Pose control method and system for air bag supporting flexible buoyant raft vibration isolation system

Abstract

The invention belongs to the technical field of vibration and noise reduction of surface ships and underwater vehicles, and discloses a pose control method and system for an air bag supporting flexible buoyant raft vibration isolation system, and the method comprises the steps: obtaining the actual observation positions of a plurality of measurement points on a raft frame; according to the actual observation position, fitting by utilizing an iterative reweighted least square method to obtain a reference datum plane of the raft frame; rigid body displacement corresponding to the measuring points is obtained based on the reference datum plane, elastic deformation displacement corresponding to the measuring points is obtained according to the actual observation positions and the rigid body displacement, and then rigid body displacement and elastic deformation displacement corresponding to the air bag are obtained; and constructing an index system according to the rigid body displacement, the elastic deformation displacement and the pressure corresponding to the air bag, and determining the to-be-adjusted air bag based on the index system. According to the method, the calculation of the reference datum plane more conforming to a novel mounting process and the rigid body displacement and the elastic deformation displacement of the approximate separation raft frame are realized, more comprehensive indexes are provided, and the accuracy of air bag selection is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field related to vibration and noise reduction of surface ships and underwater vehicles, and more specifically, relates to a posture control method and system for an airbag-supported flexible floating raft vibration isolation system. Background Art

[0002] With the development of vibration isolation technology, the application of airbag vibration isolation technology to floating rafts can significantly attenuate the transmission of equipment vibration to the hull, significantly improving the vibration isolation effect. However, as floating raft structures tend to become larger, their stiffness inevitably decreases. Due to factors such as uneven weight distribution of equipment on the raft frame, equipment startup and shutdown, slow airbag leakage, and external disturbances, the raft frame will experience significant elastic deformation, making it difficult to maintain a stable equilibrium posture. This causes uneven force on the connecting pipelines, resulting in reduced vibration isolation effect and even threatening the safety of equipment and pipelines. Therefore, to achieve high-precision raft posture control based on the consideration of the raft frame's elastic deformation, it is necessary to adopt corresponding strategies to accurately select the airbags and their inflation and deflation, and to rationally design the control strategy.

[0003] In the existing technology, the reference reference plane of the raft frame is usually determined by using the displacement sensor readings at the four corners of the raft frame to determine the displacement of the airbag supporting the raft frame, and the airbag to be adjusted is determined by combining the two parameters of the airbag pressure and displacement. Among them, the determination of the reference reference plane is not comprehensive enough and is not consistent with the new installation process of the raft frame. In addition, the elastic deformation factor of the raft frame is not considered in the process of selecting the airbag to be adjusted, and the indicators for airbag selection are relatively limited.

[0004] Furthermore, airbag selection is typically based on the direct multiplication of various parameters, making the physical meaning of the selection process unclear. Control strategies also fail to incorporate feedback corrections to airbag selection based on actual adjustment results. This can lead to significant discrepancies between the expected and actual airbag adjustment effects, impacting raft position control. Consequently, existing technologies lack accurate airbag selection and incomplete control strategy design. Summary of the Invention

[0005] In response to the above defects or improvement needs of the prior art, the present invention provides a posture control method and system for an airbag-supported flexible floating raft vibration isolation system, which is used to solve the problem that the selection of the airbag to be adjusted in the existing posture control method of the airbag-supported flexible floating raft vibration isolation system does not take into account the elastic deformation factor of the raft frame, and the indicators for airbag selection are relatively limited, resulting in insufficient accuracy in airbag selection.

[0006] To achieve the above objectives, according to one aspect of the present invention, a method for controlling the position and posture of an airbag-supported flexible floating raft vibration isolation system is provided, wherein the flexible floating raft vibration isolation system includes a raft frame, and the raft frame is supported on multiple airbags. The method for controlling the position and posture includes:

[0007] Acquiring actual observation positions of a plurality of preset measuring points on the raft, wherein the plurality of measuring points at least include a plurality of measuring points that are set in a one-to-one correspondence with the airbags;

[0008] According to the actual observed positions of the plurality of measuring points, an iterative reweighted least squares method is used to fit and obtain a reference datum plane of the raft; wherein, in the iterative reweighted least squares method, an error of the measuring point is obtained according to the deviation between the actual observed position of the measuring point and the position on the reference datum plane, an error weight is assigned to the error of each measuring point, and an optimization goal is to minimize the weighted sum of the errors of the plurality of measuring points;

[0009] Obtaining a rigid body displacement corresponding to the measuring point based on a positional distance between the measuring point on the reference datum plane and the ideal posture plane, and obtaining an elastic deformation displacement corresponding to the measuring point based on the actual observation position and the rigid body displacement, thereby obtaining a rigid body displacement and an elastic deformation displacement corresponding to the airbag;

[0010] An index system is constructed according to the rigid body displacement, elastic deformation displacement and pressure corresponding to the airbag, and the airbag to be adjusted is determined based on the index system and the posture control is performed.

[0011] According to another aspect of the present invention, a posture control system for an airbag-supported flexible floating raft vibration isolation system is provided, the system comprising a memory and a processor, the memory storing a computer program, and the processor executing any of the above-mentioned posture control methods for an airbag-supported flexible floating raft vibration isolation system when executing the computer program.

[0012] In general, compared with the prior art, the above technical solutions conceived by the present invention provide a method and system for controlling the posture of an airbag-supported flexible floating raft vibration isolation system:

[0013] 1. This paper proposes an iterative reweighted least squares method to fit the raft frame's reference datum plane based on the actual observation positions of multiple measuring points. This method takes into account the elastic deformation of the raft frame and achieves a reference datum plane that is more suitable for the new installation process. Furthermore, the reference datum plane is used to divide the displacement of the actual observation position relative to the ideal posture plane into rigid body displacement and elastic displacement. This approximate separation of the rigid body displacement and elastic deformation displacement of the raft frame provides more comprehensive indicators for airbag selection and improves the accuracy of the selection of the airbag to be adjusted.

[0014] 2. The approximate rigid body displacement and elastic deformation of the raft frame are calculated. The adjustment effect of the raft frame is then predicted. Based on the change in rigid body displacement and the inflation / deflation correction coefficient of all measuring points after any airbag is inflated / deflated, the overall change trend of the raft frame after adjusting any airbag is predicted. Finally, the approximate optimal control pressure ratio parameter of the airbag is calculated, resulting in four indicators that make airbag selection more comprehensive.

[0015] 3. The airbag selection process has been optimized. Multiple indicators in the index system are used to adopt a multi-factor comprehensive decision-making method to select airbags, which improves the accuracy of airbag selection and makes the physical meaning of the airbag selection process more reasonable and clear.

[0016] 4. The control strategy design of the airbag-supported flexible floating raft has been optimized, and the airbag selection is corrected based on the actual adjustment effect, thereby improving the control effect of the raft frame. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural schematic diagram of an airbag-supported flexible floating raft vibration isolation system of the present invention;

[0018] Figure 2 This is a schematic diagram of the raft installation process;

[0019] Figure 3 This is the flow chart for approximate solution of rigid body displacement and elastic deformation displacement;

[0020] Figure 4 This is the control flow chart of the flexible raft. DETAILED DESCRIPTION

[0021] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0022] See also Figure 1 and Figure 2 This embodiment provides a posture control method for an airbag-supported flexible floating raft vibration isolation system, wherein the flexible floating raft vibration isolation system includes a raft frame, and the raft frame is supported on multiple airbags. The posture control method includes:

[0023] Acquiring actual observation positions of a plurality of preset measuring points on the raft, wherein the plurality of measuring points at least include a plurality of measuring points that are set in a one-to-one correspondence with the airbags;

[0024] According to the actual observed positions of the plurality of measuring points, an iterative reweighted least squares method is used to fit and obtain a reference datum plane of the raft; wherein, in the iterative reweighted least squares method, an error of the measuring point is obtained according to the deviation between the actual observed position of the measuring point and the position on the reference datum plane, an error weight is assigned to the error of each measuring point, and an optimization goal is to minimize the weighted sum of the errors of the plurality of measuring points;

[0025] Based on the positional distance between the measuring point on the reference datum plane and the ideal posture plane, a rigid body displacement corresponding to the measuring point is obtained, and based on the actual observation position and the rigid body displacement, an elastic deformation displacement corresponding to the measuring point is obtained, thereby obtaining the rigid body displacement and elastic deformation displacement corresponding to the airbag; the elastic deformation displacement is the distance between the actual observation position and the position on the ideal posture plane, i.e., the displacement minus the rigid body displacement;

[0026] An index system is constructed according to the rigid body displacement, elastic deformation displacement and pressure corresponding to the airbag, and the airbag to be adjusted is determined based on the index system and the posture control is performed.

[0027] In this embodiment, the flexible floating raft vibration isolation system includes a raft frame, on which related equipment is installed, and the raft frame is supported by multiple airbags. The raft frame is provided with multiple measuring points, at least at locations corresponding to the airbags, and measuring points may also be provided at other locations. Optionally, the raft frame may have measuring points corresponding to the airbags, and multiple additional measuring points may be provided in the middle of the raft frame. The measuring points can obtain actual observed positions via sensors; the actual observed positions and displacements of the measuring points corresponding to the airbags can represent the actual observed positions and displacements of the corresponding airbags.

[0028] like Figure 1 As shown, in some specific embodiments, the airbag-supported flexible floating raft vibration isolation system of the present invention includes six parts: equipment, base, airbag, displacement sensor, pressure sensor, and raft frame. The operating equipment is located on the raft frame, and the raft frame is connected to the base via eight airbag vibration isolators evenly arranged on both sides. The system has a total of 12 sets of displacement sensors for detecting the height of each measuring point of the raft frame, with one measuring point set next to each airbag and four measuring points set in the middle of the raft frame. Each airbag is equipped with a pressure sensor to measure the actual pressure of the airbag, and two high-speed solenoid valves are connected to the high-pressure air source to realize the inflation and deflation of the airbag. The displacement sensor, airbag, and pressure sensor are all marked in the figure.

[0029] The airbag vibration isolation system effectively controls the raft frame by inflating and deflating each airbag to change the amount of gas within it, thereby changing the internal pressure and, in turn, the height of the airbag. Specifically, the airbag is made of elastic material and has a large deformability. By adjusting the air pressure within the airbag, its stiffness and load-bearing capacity can be changed to meet different vibration isolation requirements. When the airbag is inflated, the air pressure increases, causing the airbag to expand, which increases the stiffness and load-bearing capacity of the airbag. The force exerted on the airbag remains unchanged, the amount of compression on the airbag decreases, and the height of the airbag increases, corresponding to the height of the raft frame. Conversely, when the airbag is deflated, the air pressure decreases, the airbag contracts, and the stiffness and load-bearing capacity decrease. The force exerted on the airbag remains unchanged, the amount of compression on the airbag increases, and the height of the airbag decreases, corresponding to the height of the raft frame.

[0030] Due to factors such as the low rigidity of the raft structure, uneven weight distribution of equipment on the raft, equipment start-up and shutdown, slow airbag leakage, and external disturbances, the raft will not only deviate from its equilibrium position but also produce large elastic deformation. At this time, the actual displacement at each measuring point on the raft is composed of the rigid body displacement and elastic deformation displacement of the raft. The two are coupled and difficult to completely separate.

[0031] First, the system coordinate system is established:

[0032] (1) Zero point calibration of the control target and displacement sensor of the raft:

[0033] Due to the elastic deformation of the raft, the actual raft plane cannot be an ideal plane. To represent the overall posture of the raft, it is necessary to iteratively reweight the coordinates of the measurement points to fit the reference plane of the raft. Before the raft begins to adjust, it is necessary to determine the control target, that is, to adjust the reference plane of the raft to the ideal posture plane while minimizing the elastic deformation of the raft and ensuring that the pressure of each airbag is as uniform as possible. Figure 2 As shown in the figure, during the installation of the floating raft vibration isolation system, after various types of equipment are installed on the raft frame, a jack is installed between the base and the raft frame and next to each airbag. All airbags are in a completely deflated state, and the raft frame is supported by the jacks. Third-party tools (such as a level, a laser collimator, etc.) are used to check whether the raft frame is in the ideal posture plane. The height of the corresponding position of the raft frame is adjusted by changing the supporting force of each jack, thereby adjusting the raft frame to the ideal posture. At this time, the reference reference plane of the raft frame coincides with the ideal posture plane, and the height of each measuring point is its reference height. At this time, the output values ​​of all displacement sensors are set to zero.

[0034] (2) Establishment of system coordinate system and acquisition of measuring point coordinates:

[0035] When the flexible floating raft vibration isolation system has elastic deformation, in order to realize the control of the floating raft posture, the coordinate system of the system should be established first. Taking this embodiment as the research object, the raft frame is regarded as a plane and the position of the center of gravity when the raft frame is in the ideal posture plane is the origin of the coordinate system O. In order to accurately describe the position of each measuring point and the airbag in the system by mathematical methods, the direction of the airbag arrangement in the ideal posture plane of the raft frame is taken as the direction of the Y axis, the Z axis is perpendicular to the ideal posture plane of the raft frame and is in an upward direction, and the direction of the X axis is determined by the right-hand screw rule. The final result of the system coordinate system establishment is as follows: Figure 1 As shown in the figure, the detection value of the displacement sensor at this time is the displacement between the actual observation position and the initial ideal posture position.

[0036] When the raft is in an ideal position, the reference position of each measuring point on the raft, that is, the coordinate is (X j ,Y j ,0), where j = 1 to 12, X j With Y j are the distances of each measuring point along the X-axis and Y-axis relative to the center of gravity of the raft. When the raft produces rigid body displacement and elastic deformation displacement, and deviates from the ideal posture, the measuring point will deviate from its reference height, and the sensor reading h j is the result of the coupling between the rigid body displacement and the elastic deformation displacement of the raft. At this time, the coordinate of the jth measuring point along the Z axis in the system coordinate system is:

[0037] z j =h j ;

[0038] If the raft is regarded as a rigid body, only the rigid body displacement of the raft is considered. If α is the roll angle of the raft around the X axis, and β is the pitch angle of the raft around the Y axis, is the length of the jth measuring point from the X axis in the ideal position of the raft, is the length of the jth measuring point from the Y axis in the ideal position of the raft. When the position of the raft changes:

[0039]

[0040] in, is the distance from the jth measuring point to the X axis after the raft's posture changes, is the distance from the jth measuring point to the Y axis after the raft's posture changes. Limited by the adjustable range of the airbag isolator's vertical height, the raft's maximum deviation from the ideal posture is very small, near 0°, so the cosine values ​​of α and β are both approximately 1. Therefore:

[0041]

[0042] It can be seen that if only the rigid body displacement of the raft is considered, the change of the measuring point coordinates caused by the change of the raft attitude angle is very small, and the X-axis of the measuring point before and after the raft attitude change is j 、Y j The effect of the vertical elastic deformation of the raft on the horizontal coordinate X of the measuring point is very small. j 、Y j The influence of is very small and can be ignored. Therefore, if the raft generates rigid body displacement and elastic deformation displacement at the same time, the two are coupled to form the total displacement of the raft. The coordinates of each measuring point in the system coordinate system can be approximated as (X j ,Y j ,z j ), where j = 1 to 12.

[0043] Approximate separation of raft rigid body displacement and elastic deformation displacement:

[0044] The raft's reference plane not only represents the overall raft's posture but also approximately separates the raft's rigid-body displacement and elastic deformation. Displacement sensor readings only represent the coupled effect of these two displacements. Accurately adjusting the raft's posture requires precise selection of airbags and their inflation and deflation. Therefore, a reference plane—an approximate rigid-body posture plane—is required to approximately separate the two displacements. The approximate rigid-body displacement and elastic deformation at each airbag's corresponding raft position serve as the basis for airbag selection.

[0045] If only the rigid body displacement of the raft is considered, the raft is regarded as a rigid body, and the plane where it is located is the rigid body posture plane. The displacement of this plane due to deviation from the plane of the ideal posture is called the rigid body displacement of the raft.

[0046] When obtaining the approximate rigid body posture plane, if the coordinates of the 12 measuring points are directly fitted into a plane using the ordinary least squares method, when the elastic deformation displacement of the raft is not large and can be ignored, the 12 measuring points can be approximately considered to be in the same plane, and the fitting result is also more accurate.

[0047] However, due to various factors, the actual displacement measured by the displacement sensor is also affected by the elastic deformation of the raft frame. Some measuring points on the raft frame have large elastic deformations, and the positions of these measuring points deviate far from the actual raft frame rigid body posture plane, which are called outliers. Since the influence of these outliers on the fitting plane is the same as that of general points on the fitting plane, they will have a greater impact on the estimated parameters of the fitted raft frame rigid body posture plane equation when participating in the ordinary least squares method, making the estimated raft frame rigid body posture plane inaccurate. Therefore, it is necessary to reduce the influence of these outliers and introduce weights ω for the coordinate data of all measuring points. j, a smaller weight is given to the data of the measurement points that deviate farther, and a larger weight is given to the data of the measurement points that deviate closer. The iterative reweighted least squares method is used, and iteration is performed on the basis of the weighted least squares method to approximate the optimal valuation, thereby obtaining the approximate raft rigid body posture plane, and finally solving the approximate rigid body displacement and elastic deformation displacement of the raft.

[0048] Next, we use the iterative reweighted least squares method to obtain the approximate raft rigid body posture plane, i.e., the reference reference plane, and solve the approximate rigid body displacement and elastic deformation displacement of the raft. The flowchart summarizing the solution steps is as follows: Figure 3 That is, according to the actual observation positions of the plurality of measuring points, the reference plane of the raft is obtained by fitting using iterative reweighted least squares method, which specifically includes:

[0049] Setting the initial error weights of the plurality of measuring points to 1, obtaining the residual between the actual observed position of the measuring point and the regression fitting value in each iteration, and updating the error weight using the Huber weight function according to the residual of the measuring point;

[0050] When the maximum value of the residual difference of the plurality of measurement points relative to the residual difference after the previous iteration is less than the adjustment accuracy of the raft, or the number of iterations reaches the maximum number of iterations, the iteration ends and the reference datum plane is obtained. The specific process is as follows:

[0051] (1) Definition of optimization index function:

[0052] The general form of the plane equation can be written as:

[0053] Z=aX+bY+c;

[0054] Where a, b, and c are unknowns to be determined in the reference datum plane.

[0055] The plane equation has only three parameters to be estimated. If the measurement is error-free, only the displacement values ​​at three different measuring points are needed to solve a, b, and c. However, since there is always random error in each measurement, that is:

[0056] z j =Z j +v j ; or z j =aX j +bY j +c+v j ;

[0057] Where z j is the measured data, Z j is the true value, v j is a random error, j=1~12.

[0058] Obviously, adding up the measurement errors each time can form the total error:

[0059]

[0060] If the total measurement error is minimized, the criterion of minimizing the sum of squares of each measurement error is adopted. Therefore, an optimization index function is defined, that is, the minimization error function:

[0061]

[0062] Among them, ω j is the error weight of the jth measurement point.

[0063] This optimization function represents the effect of iteratively reweighted least squares (IRL) on plane fitting. IRL finds the optimal function matching the data by minimizing the sum of squared errors. The smaller J, the better the plane fitting. Therefore, the plane equation corresponding to the minimum optimization function J is the desired plane equation.

[0064] (2) Iterative reweighting to obtain the approximate rigid body posture plane equation:

[0065] The solution of plane equation can be transformed into To find the three unknowns, if J is minimized, we can use the method of finding the extreme value to get:

[0066]

[0067] Further arrangement of the above formula:

[0068]

[0069] Written in matrix form as follows:

[0070]

[0071] Solve the above matrix equation and get the solution of a, b, c as The required plane equation is:

[0072]

[0073] Therefore, the iterative calculation steps of the iterative reweighted least squares method are as follows:

[0074] a. In the first iteration, the standard least squares method is used for fitting, that is, Obtain the plane equation after the first iteration

[0075] b. After the first iteration, the X of each measuring point j 、Y jSubstitute the coordinates into the plane equation obtained after the first iteration to obtain the regression fitting value of each measuring point after the first iteration The actual observation value z of each measuring point j and regression fitted values By doing the difference, we can get the residual of each measurement point after the first iteration.

[0076]

[0077] c. The residuals of each measuring point Substituting into the Huber weight function, we can get the new weight of each measuring point Perform the next iteration and obtain the plane equation after the next iteration

[0078] The Huber weight function is defined as:

[0079]

[0080] in, is the error weight of the jth measurement point at the n+1th iteration; is the residual error of the jth measurement point after the nth iteration; γ is the clipping value, which is numerically equal to the adjustment accuracy σ of the raft.

[0081] The raft's adjustment accuracy, σ, is the minimum height change across all measurement points after a single adjustment of any airbag. This represents the minimum effect of adjusting the airbag on the raft's adjustment. Using σ as a clipping value determines whether a measurement point is an outlier, thereby reducing the weight of outliers. A single adjustment includes both inflation and deflation, and the duration of a single adjustment is a preset time. The same preset time is used when determining the airbag to be adjusted and performing posture adjustments.

[0082] If the residual of the measuring point is less than the clipping value γ, that is, the adjustment accuracy σ of the raft, it means that the residual of the measuring point is so small that it is meaningless to improve the raft posture. Therefore, it is considered that the actual observation value of the measuring point and the regression fitting value are close, and the measuring point is not an outlier. Otherwise, if the residual of the measuring point is greater than the clipping value γ, the measuring point is considered to be an outlier. It is necessary to determine the size of the weight of the measuring point in the next iteration according to the size of the residual of the measuring point to weaken the influence of the measuring point on the fitting plane.

[0083] d. Repeat steps b and c, using the new weights for each measurement point to perform iterative fitting until the maximum value of the change in the residuals of all measurement points relative to the residuals after the n-1th iteration is obtained. Less than the adjustment accuracy σ of the raft, that is:

[0084]

[0085] This indicates that the effect of the nth iteration on the plane equation fitting is no longer significant compared to improving the raft's pose. Since the iteration process is a convergent process, the plane equation obtained at this point is already close to the desired true rigid body pose plane equation. Iteration is stopped. The plane equation obtained after the nth iteration becomes the reference plane equation for the raft, i.e., the approximate rigid body pose plane equation. The number of iterations can also be used as a basis for determining the end of the iteration, but this is not a specific limitation.

[0086] (3) Calculation of the approximate rigid body displacement and elastic deformation displacement of the raft:

[0087] The rigid body displacement and elastic deformation displacement of the raft can be approximately separated by the plane equation of the rigid body posture of the raft. After the nth iteration, the plane equation is obtained That is the approximate rigid body posture plane equation of the raft, and the X of each measuring point is j 、Y j Substituting the coordinates into this plane equation can obtain the regression fitting value of each measuring point That is the approximate rigid body displacement at the measuring point, and the residual corresponding to each measuring point is This is the approximate elastic deformation displacement at the measuring point. The approximate rigid body displacement and elastic deformation displacement of each measuring point together constitute the total displacement at the measuring point, that is, the actual observation value z j ,Right now:

[0088]

[0089] Furthermore, the construction of the indicator system also includes:

[0090] Obtain the rigid body displacement change corresponding to all measuring points after any airbag inflation adjustment and the inflation correction coefficient corresponding to all measuring points after any airbag inflation adjustment, where the inflation correction coefficient is the proportional coefficient between the actual observed position change of the measuring point after any airbag inflation adjustment and the rigid body displacement change; obtain the predicted value of the overall inflation adjustment trend of any airbag by multiplying the rigid body displacement change of the measuring point by the inflation correction coefficient and summing the results for all measuring points;

[0091] Accordingly, the rigid body displacement change corresponding to all measuring points after any airbag deflation adjustment and the deflation correction coefficient corresponding to all measuring points after any airbag deflation adjustment are obtained, where the deflation correction coefficient is the proportional coefficient between the actual observed position change of the measuring point after any airbag deflation adjustment and the rigid body displacement change; the overall change trend prediction value of the deflation adjustment of any airbag is obtained by multiplying the rigid body displacement change of the measuring point by the deflation correction coefficient and summing the results at all measuring points;

[0092] According to actual observation positions of the plurality of measuring points, a predicted value of the overall change trend of the inflation adjustment of the airbag or a predicted value of the overall change trend of the deflation adjustment is selected and added to the indicator system.

[0093] The prediction of the flexible raft adjustment effect is as follows:

[0094] After approximately separating the rigid body displacement and elastic deformation of the raft, two indicators for airbag selection are obtained. However, due to the complex structural deformation of the flexible raft, the displacement trend and magnitude of each measuring point on the raft after the airbag is inflated and deflated are unknown. To further improve the accuracy of airbag selection, it is necessary to predict the effect of raft adjustment.

[0095] In the system's initial state, that is, in the ideal pose plane, the eight airbags are inflated and deflated at the same time. Before and after each adjustment, the actual observed total displacement of each measuring point is recorded, and the approximate rigid body pose plane of the raft is calculated from this. This gives the approximate rigid body displacement at each measuring point on the raft, and the change in rigid body displacement before and after the adjustment is calculated for each measuring point. That is, starting from the ideal pose, all airbags are inflated and deflated once, and the change in rigid body displacement at all measuring points before and after the adjustment is recorded, which is the rigid body displacement effect brought about by the airbag adjustment.

[0096] After the system initialization inflation and deflation process is completed, the change in the rigid body displacement of the jth measuring point before and after the i-th airbag is inflated is recorded as The change in the rigid body displacement of the jth measuring point before and after the i-th airbag is deflated is: The actual change in the total displacement of the jth measuring point before and after the i-th airbag is inflated is: The actual change in the total displacement of the jth measuring point before and after the i-th airbag is deflated is:

[0097] Therefore, the correction coefficient of the flexible raft adjustment effect prediction model is obtained:

[0098]

[0099] Where, is the inflation correction coefficient corresponding to the jth measuring point after the i-th airbag inflation adjustment; is the deflation correction coefficient corresponding to the jth measuring point after the i-th airbag is deflated and adjusted; all coefficients are written in matrix form to obtain the correction coefficient matrix:

[0100]

[0101] Where K c , K f are the correction coefficient matrices of the raft under inflation and deflation conditions respectively.

[0102] The sign and absolute value of the correction coefficient reflect the change in displacement relative to the raft's rigid body, the trend in the total displacement of the jth measuring point before and after the i-th airbag is inflated and deflated, and the degree of deformability at the raft's corresponding location. Therefore, the correction coefficient is determined by the properties of the raft itself. Therefore, the effect of pressure changes on the correction coefficient is negligible, and the correction coefficient can be approximately assumed to be constant under varying pressure changes.

[0103] So, in the more general case:

[0104]

[0105] Right now:

[0106]

[0107] in, is the predicted value of the total displacement change of the jth measuring point relative to the ideal posture plane before and after the i-th airbag is inflated / deflated in any posture of the flexible raft. is the change in rigid body displacement of the jth measuring point before and after the i-th airbag is inflated / deflated.

[0108] Define the rigid body displacement response matrix of the raft:

[0109]

[0110] Where ΔG c , ΔG f are the rigid body displacement response matrices of the raft under inflation and deflation conditions, respectively.

[0111] This matrix includes the changes in rigid body displacement under all adjustments of the raft. The i-th row represents the changes in rigid body displacement of all j measuring points on the raft after adjusting the i-th airbag.

[0112] In the initial state, when the airbag is selected for the first time, or After the kth adjustment of the airbag, if the number of the kth adjustment airbag is i k , calculate the i-th k The change in the rigid body displacement of the jth measuring point before and after the airbag is inflated / deflated is: When the airbag is selected for the k+1th time, the i-th rigid body displacement response matrix k Row update, i.e. or

[0113] That is, starting from the ideal posture plane, all airbags are inflated and deflated to adjust the experiment, and the inflation correction coefficient and the initial rigid body displacement change corresponding to all measuring points after any airbag inflation adjustment are obtained, as well as the deflation correction coefficient and the initial rigid body displacement change corresponding to all measuring points after any airbag deflation adjustment are obtained;

[0114] When the airbag to be adjusted is determined for the first time and posture control is performed, the predicted value of the overall change trend of inflation adjustment in the index system of the airbag is determined based on the inflation correction coefficients of all measuring points corresponding to the airbag and the initial rigid body displacement change, and the predicted value of the overall change trend of deflation adjustment is determined based on the deflation correction coefficients of all measuring points corresponding to the airbag and the initial rigid body displacement change;

[0115] After the airbag to be adjusted is determined and its posture is controlled, the rigid body displacement change corresponding to the adjusted airbag is updated according to the actual rigid body displacement change of all measuring points after adjustment for the next adjustment judgment.

[0116] Define the flexible raft adjustment effect prediction matrix:

[0117]

[0118] Therefore, the prediction matrix of the flexible raft adjustment effect is the dot product of the correction matrix and the rigid body displacement response matrix:

[0119]

[0120] Similarly:

[0121]

[0122] This matrix includes all adjustments of the flexible raft. The i-th row represents the predicted value of the total displacement change of all j measuring points on the raft after adjusting the i-th airbag, which predicts the adjustment effect of the flexible raft.

[0123] The predicted value of the overall change trend of the raft after adjusting the i-th airbag is the row sum of the i-th row of the flexible raft adjustment effect prediction matrix, that is:

[0124]

[0125] The airbag and its inflation and deflation selection strategies are as follows:

[0126] (1) Constructing an index system based on the pressure corresponding to the airbag specifically includes:

[0127] Performing a static analysis on the flexible floating raft vibration isolation system to obtain a moment force balance equation for the plurality of airbags, and obtaining an optimal control pressure for each airbag with the goal of minimizing the pressure variance of the plurality of airbags under the moment force balance equation;

[0128] A pressure ratio parameter of the airbag is obtained according to a ratio of the actual pressure of the airbag to the optimal control pressure, and the indicator system includes the pressure ratio parameter.

[0129] The airbag approximate optimal control pressure ratio parameters are as follows:

[0130] From the above analysis, the rigid body displacement of each airbag can be obtained Elastic deformation displacement δ i , the predicted value of the overall change trend of the raft after adjusting any airbag These three indicators only consider the changes in the raft displacement. In order to improve the long-term stability and reliability of the system and extend its service life, it is also necessary to solve the airbag approximate optimal control pressure ratio parameter. On the basis of ensuring the posture control of the raft, the balance of airbag pressure is taken into account to the maximum extent to prevent the pressure of individual airbags from being too high or too low.

[0131] If the raft is regarded as a rigid body, the moment balance equations of the airbag isolator around the X and Y axes and the force balance equation along the Z axis can be listed as follows based on the statics theory:

[0132]

[0133] Among them, P i is the pressure of the i-th airbag, X i 、Y i is the coordinate of the i-th airbag, S e is the effective area of ​​the airbag, and G is the total weight of the raft and the equipment thereon.

[0134] At this time, the system is in a hyperstatic system, P i There are countless solutions. To make the load distribution of each airbag uniform and reduce the coupling between airbags, and to increase the constraint condition of airbag pressure uniformity, the pressure variance of each airbag is required to be minimized, that is:

[0135]

[0136] The optimal control pressure P of the i-th airbag is obtained si as follows:

[0137]

[0138] The above formula is the optimal control pressure of each airbag when the raft frame is a rigid body, that is, the approximate optimal control pressure when the raft frame is a flexible raft frame.

[0139] The ratio of the actual pressure to the approximate optimal pressure is defined as the airbag pressure ratio parameter λ i To select the airbag that needs the most control:

[0140]

[0141] (2) Comprehensive decision-making based on multiple factors to select airbags:

[0142] From the above analysis, the rigid body displacement of each airbag can be obtained Elastic deformation displacement δ i , approximate optimal control pressure ratio parameter λ i , the predicted value of the overall change trend of the raft after adjusting any airbag To achieve the control of the raft's posture, it is necessary to comprehensively consider the above four indicators to select the most appropriate airbag for inflation and deflation. The selection of the airbag can be regarded as a comprehensive decision-making problem of multiple indicators, so the subjective weighting method is used to weight each indicator.

[0143] That is, based on the index system, a multi-index factor comprehensive decision-making method is used to determine the airbag to be adjusted and perform posture control, specifically including:

[0144] Each indicator in the indicator system is dimensionally processed, and then a subjective weighting method is used to assign an indicator weight to each indicator. A comprehensive evaluation index is obtained by weighted summing the dimensionless indicators, and the airbag to be adjusted is determined based on the comprehensive evaluation index.

[0145] Determining, based on actual observation positions of the plurality of measuring points, whether to inflate or deflate the airbag to be adjusted to perform posture control;

[0146] The dimensionless processing of each indicator in the indicator system specifically includes:

[0147] Based on the actual observation positions of multiple measuring points, the adjustment strategy for posture control is determined, that is, whether inflation adjustment or deflation adjustment is required; based on the favorable trends of various indicators under the adjustment strategy, the indicators with larger values ​​are considered to be positive indicators, and the indicators with smaller values ​​are considered to be negative indicators; the positive indicators and negative indicators are processed dimensionlessly. The details are as follows:

[0148] Selection of airbag inflation and deflation and dimensionless data processing:

[0149] Positive indicators are more effective when their values ​​are higher, while negative indicators are more effective when their values ​​are lower. Depending on the raft's state, there are two options: inflating or deflating the airbags. These two indicators will differ depending on the chosen state. To ensure that the airbags are adjusted in a direction that is beneficial to the raft's posture, the data must be dimensionless before assigning weights to the various indicators.

[0150] If the sum of the total displacements of the actual observation positions of all measuring points relative to the ideal posture plane position is less than 0, that is, It means that the airbag needs to be inflated to increase the overall displacement of the raft. It is hoped that the overall change trend of the raft after adjusting the airbag, that is, the predicted value of the overall change trend of the inflation adjustment, will be The larger the value, the better, so It is a positive indicator. When selecting an airbag, the smaller the values ​​of rigid body displacement, elastic deformation displacement and airbag pressure are, the better. δ i ,λ i is a reverse indicator; if the sum of the total displacements of all measuring points is greater than 0, that is The airbag needs to be deflated to reduce the overall displacement of the raft, so the situation is the opposite. is a reverse indicator, δ i ,λ i It is a positive indicator.

[0151] The dimensionless processing of positive indicators is:

[0152] The dimensionless processing of the inverse indicator is:

[0153] in:

[0154]

[0155] Where a im Represents the four indicators of the i-th airbag δ i ,λ i The value of the mth indicator in s im is the result of dimensionless processing of the mth index of the i-th airbag, a min is the minimum value among different samples under the same indicator, a max It is the maximum value among different samples under the same indicator.

[0156] The weights of each indicator are determined as follows:

[0157] The G1 method, also known as the order relationship method, is one of the commonly used subjective weighting methods. The specific steps are as follows:

[0158] The first step is to determine the order relationship:

[0159] Sort the evaluation indicators by importance: The specific analysis is as follows.

[0160] Overall change trend of the raft after adjustment of each airbag This indicator is the most important for the posture control of the raft. It reflects whether the adjustment is effective. If the overall change trend of the raft after adjustment is opposite to the expected change trend, the raft will deviate from the ideal posture as a whole, then the adjustment will have a negative effect on the posture control of the raft.

[0161] The second is the rigid body displacement of the airbag This displacement reflects the overall posture of the raft. On the basis of effective adjustment, we should first ensure that the deviation between the overall posture of the raft and the ideal posture is not too large.

[0162] Next is the elastic deformation displacement δ of each airbag i This partial displacement reflects the extent of the elastic deformation of the raft at the position of the airbag. On the basis of adjusting the overall posture of the flexible raft towards the ideal posture, the elastic deformation of the raft is adjusted.

[0163] Finally, the pressure ratio parameter λ of each airbag is i On the basis of meeting the needs of raft frame adjustment, the pressure of each airbag is made as close as possible to the approximate optimal control pressure of the airbag, so that the airbag pressure is not too high or too low, protecting the airbag and increasing the service life of the airbag, while making the system more robust.

[0164] The second step is to determine the importance ratio:

[0165] After the order relationship is determined, the importance of the evaluation indicators is assigned according to relevant standards and own experience:

[0166]

[0167] Where: r m ∈[1.0,2.0] is the importance ratio of the m-1th indicator to the mth indicator; w m 、w m-1 are the indicator weights of the mth and m-1th indicators respectively.

[0168] The reference for order relation assignment is shown in the following table.

[0169] Table 1 Reference table for sequence relation assignment

[0170]

[0171] In this embodiment, take:

[0172]

[0173] The third step is to calculate the subjective weight of each indicator:

[0174] According to the assignment, calculate the subjective indicator weight w of the mth indicator m :

[0175]

[0176] In this embodiment, the above formula is used to calculate:

[0177]

[0178] Airbag selection:

[0179] After the weights are determined, the weighted sum of the multiple indicators of each airbag is performed, and the most suitable airbag is selected for adjustment. The comprehensive evaluation index value of the i-th airbag when the system selects the airbag for the kth time is:

[0180]

[0181] If:

[0182]

[0183] Among them, Q k W is the comprehensive evaluation index value matrix of the airbag when the system selects the airbag for the kth time; k is the subjective index weight matrix of the system when selecting the airbag for the kth time; S k is the dimensionless index matrix of the system when the airbag is selected for the kth time. The above formula can be written in matrix form:

[0184] Q k =S k W k ;

[0185] When selecting the airbag for the kth time, the airbag with the largest comprehensive evaluation index value is selected to adjust the inflation and deflation.

[0186] Furthermore, the posture control method of the airbag-supported flexible floating raft vibration isolation system further includes:

[0187] Correcting the comprehensive evaluation index of the airbag by a reward and punishment factor, and determining the airbag to be adjusted by using the comprehensive evaluation index after correction of the reward and punishment factor;

[0188] The reward and punishment factors corresponding to all airbags are initially 1;

[0189] After determining the airbag to be adjusted and performing posture control, the adjustment effect index is obtained. Based on the adjustment effect index, the reward and penalty factors corresponding to the adjusted airbag are updated using the pre-defined reward and penalty function for the next adjustment judgment. The control strategy is designed as follows:

[0190] After the airbag is adjusted, in order to ensure the stability of the system, the selection of the airbag needs to be corrected based on the actual control effect. Therefore, the comprehensive evaluation index value of the airbag is corrected in combination with the actual parameter changes of the raft after adjustment. The control strategy sets a weighted feedback correction mechanism with reward and punishment factors, so that the selection of the airbag before the next adjustment is more in line with the actual situation of the raft, improving the adaptability of the control algorithm to complex working conditions and preventing the extreme situation of non-convergence of the control system. The control flow chart of the flexible floating raft is summarized as follows: Figure 4 shown.

[0191] (1) Calculation of the feedback amount after adjusting the airbag, i.e., the adjustment effect index:

[0192] Compared to the k-1th adjustment, after the kth adjustment of the airbag, the following two feedback quantities, i.e., two adjustment effect indicators, can be obtained:

[0193] Overall change trend of raft ΔZ trend k as follows:

[0194]

[0195] in, It is the sum of the total displacements of all measuring points relative to the ideal posture plane after the k-1th adjustment of the airbag; is the sum of the total displacements of all measuring points after the k-th adjustment of the airbag; is the change in the sum of the total displacements of all measuring points after the kth adjustment of the airbag. The total displacement of the measuring point is the difference between the actual observed position of the measuring point and the position in the ideal posture plane.

[0196] The maximum total displacement change Δz among all measuring points max k as follows:

[0197]

[0198] The overall change trend of the raft reflects whether the adjustment has improved the control effect of the raft's posture, and the change in the maximum total displacement of all measuring points reflects whether the adjustment has improved the control accuracy of the raft. Both indicators are inverse indicators, and the smaller the value, the better the control effect.

[0199] (2) Feedback correction of comprehensive evaluation index values ​​of each airbag:

[0200] The kth adjustment of the ik After the airbags are selected, the comprehensive evaluation index value of the k+1th airbag adjustment is required. Perform feedback correction and define the reward and punishment function as follows:

[0201]

[0202] Among them, a μ and b μ Control the influence weights of the two indicators respectively. In particular, when k = 0, is the reward and penalty factor of the i-th airbag before the k+1-th adjustment.

[0203] therefore, This is the comprehensive evaluation index value of the i-th airbag after the reward and punishment factor feedback correction. After the k-th adjustment, when the k+1-th airbag adjustment is selected, the comprehensive evaluation index values ​​of all airbags are written in matrix form:

[0204] Q' k+1 =Q k+1 ⊙M k+1 =(S k+1 W k+1 )⊙M k+1 ;

[0205] Among them, Q' k+1 M is the comprehensive evaluation index value matrix of the airbag after the reward and punishment factors are corrected when the airbag is adjusted for the k+1th time; k+1 is the reward and penalty factor matrix of the airbag when the airbag adjustment is selected for the k+1th time, and

[0206] (3) The design of the control process also includes:

[0207] After any airbag to be adjusted is determined and the posture is controlled, it is determined whether the maximum total displacement of all measuring points relative to the ideal posture plane is greater than the preset control accuracy requirement. If not, the adjustment is completed;

[0208] If so, obtain the product of all reward and penalty factors corresponding to the airbags, determine whether the minimum value of the product among all airbags is less than the preset threshold, and determine whether the change in the maximum total displacement of all measuring points relative to the displacement before adjustment is less than the adjustment accuracy of the raft. If neither of these two situations occurs, continue to adjust. If at least one of these situations occurs, reduce the control accuracy requirement and / or reset the reward and penalty factors corresponding to all airbags and readjust. The details are as follows:

[0209] After the kth adjustment and before the k+1th selected airbag adjustment, all airbags in the system have k+1 historical values ​​of reward and punishment factors, that is, from the initial reward and punishment factor 1 before the first adjustment to the reward and punishment factor after the kth adjustment, there are k+1 values ​​in total. For the i-th airbag, after the kth adjustment, the reward and punishment intensity S it receives is defined as i The product of all reward and penalty factors of the airbag:

[0210]

[0211] In this formula, t represents the number of reward and penalty factors. If after the kth adjustment, the airbag adjustment is selected for the k+1th time:

[0212] a. The maximum total displacement of all measuring points is greater than the preset control accuracy requirement, that is: max(|z j |)>ε;

[0213] Among them, ε is the preset raft control accuracy value.

[0214] b. The rewards and punishments received by all airbags in the system are within the tolerance threshold, i.e., min(S i )≥χ;

[0215] Among them, χ∈(0,1) is the tolerance threshold.

[0216] At this time, continue to use the comprehensive evaluation index value Q' after the correction of the reward and punishment factors k+1 Select the airbag to perform the k+1th adjustment until the raft frame reaches the control accuracy requirement, and then stop the adjustment.

[0217] If the raft fails to meet the control accuracy requirements and at least one of the following extreme conditions exists in the system, the preset raft control accuracy requirements are considered too high and it is necessary to lower the preset control accuracy requirements, reset the rewards and penalties for all airbags, and readjust them to make the raft converge quickly and avoid system oscillation.

[0218] a. If min(S i ) is lower than χ, that is, min(S i )<χ; This indicates that the rewards and penalties received by all airbags in the system have exceeded the tolerance threshold, and adjusting the airbag pressure is no longer able to improve the raft's posture;

[0219] b. The maximum total displacement change among all measuring points Less than the adjustment accuracy of the raft σ It is considered that this adjustment has no effect.

[0220] In another embodiment, a posture control system of an airbag-supported flexible floating raft vibration isolation system is also provided. The system includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it executes the posture control method of the airbag-supported flexible floating raft vibration isolation system described in any one of the above items.

[0221] Furthermore, the problems to be solved by this embodiment are: the existing strategy for controlling the posture of an airbag-supported flexible raft is not perfect; the reference plane is not compatible with the existing installation process; the airbag selection criteria are not comprehensive; the physical meaning of the selection process is not clear and the accuracy is insufficient; and the control strategy design has imperfections. To address the above technical problems, this embodiment provides the following technical solutions:

[0222] A method for approximate solution of raft rigid body displacement and elastic deformation displacement includes the following steps:

[0223] (1) Establish the system coordinate system O-XYZ and determine the coordinates of each measuring point. (2) Define an optimization index function to describe the effect of plane fitting. (3) Use the iterative reweighted least squares method to fit the reference plane of the raft frame, that is, the rigid body posture plane of the raft frame, and approximately separate the rigid body displacement and elastic deformation displacement of the raft frame.

[0224] A method for selecting airbags based on a multi-index comprehensive decision factor approach includes the following steps: (1) selecting the inflation and deflation of the airbags when the raft is in different positions, and performing dimensionless processing on the positive and negative index data of each airbag under different selection conditions. (2) determining the weight of each index using a subjective weighting method, performing a weighted summation on multiple airbag indexes, and calculating the comprehensive evaluation index value of each airbag to select the airbag.

[0225] A control strategy design for an airbag-supported flexible raft includes the following steps: (1) calculating the overall change trend of the raft frame and the change in the maximum total displacement of the measuring point after adjusting the airbag. (2) using the feedback amount and the reward and punishment function to calculate the reward and punishment factor of each airbag, and performing feedback correction on the comprehensive evaluation index value of each airbag. (3) calculating the maximum total displacement of all measuring points and the reward and punishment intensity of each airbag after adjusting the airbag. (4) if the maximum total displacement of all measuring points is greater than the preset control accuracy requirement, and the reward and punishment intensity of all airbags in the system is within the tolerance threshold, continue to adjust until the raft frame meets the control accuracy requirement. (5) if the reward and punishment intensity of the airbag is outside the tolerance threshold after adjustment, or the change in the maximum total displacement of all measuring points is less than the adjustment accuracy of the raft frame, in the next adjustment, reduce the control accuracy requirement and reset the reward and punishment intensity of all airbags.

[0226] This embodiment proposes a posture control strategy for an airbag-supported flexible floating raft. First, an iterative reweighted least squares method that is more in line with the new raft installation process is adopted. The reference reference plane of the raft is obtained by comprehensively considering the readings of all displacement sensors, that is, the approximate rigid body posture plane equation of the raft, thereby expressing the overall posture of the raft and solving the approximate rigid body displacement and elastic deformation displacement of the raft. Then, the adjustment effect of the raft is predicted, and the predicted value of the overall change trend of the raft after adjusting any airbag is obtained. Finally, the approximate optimal control pressure ratio parameter of the airbag is calculated, thereby obtaining four indicators, making the selection of the airbag more comprehensive; using the above indicators, a multi-indicator factor comprehensive decision-making method is adopted to obtain the comprehensive evaluation index value of the airbag to select the airbag, making the selection of the airbag more reasonable; the reward and punishment factors of each airbag are calculated based on the feedback amount after the actual adjustment, and the feedback correction of the airbag selection during the next adjustment is performed to improve the control effect of the raft.

[0227] It should also be noted that the illustrations provided in the following embodiments are merely schematic diagrams illustrating the basic concept of the present invention. The illustrations only show the components related to the present disclosure and are not drawn according to the number, shape, and size of the components in actual implementation. In actual implementation, the shape, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0228] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for controlling the position and posture of an airbag-supported flexible floating raft vibration isolation system, wherein the flexible floating raft vibration isolation system includes a raft frame supported on a plurality of airbags, characterized in that: The posture control method comprises: Acquiring actual observation positions of a plurality of preset measuring points on the raft, wherein the plurality of measuring points at least include a plurality of measuring points that are set in a one-to-one correspondence with the airbags; According to the actual observed positions of the plurality of measuring points, an iterative reweighted least squares method is used to fit and obtain a reference datum plane of the raft; wherein, in the iterative reweighted least squares method, an error of the measuring point is obtained according to the deviation between the actual observed position of the measuring point and the position on the reference datum plane, an error weight is assigned to the error of each measuring point, and an optimization goal is to minimize the weighted sum of the errors of the plurality of measuring points; Obtaining a rigid body displacement corresponding to the measuring point based on a positional distance between the measuring point on the reference datum plane and the ideal posture plane, and obtaining an elastic deformation displacement corresponding to the measuring point based on the actual observation position and the rigid body displacement, thereby obtaining a rigid body displacement and an elastic deformation displacement corresponding to the airbag; An index system is constructed according to the rigid body displacement, elastic deformation displacement and pressure corresponding to the airbag, and the airbag to be adjusted is determined based on the index system and the posture control is performed.

2. The method for controlling the position and posture of the airbag-supported flexible floating raft vibration isolation system according to claim 1, characterized in that: According to the actual observation positions of the plurality of measuring points, the reference datum plane of the raft is obtained by using an iterative reweighted least squares fitting method, specifically comprising: Setting the initial error weights of the plurality of measuring points to 1, obtaining the residual between the actual observed position of the measuring point and the regression fitting value in each iteration, and updating the error weight using the Huber weight function according to the residual of the measuring point; When the maximum value of the residual difference of the plurality of measuring points relative to the residual difference after the previous iteration is less than the adjustment accuracy of the raft, or the number of iterations reaches the maximum number of iterations, the iteration ends and the reference datum plane is obtained.

3. The method for controlling the position and posture of the airbag-supported flexible floating raft vibration isolation system according to claim 2, characterized in that: The Huber weight function is defined as: in, is the error weight of the jth measurement point at the n+1th iteration; is the residual error of the jth measurement point after the nth iteration; γ is the clipping value, which is numerically equal to the adjustment accuracy σ of the raft.

4. The method for controlling the position and posture of an airbag-supported flexible floating raft vibration isolation system according to any one of claims 1 to 3, wherein: The index system constructed according to the pressure corresponding to the airbag specifically includes: Performing a static analysis on the flexible floating raft vibration isolation system to obtain a moment force balance equation for the plurality of airbags, and obtaining an optimal control pressure for each airbag with the goal of minimizing the pressure variance of the plurality of airbags under the moment force balance equation; A pressure ratio parameter of the airbag is obtained according to a ratio of the actual pressure of the airbag to the optimal control pressure, and the indicator system includes the pressure ratio parameter.

5. The method for controlling the position and posture of an airbag-supported flexible floating raft vibration isolation system according to any one of claims 1 to 3, wherein: The construction of the indicator system also includes: Obtain the rigid body displacement change corresponding to all measuring points after any airbag inflation adjustment and the inflation correction coefficient corresponding to all measuring points after any airbag inflation adjustment, where the inflation correction coefficient is the proportional coefficient between the actual observed position change of the measuring point after any airbag inflation adjustment and the rigid body displacement change; obtain the predicted value of the overall inflation adjustment trend of any airbag by multiplying the rigid body displacement change of the measuring point by the inflation correction coefficient and summing the results for all measuring points; Accordingly, the rigid body displacement change corresponding to all measuring points after any airbag deflation adjustment and the deflation correction coefficient corresponding to all measuring points after any airbag deflation adjustment are obtained, where the deflation correction coefficient is the proportional coefficient between the actual observed position change of the measuring point after any airbag deflation adjustment and the rigid body displacement change; the overall change trend prediction value of the deflation adjustment of any airbag is obtained by multiplying the rigid body displacement change of the measuring point by the deflation correction coefficient and summing the results at all measuring points; According to actual observation positions of the plurality of measuring points, a predicted value of the overall change trend of the inflation adjustment of the airbag or a predicted value of the overall change trend of the deflation adjustment is selected and added to the indicator system.

6. The method for controlling the position and posture of the airbag-supported flexible floating raft vibration isolation system according to claim 5, characterized in that: Initially, all airbags are inflated and deflated, and the experiment of adjusting the airbags is carried out from the ideal posture plane. The inflation correction coefficients and the initial rigid body displacement changes corresponding to all measuring points after any airbag inflation adjustment are obtained, as well as the deflation correction coefficients and the initial rigid body displacement changes corresponding to all measuring points after any airbag deflation adjustment are obtained. When the airbag to be adjusted is determined for the first time and posture control is performed, the predicted value of the overall change trend of inflation adjustment in the index system of the airbag is determined based on the inflation correction coefficients of all measuring points corresponding to the airbag and the initial rigid body displacement change, and the predicted value of the overall change trend of deflation adjustment is determined based on the deflation correction coefficients of all measuring points corresponding to the airbag and the initial rigid body displacement change; After the airbag to be adjusted is determined and its posture is controlled, the rigid body displacement change corresponding to the adjusted airbag is updated according to the actual rigid body displacement change of all measuring points after adjustment for the next adjustment judgment.

7. The method for controlling the position and posture of an airbag-supported flexible floating raft vibration isolation system according to any one of claims 1 to 3, wherein: Based on the index system, a multi-index factor comprehensive decision-making method is used to determine the airbag to be adjusted and perform posture control, specifically including: Each indicator in the indicator system is dimensionally processed, and then a subjective weighting method is used to assign an indicator weight to each indicator. A comprehensive evaluation index is obtained by weighted summing the dimensionless indicators, and the airbag to be adjusted is determined based on the comprehensive evaluation index. According to the actual observation positions of the plurality of measuring points, it is determined whether to perform inflation adjustment or deflation adjustment on the airbag to be adjusted so as to perform posture control.

8. The method for controlling the position and posture of the airbag-supported flexible floating raft vibration isolation system according to claim 7, characterized in that: Also includes: Correcting the comprehensive evaluation index of the airbag by a reward and punishment factor, and determining the airbag to be adjusted by using the comprehensive evaluation index after correction of the reward and punishment factor; The reward and punishment factors corresponding to all airbags are initially 1; After determining the airbag to be adjusted and performing posture control at any time, the adjustment effect index is obtained, and the reward and punishment factors corresponding to the adjusted airbag are updated using a pre-defined reward and punishment function based on the adjustment effect index for the next adjustment judgment.

9. The method for controlling the position and posture of the airbag-supported flexible floating raft vibration isolation system according to claim 8, characterized in that: Also includes: After any airbag to be adjusted is determined and the posture is controlled, it is determined whether the maximum total displacement of all measuring points relative to the ideal posture plane is greater than the preset control accuracy requirement. If not, the adjustment is completed; If so, obtain the product of all reward and punishment factors corresponding to the airbags, determine whether the minimum value of the product in all airbags is less than the preset threshold, and determine whether the change in the maximum total displacement of all measuring points relative to the displacement before adjustment is less than the adjustment accuracy of the raft. If neither of the two situations occurs, continue to adjust. If at least one situation occurs, reduce the control accuracy requirement, and / or reset the reward and punishment factors corresponding to all airbags and readjust.

10. A posture control system for an airbag-supported flexible floating raft vibration isolation system, characterized in that: The system includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the posture control method of the airbag-supported flexible floating raft vibration isolation system according to any one of claims 1 to 9 is executed.

Images