Real-time hybrid testing loading device time delay compensation method based on fuzzy neural network

Abstract

The application relates to the technical field of vibration table control, in particular to a real-time hybrid test loading device time delay compensation method based on a fuzzy neural network. The time delay compensation method of the vibration table comprises the following steps: acquiring current time input information and current time deviation information; creating a time delay compensation algorithm based on the current time input information, the current time deviation information and a preset model; and performing time delay compensation on the vibration table based on the time delay compensation algorithm. The application combines the fuzzy neural network with the existing APE algorithm and proposes a new technical scheme. The fuzzy neural network estimates the system time delay and the amplitude overshoot coefficient in real time, realizes time delay compensation of signals of different frequency bands, improves the accuracy of parameter estimation, and overcomes the problems of signal overshoot and time delay overcompensation of high-frequency signals. The stability of the APE algorithm in vibration table time delay compensation in real-time hybrid testing is improved, and the compensation capacity of the algorithm for high-frequency signal reproduction time delay is improved.

Description

Technical Field

[0001] This application relates to the field of vibration table control technology, and in particular to a time delay compensation method for a real-time hybrid test loading device based on a fuzzy neural network. Background Technology

[0002] Real-Time Hybrid Simulation (RTHS) is a commonly used method for seismic testing of structures in the civil engineering field. In RTHS, a shaking table is used to simulate vibration loads under real-world operating conditions to evaluate the reliability and durability of structures or equipment under actual working conditions. The shaking table can generate specific vibration frequencies and amplitudes to simulate the vibration environments that might be encountered in actual operation. In this way, researchers can evaluate the performance of products or equipment in real-world usage environments. Therefore, the accuracy of the shaking table in reproducing the loaded signal, such as time delay, root mean square error, and peak error, significantly affects the stability and accuracy of RTHS.

[0003] To maintain good control performance of the vibration table when the operating frequency changes, a common approach is to combine adaptive algorithms with time delay compensation algorithms. The adaptive algorithm estimates the system's time delay and amplitude overshoot, forming the Adaptive Polynomial Extrapolation (APE) algorithm, which solves the problems of dynamic and unknown time delay compensation. However, this method is prone to overcompensation, introducing negative time delays, and can also cause amplitude overshoot with high-frequency signal inputs, resulting in poor stability in practical applications.

[0004] Therefore, improving the stability of the APE algorithm in vibration table time delay compensation in real-time hybrid tests has become an urgent problem to be solved. Summary of the Invention

[0005] This application provides a time delay compensation method for a real-time hybrid test loading device based on a fuzzy neural network, which improves the stability of the APE algorithm in vibration table time delay compensation in real-time hybrid tests.

[0006] In a first aspect, embodiments of this application provide a time delay compensation method for a real-time hybrid test loading device based on a fuzzy neural network. The method includes: acquiring current-time input information and current-time deviation information; creating a time delay compensation algorithm based on the current-time input information, current-time deviation information, and a preset model; and performing time delay compensation on the vibration table based on the time delay compensation algorithm.

[0007] In one possible implementation, the time delay compensation algorithm created based on the current input information and the current deviation information and the preset model includes: obtaining the system time delay and amplitude coefficient based on the current input information and the deviation information; and creating the time delay compensation algorithm based on the system time delay, the amplitude coefficient and the preset model.

[0008] In one possible implementation, the time delay compensation algorithm based on the system time delay and the amplitude coefficient and the preset model is as follows:

[0009]

[0010] Where, k ρ is the amplitude coefficient; r(k), r(k-1), r(k-2), and r(k-3) are the displacement signals input at times k, k-1, k-2, and k-3, respectively; ρ is the predicted value after the system time lag. i The Lagrange multipliers are determined by the system time delay.

[0011] In one possible implementation, the Lagrange coefficient ρ i The calculation formula is:

[0012]

[0013] Where θ = t(k) / h, θ is the process variable, h represents the sampling time of the system, and t(k) represents the system time delay.

[0014] In one possible implementation, the current input information includes the desired displacement, velocity, and acceleration at the current moment; the current deviation information includes the deviation between the current input signal and the output displacement.

[0015] In one possible implementation, the method further includes: training a fuzzy neural network model based on the expected displacement, velocity, and acceleration at the current moment and the deviation between the input signal and the output displacement at the current moment, to determine the amplitude coefficient k. ρ And the system time delay t(k).

[0016] In one possible implementation, the method further includes: using the deviation between the current input displacement and the actual displacement as the objective function to determine the membership function and weights of the fuzzy neural network model.

[0017] Secondly, embodiments of this application provide a time delay compensation device for a real-time hybrid test loading device based on a fuzzy neural network. The device includes: an acquisition module for acquiring current input information and current deviation information; a creation module for creating a time delay compensation algorithm based on the current input information, the current deviation information, and a preset model; and a processing module for performing time delay compensation on the vibration table based on the time delay compensation algorithm.

[0018] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in the first aspect or any of the implementations thereof.

[0019] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect or any of the implementations thereof.

[0020] Fifthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to execute the method described in the first aspect or any of its implementations.

[0021] The beneficial effects of this application embodiment compared with the prior art are: acquiring the current input information and the current deviation information; creating a time delay compensation algorithm based on the current input information, the current deviation information, and a preset model; and performing time delay compensation on the vibration table based on the time delay compensation algorithm. Compared to the prior art, the adaptive polynomial extrapolation method solves the problems of dynamic time delay and unknown time delay compensation by estimating the system's time delay and amplitude overshoot using an adaptive algorithm. However, the accuracy of the adaptive parameters in the adaptive polynomial extrapolation method is limited by empirical tuning, often leading to overcompensation and introducing negative time delay, resulting in signal overshoot when high-frequency signals are input. This application provides a time delay compensation method for a vibration table by combining a fuzzy neural network with the existing APE algorithm. By using a fuzzy neural network to estimate the system's time delay and amplitude overshoot coefficient in real time, time delay compensation for signals in different frequency bands is achieved, improving the accuracy of parameter estimation and overcoming the problems of signal overshoot and time delay overcompensation in high-frequency signal time delay compensation. This improves the algorithm's ability to compensate for time delays in high-frequency signal reproduction. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a time delay compensation method for a real-time hybrid experimental loading device based on a fuzzy neural network, provided in an embodiment of this application;

[0024] Figure 2 This is a schematic diagram of a fuzzy neural network process provided in an embodiment of this application;

[0025] Figure 3 A diagram illustrating the elongation compensation structure of a six-degree-of-freedom vibration table according to an embodiment of this application;

[0026] Figure 4 A diagram illustrating the pose compensation structure of a six-degree-of-freedom vibration table, provided in an embodiment of this application.

[0027] Figure 5 A structural block diagram of a time delay compensation device for a real-time hybrid test loading device based on a fuzzy neural network, provided in an embodiment of this application;

[0028] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0029] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0030] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0031] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0032] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0033] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0034] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0035] Real-Time Hybrid Simulation (RTHS) is a commonly used method for seismic testing of structures in the civil engineering field. In RTHS, the structure is divided into two parts: a physical substructure and a numerical substructure. The former includes parts of the structure that are difficult to model, have complex characteristics, or require experimental testing, while the latter includes parts that are easy to model but inconvenient to test or are well-established in research. The physical substructure is loaded using a shaking table, and the loading signal is calculated in real time by the numerical substructure. The numerical substructure calculates the loading signal for the next moment based on the feedback signal from the physical substructure after loading, forming a closed loop with the physical substructure. During the test, the shaking table can generate specific vibration frequencies and amplitudes to simulate vibration environments that may be encountered in actual operation. In this way, researchers can evaluate the performance of products or equipment in real-world environments. Therefore, the accuracy of the shaking table in reproducing the loading signal, such as time delay, root mean square error, and peak error, significantly affects the stability and accuracy of RTHS.

[0036] For a single specimen or operating condition, traditional control methods such as three-parameter control and PID algorithms have achieved good results after specialized tuning. However, these methods assume that the control console is a linear time-invariant system and are designed based on assumptions about specific frequencies, loads, and specimens, ignoring the changes in the vibration table and the influence of the specimen on the characteristics of the vibration table.

[0037] In RTHS tests, due to variations in the frequency of the loaded signal, conventional controllers cannot adapt to the system characteristics at different frequency bands. This makes it difficult for existing methods to accurately track reference signals at different frequency bands, resulting in varying time delays and peak errors that affect the test results. The interaction between the vibration table and the specimen prevents control algorithms designed for specific specimens from being directly applied to other tests, limiting the adaptability of the algorithms.

[0038] To maintain good control performance when the operating frequency changes, research on shaking table control mainly focuses on time delay compensation, nonlinear and robust control of earthquake simulation shaking tables, including typical algorithms such as sliding mode control, adaptive control, and iterative control. For the time delay compensation problem, researchers have proposed various time delay compensation algorithms. Polynomial extrapolation (PE) assumes a constant delay, obtains a predictive response model using a finite number of known inputs, and then applies the predicted structural response after the delay time to the controlled object to compensate for the delay. However, PE is suitable for low-frequency signals, but may introduce overshoot when handling high-frequency signals, and requires that the system time delay be known and time-delayed. To address this problem, researchers combined adaptive algorithms with time delay compensation algorithms, using adaptive algorithms to estimate the system's time delay and amplitude overshoot, forming the Adaptive Polynomial Extrapolation (APE) method, which solves the problem of dynamic time delay and unknown time delay compensation. However, existing adaptive algorithms often easily lead to overcompensation, introducing negative time delays, and also have stability issues in practical applications.

[0039] To address the aforementioned issues, this application provides a real-time hybrid test loading device time delay compensation method based on a fuzzy neural network, combining it with existing APE algorithms. The fuzzy neural network (FNN) combines the advantages of fuzzy control and neural networks, simultaneously leveraging the application of expert knowledge in fuzzy control and the parameter self-tuning capabilities of neural networks. FNN possesses strong nonlinear fitting and online learning capabilities, and is commonly used in control systems for uncertainty estimation, controller parameter tuning, and compensator construction. The time delay compensation method for a shaking table obtained by combining a fuzzy neural network with existing APE algorithms includes: acquiring current input information and current deviation information; creating a time delay compensation algorithm based on the current input information, current deviation information, and a preset model; and performing time delay compensation on the shaking table based on the time delay compensation algorithm. By using a fuzzy neural network to estimate the system time delay and amplitude overshoot coefficient in real time, time delay compensation for signals in different frequency bands is achieved, improving the accuracy of parameter estimation and overcoming the overshoot and overcompensation problems of high-frequency signal time delay compensation. This improves the algorithm's ability to compensate for time delays in high-frequency signal reproduction.

[0040] For ease of understanding, the technical solution of this application will be described in detail below with reference to the accompanying drawings.

[0041] Figure 1 A flowchart illustrating a time delay compensation method for a real-time hybrid experimental loading device based on a fuzzy neural network, as provided in an embodiment of this application, is shown below. Figure 1 As shown, it includes:

[0042] S110, obtain the current input information and the current deviation information.

[0043] For example, when creating a time delay compensation algorithm, it is necessary to obtain the input information and deviation information at the current moment, and train the algorithm based on the input information and deviation information.

[0044] Specifically, the input information at the current moment includes: the expected displacement, velocity, and acceleration at the current moment; the deviation information at the current moment is the deviation between the input signal and the output displacement at the current moment. The rise and fall of the input waveform is determined by the velocity of the input signal; the concavity and convexity of the input waveform are determined by the acceleration of the input signal; the time delay of the system is estimated by the deviation and rise and fall; and the overshoot introduced by the APE algorithm is estimated by the deviation and concavity and convexity.

[0045] S120 creates a time delay compensation algorithm based on the current input information, the current deviation information, and the preset model.

[0046] In one possible implementation, the time delay compensation algorithm, which is based on the current input information, the current deviation information, and the preset model, includes: obtaining the system time delay and amplitude coefficient based on the current input information and the current deviation information.

[0047] Specifically, the system time delay and amplitude coefficient are obtained based on the expected displacement, velocity, and acceleration at the previous moment and the deviation between the input signal and the output displacement at the current moment.

[0048] Furthermore, a time delay compensation algorithm is created based on the current input information, the current deviation information, and the preset model, including: a time delay compensation algorithm is created based on the system time delay, the amplitude coefficient, and the preset model.

[0049] As an example, a time delay compensation algorithm based on system time delay, amplitude coefficient, and a preset model is as follows:

[0050]

[0051] Where, k ρ is the amplitude coefficient, used to compensate for system overshoot; r(k), r(k-1), r(k-2) and r(k-3) are the displacement signals input at the current time k, the time before k-1, the time before k-1, and the time before k-3, respectively. ρ is the predicted value after the system time lag. i The Lagrange multipliers are determined by the system time delay.

[0052] As an example, the Lagrange coefficient ρ i The calculation formula is:

[0053]

[0054] Where θ = t(k) / h, θ is a process variable used to calculate the coefficients of the Lagrange difference polynomial, h represents the sampling time of the system, and t(k) represents the system time delay.

[0055] As an example, a fuzzy neural network model is trained based on the deviation between the expected displacement, velocity, and acceleration at the current moment and the current input signal and output displacement to determine the amplitude coefficient k. ρ And the system time delay t(k), its specific training process is by Figure 2 As shown.

[0056] Specifically, such as Figure 2 As shown, the fuzzy neural network consists of five layers: the first layer is the input layer, the second layer is the membership function layer, the third layer is the product operation layer, the fourth layer is the normalization layer, and the fifth layer is the output layer.

[0057] Furthermore, the first layer is the input layer of the network, and the inputs are the deviation e between the desired displacement and the output displacement at the current moment, and the differential of the deviation. First derivative of the input signal (Velocity), second derivative of the input signal (Acceleration). Output is 0. i (i = 1, 2, 3, 4) is:

[0058]

[0059] Furthermore, the second layer is the membership function layer. This layer uses the Gaussian membership function and the sgn(·)±1 function as the membership functions of the fuzzy neural network, σ ij and c ij Let e ​​and denot represent the center value and width of the Gaussian membership function, respectively, where the former is e and denot denoted by e. The membership function, the latter being and The membership function. The input of the second layer is the output of the first layer. i The output of the second layer Characteristic o i membership degree, e and Each correspondence has j membership functions. and Each node corresponds to only two membership functions. Therefore, the number of nodes in the second layer is 2j + 4, as shown in the specific formula:

[0060]

[0061] Furthermore, the third layer is a fuzzy rule fitness layer, and the input of the second layer is... The output of the second layer For O i The cross product of the membership functions corresponds to the fuzzy rules in fuzzy control, with the number of nodes n = 4j. 2 The specific formula is:

[0062]

[0063] Furthermore, the fourth layer is a normalization layer, and the output is... This is the normalization of the values ​​in the third layer, where l = 1, 2, 3, ..., n. The specific formula is:

[0064]

[0065] Furthermore, the fifth layer is the network output layer, outputting values. and These are the time delay t(k) increment Δt(k) and the amplitude coefficient k, respectively. p Increment Δkρ The specific formula is as follows:

[0066]

[0067] Among them, w 1i and w 2i Let t(k) and k represent t(k) and k respectively. p Update the output weights, where n represents the number of nodes in the third layer.

[0068] In one possible implementation, the deviation between the current input displacement and the actual displacement is used as the objective function to determine the membership function and weights of the fuzzy neural network model, that is, to determine the center value σ of the Gaussian membership function in the above equation (4). ij and width c ij And w in the above formula (7) l .

[0069] Specifically, the membership function nodes of fuzzy neural networks have changed compared to traditional neural networks, but the only parameter to be updated is still the weight w of the output layer. l The central value σ of the membership function ij and width c ij The parameters are updated using gradient descent, specifically including:

[0070] First, define the performance metric function of the learning algorithm as follows:

[0071]

[0072] In the formula, E(k) is the desired value, r(k) is the target value of the system at time k (i.e., the displacement signal input to the system), and y(k) is the actual output of the system at time k. To ensure that the actual output y(k) approximates the target value r(k) as closely as possible, E(k) needs to be as small as possible. The gradient descent algorithm is used to correct the weights w in the fourth and fifth layers of the fuzzy neural network online. l The central value σ of the Gaussian membership function ij and width c ij The gradient descent algorithm is as follows:

[0073]

[0074] Where: η w Let α be the learning rate, α be the momentum factor, and t, t+1 and t-1 represent the current time, the next time after the current time and the previous time, respectively. The next time after the current time is the corrected time.

[0075] Furthermore, the time delay t(k) increment Δt(k) and amplitude coefficient k are obtained using a fuzzy neural network. ρ Increment Δk ρThe amplitude coefficient k in the preset model in equation (1) is determined. p And the system time delay t(k), the adaptive coefficient k in the preset model in equation (1) p The update rate of t(k) is shown in the following formula:

[0076]

[0077] Among them, t last (k) represents the system time delay at the previous time step, and t(k) represents the updated system time delay. ρlast k represents the amplitude coefficient at the previous time step. p This is the updated amplitude coefficient.

[0078] S130, based on the time delay compensation algorithm, performs time delay compensation on the vibration table.

[0079] In one possible implementation, a time-delay compensation controller for a multi-degree-of-freedom vibration table is designed based on a time-delay compensation algorithm.

[0080] Specifically, multi-degree-of-freedom time-delay compensation controllers have two application methods in six-degree-of-freedom vibration tables: cylinder elongation compensation (…). Figure 3 ) and pose compensation ( Figure 4 ).

[0081] Specifically, such as Figure 3 As shown, after the system inputs the target pose (pose input in the figure), the expected elongation of each cylinder (L1-L6 in the figure) obtained through pose inverse kinematics is used to predict the expected elongation L1-L6 using the method provided in this application, thereby obtaining the system time delay t(k) and amplitude adaptive coefficient k. p The system also includes the time delay compensation value for the expected elongation after t(k) seconds, which is then input into the system in advance to compensate for the control time delay of each cylinder. Finally, the pose is output using a forward algorithm. The cylinder time delay t(k) and the amplitude adaptive coefficient k are also considered. p Real-time estimation is achieved through a fuzzy neural network (FNN); the time delay compensation value is calculated by the APE time delay compensation algorithm. This method enables high-precision tracking of the cylinder elongation, thereby eliminating the time delay of the system pose.

[0082] Specifically, such as Figure 4 As shown, X, Y, and Z represent the translational poses of the Stewart shaking table along the X, Y, and Z axes, respectively, while A, B, and C represent the rotational poses of the Stewart shaking table around the X, Y, and Z axes, respectively. First, the target pose (X, Y, Z, A, B, and C in the pose input diagram) is input. Time delay compensation is directly applied to the input pose. Based on the difference between the feedback pose and the desired pose, and the pose information, a fuzzy neural network (FNN) is used to obtain the time delay t(k) of the Stewart shaking table system's pose output and the amplitude adaptation coefficient k.p The desired pose is predicted by the APE algorithm after t(k) seconds and input into the vibration table system to achieve pose time delay compensation. Finally, the vibration table is controlled to output the pose. Pose compensation is a process of time delay compensation and amplitude compensation of the desired pose before inverse kinematics, which directly compensates for the pose of the vibration table surface. The feedback quantity is the pose signal output by the vibration table surface, which is a direct compensation.

[0083] The technical solution provided in this application combines a fuzzy neural network with the existing APE algorithm. First, it acquires the current input information and the current deviation information. Then, it creates a time-delay compensation algorithm based on the current input information, the current deviation information, and a preset model. Finally, it performs time-delay compensation on the vibration table based on the time-delay compensation algorithm. Compared to traditional APE algorithms, which are only suitable for low-frequency signal time-delay compensation, in high-frequency bands, due to the increased differential term of the input signal, the interpolation result will exhibit overshoot, and the overshoot value increases with increasing frequency. Traditional adaptive algorithms struggle to accurately estimate the amplitude coefficient and system time delay, easily leading to waveform amplitude distortion or even oscillation when the signal frequency changes rapidly. This application achieves time-delay compensation for signals in different frequency bands, improves the accuracy of amplitude coefficient and system time delay estimation, and overcomes the overshoot and overcompensation problems of high-frequency signal time-delay compensation. It also improves the algorithm's ability to compensate for time delays in high-frequency signal reproduction.

[0084] Figure 5 This embodiment provides a structural block diagram of a time delay compensation device for a real-time hybrid experimental loading device based on a fuzzy neural network. For ease of explanation, only the parts relevant to the embodiment of this application are shown. (Refer to...) Figure 5 The time delay compensation device 500 may include an acquisition module 501, a creation module 502, and a processing module 503.

[0085] In one implementation, the device 500 can be used to implement the above. Figure 1 The method is shown. For example, the acquisition module 501 is used to implement S110, the creation module 502 is used to implement S120, and the processing module 503 is used to implement S130.

[0086] The technical solution provided in this application combines a fuzzy neural network with the existing APE algorithm. First, it acquires the current input information and the current deviation information. Then, it creates a time-delay compensation algorithm based on the current input information, the current deviation information, and a preset model. Finally, it performs time-delay compensation on the vibration table based on the time-delay compensation algorithm. Compared to traditional APE algorithms, which are only suitable for low-frequency signal time-delay compensation, in high-frequency bands, due to the increased differential term of the input signal, the interpolation result will exhibit overshoot, and the overshoot value increases with increasing frequency. Traditional adaptive algorithms struggle to accurately estimate the amplitude coefficient and system time delay, easily leading to waveform amplitude distortion or even oscillation when the signal frequency changes rapidly. This application achieves time-delay compensation for signals in different frequency bands, improves the accuracy of amplitude coefficient and system time delay estimation, and overcomes the overshoot and overcompensation problems of high-frequency signal time-delay compensation. It also improves the algorithm's ability to compensate for time delays in high-frequency signal reproduction.

[0087] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0088] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this embodiment. Figure 6 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 6 (Only one is shown) a processor, a memory 61, and a computer program 62 stored in the memory 61 and executable on the at least one processor 60, which, when executing the computer program 62, implements the steps in any of the above method embodiments.

[0089] The electronic device 6 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. This electronic device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0090] The processor 60 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0091] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may be an external storage device of the electronic device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 6. Furthermore, the memory 61 may include both internal and external storage units of the electronic device 6. The memory 61 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0092] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0093] This application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in any of the above method embodiments.

[0094] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0095] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to a photographic device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0097] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0098] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0101] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for compensating time delay of a real-time hybrid testing loading device based on a fuzzy neural network, characterized in that, The method comprises: obtaining current time input information and current time deviation information; wherein the current time input information comprises current time expected displacement, speed and acceleration; and the current time deviation information comprises deviation of the current time expected displacement and output displacement; creating a time lag compensation algorithm based on the current time input information, the current time deviation information and a preset model; the time lag compensation algorithm created based on the current time input information and the current time deviation information and the preset model comprises: obtaining system time lag and amplitude coefficient based on the current time input information and the current time deviation information; and creating a time lag compensation algorithm based on the system time lag, the amplitude coefficient and the preset model; The time lag compensation algorithm created based on the system time lag and the amplitude coefficient and a preset model is: Wherein, is an amplitude coefficient; is respectively k, k-1, k-2 and k-3 the displacement signal input at the moment; is a predicted value of the system time lag; is a Lagrange coefficient determined by the system time lag; the lagrangian coefficient the calculation formula is: wherein, , is a process variable, characterizes the sampling time of the system, characterizes the time lag of the system; The method further comprises: Performing fuzzy neural network model training based on the current moment expected displacement, speed and acceleration and the deviation of the current moment expected displacement and the output displacement, to determine the amplitude coefficient And the system time delay ; Time delay compensation is performed on the vibration table based on the time delay compensation algorithm.

2. The method of claim 1, wherein, The method further comprises: taking the deviation of the current time expected displacement and the output displacement as a target function, determining a membership function and a weight of a fuzzy neural network model.

3. A fuzzy neural network based real-time hybrid testing loading device time delay compensation device, the device is applicable to the fuzzy neural network based real-time hybrid testing loading device time delay compensation method as claimed in any one of claims 1-2, characterized in that, The device comprises: an obtaining module configured to obtain current time input information and current time deviation information; wherein the current time input information comprises current time expected displacement, speed and acceleration; and the current time deviation information comprises deviation of the current time expected displacement and output displacement; The obtaining module comprises: an obtaining unit configured to obtain system time lag and amplitude coefficient based on the current time input information and the current time deviation information; and create a time lag compensation algorithm based on the system time lag, the amplitude coefficient and the preset model; The time lag compensation algorithm created based on the system time lag and the amplitude coefficient and a preset model is: wherein, is an amplitude coefficient; is respectively k, k-1, k-2 and k-3 a displacement signal input at the moment; is a predicted value of the system time lag; is a Lagrange coefficient determined by the system time lag; the lagrangian coefficient the calculation formula is: wherein, , is a process variable, characterizes a sampling time of the system, characterizes a time lag of the system; the device further comprises: A determining module is configured to determine the amplitude coefficient based on the fuzzy neural network model training of the current time expected displacement, speed and acceleration and the deviation of the current time expected displacement and the output displacement. and the system time delay ; a creating module configured to create a time lag compensation algorithm based on the current time input information, the current time deviation information and a preset model; and a processing module configured to perform time lag compensation on a vibration table based on the time lag compensation algorithm.

4. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The processor implements the method of any one of claims 1 to 2 when executing the computer program.

5. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 4. The computer program implements the method of any one of claims 1 to 2 when executed by the processor.

Images