A method for positioning a submersible robot for servicing a multi-oil layer medium storage tank

Abstract

The application discloses a submersible operation robot positioning method for a multi-oil-layer medium oil storage tank, and belongs to the technical field of oil storage tank operation.The method comprises the following steps: step 1001, lowering a positioning system and a submersible operation robot; step 1002, establishing a connection between the positioning system and the submersible operation robot; step 1003, acquiring the position of a positioning receiving transducer; step 1004, acquiring the position of a beacon in a multi-layer oil medium; step 1005, predicting the position of the submersible operation robot in the oil storage tank; step 1006, performing Kalman filtering processing; and step 1007, outputting the position of the submersible operation robot.In the submersible operation robot positioning method for the multi-oil-layer medium oil storage tank, the influence of oil medium physical stratification on the sound signal propagation speed is considered, and the positioning error is greatly reduced.Meanwhile, a state space model of robot operation is established, and the positioning result is optimized by means of Kalman filtering, so that the positioning result is further accurate.

Description

Technical Field

[0001] A method for positioning a submersible robot in a multi-layer oil storage tank, belonging to the field of oil storage tank operation technology. Background Technology

[0002] Oil storage tanks are extremely common storage devices in the petroleum and petrochemical industry. Due to the complexity of the internal media composition, the bottom plate of oil storage tanks is highly susceptible to corrosion, leading to equipment failure. To address this issue, the bottom plate of oil storage tanks is typically inspected and its condition assessed at regular intervals. However, because the bottom plate of oil storage tanks is often covered by various types of oil, conventional inspection methods are extremely difficult to implement smoothly. This usually requires opening the tank and shutting down for maintenance, which is time-consuming, labor-intensive, and increases costs.

[0003] Currently, online detection methods are relatively mature. The development of online inspection robots for oil tank bottom plates, utilizing magnetic flux leakage, acoustic emission, and guided wave detection technologies combined with intelligent robots, is on the agenda. However, due to the unique environment—the robot operates in a closed liquid environment—it is difficult to observe the robot's position, let alone obtain the specific location of the detection probe. Furthermore, after long-term use, due to sedimentation and other reasons, the physical properties of the oil in the tank are not uniformly distributed, resulting in different propagation speeds of the detection signal at different depths, further increasing the difficulty of locating the submersible robot.

[0004] Chinese invention patent application number 202111314099.9, filed on November 8, 2021, entitled "A Positioning Method for a Robot for Inspecting the Bottom Plate of an Oil Storage Tank," discloses a technical solution. In this solution, an inertial sensor is installed at the center of a robot operating at the bottom of the tank to obtain the robot's direction of travel. Simultaneously, two sets of four ultrasonic sensors are installed at the front and rear ends of the robot, respectively. All four sensors are located on the robot's central axis, with the two sensors at the very front and rear facing outwards, and the two in the middle positioned opposite each other. During operation, the two middle sensors, whose distances are known, are used to measure the velocity of sound waves, while the two outer sensors are used for echo ranging, measuring the straight-line distances from the front and rear ends of the robot to the tank wall. By combining the sum of these distances with the robot's direction of travel, the robot's position coordinates are calculated.

[0005] However, this technical solution requires the installation of four ultrasonic sensors along the robot's central axis. Due to the unique nature of its positioning method, the positions of these four sensors must be precisely determined; otherwise, even slight deviations in installation angle or position can significantly impact the positioning results. Furthermore, the positioning error increases with the diameter of the oil tank, making this method suitable only for small oil tanks.

[0006] The literature (Xu Yaosong. Research on Sound Source Localization Method in Enclosed Liquid Fields [D]. Tianjin University, 2012.) discloses a localization method for obtaining robot position coordinates using beacons and receiving arrays. This method involves arranging a receiving and positioning transducer array outside the tank to receive the positioning signal emitted by the robot working inside the tank, measuring the propagation time of the sound signal, and then using the different times it takes for the sound source signal to reach each positioning transducer to obtain the time difference between the robot and each positioning transducer. The time difference between the sound source signal reaching two positioning transducers can determine a pair of hyperbolas. Therefore, using three positioning transducers, two pairs of hyperbolas can be formed, and their intersection is the sound source position. For robot localization in three-dimensional space, four receiving and positioning transducers are needed to form three pairs of hyperbolas, and their intersection is the robot position.

[0007] However, in this technical solution, the positioning receiver transducer is installed outside the oil storage tank. This method causes the acoustic signal to pass through two different media—the tank wall and the oil inside—during its propagation. Due to the different propagation characteristics of acoustic signals in different media and the complex effects of the acoustic signal passing through heterogeneous interfaces, the positioning difficulty increases and the positioning accuracy is affected. Furthermore, using the signal arrival time difference method for robot positioning results in multi-valued phenomena, i.e., ambiguity, and may lead to unsolvable situations under conditions of high noise and error.

[0008] Existing technologies, including the above-mentioned technical solutions, are insufficient to meet the working environment inside oil storage tanks. Therefore, designing a submersible robot for oil storage operations, especially considering the speed of signal propagation at different depths inside the oil storage tank, to achieve precise positioning of the submersible robot, has become an urgent problem to be solved in this field. Summary of the Invention

[0009] The technical problem this invention aims to solve is to overcome the shortcomings of existing technologies and provide a method for locating submersible robots in multi-layer oil storage tanks that takes into account the influence of oil medium stratification on the propagation speed of acoustic signals, thereby significantly reducing positioning errors. Simultaneously, a state-space model of robot operation is established, and Kalman filtering is used to optimize the positioning results. This method plays a crucial role in addressing the impact of in-tank and environmental noise on acoustic positioning, resulting in more accurate positioning results.

[0010] The technical solution adopted by this invention to solve its technical problem is: a method for positioning a submersible robot in a multi-layer oil storage tank, characterized by the following steps:

[0011] Step 1001: Lower the positioning system and the submersible robot into the oil storage tank from two different manholes.

[0012] Step 1002: Establish a rectangular coordinate system at the center of the bottom of the oil storage tank and establish communication between the beacon and multiple positioning receiver transducers in the positioning system.

[0013] Step 1003: Obtain the location of the positioning receiver transducer;

[0014] Step 1004: Based on the different oil media in the oil storage tank, obtain the calculated value of the distance R between the surface beacon position of the submersible robot and the positioning receiver transducer. ;

[0015] Step 1005: While executing steps 1002 to 1004 or after executing steps 1002 to 1004, establish a state space model of the submersible robot and use the state space model to predict the predicted position of the submersible robot in the oil storage tank.

[0016] Step 1006: Optimize the positioning results using Kalman filtering based on the predicted position of the submersible robot obtained in step 1005 and the calculated position of the submersible robot obtained in step 1004.

[0017] Step 1007: Output the position of the submersible robot.

[0018] Preferably, step 1004 includes the following steps:

[0019] Step 1004-1: Based on the propagation speed of sound waves in the multi-layered oil medium at different depths inside the oil storage tank, obtain the theoretical formula for the distance R between the beacon position and the positioning receiver transducer.

[0020] Step 1004-2: The theoretical formula for the distance R between the beacon position and the positioning receiver transducer is expressed in the form of the proportional coefficient of each oil layer in the oil storage tank, and the calculation formula in the form of the proportional coefficient is obtained.

[0021] Step 1004-3: Use a neural network matrix to obtain the proportional coefficient matrix of each oil layer in the oil storage tank in the form of proportional coefficient calculation formula;

[0022] Step 1004-4 yields an improved formula for the distance between the beacon and the positioning receiver transducer on the submersible robot at time k.

[0023] Step 1004-5: Finally, define the loss function of the submersible robot under multi-layer crude oil and use the gradient descent method to obtain the values ​​of the improved formula parameters.

[0024] Step 1004-6: Using the values ​​of the improved formula parameters obtained in step 1004-5, calculate the distance R between the surface beacon position of the submersible robot and the positioning receiver transducer. .

[0025] Preferably, the positioning system includes a lowering mechanism and multiple positioning receiving transducers fixed on the lowering mechanism.

[0026] Preferably, the lowering mechanism includes a support plate, on the surface of which multiple through holes are evenly distributed along its axial direction, and a lifting rod is installed in each through hole in a way that allows it to be raised and lowered; fixed posts corresponding to the lifting rods are vertically arranged on the outer ring of the bottom of the support plate, and a hinge plate is fixed below each fixed post; a connecting rod is hinged to the bottom of each lifting rod, and a support column is hinged to the bottom of each connecting rod, and the support column is also hinged to the lower end of the hinge plate; multiple fixing holes for fixing and positioning the receiving transducer are evenly distributed along the length of each support column.

[0027] Preferably, in step 1003, the spatial coordinates of the location receiving transducer are:

[0028] ,

[0029] Where: α is the angle between the lowering mechanism and the positive x-axis of the rectangular coordinate system;

[0030] x' and y' represent the lateral and longitudinal positional offsets of the positioning receiver transducer at its lowered position, respectively. , ;

[0031] L is the distance between the fixed position of the positioning receiver transducer and the top of the support column, and β is the rotation angle of the support column after deployment. ,

[0032] Where c is the distance between the second hinge point and the third hinge point, b is the distance between the first hinge point and the second hinge point, n is the distance the lifting rod descends, the hinge point between the connecting rod and the lifting rod is the first hinge point, the hinge point between the connecting rod and the support column is the second hinge point, and the hinge point between the hinge plate and the support column is the third hinge point.

[0033] Preferably, in step 1002, the beacon's position is determined using the principle of multi-spherical intersection. The distance between the beacon and the positioning receiving transducer is calculated using sound speed and time, by means of the beacon emitting an acoustic signal and the positioning receiving transducer receiving the acoustic signal. ,

[0034] Where v is the propagation speed of the ultrasonic signal in the oil medium, and T i Let be the time from the emission of the ultrasonic signal to its reception by the i-th positioning transducer, and let the beacon coordinates be (X, Y, Z). Then, the basic mathematical model for acoustic positioning based on time delay measurement is:

[0035]

[0036] (x) i y i , z i () represents the spatial coordinates of the receiving transducer.

[0037] Preferably, in step 1005, the state-space model of the submersible robot is:

[0038]

[0039] Let the input u equal the robot's acceleration: In the formula, X, Y, and V represent the horizontal and vertical coordinates and velocity parameters of the submersible robot at the corresponding time, respectively. k To establish error compensation for the model, it can be expressed in discrete form as follows:

[0040]

[0041] The robot state at time k+1 can be derived from the state at time k, and the position of the submersible robot can be further obtained:

[0042]

[0043] In the formula, p k This represents measurement noise, which follows a Gaussian distribution.

[0044] Compared with the prior art, the beneficial effects of this invention are:

[0045] In the submersible robot positioning method for multi-layer oil storage tanks, the influence of oil stratification on the propagation speed of acoustic signals is considered, which greatly reduces the positioning error. Simultaneously, a state-space model of the robot's operation is established, and Kalman filtering is used to optimize the positioning results. This plays a crucial role in addressing the impact of tank-side and environmental noise on acoustic positioning, further improving the accuracy of the positioning results.

[0046] In the positioning method of the submersible robot for multi-layer oil storage tank, based on the stratification phenomenon of oil sludge in the tank, a neural network is designed to obtain the influence coefficient of each layer of oil medium, so as to accurately determine the distance between the beacon and the transducer after the sound signal propagates in the multi-layer medium, and finally obtain the accurate positioning result. Attached Figure Description

[0047] Figure 1 Flowchart of a method for locating a submersible robot in a multi-layer oil storage tank.

[0048] Figure 2 Front view of the mechanism for positioning and lowering the receiving transducer.

[0049] Figure 3 for Figure 2 Top view.

[0050] Figure 4 Schematic diagram of the positioning and receiving transducer lowering mechanism under different working conditions.

[0051] Figure 5 A schematic diagram of the geometric relationship of the positioning receiving transducer lowering mechanism under different working conditions.

[0052] Figure 6 A top-down view of the positioning receiving transducer.

[0053] Figure 7 This is a schematic diagram of the sound velocity distribution inside the oil tank.

[0054] Among them: 1. Lifting rod; 2. Support plate; 3. Connecting rod; 4. Hinge plate; 5. Fixed column; 6. Support column; 7. Fixed hole; 8. First hinge point; 9. Second hinge point; 10. Third hinge point. Detailed Implementation

[0055] Figures 1-7 This is the preferred embodiment of the present invention, which is described below in conjunction with the accompanying drawings. Figures 1-7 The present invention will be further described below.

[0056] like Figure 1 As shown, a method for locating a submersible robot in a multi-layer oil storage tank includes the following steps:

[0057] Step 1001: Deploy the positioning system and the submersible robot;

[0058] The positioning system and the submersible robot were lowered into the oil storage tank from two different manholes.

[0059] The positioning system includes a lowering mechanism and multiple positioning receiver transducers fixed on the lowering mechanism. For example... Figures 2-3 As shown, the lowering mechanism includes a circular support plate 2. Multiple through holes are evenly distributed along the axial direction on the surface of the support plate 2. A lifting rod 1 is installed in each through hole and can be raised and lowered. Fixed posts 5, corresponding one-to-one with the lifting rods 1, are vertically arranged on the outer ring of the bottom of the support plate 2. A hinge plate 4 is fixed below each fixed post 5.

[0060] A connecting rod 3 is hinged to the bottom of each lifting rod 1, and a support rod 6 is hinged to the bottom of each connecting rod 3. The support rod 6 is also hinged to the lower end of the hinge plate 4. For ease of description, the hinge point between the connecting rod 3 and the lifting rod 1 is defined as the first hinge point 8, the hinge point between the connecting rod 3 and the support rod 6 is defined as the second hinge point 9, and the hinge point between the hinge plate 4 and the support rod 6 is defined as the third hinge point 10.

[0061] On the surface of each support column 6, multiple fixing holes 7 are evenly distributed along its length. Positioning receiving transducers are fixed within these fixing holes 7 on the surface of the support column 6. There are four support columns 6, therefore, four positioning receiving transducers are fixed within each lowering mechanism. When the lifting rod 1 and the support plate 2 move vertically relative to each other, the corresponding support column 6 expands or contracts under the action of the connecting rod 3. Since the four positioning receiving transducers are all installed in their respective fixing holes 7 on their respective support columns 6, when the four support columns 6 expand at equal angles, the four positioning receiving transducers are at the same horizontal height.

[0062] Step 1002: Establish the connection between the positioning system and the submersible robot;

[0063] After the positioning system is lowered into the oil storage tank, all the lifting rods 1 descend a certain distance (n). At this time, the four support columns 6 expand at the same angle, and the positioning receiving transducers installed on the four support columns 6 communicate with the beacons installed on the submersible robot.

[0064] Step 1003: Obtain the location of the positioning receiver transducer;

[0065] After establishing a spatial rectangular coordinate system around the bottom of the oil storage tank, the position of each manhole can be determined. After determining the descent distance of the lifting rod 1 in the lowering mechanism, the positions of the four positioning receiving transducers can be calculated. For example... Figure 4 As shown, taking a positioning receiver transducer as an example, by moving the lifting rod 1 downward by a distance n, the positioning receiver transducer at the lowered position of the foundation will have a longitudinal positional offset y' and a lateral positional offset x'. At this time, the position of the positioning receiver transducer is ( Figure 4 The distance L between the fixed hole 7 and the top of the support column 6 is known, so L' can be approximated as follows:

[0066]

[0067] At the same time:

[0068]

[0069]

[0070] Further integration Figure 5 Based on geometric relationships, it can be finally determined that:

[0071]

[0072] Where c is the distance between the second hinge point 9 and the third hinge point 10, b is the distance between the first hinge point 8 and the second hinge point 9, and n is the distance the lifting rod 1 descends.

[0073] Further integration Figure 6 Establish a spatial rectangular coordinate system with the center of the top of the oil storage tank as the origin, and assume that the manhole where the positioning receiving transducer lowering device is located is ( Figure 6 Point A in the middle has coordinates (x, y, 0). The lowering mechanism is manually lowered to a position h meters below the manhole (i.e., the plane where the positioning receiver transducer is located is z = -h). Figure 6 Taking the positioning receiver transducer at point B as an example, if the telescopic rod is adjusted to lower the distance by n, then the spatial coordinates of the positioning receiver transducer are: , Equivalent to the above (x) i y i , z i ), where α is the angle between the lowering mechanism and the positive x-axis, which can be determined during the lowering process.

[0074] Step 1003: Obtaining the beacon position in the multi-layered oil medium;

[0075] Ideally, the positioning system of a submersible oilfield robot can employ the principle of four-spherical intersection positioning. After placing four positioning receiver transducers inside the oil storage tank, a beacon is installed on the robot's body. The beacon emits an acoustic signal, and its position within the tank is considered the robot's position. The positioning receiver transducers receive the acoustic signal, and the distance between the beacon and the transducer is obtained, thus determining the four radii R. i For each of the four spheres (i=1, 2, 3, 4), a unique point is determined by these four spheres in space; this point is the beacon's location. Where:

[0076]

[0077] Where v is the propagation speed of the ultrasonic signal in the oil medium, and T i (i=1,2,3,4) represents the time from when the ultrasonic signal is emitted to when it is received by the i-th positioning receiver transducer.

[0078] Establish a spatial rectangular coordinate system at the center of the bottom of the oil storage tank. After lowering the positioning receiving transducer, determine the spatial coordinates (x, y, z) of the positioning receiving transducer. i y i , z i Given that i = 1, 2, 3, 4, and the beacon coordinates are (X, Y, Z), the basic mathematical model for acoustic localization based on time delay measurement is:

[0079]

[0080] The spatial location of the beacon can be obtained by solving four formulas simultaneously using four positioning receiver transducers.

[0081] However, in actual oil storage tanks, due to the complexity of the medium inside large oil tanks, the physical properties of the oil are not uniformly distributed vertically. The propagation speed of sound signals varies in oil media with different properties, resulting in a stratified distribution of sound signals. Figure 7 This diagram illustrates the sound velocity distribution inside an oil storage tank. Taking a fixed positioning receiver transducer at 6 locations along a certain column as an example, the expression for the distance R between the beacon position and the positioning receiver transducer is obtained:

[0082]

[0083] Among them, h1, ... h n The thicknesses of the oil layers at the bottom of the storage tank are, in order, v1, ... vn. n The values ​​represent the propagation speed of the signal in the first, ..., nth layers of oil at the bottom of the oil storage tank, respectively, and T is the time from when the ultrasonic signal is emitted to when it is received by the positioning and receiving transducer.

[0084] The above formula (7) is the theoretical formula for the distance R between the beacon location and the positioning receiver transducer. However, due to the varying thicknesses of each layer of oil in actual oil storage tanks (h1, ... h...), the actual formula is not applicable. n Since the actual distance R between the beacon position and the positioning receiver transducer is difficult to determine, the following method is used in the positioning method for the submersible robot in this multi-layer oil storage tank:

[0085] First, rewrite the above formula (7) as follows:

[0086]

[0087] in, Each element represents a proportion coefficient of each oil layer within the oil storage tank.

[0088] Then, the proportional coefficient matrix of each oil layer in the oil storage tank in formula (8) is obtained by using a neural network matrix:

[0089]

[0090] in, The weight parameters are n×n matrix; This is the bias parameter. The elements in the formula are obtained by the following formula:

[0091]

[0092] in, For input; The mean of the radial basis functions; Let V be the variance of the radial basis functions.

[0093] An improved formula for the distance between the beacon and the positioning receiver transducer on the submersible robot at time k is obtained:

[0094]

[0095] Finally, the loss function of the submersible robot under multi-layer crude oil is defined, and the values ​​of each parameter are obtained using the gradient descent method:

[0096] Weight parameters The loss function metric of this system, obtained by training on the dataset, is as follows:

[0097]

[0098] in, Let k be the actual distance between the robot and the positioning receiver transducer at time k. The calculated value is the actual distance between the robot and the positioning receiver transducer at time k.

[0099] The gradients of each parameter are obtained as follows:

[0100]

[0101]

[0102]

[0103]

[0104] The final parameter matrices can be obtained through this gradient.

[0105] enter The propagation speed v of the acoustic signal in each oil layer i The propagation speed of the sound signal at a given oil layer can be obtained by using a positioning receiving transducer: the positioning receiving transducer is lowered to different oil layers using a positioning receiving transducer lowering device, one positioning receiving transducer emits an acoustic signal, and the other positioning receiving transducers receive the signal. The distance between the positioning receiving transducers is known, and the propagation speed of the acoustic signal at that oil layer can be obtained based on the propagation time of the signal between each positioning receiving transducer.

[0106] Substituting the parameters into formula (8), the calculated distance between the beacon and the positioning receiver transducer of the submersible robot at time k can be obtained. .

[0107] While performing steps 1002 to 1004 above, or after performing steps 1002 to 1004 above, proceed to step 1005:

[0108] Step 1005: Predict the position of the submersible robot inside the oil storage tank;

[0109] Establish a state-space model of the submersible robot and use the state-space model to predict the robot's position inside the oil storage tank.

[0110] The state-space model of the submersible robot is as follows:

[0111]

[0112] Let the input u equal the robot's acceleration: In the formula, X, Y, and V represent the horizontal and vertical coordinates and velocity parameters of the submersible robot at the corresponding time, respectively. k To establish error compensation for the model, it can be expressed in discrete form as follows:

[0113]

[0114] The robot's state at time k+1 can be derived from its state at time k. The position of the submersible robot can then be obtained:

[0115]

[0116] In the formula p k This represents measurement noise, which follows a Gaussian distribution.

[0117] Through the above steps, the robot's model-predicted position and ultrasonic positioning position were obtained. The robot position obtained through the interaction between the beacon and the transducer is a three-dimensional spatial coordinate. Since the submersible robot always operates at the bottom of the tank, its longitudinal coordinate Z is a known constant, so only its horizontal coordinate can be used as the positioning result.

[0118] Step 1006: Perform Karl von filtration.

[0119] The predicted position of the submersible working robot obtained in step 1005 and the calculated position of the submersible working robot obtained in step 1004 are further optimized by using Kalman filtering. The model prediction result is used as the state parameter, and the positioning result obtained by the positioning system is used as the observation parameter to optimize the positioning result.

[0120] The basic formula for Kalman filtering is as follows:

[0121] Prior estimation:

[0122]

[0123] Prior error covariance:

[0124]

[0125] Kalman gain:

[0126]

[0127] Posterior estimation:

[0128]

[0129] Update error covariance:

[0130]

[0131] In the above basic Kalman filter formula: x k Z represents the state parameter at time k, the predicted position of the submersible robot obtained from the state-space model; k This represents the observed parameter at time k, i.e., the calculated position of the submersible robot; u k-1 A represents the control variable; B represents the state matrix; P represents the control matrix; k Let represent the error covariance at time k; Q represent the process noise covariance matrix; R represent the measurement noise covariance matrix; and H represent the measurement coefficient matrix.

[0132] Step 1007: Output the position of the submersible robot inside the oil storage tank.

[0133] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for positioning a submersible robot in a multi-layer oil storage tank, characterized in that: Includes the following steps: Step 1001: Lower the positioning system and the submersible robot into the oil storage tank from two different manholes. Step 1002: Establish a rectangular coordinate system at the center of the bottom of the oil storage tank and establish communication between the beacon and multiple positioning receiver transducers in the positioning system. Step 1003: Obtain the location of the positioning receiver transducer; Step 1004: Based on the different oil media in the oil storage tank, obtain the calculated value of the distance R between the surface beacon position of the submersible robot and the positioning receiver transducer. ; Step 1005: While executing steps 1002 to 1004 or after executing steps 1002 to 1004, establish a state space model of the submersible robot and use the state space model to predict the predicted position of the submersible robot in the oil storage tank. Step 1006: Optimize the positioning results using Kalman filtering based on the predicted position of the submersible robot obtained in step 1005 and the calculated position of the submersible robot obtained in step 1004. Step 1007: Output the position of the submersible robot; The positioning system includes a lowering mechanism and multiple positioning receiver transducers fixed on the lowering mechanism; The lowering mechanism includes a support plate (2), on the surface of the support plate (2) are a plurality of through holes evenly opened along its axial direction, and a lifting rod (1) is installed in each through hole that can be raised and lowered; a fixed column (5) corresponding to the lifting rod (1) is vertically arranged on the outer ring of the bottom of the support plate (2), and a hinge plate (4) is fixed below each fixed column (5); a connecting rod (3) is hinged to the bottom of each lifting rod (1), and a support column (6) is hinged to the bottom of each connecting rod (3), and the support column (6) is also hinged to the lower end of the hinge plate (4); a plurality of fixing holes (7) for fixing and positioning the receiving transducer are evenly opened on the surface of each support column (6) along its length direction. In step 1003, the spatial coordinates of the location receiving transducer are: , Where: α is the angle between the lowering mechanism and the positive x-axis of the rectangular coordinate system, and h is the distance from the lowering mechanism to the manhole; x' and y' represent the lateral and longitudinal positional offsets of the positioning receiver transducer at its lowered position, respectively. , ; L is the distance between the fixed position of the positioning receiver transducer and the top of the support column (6), and β is the rotation angle of the support column (6) after deployment. , Where c is the distance between the second hinge point (9) and the third hinge point (10), b is the distance between the first hinge point (8) and the second hinge point (9), n is the distance the lifting rod (1) descends, the hinge point between the connecting rod (3) and the lifting rod (1) is the first hinge point (8), the hinge point between the connecting rod (3) and the support column (6) is the second hinge point (9), and the hinge point between the hinge plate (4) and the support column (6) is the third hinge point (10).

2. The method for positioning a submersible robot in a multi-layer oil storage tank according to claim 1, characterized in that: Step 1004 includes the following steps: Step 1004-1: Based on the propagation speed of sound waves in the multi-layered oil medium at different depths inside the oil storage tank, obtain the theoretical formula for the distance R between the beacon position and the positioning receiver transducer. Step 1004-2: The theoretical formula for the distance R between the beacon position and the positioning receiver transducer is expressed in the form of the proportional coefficient of each oil layer in the oil storage tank, and the calculation formula in the form of the proportional coefficient is obtained. Step 1004-3: Use a neural network matrix to obtain the proportional coefficient matrix of each oil layer in the oil storage tank in the form of proportional coefficient calculation formula; Step 1004-4 yields an improved formula for the distance between the beacon and the positioning receiver transducer on the submersible robot at time k. Step 1004-5: Finally, define the loss function of the submersible robot under multi-layer crude oil and use the gradient descent method to obtain the values ​​of the improved formula parameters. Step 1004-6: Using the values ​​of the improved formula parameters obtained in step 1004-5, calculate the distance R between the surface beacon position of the submersible robot and the positioning receiver transducer. .

3. The method for positioning a submersible robot in a multi-layer oil storage tank according to claim 1, characterized in that: In step 1002, the beacon's position is determined using the principle of multi-spherical intersection. The distance between the beacon and the positioning receiving transducer is calculated by using sound speed and time to obtain the sound signal emitted by the beacon and received by the positioning receiving transducer. , Where v is the propagation speed of the ultrasonic signal in the oil medium, and T i Let be the time from the emission of the ultrasonic signal to its reception by the i-th positioning transducer, and let the beacon coordinates be (X, Y, Z). Then, the basic mathematical model for acoustic positioning based on time delay measurement is: (x) i y i , z i () represents the spatial coordinates of the receiving transducer.

4. The method for positioning a submersible robot in a multi-layer oil storage tank according to claim 1, characterized in that: In step 1005, the state-space model of the submersible robot is as follows: Let the input u equal the robot's acceleration: In the formula, X, Y, and V represent the horizontal and vertical coordinates and velocity parameters of the submersible robot at the corresponding time, respectively. k To establish error compensation for the model, it can be expressed in discrete form as follows: The robot state at time k+1 can be derived from the state at time k, and the position of the submersible robot can be further obtained: In the formula, p k This represents measurement noise, which follows a Gaussian distribution.

Images