An offshore platform jacket cleaning robot and its control method

Abstract

The present invention discloses an offshore platform jacket cleaning robot, which comprises a left body and a right body. Both the left body and the right body include docking ends, and arc-shaped clamping parts are respectively arranged at the docking ends of the left body and the right body; One side of the docking ends of the left body and the right body is hinged, and the other side is connected through a locking connection mechanism; The locking connection mechanism is installed on the left body, and a snap blind hole cooperating with the locking connection mechanism is installed on the right body; A number of vertical thrusters, a number of circumferential thrusters and a number of spray guns are also arranged on the left body and the right body. The control method of the present invention proposes a basic kinematic and dynamic model, considers model uncertainty, obtains an optimized dynamic model, designs a disturbance observer to compensate for the lumped uncertainty of the model, processes external environmental disturbances and its own coupling, and improves the robustness of the control system; A fixed-time disturbance observer is designed to enable the cleaning robot to autonomously navigate to the jacket.

Description

Technical Field

[0001] The present invention relates to the field of automated robots, and particularly to a cleaning robot for a jacket of an offshore platform and a control method therefor. Background Art

[0002] With the progress of technology, the development of marine resources by people has reached a certain level. Various offshore platforms and cleaning robots have been continuously developed to explore the ocean rich in resources. An offshore platform is a structure that provides production and living facilities for activities such as offshore drilling, oil production, shipping, observation, navigation, and construction. Since its jacket is immersed in water for a long time, marine organisms such as oysters and barnacles rapidly reproduce and grow on it, which not only increases the weight of the drilling platform but also enlarges the area of the drilling platform affected by ocean currents, bringing potential safety hazards. Therefore, regularly cleaning the underwater jacket has become one of the necessary operations for offshore workers.

[0003] Currently, the main method for cleaning the jacket is to rely on operators to carry high-pressure water jet equipment and dive into the seabed to clean it. This requires high physical fitness of the operators, high cleaning costs, high labor intensity, low cleaning efficiency, and the complex marine environment will pose additional risks to the operators and even endanger their safety. Therefore, it is necessary to design an underwater jacket cleaning robot that can replace manual underwater cleaning and provide a control method to achieve automatic cleaning of the jacket by the robot, so as to reduce the cleaning cost and improve the cleaning efficiency. Summary of the Invention

[0004] Aiming at the above deficiencies of the prior art, the present invention provides a cleaning robot for a jacket of an offshore platform and a control method therefor, which can reduce the cleaning cost and improve the cleaning efficiency.

[0005] To achieve the above object of the invention, a cleaning robot for a jacket of an offshore platform of the present invention includes: a left body and a right body. Both the left body and the right body include docking ends, and arc-shaped clamping portions are respectively provided at the docking ends of the left body and the right body; one side of the docking ends of the left body and the right body is hinged, and the other side is connected by a locking connection mechanism; the locking connection mechanism is installed on the left body, and the right body is provided with a snap blind hole that cooperates with the locking connection mechanism; a plurality of vertical thrusters, a plurality of circumferential thrusters, and a plurality of spray guns are also provided on the left body and the right body.

[0006] Furthermore, the locking connection mechanism includes a buckle seat fixed on the left body and a buckle pin fixed on the buckle seat. One end of the buckle pin is fixed with a buckle pin cap, and a spring is arranged outside the buckle pin. One end of the spring is fixed with an electromagnet sleeve, which is arranged outside the buckle pin. A radial export slope is also arranged outside the electromagnet sleeve. A number of buckle protrusions are arranged radially on the buckle pin cap, and through holes matching the buckle protrusions are arranged on the buckle pin cap. An arc-shaped part matching the radial export slope is arranged on the lower side of the buckle protrusion. The buckle pin cap is fixed on the buckle pin through a buckle screw. A spring support part for installing the spring is also arranged on the buckle pin. A magnetic iron block matching the electromagnet sleeve is arranged on the inner side of the buckle pin cap. A card slot matching the buckle protrusion is arranged in the buckle blind hole.

[0007] Furthermore, two vertical thrusters are arranged on each of the left body and the right body, and vertical through holes for installing the vertical thrusters are arranged on the left body and the right body. Two circumferential thrusters are arranged on each of the left body and the right body, and the four circumferential thrusters are mutually at an angle of 45 degrees. Two spray guns are arranged on each of the left body and the right body, and spray gun brackets for installing the spray guns are arranged on both the left body and the right body.

[0008] Furthermore, a number of auxiliary guiding grooves are arranged on the arc-shaped clamping part, and auxiliary radial wheels are fixed in the auxiliary guiding grooves. The auxiliary radial wheels are fixed in the installation groove through rotating shafts, and the outer diameter of the auxiliary radial wheels protrudes from the auxiliary guiding grooves.

[0009] Furthermore, anti-collision frames are respectively arranged at the bottoms of the left body and the right body. The anti-collision frame includes two support beams fixed under the box body and an arc-shaped anti-collision beam fixed on the two support beams.

[0010] Furthermore, floating bodies are respectively arranged on the upper sides of the left body and the right body, and sealed cabins are respectively arranged on the lower sides of the left body and the right body.

[0011] The control method of the offshore platform jacket cleaning robot includes the following steps:

[0012] Step S1: Establish the basic kinematic model, basic dynamic model and uncertainty error model of the cleaning robot;

[0013] Step S2: Obtain the optimized dynamic model by using the uncertainty error model and the basic kinematic model, and obtain the motion position error η e , control law τ c and the conversion model between the lumped uncertainty d′; the lumped uncertainty d′ includes model uncertainty and external environmental disturbance;

[0014] Step S3: According to the system variable x h of the fixed-time disturbance observer and the estimated value of the system variable Design a fixed-time disturbance observer to compensate for the lumped uncertainty d′ using the fixed-time disturbance observer.

[0015] Step S4: Establish a basic sliding mode surface model, design a fixed-time sliding mode controller, and establish a control law model to enable the cleaning robot to autonomously move to the jacket.

[0016] Furthermore, in step S1, the basic kinematic model and basic dynamic model are as follows:

[0017]

[0018]

[0019] where: M is the mass inertia matrix, J(η) is the transformation matrix, C(v) is the Coriolis force and centripetal force matrix, D(v) is the fluid damping coefficient matrix, g(η) = [0, 0, -(W - W B ), 0] T is the restoring force and moment vector of the cleaning robot, W and W B are the gravity and buoyancy of the underwater cleaning robot respectively, the superscript T represents the transpose of the matrix; η = [x, y, z, ψ] T is the motion position vector of the cleaning robot in the horizontal plane under the earth coordinate system, where (x, y, z) is the position of the cleaning robot, and ψ represents the heading angle of the cleaning robot; v = [u, υ, w, r] T represents the velocity vector of the cleaning robot under the body coordinate system, where u, υ, w, and r represent the linear velocities of forward, translation, depth, and the angular velocity of yaw respectively; τ = [τ u , τ v , τ w , τ r is the control input, and the inputs τ u , τ v , τ w and τ r are provided by the actuator; τ d = [τ du , τ dv , τ dw , τ dr is the influence of wind, waves, and ocean currents in the external environment;

[0020] The uncertainty error model is:

[0021] By introducing the uncertain terms ΔM, ΔC(υ), ΔD(υ), the uncertainty error model is established as:

[0022] M = M′ + ΔM

[0023] C(v) = C′(v) + ΔC(v)

[0024] D(v) = D′(v) + ΔD(v)

[0025] In the formula: M′, C′(v), D′(v) represent the matrices of the nominal model, and ΔM, ΔC(v), ΔD(v) represent the uncertain dynamic matrices. Here, ΔM = 0.2M′, ΔC(v) = 0.2C′(v), ΔD(v) = 0.2D′(v);

[0026] In the formula:

[0027]

[0028] Among them: d 11 = -X u -X uu |u|d 22 = -Y v -Y vv |υ|, d 33 = -Z w -Z ww |w|, d 44 = -N r -N rr |r|; m is the mass of the cleaning robot, and are the added mass and added inertia of the cleaning robot respectively; X u , Y v , Z w , N r , X uu , Y vv , Z ww , N rr represent the viscous hydrodynamic parameters; I z is the moment of inertia of the cleaning robot about the Z-axis;

[0029] Furthermore, the optimized dynamic model is:

[0030]

[0031] In the formula: M η (η) = M′J -1 , D η (v, η) = D′(v)J -1 ,

[0032] The transformed model is:

[0033]

[0034] where τ c is the control law, and the control law is the transformed control input, is the lumped uncertainty,

[0035] Furthermore, the system variable x of the fixed-time disturbance observer h :

[0036]

[0037] The fixed-time disturbance observer is:

[0038]

[0039] The basic sliding mode surface model is:

[0040]

[0041]

[0042] where: λ > 0, Υ > 0 are self-defined parameters;

[0043]

[0044]

[0045] In the formula

[0046] where α represents the asymptotic law parameter of the ERL index, ε, γ0, and g are adjustable parameters; ERL is the specified reaching law.

[0047] The beneficial effects of the present invention are as follows:

[0048] 1. The current jacket cleaning method mainly relies on operators carrying high-pressure water jet equipment to clean it in the sea. The labor intensity is high, the physical fitness requirements for cleaning operators are relatively high, the flexibility in ocean operations is low, the cleaning efficiency is not high, and the complex ocean environment poses additional risks to operators and even endangers their safety. Compared with the existing jacket cleaning methods, the cleaning robot of the present invention can replace manual underwater jacket cleaning operations, reduce cleaning costs, improve cleaning efficiency, and reduce safety risks generated by underwater operations.

[0049] 2. The cleaning robot of the present invention can automatically perform underwater cleaning operations on the jacket of an offshore platform, avoid the problems faced by operators during underwater cleaning operations, reduce cleaning costs, and improve cleaning efficiency. At the same time, it also promotes the automation and intelligence of offshore development equipment.

[0050] 3. The cleaning robot of the present invention uses a locking connection mechanism at the hinge end for arm positioning, and uses multiple circumferential thrusters to stabilize the robot body for rotation in a plane. Four auxiliary radial wheels are installed on the contact surface of the body to reduce friction, enabling the robot to quickly rotate and drive the spray gun to complete efficient cleaning.

[0051] 4. The control method of the present invention takes into account external environmental disturbances, such as factors like wind, waves, and ocean currents, as well as the coupling between its own rigid bodies, enabling the cleaning robot to achieve a more stable movement to the jacket. For the uncertain factors in the model during the movement of the cleaning robot and external environmental disturbances, such as the influence of wind, waves, and ocean currents, etc.; currently, neural network algorithms are widely used, but the convergence speed of neural network algorithms is slow and a large amount of real-time calculation is required, increasing the computational amount and complexity of the system. Therefore, the disturbance observer strategy of the present invention can achieve compensation for model uncertain factors and external environmental disturbances, improve the robustness of the control system, and reduce the computational amount and complexity of the system. And a fixed-time stability strategy is adopted to enable all signals of the closed-loop control system to converge within a fixed time.

[0052] 5. The method of the present invention proposes a basic kinematic model and a basic dynamic model of the underwater cleaning robot, and then further considers model uncertainty. Through mathematical transformation, a new optimized dynamic model is obtained. A disturbance observer is designed to compensate for model uncertainty and external environmental disturbances, and to handle external environmental disturbances and its own coupling, improving the robustness of the control system. A basic sliding mode surface model is established, and a fixed-time sliding mode controller is further designed to enable the cleaning robot to autonomously navigate to the jacket. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a schematic diagram of the overall shape of the present invention;

[0054] Figure 2 is a schematic diagram of the overall shape of the present invention in another direction;

[0055] Figure 3 is Figure 1 the partial enlarged view A - A in

[0056] Figure 4 is a schematic diagram of the shape of the locking connection mechanism;

[0057] Figure 5 is the working principle of the locking connection mechanism Figure 1 ;

[0058] Figure 6 is the working principle of the locking connection mechanism Figure 2 ;

[0059] Figure 7 It is a diagram for the establishment of the geodetic coordinate system and the body coordinate system;

[0060] Figure 8 It is a schematic diagram of the closed-loop control system;

[0061] Figure 9 It is a desired tracking diagram;

[0062] Figure 10 It is an error diagram in the x direction of the desired position;

[0063] Figure 11 It is an error diagram in the y direction of the desired position.

[0064] The description of the main component symbols in the figure is as follows:

[0065] 1. Left body; 2. Right body; 3. Floating body; 4. Sealed cabin; 5. Anti-collision frame;

[0066] 6. Locking connection mechanism; 61. Buckle seat; 62. Buckle pin; 63. Spring; 64. Electromagnet sleeve; 65. Buckle convex block; 66. Buckle pin cap; 67. Buckle screw; 68. Spring support part; 69. Magnetic iron block;

[0067] 7. Auxiliary radial wheel; 8. Vertical thruster; 9. Circumferential thruster; 10. Hinge bolt; 11. Hinge nut; 12. Spray gun; 13. Spray gun bracket. Specific embodiments

[0068] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions created using the concept of the present invention are within the scope of protection.

[0069] As Figure 1-2 shown, an offshore platform jacket cleaning robot includes: a left body 1 and a right body 2. Both the left body 1 and the right body 2 include docking ends, and arc-shaped clamping parts are respectively arranged at the docking ends of the left body 1 and the right body 2; one side of the docking ends of the left body 1 and the right body 2 is hinged, and the other side is connected by a locking connection mechanism 6; the locking connection mechanism 6 is installed on the left body 1, and the right body 2 is provided with a buckle blind hole that cooperates with the locking connection mechanism 6; a plurality of vertical thrusters 8, a plurality of circumferential thrusters 9 and a plurality of spray guns 12 are also arranged on the left body 1 and the right body 2.

[0070] As Figure 3As shown, several auxiliary guiding grooves are provided on the arc-shaped clamping part. An auxiliary radial wheel 7 is fixed in the auxiliary guiding groove. The auxiliary radial wheel 7 is fixed in the installation groove through a rotating shaft. The outer diameter of the auxiliary radial wheel 7 protrudes from the auxiliary guiding groove. Preferably, there are four auxiliary radial wheels 7. The four auxiliary radial wheels 7 enable the machine body to perform fixed-axis rotation cleaning operations, reduce the lateral friction between the machine body and the column, and play an auxiliary rotation role. In addition, a rubber layer can be provided outside the auxiliary radial wheel 7 to increase the vertical friction between the auxiliary radial wheel 7 and the jacket column. The cleaning robot of the present invention uses a locking connection mechanism at the hinge end for arm positioning, uses multiple circumferential thrusters to stabilize the robot body in a plane rotation, and installs four auxiliary radial wheels on the contact surface of the machine body to reduce friction, so that the robot can quickly rotate to drive the spray gun to complete efficient cleaning.

[0071] As Figure 4 , 5 and 6 show, the locking connection mechanism 6 includes a snap seat 61 fixed on the left machine body 1 and a snap pin 62 fixed on the snap seat 61. One end of the snap pin 62 is fixed with a snap pin cap 66, and a spring 63 is arranged outside the snap pin 62; one end of the spring 63 is fixed with an electromagnet sleeve 64. The electromagnet sleeve 64 is arranged outside the snap pin 62, and a radial guiding slope is also arranged outside the electromagnet sleeve 64; several snap protrusions 65 are arranged radially on the snap pin cap 66. A through hole cooperating with the snap protrusion 65 is arranged on the snap pin cap 66, and an arc-shaped part cooperating with the radial guiding slope is arranged on the lower side of the snap protrusion 65; the snap pin cap 66 is fixed on the snap pin 62 through a snap screw 67; a spring support part 68 for installing the spring 63 is also arranged on the snap pin 62; a magnetic attraction iron block 69 cooperating with the electromagnet sleeve 66 is arranged inside the snap pin cap 66; a clamping groove cooperating with the snap protrusion 65 is arranged in the snap blind hole.

[0072] In this embodiment, two vertical thrusters 8 are respectively arranged on the left body 1 and the right body 2, and vertical through holes for installing the vertical thrusters 8 are arranged on the left body 1 and the right body 2; two circumferential thrusters 9 are respectively arranged on the left body 1 and the right body 2, and the four circumferential thrusters 9 are mutually at an angle of 45 degrees; two spray guns 12 are respectively arranged on the left body 1 and the right body 2, and spray gun brackets 13 for installing the spray guns 12 are arranged on both the left body 1 and the right body 2. In order to improve the cleaning efficiency, the four spray guns 12 on the left body and the right body are fixed to the body through the spray gun brackets 13. The nozzles of the four spray guns 12 face inwards, and the intersection point of the spraying strokes of the four spray guns 12 coincides with the center of the clamping arm hole formed by the left body and the right body. The four spray guns 12 are symmetrically distributed in a circle at an angle of 45 degrees, and all four spray guns 12 are installed and fixed on a plane, so that when the four spray guns 12 pump and spray water simultaneously, most of the impact forces are offset from each other. The vertical thrusters 8 and the circumferential thrusters 9 are preferably T200 thrusters, and all fasteners of the thrusters in contact with the medium are made of stainless steel. The propeller blades are made of elastic aluminum alloy, can withstand varying loads, the thrust is evenly distributed to the mixture, and have an optimal hydraulic design; the unique cable sealing design eliminates the risk of cable water leakage; a leakage sensor and an over-temperature protection alarm device are provided inside. The maximum working power is 453 watts. A total of eight thrusters are installed on the robot, and four of the thrusters are horizontally installed on a plane and are diagonally installed at an angle of 45 degrees with the horizontal reference plane.

[0073] In this embodiment, anti-collision frames 5 are respectively arranged at the bottoms of the left body 1 and the right body 2; the anti-collision frame 5 includes two support beams fixed to the lower part of the box body and an arc-shaped anti-collision beam fixed to the two support beams.

[0074] In this embodiment, floating bodies 3 are respectively arranged on the upper sides of the left body 1 and the right body 2, and sealed cabins 4 are respectively arranged on the lower sides of the left body 1 and the right body 2. Two flat oval-shaped floating bodies 3 are selected and assembled above the body. The flat oval shape can better reduce the resistance of the body's fixed-axis rotation, buffer the impact force brought by the spray gun, and reduce the vibration brought during the operation of the cleaning system. The floating body 3 has a three-layer structure, which are the internal composite foam, the middle protective resin layer, and the external paint coating respectively.

[0075] The control system of this cleaning robot selects STM32F103ZET6 as the main control chip of the underwater cleaning robot. This development board is equipped with 10M / 100M Ethernet, which can quickly receive and send control instructions for the cleaning robot; the working voltage is about 5V, and the selected power supply is completely sufficient; the chip also loads 16MByte SPI FLASH inside, which is completely enough to input and run the program for operating the robot's movement. And a water depth sensor, a positioning module, an underwater thruster drive module, and an infrared tracking sensor are configured for the main chip.

[0076] To enable the underwater cleaning robot of the offshore platform jacket to autonomously move to the jacket position and then start the cleaning work, a fixed-time sliding mode control method is proposed.

[0077] Step S1. Establish the basic kinematic model, basic dynamic model, and uncertainty error model of the cleaning robot;

[0078] The basic kinematic model and basic dynamic model are:

[0079]

[0080]

[0081] In the formula: M is the mass inertia matrix, J(η) is the transformation matrix, C(v) is the Coriolis force and centripetal force matrix, D(v) is the fluid damping coefficient matrix, g(η) = [0, 0, -(W - W B ), 0] T is the restoring force and moment vector of the cleaning robot, W and W B are the gravity and buoyancy of the underwater cleaning robot respectively, and the superscript T represents the transpose of the matrix; in the cleaning robot, the body coordinate system and the earth coordinate system are usually used to describe the movement of the cleaning robot, as Figure 7 shown; η = [x, y, z, ψ] T is the movement position vector of the cleaning robot in the horizontal plane under the earth coordinate system, where (x, y, z) is the position of the cleaning robot, and ψ represents the heading angle of the cleaning robot; v = [u, υ, w, r] T represents the velocity vector of the cleaning robot under the body coordinate system, where u, υ, w, and r represent the linear velocities of forward, translation, sinking, and the angular velocity of turning respectively; τ = [τ u , τ v , τ w , τ r is the control input, and the inputs τ u , τ v , τ w and τ r for forward, translation, sinking, and turning are provided by the actuator; τ d = [τ du , τ dv , τ dw , τ dr is the influence of wind, waves, and ocean currents in the external environment;

[0082] Establish the uncertainty error model:

[0083] Due to the difficulty in obtaining accurate parameters of the underwater cleaning robot, there are certain errors in the robot model. Considering the model uncertainty of the underwater cleaning robot, an uncertainty error model is established by introducing the uncertain terms ΔM, ΔC(υ), and ΔD(υ):

[0084] M = M′ + ΔM

[0085] C(v) = C′(v) + ΔC(v)

[0086] D(v) = D′(v) + ΔD(v)

[0087] where: M′, C′(v), D′(v) represent the matrices of the nominal model, and ΔM, ΔC(v), ΔD(v) represent the uncertain dynamic matrices. Here, ΔM = 0.2M′, ΔC(v) = 0.2C′(v), and ΔD(v) = 0.2D′(v);

[0088] where:

[0089]

[0090] Among them: d 11 = -X u -X uu |u|d 22 = -Y v -Y vv |υ|, d 33 = -Z w -Z ww |w|, d 44 = -N r -N rr |r|; m is the mass of the cleaning robot, and are the added mass and added inertia of the cleaning robot respectively; X u , Y v , Z w , N r , X uu , Y vv , Z ww , N rr represent the viscous hydrodynamic parameters; I z is the moment of inertia of the cleaning robot about the Z-axis;

[0091] Step S2: Use the uncertainty error model and the basic kinematic model to obtain the optimized dynamic model, and obtain the motion position error η e and the control law τ cThe conversion model between the lumped uncertainty d′; the lumped uncertainty d′ includes model uncertainty and external environmental disturbances;

[0092] According to the basic kinematic model Obtain And combine with To obtain the first derivative of the cleaning robot's velocity vector

[0093]

[0094] Substitute Into the basic dynamic model to obtain:

[0095]

[0096] Thus, the optimized dynamic model is obtained:

[0097]

[0098] In the formula: M η (η) = M′J -1 , D η (v, η) = D′(v)J -1 ,

[0099] Set the position of the offshore platform's conduit (x d , y d , z d ), set the desired heading angle ψ of the cleaning robot d , to obtain the desired position η of the cleaning robot d = [x d , y d , z d , ψ d T ; Set the desired motion speed v of the cleaning robot d = [u d , υ d , w d , r d T ,

[0100] Set the motion position error η of the cleaning robot in the earth coordinate system e , η e = η - η d ; The velocity error v of the cleaning robot in the body coordinate system e , v e = v - v d ; And set x1 = η e , ​​

[0101] Since \(x1 = \eta\) e , then

[0102] Obtained by optimizing the kinetic model:

[0103]

[0104] Furthermore, we get:

[0105]

[0106] Thus, we obtain:

[0107]

[0108] Finally, we obtain the conversion model:

[0109]

[0110] where \(\tau\) c is the control law, and the control law is the conversion control input, is the lumped uncertainty,

[0111] Step S3: Based on the system variable \(x\) h of the fixed-time disturbance observer and the estimated value of the system variable establish a fixed-time disturbance observer, use the fixed-time disturbance observer to compensate for the lumped uncertainty \(d'\), and handle the external environmental interference and its own coupling to improve the robustness of the control system;

[0112] When the cleaning robot moves, the model uncertainties and external environmental interferences, such as the effects of wind, waves, and ocean currents, etc., the formed lumped uncertainty \(d'\) will affect the state of the underwater cleaning robot. If not processed in time, it will lead to the accumulation of errors and affect the motion performance of the underwater cleaning robot; through the above mathematical transformation, the lumped uncertainty \(d'\) is obtained. In order to quickly and accurately compensate for the lumped uncertainty \(d'\), a fixed-time disturbance observer is designed based on the kinematics and dynamics of the underwater cleaning robot; the process of designing the fixed-time disturbance observer is as follows:

[0113] Define the system variable \(x\) h of the fixed-time disturbance observer:

[0114]

[0115] where \(l1>0\) is a user-defined positive constant, and the first derivative h of \(x\)

[0116]

[0117] Combined with we get:

[0118]

[0119] Thus, we can obtain:

[0120]

[0121] y1 = l2x h

[0122] Set as the estimated value of the system variable x h and thus the fixed-time disturbance observer is:

[0123]

[0124] where: α1 > 0, β1 > 0, 0 < p < 1, q > 1; l3 > 0 is a custom positive constant for the disturbance observer gain, is the approximation error, and at the same time e h is also the compensation for the lumped uncertainty d′ by the fixed-time disturbance observer;

[0125] Let the estimated value of the lumped uncertainty d′ be set to satisfy:

[0126]

[0127] where: l1 > 0, l2 > 0 are custom positive constants;

[0128] Further combined with i.e., Define d′ e the approximation error of the lumped uncertainty, we get:

[0129]

[0130] Combined with we get y1 = l2x h ; Prove the asymptotic stability of the disturbance observer:

[0131] Construct the Lyapunov function Take the first derivative of V1 to get:

[0132]

[0133] According to the conditions for the asymptotic stability of the disturbance observer Obtained:

[0134]

[0135] Combined with y1 = l2x h Obtained:

[0136]

[0137] When k h > 0, and k h = l2l3, let then the asymptotic stability condition of the disturbance observer is satisfied; further combined with y1 = l2x h , that is Obtained

[0138] Further considering the fixed-time stability of the disturbance observer to make it satisfy the conditions in Lemma 1,

[0139] Obtained the fixed-time disturbance observer:

[0140]

[0141] where sig i (x) = sign(x)|x| i ;.

[0142] Stability analysis of the fixed-time disturbance observer:

[0143] Establish the Lyapunov function Take the first derivative of V1:

[0144]

[0145] Combined with Lemma 1, in the fixed-time disturbance observer, all signals are fixed-time stable; therefore, the convergence time T of all signals of the fixed-time disturbance observer satisfies:

[0146]

[0147] It can be seen from Lemma 1 that the lumped uncertainty of the underwater cleaning robot can be compensated by the observer within a fixed time, and we do not need to know any properties of the uncertainty during this process, which means that this method can be better applied in practice.

[0148] Step S4: Establish the basic sliding mode surface model, design the fixed-time sliding mode controller, and establish the control law model; finally, through the fixed-time disturbance observer, the fixed-time sliding mode controller and combined with the kinematic model and the optimized dynamic model, the cleaning robot is enabled to autonomously move to the jacket; asFigure 8 As shown in the figure, the schematic diagram of the closed-loop control system;

[0149] A fixed-time sliding mode controller is designed to address the uncertainties and nonlinearities of the underwater cleaning robot model for offshore platform jacket, enabling the cleaning robot to autonomously move to the jacket.

[0150] Establish the basic sliding mode surface model:

[0151]

[0152]

[0153] where: λ > 0, Υ > 0 are self-defined parameters;

[0154]

[0155]

[0156] The exponential variable h(s) satisfies:

[0157]

[0158] where α represents the asymptotic law parameter of the ERL index, ε, γ0, and g are adjustable parameters; ERL is the specified reaching law;

[0159] Prove the stability of the sliding mode controller:

[0160] Construct a Lyapunov function V2:

[0161]

[0162] Through in the basic sliding mode surface model, take the first derivative to obtain and combine with the transformation model to obtain the optimized sliding mode surface model:

[0163]

[0164] Take the first derivative of V2 to obtain:

[0165]

[0166] Combine the basic sliding mode surface model and the optimized sliding mode surface model to obtain the control law model:

[0167]

[0168] That is

[0169] Thus, the control law model is obtained, and the control law model is:

[0170]

[0171] Prove the stability of the fixed-time sliding mode controller, and prove that the entire fixed-time sliding mode controller is fixed-time stable;

[0172] Prove the stability of the fixed-time sliding mode controller:

[0173] Establish a Lyapunov function:

[0174]

[0175] Combining the stability conditions of the fixed-time sliding mode controller and the fixed-time stability conditions of the disturbance observer, and combining Lemma 3, we get:

[0176]

[0177] Further combining and we get:

[0178]

[0179] And under the constraints of the stability conditions, the following equation always holds:

[0180]

[0181] where:

[0182]

[0183] k1 = min{α1, α2}

[0184] k2 = min{β1, β2}

[0185] Δ = s T d′ e

[0186] Then, according to Lemma 2, in the fixed-time sliding mode controller, all signals will converge to the region set and approach the origin within a fixed time;

[0187] Therefore, the convergence time T of all signals in the closed-loop system N satisfies:

[0188]

[0189] So the fixed-time sliding mode controller is stable.

[0190] Conduct a simulation experiment to verify the effectiveness of the control strategy:

[0191] Model parameters of the underwater cleaning robot: d 11 =-X u -X uu |u|d 22 =-Y v -Y vv |v|, d 33 =-Z w -Z ww |w|d 44 =-N r -N rr |r|; m = 95, W = 932N, W B = 1092N, X u =-30.6, X uu =-43, Y v =-91, Y vv =-54, Z w =-54.5, Z ww =-12,

[0192] Parameters of the control system of the underwater cleaning robot: α1 = 0.8, β1 = 1.2, p = 0.8, q = 1.2, l1 = 18, l2 = 12, l3 = 12, α2 = 0.3, β2 = 0.3, λ = diag(0.5, 0.2, 0.01, 0.01), Υ = 0.002, γ0 = 0.1, α = 0.02, g = 0.2, ε = 1.2.

[0193] External disturbance received: τ fu =-36 - 0.0429x - 0.01314y 2 -0.025z 2 -8.4706z, τ fv =-6.2134y, τ fw =-12 - 0.0684x 2 +0.07493x - 0.01125y 2 -0.693z 2 -0.2529z, τ fr =(τ fu +τ fv +τ fw ) / 10, τ f =[τ fu , τ fv , τ fw , τ fr T , τ g ​= [-15sin(0.1t), -8sin(0.2t), -2cos(0.1t), -12sin(0.1t)] T , τ d = τ f + τ g .

[0194] Assume the position of the jacket of the offshore platform is η d = [27, 2, 0, 0] T , and the initial position of the underwater cleaning robot is η0 = [23, -2, 0, 0] T , and the initial velocity is v0 = [0, 0, 0, 0] T .

[0195] The simulation results are as Figure 9-11 shown, where Figure 9 represents the desired tracking diagram. According to Figure 9 it is shown that the underwater cleaning robot can achieve autonomous movement to the desired position, Figure 10-11 and is the error diagram with the desired position. Therefore, the designed controller can enable the underwater cleaning robot to autonomously move to the desired position, indicating the effectiveness of the control strategy, and even under the influence of a complex marine environment, the closed-loop control system also has good robustness.

[0196] Principle of the modeling part:

[0197] Assumption 1. The lumped uncertainty d′ is bounded, that is, there exists a constant d′ s such that ||d′|| ≤ d′ s and d′ s > 0.

[0198] Lemma 1: If the nonlinear system satisfies x(0) = x0, there is the following inequality:

[0199]

[0200] where α > 0, β > 0, 0 < p < 1, q > 1, then the nonlinear system satisfies fixed-time stability, and the upper limit of the convergence time reached satisfies:

[0201]

[0202] Lemma 2: If the nonlinear system satisfies x(0) = x0, there is the following inequality:

[0203]

[0204] where α > 0, β > 0, 0 < θ < ∞, 0 < p < 1, q > 1, then the nonlinear system is fixed-time stable, and the upper bound of the reaching convergence time is:

[0205]

[0206] In the formula: The residual set of the solution of the nonlinear system can be expressed as:

[0207]

[0208] Lemma 3: For any there exists

[0209]

[0210]

Claims

1. An offshore platform jacket cleaning robot, characterized in that, It includes a left body (1) and a right body (2). Both the left body (1) and the right body (2) include docking ends, and arc-shaped clamping parts are respectively arranged at the docking ends of the left body (1) and the right body (2). One side of the docking ends of the left body (1) and the right body (2) is hinged, and the other side is connected by a locking connection mechanism (6). The locking connection mechanism (6) is installed on the left body (1), and the right body (2) is equipped with a snap blind hole that cooperates with the locking connection mechanism (6). A number of vertical thrusters (8), a number of circumferential thrusters (9) and a number of spray guns (12) are also arranged on the left body (1) and the right body (2). The control method of the offshore platform jacket cleaning robot includes the following steps: Step S1: Establish the basic kinematic model, basic dynamic model and uncertainty error model of the cleaning robot. Step S2: An optimized dynamic model is obtained by using an uncertainty error model and a basic kinematic model, and the motion position error of the cleaning robot is obtained through the optimized dynamic model , control law and lumped uncertainty between transformation models; lumped uncertainty includes model uncertainty and external environmental disturbances; Step S3: Based on the system variables of the fixed-time disturbance observer and the estimated values of the system variables , design a fixed-time disturbance observer, and use the system variable approximation error of the fixed-time disturbance observer to compensate for the lumped uncertainty ; The fixed-time disturbance observer is: ； wherein, , , , ; is the disturbance observer gain which is a user-defined positive constant, is the approximation error, and at the same time is also the compensation of the fixed-time disturbance observer for the lumped uncertainty ; , is a user-defined positive constant; Step S4: Establish the basic sliding mode surface model, design the fixed-time sliding mode controller, and establish the control law model to enable the cleaning robot to autonomously move to the jacket. The control law model is: ； wherein, is the control law, is the estimated value of the lumped uncertainty ; , is a self-defined parameter.

2. The offshore platform jacket cleaning robot according to claim 1, characterized in that, The locking connection mechanism (6) includes a snap seat (61) fixed on the left body (1) and a snap pin (62) fixed on the snap seat (61). A snap pin cap (66) is fixed at one end of the snap pin (62), and a spring (63) is arranged outside the snap pin (62). One end of the spring (63) is fixed with an electromagnet sleeve (64). The electromagnet sleeve (64) is arranged outside the snap pin (62), and a radial export inclined surface is also arranged outside the electromagnet sleeve (64). A number of snap protrusions (65) are arranged radially on the snap pin cap (66). Through holes that cooperate with the snap protrusions (65) are arranged on the snap pin cap (66). An arc-shaped part that cooperates with the radial export inclined surface is arranged on the lower side of the snap protrusion (65). The snap pin cap (66) is fixed on the snap pin (62) by a snap screw (67). A spring support part (68) for installing the spring (63) is also arranged on the snap pin (62). A magnetic attraction iron block (69) that cooperates with the electromagnet sleeve (66) is arranged on the inner side of the snap pin cap (66). A card slot that cooperates with the snap protrusion (65) is arranged in the snap blind hole.

3. The cleaning robot for the jacket of an offshore platform according to claim 1, characterized in that, Two vertical thrusters (8) are arranged on each of the left body (1) and the right body (2). Vertical through holes for installing the vertical thrusters (8) are arranged on the left body (1) and the right body (2). Two circumferential thrusters (9) are arranged on each of the left body (1) and the right body (2). The four circumferential thrusters (9) are 45 degrees apart from each other. Two spray guns (12) are arranged on each of the left body (1) and the right body (2). Spray gun brackets (13) for installing the spray guns (12) are arranged on both the left body (1) and the right body (2).

4. The cleaning robot for the jacket of an offshore platform according to claim 1, wherein A plurality of auxiliary guiding grooves are provided on the arc-shaped clamping portion, and auxiliary radial wheels (7) are fixed in the auxiliary guiding grooves. The auxiliary radial wheels (7) are fixed in the mounting grooves through rotating shafts, and the outer diameters of the auxiliary radial wheels (7) protrude from the auxiliary guiding grooves.

5. The cleaning robot for the jacket of an offshore platform according to claim 1, characterized in that Anti-collision frames (5) are further respectively provided at the bottoms of the left body (1) and the right body (2); the anti-collision frames (5) include two support beams fixed to the lower part of the box body and an arc-shaped anti-collision beam fixed to the two support beams.

6. The cleaning robot for the jacket of an offshore platform according to claim 1, characterized in that, Float bodies (3) are respectively provided on the upper sides of the left body (1) and the right body (2), and sealed cabins (4) are respectively provided on the lower sides of the left body (1) and the right body (2).

7. The control method of the cleaning robot for the jacket of an offshore platform according to claim 1, characterized in that, In step S1, the basic kinematic model and the basic dynamic model are: Wherein: is the mass inertia matrix, is the transformation matrix, is the Coriolis force and centripetal force matrix, is the fluid damping coefficient matrix, is the restoring force and moment vector of the cleaning robot, , are respectively the gravity and buoyancy of the underwater cleaning robot, and the superscript represents the transpose of the matrix; is the movement position vector of the cleaning robot in the horizontal plane under the earth coordinate system, where is the position of the cleaning robot, represents the heading angle of the cleaning robot; represents the velocity vector of the cleaning robot under the body coordinate system, where , , , respectively represent the linear velocities of forward, translation, depth, and the angular velocity of yaw; is the control input, and the inputs of forward, translation, depth, and yaw , , and are provided by the actuator; is the influence of wind, waves, and ocean currents in the external environment; The uncertainty error model is: By reference to uncertain terms , , Establish as an uncertainty error model: In the formula: , , represent the matrices of the nominal model, , , represent the uncertain dynamics matrix, where we take , , ; Wherein: , ， Wherein: , , , , , , ; , , ; is the mass of the cleaning robot, , , and are the additional mass and additional inertia of the cleaning robot, respectively; , , , , , , , represent the viscous hydrodynamic parameters; is the moment of inertia of the cleaning robot about the Z axis.

8. The control method of the cleaning robot for the jacket of an offshore platform according to claim 7, characterized in that, The optimized dynamic model is: Wherein: , , , ; The conversion model is: wherein is the control law, and the control law is the transformed control input, is the lumped uncertainty, .

9. The control method of the cleaning robot for the jacket of an offshore platform according to claim 7, characterized in that, System variables of the fixed-time disturbance observer : In the formula, is the system variable of the fixed-time disturbance observer; The basic sliding mode surface model is: Wherein: In the formula: ; Among them, is the asymptotic law parameter of the ERL index, , and are adjustable parameters; ERL is the specified arrival law.

Images