the storage medium to be tested, the propagation range of the acoustic emission signal on the bottom plate of the tank is denoted by R ac , which can be determined through experimental research before testing.

[0075] After receiving the parameter command sent by the host computer 5, the ultrasonic detection circuit will send out an excitation signal according to the configured parameters and complete the sampling; the parameters that need to be configured include: excitation voltage, filter frequency band, sampling rate, number of sampling points N ut The value of the ultrasonic signal is the voltage value after taking the absolute value, and the single-packet ultrasonic data has the same data packet format as the magnetic flux leakage detection circuit 403. The host computer 5 can process the ultrasonic detection data in steps 201 to 205.

[0076] Step 201: intercepting the base value Q0, specifically including: the data unpacking algorithm for ultrasonic signal processing is the same as that for magnetic flux leakage signal processing. After obtaining the single-channel ultrasonic signal data, firstly clear the ultrasonic signal data points whose logarithmic values ​​are less than the set voltage value to eliminate the influence of noise and obtain the denoised ultrasonic signal.

[0077] Step 202: Finding the oscillation peak point H1, specifically including: determining each peak point of the oscillation signal by comparing the voltage values ​​between adjacent sampling points in the denoised ultrasonic signal, such as Fig.16 As shown in (a) in the figure, in order to determine the peak spacing of the echo signals, the peak position of each echo signal needs to be determined. Fig.16 Part (a) shows the base value and echo sampling point changing with time. Fig.16 Part (b) shows the echo signal changing with time.

[0078] Step 203: Segmented envelope interpolation, specifically including: after discarding the excitation pulse signal Q1 of the initial segment of the denoised ultrasonic signal, the echo signal Q2 is obtained, and the oscillation peak point of the echo signal is segmented; when it exceeds half of the normal peak spacing, if there is no new oscillation peak point, it is segmented into one segment to obtain multiple segmented echo signals, and the normal peak spacing is the set peak spacing; the envelope of each segment of the echo signal is calculated. In order to improve the calculation efficiency, the spline interpolation method can be used to calculate the envelope of each segment of the echo signal, and the control point of the interpolation algorithm is the oscillation peak point.

[0079] Step 204: Finding the envelope peak point H2, specifically including: after increasing the density of discrete sampling points in each segment echo signal, calculating the position of the maximum value of each interpolation curve in each segment echo signal, and obtaining the envelope peak point in the envelope of each segment echo signal. is the distance between two envelope peak points. If the required ultrasonic thickness measurement error is smaller than the corresponding thickness value between adjacent oscillation peak points, the envelope peak point can be directly replaced by the maximum value of the oscillation peak point in each segment; this method can improve signal processing efficiency.

[0080] Step 205: Calculate the average interval. Calculate the average of the interval distances between the envelope peak points in adjacent segments and convert it into the thickness of the tank floor being tested, so that the position where the tank floor thickness is less than the set thickness value is marked as a defect position.

[0081] The signal processing method of the acoustic emission signal is similar to that of the ultrasonic signal. The key waveform parameters extracted are the rise time, peak value, fall time, tail length, etc. of the acoustic emission signal, which are used to determine whether it is a corrosion cracking signal of the tank bottom plate.

[0082] The main control board 407 sends instructions to the following devices: temperature sensor and inertial navigation sensor, multiple motor drivers 401, multiple obstacle avoidance sensors 108, sonar excitation circuit 402, magnetic flux leakage detection circuit 403, network port switch 404, switch state monitor, ultrasonic and acoustic emission detection circuit 406 and electromagnetic relay, and is connected to multiple obstacle avoidance sensors to obtain data. The power supply of the main control board 407 is 24V, and the 24 power supply is converted into a 5V power supply for use by other devices. The inertial navigation sensor is used to monitor the robot posture. Monitoring acceleration can determine whether the robot falls to the bottom and lies flat during the descent of the robot from the manhole 10. It can also monitor the robot to cross obstacles during the bottom plate detection process. Monitoring angular acceleration can calculate the robot's rotation angle through double integrals, which can realize the robot's rotation at a fixed angle in place. The main control board 407 is connected to the host computer 5, and the instructions issued by the host computer 5 include: motor movement, detection module parameter configuration and acquisition, sonar parameter configuration and excitation, and inertial navigation sensor parameter configuration. The instructions automatically executed by the main control board 407 include: obstacle avoidance, temperature sensor collection and uploading of information, etc.

[0083] The function of the electromagnetic relay switch 408 is to control the power switch of the high-power electrical equipment in the oil detection robot 1, such as the power switch of each motor, through the electromagnetic relay. The electromagnetic relay switch 408 can control the power supply line switch of the three motor drivers 401 and a hydraulic pump 301, and send the switch command to the main control board 407 through the serial port line. The power supply is 5V. The function of the optical fiber network port converter 409 is to convert the optical fiber signal line into an Ethernet network cable inside the oil detection robot 1, one end is connected to the composite cable 7, and the other end is connected to the network card switch. The power supply is 5V. The function of the power management module 410 is to convert the 220V AC voltage input at the winch 6 into a constant 24V DC voltage.

[0084] The positioning unit of this application consists of three parts: internal and external tank hardware and positioning algorithm software. Fig.12As shown, the positioning unit belongs to the navigation module. The hardware part inside the tank includes: a first positioning sonar rod 1011, a first positioning sonar rod driving motor, a motor driver of the first positioning sonar rod, a second positioning sonar rod 1012, a second positioning sonar rod driving motor, a motor driver of the second positioning sonar rod and a sonar excitation circuit 402, and the motor driver of the first positioning sonar rod and the motor driver of the second positioning sonar rod are connected to the host computer 5; the hardware part outside the tank includes: a positioning signal receiver 2, a positioning signal wireless transceiver 3, a positioning signal wireless transceiver host 4 and a host computer 5.

[0085] The sonar excitation circuit 402 is composed of a central processing unit (CPU), a pulse signal transmitter and a signal conditioning circuit. The CPU receives the signal excitation command sent by the main control board 407, including the excitation voltage A imp , pulse frequency f imp 、Number of continuous pulses imp , continuous pulse interval t imp , Interval excitation time t gap . Excitation voltage A imp is the voltage value of the pulse signal. Pulse frequency f imp The number of voltage vibrations per unit time. imp Determines the number of repetitions of the same location positioning point. Continuous pulse interval t imp It is the interval time between two consecutive pulses. Interval excitation time t gap The interval time between two positioning points, together with the travel speed of the oil detection robot 1, determines the travel distance between the two positions.

[0086] The positioning method of the positioning unit includes: obtaining the voltage amplitude generated by each positioning signal receiver 2 of the multiple positioning signal receivers 2 installed on the outer wall of the storage tank when receiving the sound signal. For each positioning signal receiver 2, it is determined whether the voltage amplitude generated by the positioning signal receiver 2 when receiving the sound signal is greater than the preset voltage amplitude. If so, the positioning signal receiver 2 is selected as an available receiver; the sound signal is emitted by the positioning signal transmitter (first positioning sonar and second positioning sonar) installed on the oil detection robot 1 on the bottom plate of the storage tank. Obtain the absolute arrival time of the sound signal to each available receiver, and obtain the absolute arrival time corresponding to each available receiver; sort all available receivers in order from small to large according to the absolute arrival time, and for each available receiver, calculate the difference between the absolute arrival time corresponding to the available receiver and the absolute arrival time corresponding to the available receiver ranked first, and obtain the relative arrival time corresponding to the available receiver. The tank bottom plate is positioned on the oil detection robot 1 according to the installation positions and relative arrival times corresponding to all available receivers.

[0087] In an exemplary embodiment, the positioning of the tank bottom plate on the oil detection robot 1 adopts a multi-point overdetermined optimization solution algorithm, that is, the propagation time collected by all positioning signal receivers 2 that exceed the sampling threshold is brought into the objective function for optimization iterative calculation. The positioning step includes steps 301 to 305.

[0088] Step 301: Determine the coordinates. Establish the coordinate system XOY with the center of the tank, the coordinate of the center O of the tank is (0, 0), the radius of the tank is R, and the coordinate of the positioning sonar of the oil detection robot 1 is A ( x , y ), with coordinates A ( x , y ) is the first positioning sonar or the second positioning sonar, x is the horizontal axis, y is the vertical axis, Fig.13 Point A is A( x , y ); There are N positioning signal receivers 2 with equal angles around the tank. The coordinates of the i-th positioning signal receiver 2 are P i ( , ) is determined by the following formula: ; ; ,in, is the angle between the i-th positioning signal receiver 2 and the center of the circle, is the horizontal coordinate of the i-th positioning signal receiver 2, is the ordinate of the i-th positioning signal receiver 2. For example, Fig.13There are 8 positioning signal receivers 2 in total, that is, N=8.

[0089] Step 302: Activate the sonar. The positioning sonar activates a signal at point A according to the set parameters, and sends out N imp After an unknown time, the nearest positioning signal receiver 2P i The first voltage signal detected to exceed the threshold is denoted as P1'. After that, the remaining positioning signal receivers 2 are sorted in order according to the time when they exceed the threshold, and are recorded as P2' to P M ', M-1 is the number of positioning signal receivers 2 whose voltage signals exceed the threshold; wherein, the number of positioning signal receivers 2 that do not exceed the threshold is NM. Fig.13 Middle, D i For the i The distance between the positioning signal receiver 2 and point A, P2 is closest to point A, recorded as P1', and the coordinates are from P2 (x P2 ,y P2 ) is rewritten as P1'(x P1 ',y P1 '); among the remaining points, P3 is recorded as P2', P1 is recorded as P3', P4 is recorded as P4', and so on; the time from when P1' receives the voltage exceeding the threshold value to when P2' receives the voltage exceeding the threshold value is t 1,2 '、t 1,3 '、t 1,4 ', and so on. If N imp When it is greater than 1, multiple oscillation signals appear continuously in the signals collected by each positioning signal receiver 2. At this time, the time interval t imp The obtained signal is segmented, and each segment is recorded as a group of data points for independent data processing.

[0090] Step 303: Establish equations. Since the time when the pulse signal is sent is unknown, the relative time solution method is adopted, and the positioning equations are as follows: .

[0091] in, , For the i Point P receives the pulse signal i 'The horizontal and vertical coordinates; The time when the first positioning signal receiver 2 receives a voltage exceeding the threshold value, which is an unknown quantity; is the difference between the time when the first positioning signal receiver 2 receives a voltage exceeding the threshold value and the time when the i-th positioning signal receiver 2 receives a voltage exceeding the threshold value, and =0; c is the speed of sound in the tank medium.

[0092] Step 304: Particle swarm optimization. To improve the calculation speed, the particle swarm optimization method is used. The particle swarm algorithm assumes that there are a number of evenly dispersed particles within the range of the tank, and each particle searches for the minimum value of the target equation in inertial motion; in one round, if the target equation value found by a particle is the smallest among all particles, the current position is broadcast to the remaining particles. In the next round, the remaining particles receive the position and move toward it; when the target equation values ​​of all particles converge within a very small range, the optimization ends. The optimized target equation for: .

[0093] in, To introduce the currently unknown error of the particle, M The number of signal receivers 2 is located for all signals whose voltage exceeds the threshold. The optimization goal is to minimize the sum of errors of all valid sampling points. The detailed steps of particle swarm optimization are the same as those of the existing method.

[0094] Step 305: The first step is to calculate the number of consecutive N pulses in a group of pulses. imp The average position of the positioning pulse signals is used to determine a more accurate positioning point. The second step is to calculate the distance between the positioning point of this group of data and the positioning point of the previous group of data. If it is far beyond the position that the current oil detection robot 1 can reach, it is recorded as abnormal data and the data of this group is discarded.

[0095] The present application discloses a detection method, overall control algorithm and path tracking algorithm of an oil detection robot 1 in an invisible environment. The oil detection robot 1 for the tank bottom can adopt an automatic control mode to perform full coverage operation on the tank bottom. The detection mode adopts leakage magnetic field, ultrasound and acoustic emission detection to work in coordination, such as Fig.17 As shown, Fig.17 Part (a) is a schematic diagram of the detection width of the combined detection of magnetic flux leakage and ultrasound. Fig.17 Part (b) is a schematic diagram of the detection width of ultrasonic testing. When thick sludge exists on the bottom plate, ultrasonic testing is used to detect the thickness of the sludge. Fig.17 Part (c) is a schematic diagram of the detection range of acoustic emission detection.

[0096] Under normal circumstances, that is, when the thickness of the bottom plate deposit is less than the set thickness, the full-channel magnetic flux leakage probe and the dual-channel ultrasonic probe 505 are activated for scanning; when the bottom plate deposit is thicker, the full-channel ultrasonic probe 505 is activated for scanning; for areas with obstacles, the acoustic emission probe 506 is activated to monitor the corrosion acoustic emission signal under static conditions to estimate the degree of corrosion. The route map and schematic diagram of the above operation method are shown in Fig.18 and Fig.19 .

[0097] In terms of defect detection of the tank bottom plate based on the magnetic field signal collected by the leakage magnetic detection module 105 and the ultrasonic echo signal and acoustic reflection signal collected by the ultrasonic and acoustic emission detection module 106, the upper computer 5 is specifically used for: 1) During the movement of the oil detection robot 1, the thickness of the bottom plate sediment is detected by the leakage magnetic detection module 105 or the ultrasonic detection unit.

[0098] 2) When the thickness of the bottom plate deposit is less than the set thickness, that is, it belongs to the leakage magnetic detection area 18, the leakage magnetic detection module 105 is enabled to perform tank bottom plate defect detection, and the dual-channel ultrasonic probe 505 in the ultrasonic detection unit is started to detect the thickness of the bottom plate deposit. When a defect is detected, the defect position is marked, and the defect position mark is the coordinate corresponding to the defect.

[0099] 3) When the thickness of the bottom plate sediment is greater than or equal to the set thickness, that is, it belongs to the ultrasonic detection area 17, the leakage magnetic flux lifting mechanism 109 is controlled to lift the leakage magnetic flux detection module 105 until the distance between the leakage magnetic flux detection module 105 and the tank bottom plate is the set distance threshold, and the leakage magnetic flux detection module 105 and the full-channel ultrasonic probe 505 in the ultrasonic detection unit are enabled to perform tank bottom plate defect detection, and the defect position is marked when a defect is detected.

[0100] 4) When it is detected that there is an obstacle within a set distance in front of the oil detection robot 1 moving along the reference path 11, the oil detection robot 1 is controlled to stop moving, the positioning function is turned off, the acoustic emission detection unit is started to perform tank bottom plate defect detection, and the defect position is marked when a defect is detected.

[0101] 5) When the acoustic emission detection unit finishes the detection, an obstacle avoidance path 16 is generated according to the size of the obstacle, and the oil detection robot 1 is controlled to move to the reference path 11 according to the obstacle avoidance path 16, and returns to the step of detecting the thickness of the bottom plate sediment through the magnetic flux leakage detection module 105 or the ultrasonic detection unit until it reaches the end of the reference path 11; the obstacle avoidance path 16 is composed of three obstacle avoidance branches connected in sequence, the angle between each connected obstacle avoidance branch is 90 degrees, and the distance between each obstacle avoidance branch and the obstacle is ,in, L blk is the maximum side length of the obstacle (known quantity), k tr is the coefficient, W mfl is the path width of the reference path 11.

[0102] The interior of the fuselage 111 includes a dredging wheel driving motor 201 and a track wheel driving motor 202 . The dredging wheel driving motor 201 is used to drive the dredging wheel 104 , and the track wheel driving motor 202 is used to drive the track motion mechanism 107 .

[0103] In an exemplary embodiment, the specific steps of the overall control algorithm of the oil detection robot 1 are steps 401 to 408.

[0104] Step 401: Initialization configuration, specifically including: after the oil detection robot 1 enters the tank, it initializes its own state and loads the initial path. The initial path refers to the initial passable area in the tank that is traversed by a "Z"-shaped broken line. The width that a single path can cover is W. mfl The initial passable area refers to the area that the oil detection robot 1 can detect based on the precise obstacle position in the tank. The obstacle is called a preset obstacle. The distance from the edge of the initial passable area to the preset obstacle is the width of the oil detection robot 1 multiplied by a safety factor k. tr In order to bypass obstacles, the preset obstacle avoidance path 16 is three vertical broken lines, and the distance between each broken line and the preset obstacle 12 is (1+k tr )W mfl .

[0105] Step 401: Determine the thickness of the sediment layer, specifically including: preferentially determine the thickness of the sediment layer on the bottom plate of the storage tank, thereby determining the detection mode. There are two methods of determination. The first is to determine whether there is a large fluctuation in the base value of the leakage magnetic detection result when the leakage magnetic detection module 105 is enabled. When the bottom plate is in normal condition, the measured voltage value should be approximately a straight line, that is, the base value. When there is sediment on the bottom plate of the storage tank, the lift-off will increase, and the detected magnetic field intensity will fluctuate greatly, irregularly, and over a long span with the spatial position. This fluctuation will affect the detection accuracy or cause defects to be missed. To this end, the variance of the voltage value measured by a single magnetic field probe is calculated to determine the degree of fluctuation of the base value, thereby determining whether there is a sediment layer. The second is when the ultrasonic detection module is enabled, that is, the previous state is abnormal, if the amplitude of the ultrasonic echo signal increases significantly, the lift-off returns to the normal level. For normal conditions, enter step 403; for abnormal conditions, enter step 404.

[0106] Step 403: Detection under normal conditions, specifically including: when each detection module determines that the thickness of the current bottom plate deposition layer is low, the full channel of the magnetic field array probe of the magnetic flux leakage detection module 105 is activated for detection. At the same time, in order to quantify the bottom plate thickness, the two ultrasonic probes 505 near the edge of the ultrasonic detection module are activated for detection.

[0107] Step 404: Detection of abnormal conditions, specifically including: when the magnetic flux leakage detection module 105 determines that the thickness of the current bottom plate deposition layer is high, all probes of the ultrasonic detection array are activated for detection.

[0108] Step 405: Detecting obstacles, specifically including: when the oil detection robot 1 moves along the initial path, using the obstacle avoidance sensor 108 to monitor whether there are impassable obstacles in the front area. If there are no obstacles within 30 cm in front, continue to move forward, that is, step 408; otherwise, enter the obstacle avoidance mode, that is, step 406.

[0109] Step 406: Acoustic emission detection, specifically including: since the oil detection robot 1 does not pass through the surrounding area of ​​the obstacle, the acoustic emission detection module is used to cover the obstacle and its surrounding area, and the corrosion condition in the area is judged by continuously monitoring the acoustic emission signal. When an obstacle is found, the oil detection robot 1 stops at the current position and remains silent, and the positioning unit and obstacle avoidance module need to be turned off, and the ultrasonic and acoustic emission detection circuit 406 is started, and the acoustic emission signal in the environment is monitored by the acoustic emission probe 506. The signal listening range is R ac If a signal matching the characteristics of the corrosion acoustic emission signal is identified within a specific time, the monitoring time is extended; otherwise, the acoustic emission detection is terminated and the process proceeds to step 407 .

[0110] Step 407: obstacle avoidance movement, specifically including: to complete the obstacle avoidance movement, first generate an obstacle avoidance path 16 at the current position, the path consists of four obstacle avoidance path points 14 arranged in a square and three obstacle avoidance paths 16. Since the exact range of the obstacle is unknown, it is necessary to make a worst-case assumption based on the approximate size of the obstacle in the tank design drawing: assuming that the maximum side length of the obstacle in the current area is L blk , then the distance between obstacle avoidance path points 14 is After that, the oil detection robot 1 moves along the obstacle avoidance path 16, and each time at the obstacle avoidance path point 14, it rotates 90 degrees in situ according to the angular relationship of the path according to the angular acceleration measured by the inertial navigation sensor. When the oil detection robot 1 passes the last obstacle avoidance path point 14 and completes the turn, it enters step 408.

[0111] Step 408: During the traveling process, detection is performed through the enabled detection module. If the current target path point of the route is the last initial path point 13, that is, the end point, the detection ends; otherwise, detection continues and the process returns to step 402.

[0112] In summary, the path planning algorithm of the oil detection robot 1 can observe obstacles that cannot be crossed by the oil detection robot 1 at a long distance. During the movement, the range of the obstacles needs to be estimated.

[0113] The difficulty of the path tracking of the oil detection robot 1 in the tank bottom plate in the present application is that the positioning accuracy and refresh rate in the invisible environment are lower than the accuracy that can be achieved by the detection methods such as visible light, laser radar, GPS, etc. in the conventional and visible environment. Since the positioning signal receiver 2 is installed on the outer wall of the tank, the acoustic positioning accuracy will be affected by factors such as the environmental noise inside and outside the tank, the reflection and echo of the positioning signal, and the geometric error of the tank wall, resulting in centimeter-level random errors. This means that the positional relationship between two adjacent positioning points has a lower credibility, and only a large number of data points generated by the oil detection robot 1 during the movement have a higher credibility in statistics. Therefore, when calculating the path deviation, it is necessary to estimate the direction and position of the oil detection robot 1 at the current moment based on the set of positioning points collected from a previous moment to the current moment, so as to make a more reasonable control decision. The present application proposes a path tracking algorithm based on linear regression under non-precise positioning conditions.

[0114] like Fig. 20 and Fig.21 As shown, in terms of using a linear regression-based path tracking algorithm to track the path of the in-oil detection robot 1 according to the collected positioning points and the reference path 11, the navigation module is used to: 1) determine whether the in-oil detection robot 1 turns or stops on the spot, and if so, stop the path tracking, otherwise initialize the positioning point set to be empty. 2) Sequentially determine the distance between the positioning points collected at two adjacent moments, and if the distance between the positioning points collected at two adjacent moments is less than or equal to a first set value, then put the positioning point at the latter of the two adjacent moments into the positioning point set. 3) When the number of positioning points in the positioning point set is greater than the set number and the radius of the positioning point set is greater than a second set value =3 , perform univariate linear regression on the set of positioning points to obtain the first straight line; the radius of the set of positioning points is the maximum value of the distance between the average value of the positioning points in the set of positioning points and any positioning point. 4) Calculate the linear fitting correlation coefficient of the first straight line. 5) If the linear fitting correlation coefficient is less than the threshold, stop path tracking, otherwise determine the current direction and position of the oil detection robot 1 based on the first straight line. 6) Based on the reference path 11, the current direction and position of the oil detection robot 1, use the PID control path planning algorithm to control the oil detection robot 1.

[0115] Wherein, determining the direction and position of the oil detection robot 1 at the current moment according to the first straight line specifically includes: according to the formula Calculate the orientation of the oil detection robot 1 at the current moment; is the direction at the current moment, The straight line passing through the last positioning point in the positioning point set and perpendicular to the first straight line is taken as the second straight line, and the intersection of the first straight line and the second straight line is taken as the position of the oil detection robot 1 at the current moment. Fig. 20 Part (a) is the position of the oil detection robot 1 at the current moment determined by the first straight line and the second straight line during the path tracking process. Fig. 20 Part (b) is a schematic diagram of the situation where the positioning point set fails. Fig. 20 Part (c) is a schematic diagram of the path tracking correction principle.

[0116] In an exemplary embodiment, the path tracking algorithm based on linear regression specifically includes steps 501 to 505 .

[0117] Step 501: Determine the start conditions, specifically including: the confidence level determined by experiment is 95% and the positioning error radius is , if the distance between the front and rear positioning points is greater than 2 , it is considered that the current positioning point is affected by abnormal noise, and the positioning point is discarded in the path tracking algorithm. The first setting value is 2 In each round of control, the latest K positioning points are taken as data input, K is the set number, recorded as the positioning point set ( x i , y i ). If the radius of the anchor point set Greater than 3 , then path tracking control starts, and the second setting value is 3 ; The anchor point set The average value ( , ) to any of its points.

[0118] The radius of the anchor point set is expressed as: .

[0119] Step 502: Estimating the current direction and positioning point, specifically including: Perform a univariate linear regression and get the first straight line , the first straight line equation is: .in, and All are fitting coefficients.

[0120] Step 503: Failure conditions, specifically including: when the positioning point set When the distribution diverges significantly, the current set of positioning points should be abandoned. The judgment method is to calculate the set of positioning points If the linear fitting correlation coefficient R value is less than the threshold value of 0.90, it is considered that the current positioning point set has an error and the path tracking control at the current moment should be abandoned. For example, Fig. 20 Part (b) The first straight line obtained at the moment If R is less than the threshold value of 0.90, the path tracking control at the current moment is abandoned. The time is the time when there are K+S positioning points in the positioning point set.

[0121] Step 504: estimating the current direction and positioning point, specifically including: the current moving direction of the oil detection robot 1 is recorded as: .

[0122] Current location For the last positioning point ( x K , y K ) and with the first straight line The second vertical line With the first straight line The intersection of is the horizontal coordinate of the current position, is the ordinate of the current position; the second straight line l 2 is written as: .

[0123] Step 505: PID control, specifically including: comparing the current position and orientation of the oil detection robot 1 with the pre-planned path, calculating the deflection angle between the two, and substituting the path planning algorithm based on PID control to calculate the left and right speeds of the crawler drive motor. The orientation is the direction of travel, for example, Fig. 20 Part (c) Moment and The deflection angle at time is .

[0124] Tracking of the deflection angle. The scope of application of the above method is the movement mode of the oil detection robot 1 moving in a straight line or lateral during movement, and when the oil detection robot 1 turns in place, the path tracking needs to be stopped. Since the oil detection robot 1 rotates slowly in place, turning in place will cause a large number of positioning points to accumulate in place, resulting in the failure of the above path tracking algorithm. There are two methods for tracking the steering angle when the oil detection robot 1 turns in place. The first method is to calculate the integral of the angular velocity based on the inertial navigation sensor carried by the oil detection robot 1 to determine the deflection angle; the disadvantage of this method is that when the angular acceleration sensor is used to calculate the deflection angle, mechanical vibration will produce accumulated errors in the calculation results. Therefore, after turning in place, the direction of travel needs to be corrected during the movement. The second method is to determine the direction of the oil detection robot 1 according to the two positioning points generated by stimulating two positioning sonars one after the other during the rotation process; the disadvantage of this method is that the error of a single positioning point will also produce an error in the deflection angle. In actual application, it is necessary to determine a more accurate deflection angle monitoring method based on the on-site conditions.

[0125] In an exemplary embodiment, the complete operation flow of the tank bottom plate oil inspection robot 1 during the inspection of the large floating roof tank bottom plate is as follows: Step 601 to Step 605.

[0126] Step 601: Preparation.

[0127] Before the inspection team enters the tank farm, they need to complete the following preparations: calibrate the inspection module, obtain relevant information about the tank to be tested, preset inspection parameters, contact the crane, and prepare for air replacement.

[0128] The obtained information of the tank to be tested includes: the diameter of the tank, the location of the external appendages of the tank (pipelines, brackets, stairs, etc.), and the location of obstacles inside the tank.

[0129] The preset detection parameters include: the installation position of the positioning signal receiver 2, the physical properties of the storage medium, and the working parameters of each detection module.

[0130] The reason for contacting the crane is that the weight of the tank bottom plate of the oil detection robot 1 can reach 200kg, so a crane is needed to complete the lifting and lowering operations. In addition, it should be determined whether the operating position of the crane and the length of the boom can meet the lifting and lowering requirements of the oil detection robot 1.

[0131] The purpose of obtaining the information of the tank to be tested and determining the installation position of the positioning signal receiver 2 is to avoid interference between the positioning signal receiver 2 and the external appendages of the tank; if there are obstacles, the coordinates P of each positioning signal receiver 2 in the software should be appropriately adjusted. i .

[0132] The purpose of determining the internal obstacles of the storage tank is to complete the path planning in advance. If the position of the internal obstacles is uncertain, the obstacles may not be marked and the in-oil detection robot 1 may detect them by itself instead.

[0133] The physical properties of the storage medium are determined in order to obtain the sound velocity and attenuation coefficient, where the sound velocity c is related to the propagation distance of the positioning signal in the tank, and the attenuation coefficient determines whether there are at least three positioning signal receivers 2 that can obtain voltage signals exceeding the threshold when the oil detection robot 1 is at any position in the tank; otherwise, a positioning sonar and sonar excitation circuit 402 with a lower operating frequency must be selected.

[0134] In order to remove the flammable gas between the floating roof and the medium in the floating roof tank, the accumulated gas needs to be replaced with a chemically stable gas. To do this, it is necessary to prepare an explosion-proof air pump and replacement gas storage tank (such as carbon dioxide, nitrogen, etc.) required for the replacement gas, and transport the facilities and equipment to the tank top in advance.

[0135] Step 602: On-site arrangement.

[0136] a) After entering the tank area, the inspection team should be divided into two groups: one group is the tank outside group, which goes to the bottom of the tank outside and is responsible for the work related to the control of the oil detection robot 1; the other group is the tank top group, which goes to the top of the tank and is responsible for the work related to the retraction and deployment of the oil detection robot 1. b) The tank outside group needs to complete the following arrangements: install and debug the positioning signal receiver 2, the positioning signal wireless transceiver 3 and the positioning signal wireless transceiver host 4, connect the cables required by the host computer 5, cooperate with the crane to hang the hook on the two pull rings on the rear side of the oil detection robot 1 body 111, and debug the basic functions of the oil detection robot 1. c) The tank top group needs to complete the following arrangements: first use the pre-installed air pump to extract the accumulated gas in the floating roof and replace it with non-flammable gas such as nitrogen, and then fix the guide frame 8 and tripod 9 at the manhole 10 on the tank top. d) The tank outside group can set the host computer 5 in a non-explosion-proof area; while the tank top group works in the tank top explosion-proof zone 2, and needs to carry a full set of protective measures and explosion-proof communication equipment to maintain communication with the tank bottom group.

[0137] Step 603: The oil detection robot 1 enters the tank.

[0138] a) After the arrangement is completed, the tank outside group cuts off the power supply of the oil detection robot 1 and removes the composite cable 7. The crane pulls up the oil detection robot 1 and moves it to the tank top. b) The tank top group cooperates with the crane. When the oil detection robot 1 falls above the manhole 10 and on the side of the tripod 9, the pull ring on the rear side of the oil detection robot 1 is hung on the hook of the tripod 9, and then the link between the pull ring and the crane hook is untied. c) After the oil detection robot 1 is hoisted, the crane lifts the sample roll from the ground and transfers it to the tank top. d) The tank top group first connects the power supply and Ethernet cable required by the winch 6, then connects the composite cable 7 on the winch 6 to the oil detection robot 1, and puts the composite cable 7 into the guide frame 8, and finally starts the winch 6 to tighten the composite cable 7. e) When the oil detection robot 1 starts to enter the tank, the tank top group operates the tripod 9, and after the oil detection robot 1 is lowered into the copper sleeve of the manhole 10, the tripod 9 hook is released, and the weight of the oil detection robot 1 is borne by the composite cable 7 instead. f) The tank top group controls the winch 6 to release the cable, so that the oil detection robot 1 enters the area 1 between the floating roof and the medium of the storage tank. When the cable release length is greater than the height of area 1, the oil detection robot 1 begins to enter the storage medium. g) The tank bottom group turns on the main power supply of the oil detection robot 1, connects the oil detection robot 1 control software of the upper computer 5, and reads the pitch angle of the oil detection robot 1 output by the inertial navigation sensor. During the process of the winch 6 releasing the line, when the pitch angle starts to increase from -90°, it means that the pull cylinder in front of the oil detection robot 1 begins to touch the ground. At this time, the release speed is reduced until the pitch angle changes to close to 0°, the oil detection robot 1 is located at the bottom of the tank and is in a flat state, and the winch 6 stops releasing the line.

[0139] Step 604: Start detection.

[0140] a) Before the oil detection robot 1 starts working, the following work needs to be done to initialize the oil detection robot 1: configure the parameters of the sensor, detection module, navigation module, and control module and test them, turn on the power of the dredging wheel drive motor 201, the track drive motor, and the positioning sonar motor, raise the positioning sonar, and stimulate the two positioning sonars in sequence to determine the current direction of the oil detection robot 1. b) After completing the initialization, turn on the dredging wheel drive motor 201, turn on the automatic control function for detection, and the automatic control realizes the coordinated detection of multiple detection modules according to the overall control algorithm of this application. c) During the operation of the control module, the process of moving from one path point to another path point follows the path tracking algorithm of this application to complete the correction of the motion trajectory. d) During the detection process, all detection data are stored in the database of the host computer 5, and the stored data includes positioning point data, detection data, and defect data. e) The positioning point data includes: time, positioning point, the number of the positioning signal receiver 2 enabled when calculating the positioning point, and the characteristic parameters of the received signal (relative receiving time, peak voltage). f) The detection data includes: time, positioning point, original detection data, and detection data after data processing. Defect data includes: defect location and defect size.

[0141] Step 605: End the detection.

[0142] a) After the tank bottom group determines that the detection data meets the user's needs, perform the following work to end this bottom plate detection. b) Control the oil detection robot 1 to return to the coordinate point corresponding to the manhole 10, lower the positioning sonar, and determine that the two positioning sonars are completely lowered according to the output status of the switch status monitor, and then turn off the power of the oil detection robot 1. c) The crane pulls up the oil detection robot 1 and stops when the oil detection robot 1 is about to enter the manhole 10. d) The tank top group needs to confirm whether the composite cable 7 is concentric with the manhole 10. e) After the oil detection robot 1 pulls out the copper sleeve, the tank top group connects the hook of the tripod 9 to the pull ring at the tail of the oil detection robot 1. f) After pulling up the oil detection robot 1 through the tripod 9, perform a simple cleaning of the oil detection robot 1.

[0143] After the cleaning is completed, the crane pulls up the oil detection robot 1 and moves it to the bottom of the tank and lays it flat. This inspection is completed.

[0144] The present application realizes the detection of the oil bottom plate of a large floating roof crude oil storage tank without opening the tank. The error caused by the system response is eliminated by statistically analyzing the relative time when the positioning signal arrives at each positioning signal receiver 2, and the signal received by all positioning signal receivers 2 is solved by particle swarm optimization to reduce the influence of the single receiver error on the overall positioning accuracy. The positioning points at the current time and the previous time are statistically analyzed based on the linear regression method, and the position of the in-oil detection robot 1 with the smallest error is estimated, thereby further reducing the positioning error and making automatic control under non-precise positioning conditions possible.

[0145] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A tank bottom plate detection device based on an oil detection robot, characterized in that: The tank bottom plate detection device based on the in-oil detection robot comprises: an in-oil detection robot, a navigation module and a host computer; The navigation module is used to collect the positioning points of the in-oil detection robot at each moment, and to track the path of the in-oil detection robot using a path tracking algorithm based on linear regression according to the collected positioning points and the reference path; The oil detection robot comprises a magnetic flux leakage detection module and an ultrasonic and acoustic emission detection module connected to the host computer, and the host computer is used to perform defect detection on the tank bottom plate according to the magnetic field signal collected by the magnetic flux leakage detection module and the ultrasonic echo signal and acoustic reflection signal collected by the ultrasonic and acoustic emission detection module; In terms of using a linear regression-based path tracking algorithm to track the path of the oil detection robot according to the collected positioning points and reference paths, the navigation module is used to: Determine whether the oil detection robot turns or stops in place, if so, stop path tracking, if not, initialize the positioning point set to be empty; The distance between the positioning points collected at two adjacent moments is determined in sequence. If the distance between the positioning points collected at two adjacent moments is less than or equal to a first set value, the positioning point at the latter of the two adjacent moments is added to the positioning point set. When the number of positioning points in the positioning point set is greater than the set number and the radius of the positioning point set is greater than the second set value, a univariate linear regression is performed on the positioning point set to obtain a first straight line; the radius of the positioning point set is the maximum value of the distance between the average value of the positioning points in the positioning point set and any positioning point; Calculating the linear fitting correlation coefficient of the first straight line; If the linear fitting correlation coefficient is less than a threshold, path tracking is stopped; otherwise, the direction and position of the oil detection robot at the current moment are determined according to the first straight line; According to the reference path, the orientation and position of the in-oil detection robot at the current moment, a PID controlled path planning algorithm is adopted to control the in-oil detection robot.

2. The tank bottom plate detection device based on the oil detection robot according to claim 1 is characterized in that: Determining the orientation and position of the oil detection robot at a current moment according to the first straight line specifically includes: According to the formula Calculate the orientation of the oil detection robot at the current moment, where: is the direction at the current moment, is the slope of the first straight line; A straight line passing through the last positioning point in the positioning point set and perpendicular to the first straight line is taken as the second straight line, and an intersection of the first straight line and the second straight line is taken as the position of the oil detection robot at the current moment.

3. The tank bottom plate detection device based on the oil detection robot according to claim 1 is characterized in that: The tank bottom plate detection device based on the in-oil detection robot also includes a host computer, and the in-oil detection robot includes a fuselage, an obstacle avoidance sensor and a magnetic flux leakage lifting mechanism; the obstacle avoidance sensor and the magnetic flux leakage lifting mechanism are both connected to the host computer; The magnetic flux leakage detection module is installed in the first groove at the bottom of the fuselage through the magnetic flux leakage lifting mechanism, and the magnetic flux leakage lifting mechanism is used to adjust the distance between the magnetic flux leakage detection module and the bottom plate of the tank. The ultrasonic and acoustic emission detection module is located in the second groove at the bottom of the fuselage; the obstacle avoidance sensor is located in front of the fuselage, and the obstacle avoidance sensor is used to detect whether there is an obstacle in front of the oil detection robot and the distance to the obstacle.

4. The tank bottom plate detection device based on the oil detection robot according to claim 3 is characterized in that: The oil detection robot further includes a top cover, a first positioning sonar, a second positioning sonar, a sound speed calibration sonar, a cable connector, a dredging wheel and a crawler motion mechanism; the first positioning sonar, the second positioning sonar and the sound speed calibration sonar are all connected to the host computer; The dredging wheel is located in front of the fuselage and below the obstacle avoidance sensor, the crawler motion mechanism is located on both sides of the fuselage, the top cover is located above the fuselage, the first positioning sonar is located in the third groove of the top cover, the second positioning sonar is located in the fourth groove of the top cover, the third groove and the fourth groove are at different distances from the front end of the top cover, and the sound speed calibration sonar is located at the rear end of the top cover; the first positioning sonar is used to determine the first positioning coordinate, the second positioning sonar is used to determine the second positioning coordinate, the host computer is used to determine the steering angle of the oil detection robot according to the first positioning coordinate and the second positioning coordinate, and the sound speed calibration sonar is used to collect the sound speed of the medium in the storage tank at the current position; the cable connector is fixed at the rear end of the fuselage, and the cable connector is used to connect the magnetic flux leakage detection module, the ultrasonic and acoustic emission detection module, the obstacle avoidance sensor, the magnetic flux leakage lifting mechanism, the first positioning sonar, the second positioning sonar and the sound speed calibration sonar to the host computer through a composite cable.

5. The tank bottom plate detection device based on the oil detection robot according to claim 3 is characterized in that: The ultrasonic and acoustic emission detection module includes an ultrasonic detection unit and an acoustic emission detection unit; The magnetic flux leakage detection module includes a magnetic field probe array, the ultrasonic detection unit includes a plurality of ultrasonic probes, and the acoustic emission detection unit includes an acoustic emission probe; In terms of performing defect detection on the bottom plate of the storage tank according to the magnetic field signal collected by the magnetic flux leakage detection module, the ultrasonic echo signal and the acoustic reflection signal collected by the ultrasonic and acoustic emission detection module, the host computer is specifically used for: During the movement of the oil detection robot, the thickness of the bottom plate sediment is detected by the magnetic flux leakage detection module or the ultrasonic detection unit; When the thickness of the bottom plate sediment is less than the set thickness, the magnetic flux leakage detection module is activated to detect the defects of the tank bottom plate, and the dual-channel ultrasonic probe in the ultrasonic detection unit is started to detect the thickness of the bottom plate sediment. When a defect is detected, the defect position is marked, and the defect position mark is the coordinate corresponding to the defect; When the thickness of the bottom plate sediment is greater than or equal to the set thickness, the magnetic flux leakage lifting mechanism is controlled to lift the magnetic flux leakage detection module to a distance between the magnetic flux leakage detection module and the tank bottom plate that is a set distance threshold, the magnetic flux leakage detection module and the full-channel ultrasonic probe in the ultrasonic detection unit are enabled to perform tank bottom plate defect detection, and the defect position is marked when a defect is detected; When it is detected that there is an obstacle within a set distance ahead of the oil detection robot traveling along the reference path, the oil detection robot is controlled to stop moving, the acoustic emission detection unit is started to perform tank bottom plate defect detection, and the defect position is marked when a defect is detected; When the acoustic emission detection unit finishes the detection, an obstacle avoidance path is generated according to the size of the obstacle, the in-oil detection robot is controlled to move to the reference path according to the obstacle avoidance path, and returns to the step of detecting the thickness of the bottom plate sediment through the magnetic flux leakage detection module or the ultrasonic detection unit until the end of the reference path is reached; the obstacle avoidance path is composed of three obstacle avoidance branches connected in sequence, the angle between each connected obstacle avoidance branch is 90 degrees, and the distance between each obstacle avoidance branch and the obstacle is l ,in, L blk is the maximum side length of the obstacle, k tr is the coefficient, W mfl is the path width of the reference path.

6. The tank bottom plate detection device based on the oil detection robot according to claim 4 is characterized in that: The fuselage includes a dredging wheel drive motor and a track wheel drive motor, wherein the dredging wheel drive motor is used to drive the dredging wheel, and the track wheel drive motor is used to drive the track motion mechanism; The leakage magnetic lifting mechanism is located inside the fuselage, and includes a hydraulic pump, an oil pipe, a drawing shaft, a lifting beam and two hydraulic cylinders; the two hydraulic cylinders are connected to the hydraulic pump through the oil pipe, and one hydraulic cylinder is connected to each of the two ends of the lifting beam. The drawing shaft is located on the lifting beam, and the drawing shaft is connected to the leakage magnetic detection module. The lifting beam is used to transmit the lifting force of the hydraulic cylinder to the drawing shaft. The in-oil detection robot also includes four lifting guide shafts, which are respectively connected to the four corners of the magnetic flux leakage detection module, one end of each lifting guide shaft is connected to the fuselage, and the other end is connected to the magnetic flux leakage detection module.

7. The tank bottom plate detection device based on the oil detection robot according to claim 4 is characterized in that: There is a sealed cavity between the top cover and the fuselage, and the sealed cavity includes a motor driver, a sonar excitation circuit, a magnetic flux leakage detection circuit, an ultrasonic and acoustic emission detection circuit and a main control board; The motor driver is used to set the direction and speed of the motor. The motor driver is also used to collect the power and actual speed of the motor. The motor driver is connected to the main control board via a bus. The sonar excitation circuit is used to send pulse signals to the first positioning sonar, the second positioning sonar and the sound speed calibration sonar respectively, and the sonar excitation circuit is connected to the main control board through a serial port line; The magnetic flux leakage detection circuit is used to collect the voltage signal of each magnetic field probe in the magnetic field probe array, and convert the voltage signal of each magnetic field probe into a digital signal and upload it to the host computer through the Ethernet in the composite cable; The ultrasound and acoustic emission detection circuit is used to collect the voltage signal of each ultrasound probe and the voltage signal of the acoustic emission probe, and upload the voltage signal of each ultrasound probe and the voltage signal of the acoustic emission probe to the host computer through the Ethernet in the composite cable; The main control board is connected to the host computer.

8. The tank bottom plate detection device based on the oil detection robot according to claim 4 is characterized in that: The tank bottom plate detection device based on the oil detection robot comprises a first positioning sonar rod, a second positioning sonar rod, a first positioning sonar rod driving motor and a second positioning sonar rod driving motor, one end of the first positioning sonar rod is movably connected to the third groove, and the other end is provided with the first positioning sonar, one end of the second positioning sonar rod is movably connected to the fourth groove, and the other end is provided with the second positioning sonar, the first positioning sonar rod driving motor is used to adjust the posture of the first positioning sonar rod, and the second positioning sonar rod driving motor is used to adjust the posture of the second positioning sonar rod; When the first positioning sonar and the second positioning sonar are working, the first positioning sonar rod rotates from a horizontal posture to a vertical posture in the third groove, and the second positioning sonar rod rotates from a horizontal posture to a vertical posture in the fourth groove. The horizontal posture is a posture parallel to the third groove or the fourth groove, and the vertical posture is a posture perpendicular to the plane where the top cover is located.

9. The tank bottom plate detection device based on the oil detection robot according to claim 5 is characterized in that: The magnetic field probe array includes 1 row of 21 columns of magnetic field probes, each magnetic field probe is a Hall element, the detection width of the leakage magnetic detection module is the width of the 21 columns of magnetic field probes, and the detection width is the path width of the reference path.

Images