Multi-channel iteration method based on multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system decoupling method and storage medium

Abstract

The application discloses a multi-channel iteration method based on a multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system decoupling method and a storage medium, and belongs to the technical field of electro-hydraulic servo control. In order to solve the problems of multi-degree-of-freedom coupling interference and insufficient waveform reproduction accuracy of a multi-channel electro-hydraulic servo vibration test bench, the application adopts an H1 non-parametric identification method to perform system identification to obtain a multi-channel transfer function matrix, and then designs a decoupling controller based on the transfer function matrix to eliminate the coupling interference between the multi-channels; finally, the control input is dynamically optimized through correction iteration, so that the purpose of improving the control accuracy of the multi-degree-of-freedom electro-hydraulic servo vibration table and accurately simulating the structural response under the action of an earthquake is achieved. The test data of the application is accurate and reliable. The application is suitable for the fields of civil engineering, traffic and machinery.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of electro-hydraulic servo control, and particularly relates to a multi-channel iteration method based on a multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system decoupling method and a storage medium. BACKGROUND

[0002] In the field of structural seismic test, the earthquake simulation shaking table is the core equipment for reproducing earthquake action and studying structural dynamic response. With the rapid development of engineering technology itself, the environmental requirements for vibration test simulation and simulation are becoming more and more complex and strict, especially the dynamic test environment, so the single-axis single-point excitation vibration test using only one axis and a single excitation system cannot be used. Compared with this, the advantages of the multi-degree-of-freedom electro-hydraulic vibration test bench mainly include the following:

[0003] (1) The volume and weight of the tested piece are very large, and the force provided by the single-axis method cannot drive the equipment, so a multi-axis multi-excitation vibration table is used to achieve it;

[0004] (2) The structure of the tested piece is complex, or has certain requirements for temperature, humidity and other comprehensive environment, and multi-dimensional and multi-degree-of-freedom vibration needs to be simulated, so a multi-axis vibration table needs to be used to achieve it;

[0005] (3) For some special test pieces of structure, such as slender body structure and flexible structure, a multi-axis vibration table needs to be used to meet the test conditions and prevent the test from causing damage to the test piece structure;

[0006] (4) Compared with the single-axis vibration test table, the multi-axis vibration table is often more efficient in acceleration load test.

[0007] However, there are two key technical problems in the multi-degree-of-freedom electro-hydraulic vibration table: one is the low waveform reproduction accuracy caused by the low frequency bandwidth of the hydraulic vibration table system and various nonlinear factors; the other is the coupling interference between the degrees of freedom in the multi-degree-of-freedom vibration table servo system.

[0008] In the prior art, the iterative learning control technology is used to improve the accuracy of the single-channel vibration table, which corrects the control input through repeated iteration, and the output gradually approaches the expected trajectory. However, the traditional iteration method does not design a decoupling mechanism for the multi-channel coupling problem, and in the multi-axis scene, the iterative correction of a single degree of freedom will interfere with the response of other degrees of freedom due to the coupling effect, resulting in limited overall control accuracy improvement.

[0009] Therefore, there is an urgent need for a control algorithm that can balance multi-channel decoupling and high-precision iterative correction to solve the multi-degree-of-freedom coupling interference and insufficient waveform reproduction accuracy of the multi-channel electro-hydraulic servo vibration test table. SUMMARY

[0010] The present application is to solve the problem of multi-degree-of-freedom coupling interference and insufficient waveform reproduction accuracy of a multi-channel electro-hydraulic servo vibration test bench.

[0011] The multi-channel iteration method based on the decoupling method of the servo system of the multi-degree-of-freedom electro-hydraulic servo vibration test bench comprises the following steps:

[0012] Obtaining the transfer function matrix G(s) of the multi-degree-of-freedom electro-hydraulic servo vibration test bench; based on the transfer function matrix G(s), using the relative gain array technique, calculating the relative gain array between each output and input of the electro-hydraulic servo vibration test bench system, and identifying the best input-output pairing by relative gain analysis; after determining the best pairing interaction, using the LMI algorithm to minimize the relative gain values of other interactions; expressing the obtained decoupling gain values in the form of a decoupling matrix to design a decoupling system G N (s) and decoupling;

[0013] For a certain SISO system after decoupling, according to the expected output command y d (t), obtaining the first round command input u0(t) in the through mode of the SISO system, inputting the command input to the input end of the SISO system to obtain the response output, calculating the transfer function h k (f) of the current SISO system according to the Fourier spectrum of the command input and the response output, f representing frequency; using the Fourier transform of the difference between the expected output command and the response output Δy k (f) to solve the input difference Δu k (f) in the frequency domain, and then performing inverse Fourier transform to obtain Δu k (t) in the time domain, and obtaining a new input command u k (t) based on Δu k+1 (t), continuing to iterate by inputting the command input to the input end of the SISO system until the iteration converges, ending the iteration and saving the current round input command u i (t);

[0014] The other SISO systems after decoupling are processed in the same way.

[0015] Further, in the process of obtaining the transfer function matrix G(s) of the multi-degree-of-freedom electro-hydraulic servo vibration test bench, the H1 non-parametric identification method is used to obtain the transfer function matrix.

[0016] Further, the process of expressing the obtained decoupling gain values in the form of a decoupling matrix to design a decoupling system G N (s) comprises:

[0017] The decoupling matrix structure is constructed for the multi-degree-of-freedom electro-hydraulic servo vibration test bench, the optimal decoupling parameters are solved by the interior point algorithm, and the determined decoupling matrix D(s) is obtained by substituting the optimal decoupling parameters into the decoupling matrix, so as to design the decoupling system G N (s) = G(s) · D(s).

[0018] Further, the command input is introduced into the input end of the SISO system to obtain the response output, and the transfer function h k (f) of the current SISO system is calculated according to the Fourier spectrum of the command input and the response output.

[0019] The command input u k (f) is introduced into the input end of the SISO system in the kth iteration. k (f), and the response output y k (f) is obtained; the command input u k (f) and the response output y k (f) are calculated according to the Fourier spectrum of the command input and the response output. k (f) = y k (f) / u k (f).

[0020] Further, the Fourier transform Δy k (f) of the difference between the expected output command and the response output is used to solve the input difference in the frequency domain.

[0021] Further, the new input command is obtained based on Δu k (t). K is an influence factor, and the value is between 0 and 1; and γ is a step adjustment coefficient.

[0022] Preferably, the influence factor K adopts an adaptive adjustment strategy: when the iteration is in the initial stage, K is 0.6-0.8; and when the iteration is in the convergence stage, K is adjusted to 0.2-0.4.

[0023] Further, the command input is introduced into the input end of the SISO system for continuous iteration until the iteration converges, and the iteration convergence condition is judged according to the correlation coefficient of the response output y i (t) and the expected output command y d (t) in the current time domain.

[0024] Preferably, when the correlation coefficient of the response output y i (t) and the expected output command y d (t) is greater than or equal to 0.9, it is considered that the iteration converges.

[0025] A computer storage medium, the storage medium has a computer program, the computer program is loaded and executed by the processor to realize the multi-channel iteration method of the decoupling method of the multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system.

[0026] Beneficial effects:

[0027] The present application solves the problem of low precision and poor robustness of the existing decoupling method based on the coupling problem between each degree of freedom of the multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system. The MD-EHSV-DS decoupling method adopted in the present application optimally pairs through the relative gain array (RGA) technology and optimizes through the linear matrix inequality (LMI) algorithm, eliminates the coupling interference between multiple degrees of freedom, optimizes the control accuracy of each degree of freedom independently, and reduces the interference of other degrees of freedom on the single degree of freedom response to below 5% after decoupling, greatly improving the decoupling accuracy.

[0028] The present application solves the problem of insufficient waveform reproduction accuracy of the multi-degree-of-freedom electro-hydraulic servo vibration test bench. In the present application, the multi-channel transfer function matrix is accurately obtained by H1 non-parametric identification method, and the input is dynamically corrected by combining iterative learning control, which effectively reduces the amplitude attenuation and phase lag, and the single-channel sinusoidal waveform bandwidth can reach 200Hz, and the multi-channel sinusoidal waveform bandwidth can reach 100Hz, and the response / command amplitude ratio is not less than -3dB.

[0029] The present application solves the problem of low efficiency of the existing iterative method based on the multi-channel electro-hydraulic servo vibration test bench system. In the present application, the iteration is terminated by error dynamic judgment, which avoids blind increase of iteration times, shortens the test time under the premise of ensuring the accuracy, and improves the efficiency by more than 30% compared with the traditional fixed iteration number method. The present application is suitable for civil engineering, transportation, aerospace and mechanical fields, etc. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 The flow chart of the multi-channel iterative method based on the decoupling method of the multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system.

[0031] Figure 2 The iteration principle diagram of the multi-channel iterative method based on the multi-degree-of-freedom electro-hydraulic servo vibration test bench servo system.

[0032] Figure 3 The displacement-time history curve comparison diagram before displacement-displacement iteration.

[0033] Figure 4 The displacement-time history curve comparison diagram after displacement-displacement iteration.

[0034] Figure 5 The acceleration-time history curve comparison diagram before displacement-acceleration iteration.

[0035] Figure 6 is a displacement-acceleration iteration post-acceleration time history curve comparison chart. DETAILED DESCRIPTION

[0036] The present application firstly adopts H1 non-parametric identification method to obtain the transfer function matrix G(s) of the multi-degree-of-freedom electro-hydraulic servo vibration test bench, then adopts MD-EHSV-DS decoupling method based on the transfer function matrix G(s) to eliminate the coupling interference between the degrees of freedom, and finally realizes the purpose of improving the control precision of the multi-degree-of-freedom electro-hydraulic servo vibration test bench and accurately simulating the structural response under the action of earthquake by iteratively optimizing the control input. The present application will be further described in combination with the drawings and specific embodiments.

[0037] Specific embodiment one: see Figure 1 The multi-channel iteration method based on the servo system decoupling method of the multi-degree-of-freedom electro-hydraulic servo vibration test bench in the present embodiment includes the following steps:

[0038] Step one: adopting H1 non-parametric identification method to obtain the transfer function matrix G(s) of the multi-degree-of-freedom electro-hydraulic servo vibration test bench, the elements of the transfer function matrix G(s) being the transfer functions between the degrees of freedom;

[0039] Step two: based on the transfer function matrix G(s), adopting MD-EHSV-DS decoupling method to design a decoupling controller for decoupling processing, converting the MIMO system of the multi-degree-of-freedom electro-hydraulic servo vibration test bench into multiple SISO systems;

[0040] The specific process of the MD-EHSV-DS decoupling method includes:

[0041] S201: analyzing the transfer function matrix G(s) of the identified multi-degree-of-freedom electro-hydraulic vibration table servo system, calculating the relative gain array Λ between the outputs and inputs of the electro-hydraulic servo vibration test bench system by using the relative gain array (RGA) technology, and identifying the best input-output pairing from the relative gain analysis;

[0042] The input and output of the electro-hydraulic servo vibration test bench system are respectively u i and y i , i represents a degree of freedom of the multi-degree-of-freedom electro-hydraulic vibration table, such as i=1, 2, 3 for a three-degree-of-freedom electro-hydraulic vibration table, having three inputs and three outputs. Taking three degrees of freedom (represented by 1, 2, and 3) as an example, the transfer function matrix of the three-degree-of-freedom electro-hydraulic vibration table is in the following form:

[0043]

[0044] Wherein G 11 (s), G 12(s)…G 33 (s) is the transfer function between each degree of freedom, then the output y1, y2, y3 and input u1, u2, u3 of the system can be written as the following relationship:

[0045]

[0046] S202: In order to make the application of linear matrix inequality (LMI) algorithm more convenient, the transfer function matrix G(s) of the electro-hydraulic servo vibration test bench system is converted into a state space form, and the state matrix, input matrix, output matrix and direct transfer matrix are determined;

[0047] S203: After determining the optimal pair interaction, the LMI algorithm is used to minimize the relative gain value of other interactions, reduce the influence of non-optimal pair interactions in the electro-hydraulic servo vibration test bench system, and thus improve the overall control performance;

[0048] S204: After following steps S201-S203, the obtained decoupling gain value is expressed in the form of a decoupling matrix, and a decoupling system G N (s) is designed and decoupled;

[0049] Step three: for a certain SISO system (one single channel), set the sampling frequency and signal output time, and import the expected output command y d (t);

[0050] Step four: load y d (t) in the SISO system direct mode to obtain the real response y x (t), and generate the first round command input u0(t) according to y x (t);

[0051] Step five: import the command input u0(t) into the input end of the SISO system to obtain the response output y0(t), and calculate the transfer function of the current SISO system according to the Fourier spectrum of the command input u0(t) and the response output y0(t), the calculation formula is h k (f) = y k (f) / u k (f), where u k (f), y k (f) are the command input and response output expressed in the corresponding frequency domain in the kth iteration, and f represents the frequency;

[0052] u0(t) is the initial value at the beginning of iteration, and the initial value of the response output is used for the first iteration. In the subsequent iteration process, the command input u k+1 (t) is imported into the input end of the SISO system to obtain the corresponding response output, so as to complete the iteration.

[0053] Step six: use the difference Δy between the desired output command and the response output k (t) = y d (t) - y k Fourier transform of Δy(t) Δy(f) k (f) solve the input difference in the frequency domain k (t); e(f) represents the error function between the actual output and the desired output of the SISO system.

[0054] If the error is directly corrected in the time domain, it is difficult to distinguish the influence of different frequency components and it is not possible to compensate for the nonlinearity of the system at a specific frequency. This may result in the introduction of new distortion in the corrected input signal. Based on the limitations of the above time domain error processing, the present embodiment uses the principle of inverse filtering in the frequency domain to obtain the time domain input error Δu(t) by inverting the frequency domain error. k This embodiment not only improves the convergence speed and accuracy of the iteration (the time domain input error obtained by inverse filtering in the frequency domain has been compensated for the characteristics of the system, so the input signal corrected based on this error can more directly reduce the output error and reduce the number of iterations, and since the interference of system distortion is eliminated, the final output signal is more consistent with the desired signal, i.e. the iteration accuracy is higher), but also reduces the influence of noise and interference (noise is effectively suppressed in the frequency domain by filtering, avoiding the amplification of noise in the time domain correction, which makes the time domain input error obtained by inversion more pure and the corrected input signal more stable).

[0055] Step seven: introduce an influence factor K for correcting the iteration, which is between 0 and 1, to obtain a new input command In the formula, γ is the step size adjustment coefficient, and the step size refers to the difference between u k+1 (t) and u k (t); preferably, the influence factor K uses an adaptive adjustment strategy: when the error is large at the beginning of the iteration, K takes 0.6-0.8 to speed up the convergence; when in the iteration convergence stage, K is adjusted to 0.2-0.4 to improve the control accuracy.

[0056] Step eight: repeat steps five to seven until the response output y i (t) in the current round time domain and the desired output command y d (t) meet the requirements, i.e. the iteration converges, at which point the iteration can be ended and the current round input command u i (t) is saved.

[0057] Here the concept of relative coefficient is used to determine whether the iteration converges, assuming that the desired and output signals in the time period (N1, N2) are X n , Y​n n = N1, N2, relative coefficient When K xy ≥0.90, the output signal is in good agreement with the expected signal, which can meet the precision requirement of the vibration table test, that is, the iteration is judged to be converged.

[0058] Step nine: the iteration process of the remaining SISO system (i.e. other single channels) is the same as steps three to eight.

[0059] In the embodiment, the H1 non-parametric identification method is used to obtain the transfer function matrix G(s) of the multi-degree-of-freedom electro-hydraulic servo vibration test table, and then a decoupling controller is designed based on the transfer function matrix of the multi-degree-of-freedom (multi-channel) electro-hydraulic servo vibration test table system, thereby solving the multi-degree-of-freedom coupling interference problem of the multi-channel electro-hydraulic servo vibration test table. Finally, the control input is dynamically optimized by iteration, thereby solving the problem of insufficient waveform reproduction accuracy of the multi-channel electro-hydraulic servo vibration test table.

[0060] The MD-EHSV-DS decoupling method described in the embodiment uses the relative gain array (RGA) technology to perform input-output pairing analysis and calculate the relative gain array Λ. The optimal pairing relationship between the input and output of the electro-hydraulic servo vibration test table system is determined according to the relative gain array Λ. Meanwhile, the linear matrix inequality (LMI) algorithm is also used to optimize the design of the decoupling matrix, construct the decoupling matrix, convert the decoupling problem into a convex optimization problem, establish the LMI constraint condition, and solve the optimal decoupling matrix parameters to minimize the coupling effect of the electro-hydraulic servo system.

[0061] In addition, the MD-EHSV-DS decoupling method of the present application uses the Kharitonov theorem and the BFO algorithm to design the PID controller. The robust stability interval of the PID controller parameters is determined by using the Kharitonov theorem, and the BFO algorithm is used to optimize the PID controller parameters in the interval, which can realize the robustness analysis and disturbance control of the decoupled electro-hydraulic servo vibration test table system. Specific implementation method two:

[0063] The embodiment is a computer storage medium, which stores a computer program. The computer program is loaded and executed by a processor to realize the multi-channel iteration method based on the multi-degree-of-freedom electro-hydraulic servo vibration test table servo system decoupling method.

[0064] It should be understood that the storage medium described in the embodiment includes but is not limited to magnetic storage media and optical storage media. The magnetic storage medium includes but is not limited to RAM, ROM, and other hard disks, U disks, and other storage media.

[0065] The storage medium described in the embodiment can be used for a device comprising a processor to run a computer program, and it should also be understood that such a device can also comprise other devices corresponding to units or modules with information collection, information interaction, control functions, for example, the device can also comprise a signal collection device, etc. The device includes but is not limited to a PC, a workstation, a mobile device, etc.

[0066] Embodiment

[0067] This embodiment takes a certain three-degree-of-freedom (X, Y, Z axis) electro-hydraulic servo vibration test bench as an example, and uses an El Centro wave with an acceleration peak of 35 gal to perform simulation simulation of the empty vibration table iterative learning control, and the basic principle and use steps of the method are described.

[0068] The following will be combined with the attached Figure 2 to the attached Figure 6 The present application will be described in detail. The iteration principle diagram of the multi-channel iteration method based on the servo system of the multi-degree-of-freedom electro-hydraulic servo vibration test bench is shown in Figure 2 The displacement-time history curve comparison chart before and after displacement-displacement iteration is shown in Figure 3 and Figure 4 The acceleration-time history curve comparison chart before and after displacement-acceleration iteration is shown in Figure 5 and Figure 6 .

[0069] The method of the embodiment is a multi-channel iterative test based on the decoupling method of the servo system of the multi-degree-of-freedom electro-hydraulic servo vibration test bench, which specifically includes the following steps:

[0070] Step one, obtain the transfer function matrix of the three-degree-of-freedom electro-hydraulic servo vibration test bench by H1 non-parametric identification method:

[0071]

[0072] Step two, use MD-EHSV-DS decoupling method to decouple the system, and the specific process is as follows:

[0073] First, calculate the relative gain array Lambda using RGA technology;

[0074]

[0075] From lambda 11 = 0.89 (closest to 1), lambda 22 = 0.92 (closest to 1), the relative gain analysis can determine the optimal pairing relationship: the X-axis input only dominates the X-axis output, the Y-axis input only dominates the Y-axis output, and the Z-axis input only dominates the Z-axis output;

[0076] The system transfer function matrix G(s) is converted into a state space form to determine a state matrix, an input matrix, an output matrix, and a direct transfer matrix;

[0077] Then the decoupling matrix structure is constructed as follows:

[0078]

[0079] Where d 12 , d 13 are decoupling parameters for suppressing the interference of the Y and Z axes on the X axis, d 21 , d 23 are decoupling parameters for suppressing the interference of the X and Z axes on the Y axis, and d 31 , d 32 are decoupling parameters for suppressing the interference of the X and Y axes on the Z axis.

[0080] The optimal decoupling parameters are solved by the interior point algorithm and substituted into the decoupling matrix as follows:

[0081]

[0082] Thus, the decoupled system is designed as follows:

[0083]

[0084] Step three: After the system is decoupled, the El Centro wave with an acceleration peak of 35 gal is used for the simulation of the empty vibration table iterative learning control. Taking the X axis channel as an example, the sampling frequency and signal output time are set, and the expected output command y d (t) is imported.

[0085] Step four: The expected output command y d (t) is loaded in the SISO system in a direct-through mode to obtain the real response y x (t), and the first round of command input u0(t) is generated according to y x (t).

[0086] Step five: The command input u0(t) is imported into the input end of the SISO system to obtain the response output y0(t). The transfer function of the current SISO system is calculated according to the Fourier spectrum of the command input u0(t) and the response output y0(t), and the calculation formula is h k (f) = y k (f) / u k (f), where u k (f) and y k (f) are the command input and the response output represented in the frequency domain at the kth iteration, and f represents the frequency.

[0087] Step six: The difference Δy k(t)=y d (t)-y k Fourier transform Δy of (t) k (f) Solving for the input difference in the frequency domain Then, the inverse Fourier transform is used to obtain Δu in the time domain. k (t);

[0088] Step 7: Introduce the influence factor K, whose value is between 0 and 1, to obtain the new input command. In the formula, γ is the step size adjustment coefficient. The larger the error, the more conservative the step size, to avoid iteration divergence.

[0089] Step 8: Repeat steps 5 through 7 until the current round y. i (t) and the expected output command y d The absolute error of (t) and its correlation coefficient K xy If the requirements are met and the iteration converges, the iteration can be terminated, and the input command u of the current round can be saved. i (t), u i (t) It serves as a new driving command and is input into the shaking table to complete the shaking table test;

[0090] Step 9: The iteration process for the remaining single channels is the same as steps 3 to 8.

[0091] This invention solves the problems of low accuracy and poor robustness of existing decoupling methods for servo systems based on multi-degree-of-freedom electro-hydraulic servo vibration test benches, which address the coupling issues between degrees of freedom. The MD-EHSV-DS decoupling method employed in this invention utilizes relative gain array (RGA) technology for optimal pairing and linear matrix inequality (LMI) algorithm optimization to eliminate coupling interference between multiple degrees of freedom. This allows for independent optimization of the control accuracy of each degree of freedom, reducing the interference from other degrees of freedom in the single-degree-of-freedom response to below 5% after decoupling, significantly improving decoupling accuracy.

[0092] This invention solves the problem of insufficient waveform reproduction accuracy in multi-degree-of-freedom electro-hydraulic servo vibration test benches. The invention accurately obtains the multi-channel transfer function matrix using the H1 nonparametric identification method, and combines iterative learning control with dynamic input correction, effectively reducing amplitude attenuation and phase lag. The bandwidth of a single-channel sine wave can reach 200Hz, and the bandwidth of a multi-channel sine wave can reach 100Hz, with a response / command amplitude ratio of no less than -3dB.

[0093] This invention solves the problem of low efficiency in existing iterative methods based on multi-channel electro-hydraulic servo vibration test bench systems. In this invention, iteration is terminated by dynamic error judgment, avoiding blindly increasing the number of iterations. This shortens the test time while ensuring accuracy, improving efficiency by more than 30% compared to traditional methods with a fixed number of iterations.

[0094] Although the present application has been disclosed in its preferred embodiments as above, it is not intended to limit the present application, and any person skilled in the art can make various modifications and modifications without departing from the spirit and scope of the present application, and the protection scope of the present application should be defined by the claims. At the same time, it should be pointed out that the specification and drawings of the present application give the preferred embodiments of the present application and the drawings of the present application, but the present application can be realized in many different forms and is not limited to the embodiments described in the specification, and these embodiments are not additional limitations on the content of the present application, and the purpose of providing this embodiment is to make the understanding of the disclosure of the present application more thorough and comprehensive. And, the above technical features continue to combine, form various embodiments not listed above, which are considered to be within the scope of the present application; further, for those skilled in the art, the above description can be improved or changed, and all these improvements and changes shall belong to the protection scope of the claims of the present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench, characterized in that, Includes the following steps: Obtain the transfer function matrix of a multi-degree-of-freedom electro-hydraulic servo vibration test bench Based on the transfer function matrix Using relative gain array technology, the relative gain array between each output and input of the electro-hydraulic servo vibration test bench system is calculated, and the optimal input-output pairing is identified through relative gain analysis. After determining the optimal pairing interaction, the LMI algorithm is used to minimize the relative gain values ​​of other interactions. The obtained decoupling gain values ​​are expressed in the form of a decoupling matrix, and a decoupling system is designed. And decouple them; For a decoupled SISO system, output commands according to expectations. The first round of command input is obtained in the SISO system pass-through mode. The SISO system inputs a command and receives a response. The transfer function of the current SISO system is then calculated based on the Fourier spectra of the command and response outputs. f represents the frequency; the Fourier transform of the difference between the desired output command and the response output. Solving for the input difference in the frequency domain Then, through inverse Fourier transform, the time domain is obtained. and based on Get new input commands Import command inputs into the SISO system input terminal and continue iterating until the iteration converges, then end the iteration and save the input commands for the current round. ; Other decoupled SISO systems are handled in the same way.

2. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 1, characterized in that, Obtain the transfer function matrix of a multi-degree-of-freedom electro-hydraulic servo vibration test bench During the process, the H1 nonparametric identification method is used to obtain the transfer function matrix.

3. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 1, characterized in that, The obtained decoupling gain value is expressed in the form of a decoupling matrix, and a decoupling system is designed. The process includes: A decoupling matrix structure was constructed for a multi-degree-of-freedom electro-hydraulic servo vibration test bench. The optimal decoupling parameters were obtained by solving the interior-point algorithm and then substituted into the decoupling matrix to obtain the determined decoupling matrix. Design a decoupling system .

4. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to any one of claims 1 to 3, characterized in that, Input a command into the SISO system and obtain a response output. Calculate the transfer function of the current SISO system based on the Fourier spectrum of the command input and the response output. The process includes: During the k-th iteration, command input is imported into the SISO system input terminal. Get the response output The command input during the first iteration is The response output is Enter according to the command and response output The Fourier spectrum is used to calculate the transfer function of the current SISO system. .

5. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 4, characterized in that, Fourier transform of the difference between the expected output command and the response output Solving for the input difference in the frequency domain .

6. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 5, is characterized in that, based on Get new input commands K is the influence factor, with a value between 0 and 1; This is the step size adjustment coefficient.

7. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 6, characterized in that, The influence factor K adopts an adaptive adjustment strategy: K is set to 0.6~0.8 in the early stage of iteration; and K is adjusted to 0.2~0.4 in the iteration convergence stage.

8. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 6, characterized in that, During the process of importing command inputs into the SISO system input terminal and continuing iteration until convergence, the convergence condition is based on the response output in the current time domain. and expected output command The correlation coefficient is used to make a judgment.

9. The multi-channel iterative method based on the decoupling method of the servo system of a multi-degree-of-freedom electro-hydraulic servo vibration test bench according to claim 8, characterized in that, Response output and expected output command The iteration is considered to be convergent when the correlation coefficient is greater than or equal to 0.

9.

10. A computer storage medium, characterized in that, The storage medium stores a computer program, which is loaded and executed by a processor to implement the multi-channel iterative method based on the decoupling method of the servo system of the multi-degree-of-freedom electro-hydraulic servo vibration test bench as described in any one of claims 1 to 9.

Images