Left eigenvector configuration optimization method for controlling the excitation intensity of ship frame structures
Through displacement-velocity feedback control and feature vector optimization, the excitation and noise control problems in ships are solved, and the low-frequency vibration suppression and real-time control of excitation intensity is realized, reducing maintenance and manufacturing costs.
Patent Information
- Application Number
- CN202211307816.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-10-24
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-10-24
AI Technical Summary
When the prior art combines passive and active control methods to suppress vibration and noise in ships, it is difficult to implement, increases manufacturing cost and manufacturing cycle, and requires a lot of maintenance, making it difficult to effectively control the impact of ship excitation and noise on crew health. .
By applying displacement-speed feedback control, the left eigenvector configuration of the ship mount structure is optimized, and the theory of eigenvalue and eigenvectors is used, combined with matlab and isight for optimization, the optimal feature structure configuration node is selected, and the feedback gain matrix is adjusted in real time. Control the excitation intensity.
Low-frequency vibration suppression of the ship mount structure is achieved, the optimal configuration nodes are quickly screened out, the excitation intensity is reduced, the ship operates in a low excitation state, and maintenance needs and manufacturing costs are reduced.
Smart Images

Figure CN115639753B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of ship vibration control, in particular to a left eigenvector configuration optimization method for controlling the excitation intensity of a ship plate frame structure. Background Art
[0002] When a ship is sailing on the ocean, it is bound to produce vibrations due to external excitation, which will not only damage the ship's structure, but also cause adverse effects on the physical and mental health of the crew due to the noise caused by the vibrations.
[0003] With the advancement of manufacturing technology, the design of vibration and noise suppression of ships has gradually been put on the agenda. In the prior art, passive control methods and active control methods are adopted at the same time to suppress vibration and noise of ships in navigation. The passive control method is more effective in eliminating high-frequency vibrations. Its specific implementation plan is: adding a damping vibration reduction device in the cabin, or optimizing the steel used to prepare the hull. The active control method is more effective in eliminating low-frequency vibrations. Its specific implementation plan is: the force generating device applies active intervention force to the local area of the cabin side wall to eliminate low-frequency vibrations. With the combination of passive control method and active control method, the vibration excitation problem of the ship can be effectively solved. However, the implementation difficulty is relatively large, which not only greatly increases the manufacturing cost of the ship, but also prolongs the manufacturing cycle. In addition, after the ship is put into actual operation, a large amount of manpower and material resources are required to perform maintenance operations on the damping vibration reduction device or the force generating device. Therefore, it is urgent for technicians to solve the above problems. Summary of the invention
[0004] Therefore, in view of the above-mentioned existing problems and defects, the research team of the present invention collected relevant information, conducted multi-faceted evaluations and considerations, and continuously explored and designed improvements by the research team members, which ultimately led to the emergence of the left eigenvector configuration optimization method for controlling the excitation intensity of the ship frame structure.
[0005] In order to solve the above technical problems, the present invention relates to a left eigenvector configuration optimization method for controlling the excitation strength of a ship frame structure, assuming that a typical ship frame structure is a structural system with n-order degrees of freedom, and its mass, damping and stiffness matrices are M, C, K, respectively, and are all n×n symmetric matrices, where M is a positive definite matrix, and C and K are semi-positive definite matrices;
[0006] The left eigenvector configuration optimization method for controlling the excitation intensity of the ship frame structure includes the following steps:
[0007] S1. Apply displacement-velocity feedback control to the ship frame structure and establish its system structure equation;
[0008] The force matrix of displacement-velocity feedback control is set to B, the velocity feedback matrix is set to S, and the displacement feedback matrix is set to R. When the control ship frame structure is subjected to displacement-velocity feedback control, its system equation is expressed as follows:
[0009]
[0010] S2, introduce the stiffness coefficient, and configure the eigenvalue and eigenvector of the ship frame structure;
[0011] According to the stiffness coefficient, the modal vibration shape of the control ship frame structure is defined as r;
[0012] Before the displacement-velocity feedback control is applied to the control ship frame structure, the eigenvalue of the left eigenvector is expressed by the following equation:
[0013]
[0014] When displacement-velocity feedback control is applied to the control ship frame structure, the eigenvalue of its left eigenvector is expressed by the following equation:
[0015]
[0016] The left eigenvector configured here is the variable to be optimized next. The left eigenvector of the corresponding node is configured according to the location and size of the excitation force.
[0017] When configuring the left eigenvector partial eigenstructure of the control ship frame structure, it is assumed that the first n eigenvalue configurations of the system are expressed by equations:
[0018]
[0019] The remaining eigenvalues remain unchanged, and the eigenvalues of each system are different after applying displacement-velocity feedback control to control the ship frame structure;
[0020] S3, measuring and obtaining the admittance matrix of the ship frame structure;
[0021] S4, screening out the optimal characteristic structure configuration nodes of the ship frame structure;
[0022] S5. Solve the feedback gain matrix of the ship frame structure and use it to determine its excitation intensity.
[0023] As a further improvement of the technical solution of the present invention, in step S3, the admittance matrix of the ship frame structure is as follows:
[0024]
[0025] M iis a variable, and according to the set stiffness coefficient range, it is combined with other variables to optimize together and select the optimal configuration result;
[0026] H(η k )=H(x) (5);
[0027] Multiplying both sides of equation (3) by the transpose of H(x) yields the following expression:
[0028] r k =H(x) T (η k S+R)B T r k (6).
[0029] As a further improvement of the technical solution of the present invention, step S4 includes the following sub-steps:
[0030] S41, assuming that the excitation force of displacement-velocity feedback control is δ i,j , where i represents the modal order and j represents the number of nodes;
[0031] S42, simplifying equation (6) into equation (7) and equation (8) respectively;
[0032] γ k =(η k S+R)B T r k (7)
[0033] r k =H(x) T γ k (8)
[0034] Combine equations (7) and (8) to solve the left eigenvector of the pre-configured ship frame structure: ;
[0035] S43. Through the joint optimization of MATLAB and ISHTT, the frequency response function comparison diagram of the ship frame structure before and after the displacement-velocity feedback control is simulated.
[0036] As a further improvement of the technical solution of the present invention, in step S43, the position of the external excitation, the position of the configured left eigenvector, the stiffness coefficient, and the seawater temperature and weather temperature of the ship sailing on that day are set as variables in MATLAB. Each time an excitation is generated, the corresponding node is used to configure the left eigenvector at the position of this node. In isight, a genetic algorithm is used for optimization. The position range of the excitation force is set to different values according to the degrees of freedom of the controlled system, and it is set as an input variable. The range of the position of the configured left eigenvector is set to the same interval range according to the size of the excitation force, and is also set as an input variable. Each time an excitation is generated, the frequency response function obtained by configuring the corresponding left eigenvector is set as an output variable; the stiffness coefficient, seawater temperature, and weather temperature are all set as output variables, the population size in the genetic algorithm is set to a multiple of the number of variables, and the number of generations can be set according to the system default; when the optimization is performed, the number of iterations is the population size multiplied by the number of generations, and finally the optimal result is screened out according to the data at the iteration.
[0037] As a further improvement of the technical solution of the present invention, in step S42, assuming that the ship frame structure is subjected to displacement-velocity feedback control, in any area under the action of the excitation force, the generated left eigenvector is orthogonal to the excitation force vector, then .
[0038] As a further improvement of the technical solution of the present invention, in step S5, it is assumed that the modal vibration type η of the closed-loop system of the ship frame structure is k is any left eigenvector γ k A linear combination of k is an arbitrarily selected coefficient; the above expression is expressed as follows:
[0039] P k Y=γ k (9)
[0040] For the remaining 2n-p equations before control, they can be simplified as:
[0041] Q k Y=0 (10)
[0042] Equation (9) and equation (10) are combined to obtain the feedback gain matrices S and R of the ship frame structure.
[0043] According to common knowledge, the right eigenvector is an inherent property of the structure and has nothing to do with whether it is acted upon by an excitation force or the magnitude of the excitation force, while the left eigenvector represents the ability of the structure to resist external excitation. Under a specific excitation force, the excitation intensity of the structure can be precisely controlled by configuring the left eigenvector.
[0044] In the technical solution disclosed in the present invention, the left eigenvector configuration theory is innovatively applied to the multi-input and multi-output excitation model of the ship frame structure, and the left eigenvector configuration of some poles is demonstrated under the condition of multi-input control. After actual numerical deduction, the left eigenvector configuration optimization method can perfectly match and reflect various external excitations, not only realizing the optimal configuration of the eigenvalues (natural frequency and damping ratio) of the ship frame structure, but also controlling the magnitude or direction of the excitation force output to the ship frame structure in real time according to the derived left eigenvector, ensuring that the ship frame structure is always maintained in a low excitation intensity state. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0046] Figure 1 It is a schematic diagram comparing the amplitudes of the frequency response functions of the ship frame structure before and after the displacement-velocity feedback control in the present invention. DETAILED DESCRIPTION
[0047] In this embodiment, it is assumed that the typical ship frame structure is a structural system with n degrees of freedom, and its mass, damping and stiffness matrices are M, C, K, respectively, and are all n×n symmetric matrices, where M is a positive definite matrix, and C and K are semi-positive definite matrices.
[0048] The left eigenvector configuration optimization method for controlling the excitation intensity of the ship frame structure includes the following steps:
[0049] S1. Apply displacement-velocity feedback control to the ship frame structure and establish its system structure equation;
[0050] The force matrix of displacement-velocity feedback control is set to B, the velocity feedback matrix is set to S, and the displacement feedback matrix is set to R. When the control ship frame structure is subjected to displacement-velocity feedback control, its system equation is expressed as follows:
[0051]
[0052] S2. According to the stiffness coefficient of the structure, the eigenvalue and eigenvector configuration of the ship frame structure is performed;
[0053] Different structures have different stiffness coefficients, so the influence of the excitation force is also different. The stiffness coefficient is set as the optimization input variable, so that the magnitude of the excitation can be controlled more accurately. According to the stiffness coefficient, the modal vibration shape of the ship frame structure is defined as r;
[0054] Before the displacement-velocity feedback control is applied to the control ship frame structure, the eigenvalue of the left eigenvector is expressed by the following equation:
[0055]
[0056] When displacement-velocity feedback control is applied to the control ship frame structure, the eigenvalue of its left eigenvector is expressed by the following equation:
[0057]
[0058] When configuring the left eigenvector partial eigenstructure of the control ship frame structure, it is assumed that the first n eigenvalue configurations of the system are expressed by equations:
[0059]
[0060] The remaining eigenvalues remain unchanged, and the eigenvalues of each system are different after applying displacement-velocity feedback control to control the ship frame structure;
[0061] S3, measuring and obtaining the admittance matrix of the ship frame structure;
[0062] The admittance matrix of the ship frame structure is as follows:
[0063]
[0064] M i is a variable, and according to the set stiffness coefficient range, it is combined with other variables to optimize together and select the optimal configuration result;
[0065] H(η k )=H(x) (5);
[0066] Multiplying both sides of equation (3) by the transpose of H(x) yields the following expression:
[0067] r k =H(x) r (η k S+R)B T r k (6);
[0068] S4, screening out the optimal characteristic structure configuration nodes of the ship frame structure;
[0069] It includes the following sub-steps:
[0070] S41, assuming that the excitation force of displacement-velocity feedback control is δ i,j , where i represents the modal order and j represents the number of nodes;
[0071] S42, simplifying equation (6) into equation (7) and equation (8) respectively;
[0072] γ k =(η k S+R)B T r k (7)
[0073] r k =H(x) T γ k (8)
[0074] Combine equations (7) and (8) to solve the left eigenvector of the pre-configured ship frame structure:
[0075] Assuming that the ship frame structure is subject to displacement-velocity feedback control, the left eigenvector generated in any area under the action of the excitation force is is orthogonal to the excitation force vector, then
[0076] S43, through the joint optimization of MATLAB and ISHTT, to simulate the frequency response function comparison diagram of the ship frame structure before and after the displacement-velocity feedback control;
[0077] In MATLAB, the position of the external excitation, the position of the configured left eigenvector, the stiffness coefficient, and the seawater temperature and weather temperature of the ship on that day are set as variables. For each excitation, the corresponding node is used to configure the left eigenvector at the position of this node. In ISEIGHT, the genetic algorithm is used for optimization. The position range of the excitation force is set to different values according to the degrees of freedom of the controlled system and set as the input variable. The range of the configured left eigenvector position is set to the same interval range according to the magnitude of the excitation force and is also set as the input variable. For each excitation, the frequency response function obtained by configuring the corresponding left eigenvector is set as the output variable; the stiffness coefficient, seawater temperature, and weather temperature are all set as output variables, the population size in the genetic algorithm is set to a multiple of the number of variables, and the number of generations can be set according to the system default; during the optimization, the number of iterations is the population size multiplied by the number of generations, and finally the optimal result is screened out based on the data at the iteration.
[0078] S5. Solve the feedback gain matrix of the ship frame structure and use it to determine its excitation intensity;
[0079] Assuming that the modal vibration shape of the closed-loop system of the ship frame structure is k is any left eigenvector γ k A linear combination of k is an arbitrarily selected coefficient; the above expression is expressed as follows:
[0080] P k Y=γ k (9)
[0081] For the remaining 2n-p equations before control, they can be simplified as:
[0082] Q k Y=0 (10)
[0083] Equation (9) and equation (10) are combined to obtain the feedback gain matrices S and R of the ship frame structure.
[0084] In the technical solution disclosed in the present invention, the configuration left eigenvector theory is innovatively applied to the multi-input and multi-output excitation model of the ship frame structure, and under the condition of multi-input control, the left eigenvector configuration of some poles is demonstrated.
[0085] The application of the above scheme can effectively suppress the low-frequency vibration of the ship frame structure, and can quickly and accurately screen the location of the optimal configuration node. For the control system, the eigenvector of the closed-loop system is not arbitrarily configured. In the actual active control application, usually only a few orders of eigenvalues or eigenvectors of the structure need to be reconfigured, so the partial eigenstructure configuration method that only configures some eigenvalues and eigenvectors while keeping other orders unchanged is meaningful in actual vibration active control.
[0086] Since the positions of different structural systems are excited will have different effects on the system, when the position of the excitation force is known, the left eigenvector of the node position is controlled to be zero, so that it is orthogonal to the external excitation force vector, and the vibration mode of the left eigenvector of this order will not be excited. Controlling different nodes will have different effects, so how to effectively and accurately control and produce the effect of maximizing vibration reduction is a process that requires continuous attempts. The present invention uses matlab and isight to jointly optimize, saves the trouble of manual screening, and improves accuracy and efficiency.
[0087] Here, it is important to note that, since the optimal configuration node of the ship frame structure has been selected through numerical optimization, it means that the action area of the intervention excitation force has been known and the optimal configuration node has been identified. Therefore, in the process of actively applying excitation force control to the ship frame structure, it is necessary to always pay attention to the difference in the natural frequency and damping ratio of the ship frame structure before and after the excitation force acts, and adjust the magnitude and direction of the intervention excitation force acting at the optimal configuration node in real time until its left eigenvector approaches zero. When the multiplication of the intervention excitation force and the left eigenvector is equal to zero or close to zero, it means that the vibration intensity of the ship frame structure is reduced, that is, the active control of the excitation intensity of the ship frame structure is finally achieved.
[0088] The specific implementation example is as follows: the ship frame structure is simplified into a simple 10-degree-of-freedom multi-input control system.
[0089] As shown in the matrix below, n=10, the position where the excitation force acts is set as the input variable, the range is [1, 9], the corresponding position where the left eigenvector is zero is set as the input variable, the range is [1, 9], and the amplitude of the frequency response function is used as the output variable.
[0090] As shown in the matrix below, n = 10, m = 2, the position of the excitation force is set to the input variable range is [1, 9], the sea water temperature range is Te sea [10-30], weather temperature Te sky [0-40] Set them as integer data. Set the corresponding left eigenvector zero position as input variable, the range is [1, 9], B = I n , M=4I n , C=4I n , K is as follows i∈[1,9]:
[0091]
[0092] In the above embodiment, in the specific deduction, two eigenvalues are configured, and other eigenvalues and eigenvectors remain unchanged. The first two eigenvalues {0.0000, -0.0251} are configured to {-0.4+0.7i, -0.4-0.7i}. The other eigenvalues remain unchanged. After the isight iteration is completed, the optimal temperature and stiffness coefficient are obtained, and finally the optimal node position is obtained, and the corresponding modal vibration shape is configured to make it orthogonal to the excitation force, so as to achieve the purpose of eliminating the excitation force or minimizing the impact thereof.
[0093] Figure 1The figure shows a schematic diagram of the comparison of the frequency response function amplitudes of the ship frame structure before and after the displacement-velocity feedback control in the present invention. It can be seen that when the excitation force acts on the fifth node, the amplitude of the frequency response function after control is greatly reduced, which controls the vibration acting on the fifth node of the ship frame, achieves our goal, and verifies the correctness and effectiveness of the theory of the present invention.
[0094] After actual numerical simulation, it is found that the left eigenvector configuration optimization method can perfectly match and reflect various external excitations. It can not only realize the optimal configuration of the eigenvalues (natural frequency and damping ratio) of the ship frame structure, but also can control the magnitude or direction of the exciting force output to the ship frame structure in real time according to the derived left eigenvector, ensuring that the ship frame structure always maintains a low excitation intensity state.
[0095] The definitions of the parameters in the above text are summarized as follows: M is the mass matrix, C is the damping matrix, K is the stiffness matrix, z(t): node displacement vector, B: control force matrix, S is the velocity feedback gain matrix, R is the displacement feedback gain matrix, : kth eigenvalue, η k : The kth eigenvector, H(η k ):admittance response. δ:external excitation, δ 1,1 : Acts on the first node of the first-order mode vibration shape, Eigenvector after control at the corresponding node, ε: damping ratio, Te sea Sea water temperature, Te sky Air temperature.
[0096] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A left eigenvector configuration optimization method for controlling the excitation intensity of a ship frame structure, characterized in that: Assume that the typical ship frame structure is a structural system with n-degrees of freedom, and its mass, damping and stiffness matrices are M, C, K, respectively, and are all n×n symmetric matrices, where M is a positive definite matrix, and C and K are semi-positive definite matrices; The left eigenvector configuration optimization method for controlling the excitation intensity of the ship frame structure comprises the following steps: S1, applying displacement-velocity feedback control to the ship frame structure and establishing its system structure equation; The force matrix of displacement-velocity feedback control is set to B, the velocity feedback matrix is set to S, and the displacement feedback matrix is set to R; when the control ship frame structure is subjected to displacement-velocity feedback control, its system equation is expressed as follows: S2. Introducing stiffness coefficients, and configuring eigenvalues and eigenvectors for the ship frame structure; According to the stiffness coefficient, the modal vibration shape of the control ship frame structure is defined as r; Before the displacement-velocity feedback control is applied to the control ship frame structure, the eigenvalue of the left eigenvector is expressed by the following equation: When displacement-velocity feedback control is applied to the control ship frame structure, the eigenvalue of the left eigenvector is expressed by the following equation: The left eigenvector configured here is the variable to be optimized next. The left eigenvector of the corresponding node is configured according to the location and size of the excitation force. When the left eigenvector partial eigenstructure configuration is performed on the control ship frame structure, it is assumed that the first n eigenvalue configurations of the system are expressed by equations as follows: The remaining eigenvalues remain unchanged, and the eigenvalues of each system are different after applying displacement-velocity feedback control to the control ship frame structure; S3. Measure and obtain the admittance matrix of the ship frame structure; the admittance matrix of the ship frame structure is as follows: M i is a variable, and according to the set stiffness coefficient range, it is combined with other variables to optimize together and select the optimal configuration result; H(η k )=H(x) (5); Multiplying both sides of equation (3) by the transpose of H(x) yields the following expression: r k =H(x) T (η k S+R)B T r k (6); S4, screening out the optimal characteristic structure configuration node of the ship frame structure; comprising the following sub-steps: S41, assuming that the excitation force of the displacement-velocity feedback control is δ i,j , where i represents the modal order and j represents the number of nodes; S42, simplifying equation (6) into equation (7) and equation (8) respectively; c k =(the k S+R)B T r k (7) r k =H(x) T c k (8) Combine equations (7) and (8) to solve the preconfigured left eigenvector of the ship frame structure: S43, through the joint optimization of MATLAB and ISHIGHT, to simulate the frequency response function comparison diagram of the ship frame structure before and after the displacement-velocity feedback control; in MATLAB, the position of the external excitation, the position of the left eigenvector, the stiffness coefficient, and the seawater temperature and weather temperature of the ship sailing on that day are set as variables, and each time an excitation is generated, the corresponding node is used to configure the left eigenvector at the position of this node. In ISHIGHT, the genetic algorithm is used for optimization, and the position range of the excitation force is set to different values according to the degree of freedom of the controlled system, and it is set as an input variable. The range of the position of the configured left eigenvector is set to the same interval range according to the magnitude of the excitation force, and is also set as an input variable. Each time an excitation is generated, the frequency response function obtained by configuring the corresponding left eigenvector is set as an output variable; the stiffness coefficient, seawater temperature, and weather temperature are all set as output variables, the population size in the genetic algorithm is set to a multiple of the number of variables, and the number of generations is set according to the system default; when the optimization is in progress, the number of iterations is the population size multiplied by the number of generations, and finally the optimal result is screened out according to the data at the iteration; S5. Solve the feedback gain matrix of the ship frame structure and use it to determine its excitation intensity.
2. The left eigenvector configuration optimization method for controlling the excitation intensity of a ship frame structure according to claim 1 is characterized in that: In step S42, assuming that the ship frame structure is subjected to displacement-velocity feedback control, the left eigenvector generated in any region under the action of the excitation force is is orthogonal to the excitation force vector, then 3. The left eigenvector configuration optimization method for controlling the excitation intensity of a ship frame structure according to claim 2 is characterized in that: In step S5, it is assumed that the modal vibration type η of the closed-loop system of the ship frame structure is k is any left eigenvector γ k A linear combination of k is an arbitrarily chosen coefficient; the expression is: P k Y=γ k (9) For the remaining 2n-p equations before control, they can be simplified as: Q k Y=0 (10) Equation (9) and equation (10) are combined to obtain the feedback gain matrices S and R of the ship frame structure.
