An intelligent vibration isolation control method and system based on vibration monitoring

Abstract

The present invention discloses an intelligent vibration isolation control method and system based on vibration monitoring, comprising collecting vibration data for preprocessing, performing fuzzy conversion on the preprocessed data, and then calculating a preliminary control quantity; constructing an optimization model and defining an objective function, iteratively optimizing the population position using a locust optimization algorithm until the objective function value converges, and using the preliminary control quantity as the input of the final optimization model to obtain a final control quantity; calculating a driving force signal based on the final control quantity, and further constructing a differential equation based on the accumulation of the driving force signal to output a predicted value, and performing conversion based on the obtained predicted value to obtain instantaneous data. This method achieves efficient adaptive control of complex nonlinear vibration systems and enables rapid convergence of the optimization process. The objective function optimization result is directly used for dynamic adjustment of the control quantity, significantly shortening the time for calculating the control quantity. Furthermore, the future vibration trend is predicted through the differential equation, further enhancing the dynamic response capability of the present invention.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent vibration isolation control, and in particular relates to an intelligent vibration isolation control method and system based on vibration monitoring. Background Art

[0002] In recent years, with the rapid development of sensor technology, signal processing technology and intelligent control algorithms, active vibration isolation technology based on vibration monitoring has gradually become a research hotspot. Active vibration isolation technology monitors the vibration signal of the system in real time, combines intelligent algorithms to calculate the optimal control quantity, drives the actuator to generate reverse force to offset the vibration, and realizes adaptive isolation in complex vibration environments.

[0003] However, existing vibration isolation control methods usually find it difficult to achieve efficient control in multivariable, complex nonlinear systems, and it is difficult to complete complex control quantity calculations in a short time, which affects the real-time performance and dynamic response capabilities of the vibration isolation control system. Summary of the Invention

[0004] In response to the above technical problems, the present invention provides an intelligent vibration isolation control method and system based on vibration monitoring, which solves the problem that existing vibration isolation control methods are generally difficult to achieve efficient control in multi-variable, complex nonlinear systems, and difficult to complete complex control quantity calculations in a short time, affecting the real-time and dynamic response capabilities of the vibration isolation control system.

[0005] The technical solution adopted by the present invention to solve the technical problem is:

[0006] An intelligent vibration isolation control method based on vibration monitoring, the method comprising the following steps:

[0007] S100: collecting vibration data, preprocessing the vibration data, and obtaining preprocessed data;

[0008] S200: Calculate the target signal and vibration deviation based on the preprocessed data, analyze the vibration deviation change rate, perform fuzzy conversion on the vibration deviation and change rate and integrate the membership values, infer and merge the control quantity fuzzy set using the Mamdani fuzzy inference method, divide the discrete points and calculate the membership values, screen the discrete points and calculate the preliminary control quantity;

[0009] S300: Build an optimization model and define the objective function, randomly initialize membership values and generate an initial population, iteratively optimize the population position using the locust optimization algorithm until the objective function value converges, output the final optimization model, and use the preliminary control quantity as the input of the final optimization model to obtain the final control quantity;

[0010] S400: Calculating a driving force signal based on the final controlled variable, further constructing a differential equation based on the accumulation of the driving force signal, and outputting a predicted value, performing conversion based on the obtained predicted value to obtain instantaneous data;

[0011] S500: Calculating a control error based on instantaneous data, performing control adjustments based on the control error, and storing the adjustment results in a database.

[0012] Preferably, S100 includes:

[0013] S110: collecting vibration data by deploying acceleration sensors;

[0014] S120: Decomposing and reconstructing the collected data using wavelet decomposition technology to obtain denoised data;

[0015] S130: Based on the denoised data, a normalization operation is performed on the denoised data using a normalization method to obtain normalized data.

[0016] Preferably, S200 includes:

[0017] S210: Calculate the mean value using the mean formula based on the denoised data, define the obtained mean value as the target signal, and subtract the normalized data from the target signal to obtain the vibration deviation;

[0018] S220: Calculate the vibration deviation of the previous time using a time series analysis method, and subtract the vibration deviation from the vibration deviation of the previous time to obtain a vibration deviation change rate;

[0019] S230: defining the vibration deviation and the vibration deviation change rate as input variables, performing fuzzy conversion based on the obtained input variables using a triangular membership function to obtain converted membership values, and integrating the obtained membership values to obtain a fuzzy value set of the vibration deviation and a fuzzy value set of the vibration deviation change rate;

[0020] S240: defining a fuzzy control rule table, taking the fuzzy value set of the vibration deviation and the fuzzy value set of the vibration deviation change rate as inputs of the Mamdani fuzzy inference method, and using the Mamdani fuzzy inference method in combination with the rule activation strength and the fuzzy set truncation formula to infer the fuzzy control rule table to obtain an inferred control quantity fuzzy set, and merging the control quantity fuzzy sets of all rules using the maximum membership method based on the inferred control quantity fuzzy set to obtain a merged fuzzy set;

[0021] S250: Divide the merged fuzzy set into M discrete points by an equal interval discretization method, and calculate the membership value of each discrete point by using a triangular membership function;

[0022] S260: Set a selection threshold, compare the membership value corresponding to the discrete point with the threshold, and retain the discrete point when the membership value corresponding to the discrete point is greater than or equal to the threshold; otherwise, remove the discrete point; and calculate the preliminary control amount based on the retained discrete point using the discretization formula.

[0023] Preferably, in S230, based on the obtained input variables, a triangular membership function is used to perform fuzzy conversion to obtain a converted membership value, specifically:

[0024] The fuzzy formula of vibration deviation is:

[0025] ;

[0026] Where, Indicates the vibration deviation of the input variable Membership value of the set of fuzzy values belonging to the vibration deviation, Indicates vibration deviation, Indicates the maximum absolute value of vibration deviation;

[0027] The fuzzy value set of vibration deviation is expressed as:

[0028] ;

[0029] Where, Indicates a large negative. Indicates a small negative. represents the center value, Indicates positive small, It means upright;

[0030] The fuzzy formula of the vibration deviation change rate is:

[0031] ;

[0032] Where, Indicates the input variable vibration deviation change rate The membership value of the fuzzy value set belonging to the vibration deviation change rate, Indicates the vibration deviation change rate, Indicates the maximum absolute value of the vibration deviation change rate;

[0033] The fuzzy value set of the vibration deviation change rate is expressed as:

[0034] ;

[0035] Where, Indicates negative speed, Indicates zero speed, Indicates positive speed;

[0036] The rule activation formula in the Mamdani fuzzy inference method in S240 is:

[0037] ;

[0038] Where, represents the rule activation strength, Indicates taking the minimum value of the membership value;

[0039] The truncation formula of the rule output fuzzy set is:

[0040] ;

[0041] Where, Indicates the activation intensity Truncated fuzzy set membership function, Represents the output variable obtained through Mamdani fuzzy inference, The original membership function of the fuzzy set representing the output of the fuzzy rule;

[0042] Based on the inferred control quantity fuzzy set, the maximum membership method is used to merge the control quantity fuzzy sets of all rules to obtain the merged fuzzy set, which is specifically:

[0043] ;

[0044] Where, represents the merged fuzzy set, Indicates taking the maximum membership value, represents the fuzzy set output by the nth rule;

[0045] The discretization formula used in S250 to calculate the preliminary control quantity is as follows:

[0046] ;

[0047] Where, represents the initial control quantity at time t, represents the weight of the discrete points, Represents the i-th discrete point The membership value of , S represents the total number of discrete points retained.

[0048] Preferably, S300 includes:

[0049] S310: Based on the obtained preliminary control amount, use a recursive estimation method to calculate the preliminary control amount of the previous state, subtract the preliminary control amount of the previous state from the preliminary control amount, and obtain the rate of change of the control amount;

[0050] S320: Randomly initialize the membership values of all retained discrete points, generate multiple groups of initial membership value combinations as initial populations, and perform offset adjustment on the total parameter value of the initial population according to the preliminary control amount;

[0051] S330: Construct an optimization model and define the objective function of the optimization model based on the rate of change of the control variable:

[0052] ;

[0053] Where, represents the objective function value, represents the rate of change of the control quantity at time t, represents the vibration deviation at time t, Represents the weight factor of the control quantity change, represents the total number of time steps;

[0054] S340: Use the locust optimization algorithm to calculate the interaction force between individuals in the population, and calculate the objective function value corresponding to the individual after updating the position of the individual in the population. During the iteration process, the population position is continuously updated. When the objective function value converges, the iteration is stopped and the final optimization model is output;

[0055] S350: Input the preliminary control amount into the final optimization model and output the final control amount.

[0056] Preferably, S400 includes:

[0057] S410: Calculating a driving force signal according to the final control amount, and using a sliding window method to obtain the vibration data and driving force signal of the previous moment based on the normalized vibration data and driving force signal;

[0058] S420: constructing a time series set at each time point based on the driving force signal at the previous moment and the corresponding vibration data at the previous moment, and accumulating the driving force signal and vibration data in the time series set to obtain an accumulated driving force signal and accumulated vibration data;

[0059] S430: Construct a rectangular matrix based on the accumulated driving force signal and the accumulated vibration data, construct a target vector based on the accumulated vibration data, and perform a transposition operation on the rectangular matrix and the target vector to obtain a parameter vector Then, the matrix decomposition technology is further used to decompose the vibration data to obtain the coefficients , coefficient of driving force signal and static offset ;

[0060] S440: Based on the accumulated driving force signal, accumulated vibration data, coefficients of the vibration data, coefficients of the driving force signal and the static offset, a differential equation is constructed to output the accumulated value at the future moment, and the accumulated value at the future moment is converted using the cumulative subtraction generation method to obtain instantaneous data.

[0061] Preferably, the driving force signal is calculated according to the final control amount in S410, specifically:

[0062] ;

[0063] Where, Indicates the driving force signal, represents the stiffness parameter, set through experiments, is the final control quantity;

[0064] In S430, a rectangular matrix is constructed based on the accumulated driving force signal and the accumulated vibration data, specifically:

[0065] ;

[0066] Where, represents a rectangular matrix, represents the total number of data points in the time series, Indicates the accumulated vibration data, Indicates the cumulative driving force signal;

[0067] In S430, a target vector is constructed based on the accumulated vibration data, specifically:

[0068] ;

[0069] Where, represents the target vector, Indicates the total length of the time series;

[0070] In S430, the rectangular matrix and the target vector are transposed using a transposition operation to obtain a parameter vector, which is further decomposed using a matrix decomposition technique to obtain coefficients of the vibration data, coefficients of the driving force signal, and a static offset, specifically:

[0071] ;

[0072] ;

[0073] Where, represents the transpose operation, represents the parameter vector, The coefficients representing the vibration data, represents the coefficient of the driving force signal, Indicates static offset;

[0074] In S440 , a differential equation is constructed based on the accumulated driving force signal, accumulated vibration data, coefficients of the vibration data, coefficients of the driving force signal, and the static offset, and the accumulated value at the future moment is output. Specifically, it is:

[0075] ;

[0076] Where, Indicates the future moment The accumulated value of

[0077] In S440, the accumulated value at the future moment is converted using the cumulative-subtraction generation method to obtain instantaneous data, specifically:

[0078] ;

[0079] Where, Represents instantaneous data, Represents the accumulated value at the current time t.

[0080] Preferably, calculating the control error based on the instantaneous data in S500 and performing control adjustment based on the control error include:

[0081] S510: Subtracting the instantaneous data from the normalized vibration data to obtain a control error;

[0082] S520: Setting an optimization threshold, comparing the control error with the threshold, when the control error is greater than or equal to the threshold, re-optimizing the control amount through the final optimization model, and when the control error is less than the threshold, using the control error as the basis for vibration isolation control.

[0083] Preferably, using a database to store the adjustment result in S500 includes:

[0084] Based on the obtained control error, a database is used for storage. During the storage process, the control error is recorded in time series, and the vibration data and time series set corresponding to the error are stored in association. The historical error data is classified and organized through the database, and trend analysis data is generated.

[0085] An intelligent vibration isolation control system based on vibration monitoring, comprising a data preprocessing module, a preliminary control quantity calculation module, a model optimization module, an integration and conversion module, and an adjustment and storage module;

[0086] A data preprocessing module is used to collect vibration data, preprocess the vibration data, and obtain preprocessed data;

[0087] The preliminary control quantity calculation module is used to calculate the target signal and vibration deviation based on the preprocessed data, analyze the vibration deviation change rate, perform fuzzy conversion on the vibration deviation and change rate and integrate the membership values, infer and merge the control quantity fuzzy sets through the Mamdani fuzzy inference method, divide the discrete points and calculate the membership values, screen the discrete points and calculate the preliminary control quantity;

[0088] The model optimization module is used to build the optimization model and define the objective function, randomly initialize the membership value and generate the initial population, iteratively optimize the population position using the locust optimization algorithm until the objective function value converges, output the final optimization model, and use the preliminary control quantity as the input of the final optimization model to obtain the final control quantity;

[0089] The integration and conversion module is used to calculate the driving force signal based on the final control quantity, further construct a differential equation based on the accumulation of the driving force signal, and output a predicted value. The predicted value is converted to obtain instantaneous data;

[0090] The adjustment and storage module is used to calculate the control error based on the instantaneous data, perform control adjustment based on the control error, and use the database to store the adjustment results.

[0091] The above-mentioned intelligent vibration isolation control method and system based on vibration monitoring activates the rules through the Mamdani fuzzy reasoning method, generates preliminary control quantities and further optimizes them, thereby realizing efficient adaptive control of complex nonlinear vibration systems. By combining the locust optimization algorithm to randomly initialize the membership value combination as a population, and adjusting the parameter range of the initial population based on the preliminary control quantity, the optimization process converges quickly, and the objective function optimization result is directly used for the dynamic adjustment of the control quantity, which significantly shortens the time for calculating the control quantity, and predicts the future vibration trend through the differential equation, so that the dynamic response capability of the present invention is further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 This is a flow chart of an intelligent vibration isolation control method based on vibration monitoring in one embodiment of the present invention;

[0093] Figure 2 The figure is a schematic diagram of the principle of an intelligent vibration isolation control system based on vibration monitoring in one embodiment of the present invention. DETAILED DESCRIPTION

[0094] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below with reference to the accompanying drawings.

[0095] In one embodiment, Figure 1 As shown, an intelligent vibration isolation control method based on vibration monitoring includes the following steps:

[0096] S100: collecting vibration data, preprocessing the vibration data, and obtaining preprocessed data;

[0097] S200: Calculate the target signal and vibration deviation based on the preprocessed data, analyze the vibration deviation change rate, perform fuzzy conversion on the vibration deviation and change rate and integrate the membership values, infer and merge the control quantity fuzzy set using the Mamdani fuzzy inference method, divide the discrete points and calculate the membership values, screen the discrete points and calculate the preliminary control quantity;

[0098] S300: Build an optimization model and define the objective function, randomly initialize membership values and generate an initial population, iteratively optimize the population position using the locust optimization algorithm until the objective function value converges, output the final optimization model, and use the preliminary control quantity as the input of the final optimization model to obtain the final control quantity;

[0099] S400: Calculating a driving force signal based on the final controlled variable, further constructing a differential equation based on the accumulation of the driving force signal, and outputting a predicted value, performing conversion based on the obtained predicted value to obtain instantaneous data;

[0100] S500: Calculating a control error based on instantaneous data, performing control adjustments based on the control error, and storing the adjustment results in a database.

[0101] In one embodiment, S100 includes:

[0102] S110: collecting vibration data by deploying acceleration sensors;

[0103] S120: Decomposing and reconstructing the collected data using wavelet decomposition technology to obtain denoised data;

[0104] S130: Based on the denoised data, a normalization operation is performed on the denoised data using a normalization method to obtain normalized data.

[0105] Specifically, vibration data is collected by acceleration sensors, and combined with wavelet decomposition technology and normalization methods, an efficient process from vibration signal acquisition to preprocessing is achieved. The wavelet decomposition technology is used to decompose and reconstruct the signal, effectively removing noise and retaining key features, thereby improving data accuracy. The normalization operation unifies the data scale and enhances the stability and efficiency of subsequent fuzzy control and optimization algorithm calculations.

[0106] In one embodiment, S200 includes:

[0107] S210: Calculate the mean value using the mean formula based on the denoised data, define the obtained mean value as the target signal, and subtract the normalized data from the target signal to obtain the vibration deviation;

[0108] S220: Calculate the vibration deviation of the previous time using a time series analysis method, and subtract the vibration deviation from the vibration deviation of the previous time to obtain a vibration deviation change rate;

[0109] S230: defining the vibration deviation and the vibration deviation change rate as input variables, performing fuzzy conversion based on the obtained input variables using a triangular membership function to obtain converted membership values, and integrating the obtained membership values to obtain a fuzzy value set of the vibration deviation and a fuzzy value set of the vibration deviation change rate;

[0110] S240: defining a fuzzy control rule table, taking the fuzzy value set of the vibration deviation and the fuzzy value set of the vibration deviation change rate as inputs of the Mamdani fuzzy inference method, and using the Mamdani fuzzy inference method in combination with the rule activation strength and the fuzzy set truncation formula to infer the fuzzy control rule table to obtain an inferred control quantity fuzzy set, and merging the control quantity fuzzy sets of all rules using the maximum membership method based on the inferred control quantity fuzzy set to obtain a merged fuzzy set;

[0111] S250: Divide the merged fuzzy set into M discrete points by an equal interval discretization method, and calculate the membership value of each discrete point by using a triangular membership function;

[0112] S260: Set a selection threshold based on domain knowledge, compare the membership value corresponding to the discrete point with the threshold, and retain the discrete point when the membership value corresponding to the discrete point is greater than or equal to the threshold; otherwise, remove the discrete point. Calculate the preliminary control amount based on the retained discrete points using the discretization formula.

[0113] In one embodiment, in S230, a fuzzy transformation is performed based on the obtained input variables using a triangular membership function to obtain a transformed membership value, specifically:

[0114] The fuzzy formula of vibration deviation is:

[0115] ;

[0116] Where, Indicates the vibration deviation of the input variable Membership value of the set of fuzzy values belonging to the vibration deviation, Indicates vibration deviation, Indicates the maximum absolute value of vibration deviation;

[0117] The fuzzy value set of vibration deviation is expressed as:

[0118] ;

[0119] Where, Indicates a large negative. Indicates a small negative. represents the center value, Indicates positive small, It means upright;

[0120] The fuzzy formula of the vibration deviation change rate is:

[0121] ;

[0122] Where, Indicates the input variable vibration deviation change rate The membership value of the fuzzy value set belonging to the vibration deviation change rate, Indicates the vibration deviation change rate, Indicates the maximum absolute value of the vibration deviation change rate;

[0123] The fuzzy value set of the vibration deviation change rate is expressed as:

[0124] ;

[0125] Where, Indicates negative speed, Indicates zero speed, Indicates positive speed;

[0126] The rule activation formula in the Mamdani fuzzy inference method in S240 is:

[0127] ;

[0128] Where, represents the rule activation strength, Indicates taking the minimum value of the membership value;

[0129] The truncation formula of the rule output fuzzy set is:

[0130] ;

[0131] Where, Indicates the activation intensity Truncated fuzzy set membership function, Represents the output variable obtained through Mamdani fuzzy inference, The original membership function of the fuzzy set representing the output of the fuzzy rule;

[0132] Based on the inferred control quantity fuzzy set, the maximum membership method is used to merge the control quantity fuzzy sets of all rules to obtain the merged fuzzy set, which is specifically:

[0133] ;

[0134] Where, represents the merged fuzzy set, Indicates taking the maximum membership value, represents the fuzzy set output by the nth rule;

[0135] The discretization formula used in S250 to calculate the preliminary control quantity is as follows:

[0136] ;

[0137] Where, represents the initial control quantity at time t, represents the weight of the discrete points, Represents the i-th discrete point The membership value of , S represents the total number of discrete points retained.

[0138] Specifically, by calculating the vibration deviation and its rate of change, the deviation amplitude and change trend of the vibration system are reflected in real time, providing high-precision input data for subsequent fuzzy conversion, and the calculation of the vibration deviation is based on the difference between the normalized data and the target signal, and the vibration deviation change rate is obtained by the deviation difference between adjacent time points. This can ensure that the present invention responds sensitively to dynamic vibration signals and adapts to complex vibration environments. The introduction of fuzzy conversion greatly enhances the present invention's ability to handle uncertainty and complexity. The triangular membership function is used to convert the vibration deviation and its rate of change into a membership value, which effectively simplifies the modeling process and lays the foundation for the reasoning of the fuzzy control rule table. The Mamdani fuzzy reasoning method is concluded According to the rule activation intensity and fuzzy set truncation formula, the fuzzy set is inferred and adjusted, so that the control quantity can be dynamically generated according to the real-time changes of the vibration deviation. The fuzzy sets are merged through the maximum membership method to quickly generate clear control quantities, which improves the dynamic response speed and real-time performance of the present invention. Secondly, the fuzzy set is optimized by equal-interval discretization and threshold screening, which significantly reduces the computational complexity and retains the key discrete points that have the greatest impact on the control effect. This not only improves the real-time performance of the system, but also ensures the accuracy and stability of the calculation results. In combination with the calculation formula of the preliminary control quantity, the present invention realizes the close coupling from the vibration signal input to the preliminary control quantity output, laying a solid foundation for subsequent optimization and execution.

[0139] In one embodiment, S300 includes:

[0140] S310: Based on the obtained preliminary control amount, use a recursive estimation method to calculate the preliminary control amount of the previous state, subtract the preliminary control amount of the previous state from the preliminary control amount, and obtain the rate of change of the control amount;

[0141] S320: Randomly initialize the membership values of all retained discrete points, generate multiple groups of initial membership value combinations as initial populations, and perform offset adjustment on the total parameter value of the initial population according to the preliminary control amount;

[0142] S330: Construct an optimization model and define the objective function of the optimization model based on the rate of change of the control variable:

[0143] ;

[0144] Where, represents the objective function value, represents the rate of change of the control quantity at time t, represents the vibration deviation at time t, Represents the weight factor of the control quantity change, represents the total number of time steps;

[0145] S340: Use the locust optimization algorithm to calculate the interaction force between individuals in the population, and calculate the objective function value corresponding to the individual after updating the position of the individual in the population. During the iteration process, the population position is continuously updated. When the objective function value converges, the iteration is stopped and the final optimization model is output;

[0146] S350: Input the preliminary control amount into the final optimization model and output the final control amount.

[0147] Specifically, by introducing a recursive estimation method and a calculation mechanism for the control quantity change rate, the present invention can capture the dynamic change trend of the control quantity in real time, ensure rapid response in a complex vibration environment, and by quantifying the difference between the current control quantity and the previous state, dynamically adjust the control strategy to achieve more accurate vibration suppression. The vibration deviation and the control quantity change rate are combined together through the objective function, and the importance of the two is dynamically balanced through the weight factor, so that the optimization result not only has high-precision vibration control capability, but also ensures the smoothness of the control quantity. Secondly, the locust optimization algorithm is used to optimize the control quantity, and global search is achieved by simulating the interaction force of the population, which can effectively avoid falling into local optimality and adapt to the needs of complex nonlinear vibration systems. By randomly initializing the population, dynamically adjusting the parameter range and rapid iterative optimization based on the objective function, the present invention can output the final control quantity in a short time and meet the real-time control requirements in a high-frequency vibration environment.

[0148] In one embodiment, S400 includes:

[0149] S410: Calculating a driving force signal according to the final control amount, and using a sliding window method to obtain the vibration data and driving force signal of the previous moment based on the normalized vibration data and driving force signal;

[0150] S420: constructing a time series set at each time point based on the driving force signal at the previous moment and the corresponding vibration data at the previous moment, and accumulating the driving force signal and vibration data in the time series set to obtain an accumulated driving force signal and accumulated vibration data;

[0151] S430: Construct a rectangular matrix based on the accumulated driving force signal and the accumulated vibration data, construct a target vector based on the accumulated vibration data, and perform a transposition operation on the rectangular matrix and the target vector to obtain a parameter vector Then, the matrix decomposition technology is further used to decompose the vibration data to obtain the coefficients , coefficient of driving force signal and static offset ;

[0152] S440: Based on the accumulated driving force signal, accumulated vibration data, coefficients of the vibration data, coefficients of the driving force signal and the static offset, a differential equation is constructed to output the accumulated value at the future moment, and the accumulated value at the future moment is converted using the cumulative subtraction generation method to obtain instantaneous data.

[0153] In one embodiment, the driving force signal is calculated according to the final control amount in S410, specifically:

[0154] ;

[0155] Where, Indicates the driving force signal, represents the stiffness parameter, set through experiments, is the final control quantity;

[0156] In S430, a rectangular matrix is constructed based on the accumulated driving force signal and the accumulated vibration data, specifically:

[0157] ;

[0158] Where, represents a rectangular matrix, represents the total number of data points in the time series, Indicates the accumulated vibration data, Indicates the cumulative driving force signal;

[0159] In S430, a target vector is constructed based on the accumulated vibration data, specifically:

[0160] ;

[0161] Where, represents the target vector, Indicates the total length of the time series;

[0162] In S430, the rectangular matrix and the target vector are transposed using a transposition operation to obtain a parameter vector, which is further decomposed using a matrix decomposition technique to obtain coefficients of the vibration data, coefficients of the driving force signal, and a static offset, specifically:

[0163] ;

[0164] ;

[0165] Where, represents the transpose operation, represents the parameter vector, The coefficients representing the vibration data, represents the coefficient of the driving force signal, Indicates static offset;

[0166] In S440 , a differential equation is constructed based on the accumulated driving force signal, accumulated vibration data, coefficients of the vibration data, coefficients of the driving force signal, and the static offset, and the accumulated value at the future moment is output. Specifically, it is:

[0167] ;

[0168] Where, Indicates the future moment The accumulated value of

[0169] In S440, the accumulated value at the future moment is converted using the cumulative-subtraction generation method to obtain instantaneous data, specifically:

[0170] ;

[0171] Where, Represents instantaneous data, Represents the accumulated value at the current time t.

[0172] Specifically, by adopting a precise calculation method of the driving force signal, the final control quantity is combined with the stiffness parameter to dynamically generate the driving force signal, and the sliding window method is used to extract the vibration data and driving force signal of the previous moment, ensuring that the input data has high timeliness and accuracy, providing a reliable basis for the construction of subsequent time series sets, and using the cumulative formula to calculate the cumulative driving force signal and the cumulative vibration data respectively, extracting the global characteristics of the time series, analyzing the long-term change trend of the data, and the rectangular matrix and target vector constructed based on the cumulative signal can fully express the multi-dimensional correlation between the vibration data and the driving force signal, providing an accurate mathematical model for the solution of the parameter vector. Secondly, by the rectangular matrix The present invention uses linear algebra methods to calculate and decompose parameter vectors to obtain vibration data coefficients, driving force signal coefficients and static offsets. These parameters serve as core factors when constructing differential equations. The dynamic prediction capability of differential equations further enhances the control accuracy of the system and its response capability to future vibration trends. The instantaneous data is restored from the accumulated values at future moments through the cumulative subtraction generation method, providing an efficient and accurate basis for real-time calculation. The instantaneous data directly reflects the vibration state at the current moment, providing an important reference for error correction and optimization adjustment of vibration isolation control. Combined with dynamic prediction and real-time adjustment, the present invention realizes efficient and accurate vibration isolation control of the vibration system.

[0173] In one embodiment, calculating the control error based on the instantaneous data in S500 and performing control adjustment based on the control error includes:

[0174] S510: Subtracting the instantaneous data from the normalized vibration data to obtain a control error;

[0175] S520: Set an optimization threshold based on accuracy requirements and personal experience, and compare the control error with the threshold. When the control error is greater than or equal to the threshold, re-optimize the control amount through the final optimization model. When the control error is less than the threshold, the control error is used as the basis for vibration isolation control.

[0176] Specifically, the calculation of instantaneous data can quickly capture the system vibration state, and after comparison with the normalized vibration data, an accurate control error can be generated. By setting the optimization threshold, adaptive adjustment of the control strategy is achieved, avoiding the waste of resources caused by frequent optimization, while ensuring the ability to quickly correct large errors.

[0177] In one embodiment, using a database to store the adjustment result in S500 includes:

[0178] Based on the obtained control error, a database is used for storage. During the storage process, the control error is recorded in time series, and the vibration data and time series set corresponding to the error are stored in association. The historical error data is classified and organized through the database, and trend analysis data is generated.

[0179] Specifically, through time series recording and associative storage, accurate data support is provided for real-time optimization, and through trend analysis data, future vibration behavior is predicted, and the control strategy is dynamically adjusted to adapt to complex nonlinear vibration environments.

[0180] This embodiment also provides an intelligent vibration isolation control system based on vibration monitoring, such as Figure 2 As shown, it includes a data preprocessing module, a preliminary control quantity calculation module, a model optimization module, an integration and transformation module, and an adjustment and storage module;

[0181] A data preprocessing module is used to collect vibration data, preprocess the vibration data, and obtain preprocessed data;

[0182] The preliminary control quantity calculation module is used to calculate the target signal and vibration deviation based on the preprocessed data, analyze the vibration deviation change rate, perform fuzzy conversion on the vibration deviation and change rate and integrate the membership values, infer and merge the control quantity fuzzy sets through the Mamdani fuzzy inference method, divide the discrete points and calculate the membership values, screen the discrete points and calculate the preliminary control quantity;

[0183] The model optimization module is used to build the optimization model and define the objective function, randomly initialize the membership value and generate the initial population, iteratively optimize the population position using the locust optimization algorithm until the objective function value converges, output the final optimization model, and use the preliminary control quantity as the input of the final optimization model to obtain the final control quantity;

[0184] The integration and conversion module is used to calculate the driving force signal based on the final control quantity, further construct a differential equation based on the accumulation of the driving force signal, and output a predicted value. The predicted value is converted to obtain instantaneous data;

[0185] The adjustment and storage module is used to calculate the control error based on the instantaneous data, perform control adjustment based on the control error, and use the database to store the adjustment results.

[0186] For the specific definition of an intelligent vibration isolation control system based on vibration monitoring, please refer to the definition of an intelligent vibration isolation control method based on vibration monitoring above, which will not be repeated here. The various modules in the above-mentioned intelligent vibration isolation control system based on vibration monitoring can be implemented in whole or in part through software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0187] This embodiment also provides a computer device, which is suitable for the case of an intelligent vibration isolation control method based on vibration monitoring, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent vibration isolation control method based on vibration monitoring proposed in the above embodiment.

[0188] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0189] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent vibration isolation control method based on vibration monitoring proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0190] The above is a detailed introduction to the intelligent vibration isolation control method and system based on vibration monitoring provided by the present invention. This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the core idea of the present invention. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, the present invention can also be improved and modified in several ways, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. An intelligent vibration isolation control method based on vibration monitoring, characterized in that: The method comprises the following steps: S100: collecting vibration data, preprocessing the vibration data, and obtaining preprocessed data; S200: Calculate the target signal and vibration deviation based on the preprocessed data, analyze the vibration deviation change rate, perform fuzzy conversion on the vibration deviation and change rate and integrate the membership values, infer and merge the control quantity fuzzy set using the Mamdani fuzzy inference method, divide the discrete points and calculate the membership values, screen the discrete points and calculate the preliminary control quantity; S300: Build an optimization model and define the objective function, randomly initialize membership values and generate an initial population, iteratively optimize the population position using the locust optimization algorithm until the objective function value converges, output the final optimization model, and use the preliminary control quantity as the input of the final optimization model to obtain the final control quantity. S300 includes: S310: Based on the obtained preliminary control amount, use a recursive estimation method to calculate the preliminary control amount of the previous state, subtract the preliminary control amount of the previous state from the preliminary control amount, and obtain the rate of change of the control amount; S320: Randomly initialize the membership values of all retained discrete points, generate multiple groups of initial membership value combinations as initial populations, and perform offset adjustment on the total parameter value of the initial population according to the preliminary control amount; S330: Construct an optimization model and define the objective function of the optimization model based on the rate of change of the control variable: ； Where, represents the objective function value, represents the rate of change of the control quantity at time t, represents the vibration deviation at time t, Represents the weight factor of the control quantity change, represents the total number of time steps; S340: Use the locust optimization algorithm to calculate the interaction force between individuals in the population, and calculate the objective function value corresponding to the individual after updating the position of the individual in the population. During the iteration process, the population position is continuously updated. When the objective function value converges, the iteration is stopped and the final optimization model is output; S350: Inputting the preliminary control amount into the final optimization model and outputting the final control amount; S400: Calculating a driving force signal based on the final controlled variable, further constructing a differential equation based on the accumulation of the driving force signal, and outputting a predicted value, performing conversion based on the obtained predicted value to obtain instantaneous data; S500: Calculating a control error based on instantaneous data, performing control adjustments based on the control error, and storing the adjustment results in a database.

2. The method according to claim 1, characterized in that S100 includes: S110: collecting vibration data by deploying acceleration sensors; S120: Decomposing and reconstructing the collected data using wavelet decomposition technology to obtain denoised data; S130: Based on the denoised data, a normalization operation is performed on the denoised data using a normalization method to obtain normalized data.

3. The method according to claim 2, characterized in that S200 includes: S210: Calculate the mean value using the mean formula based on the denoised data, define the obtained mean value as the target signal, and subtract the normalized data from the target signal to obtain the vibration deviation; S220: Calculate the vibration deviation of the previous time using a time series analysis method, and subtract the vibration deviation from the vibration deviation of the previous time to obtain a vibration deviation change rate; S230: defining the vibration deviation and the vibration deviation change rate as input variables, performing fuzzy conversion based on the obtained input variables using a triangular membership function to obtain converted membership values, and integrating the obtained membership values to obtain a fuzzy value set of the vibration deviation and a fuzzy value set of the vibration deviation change rate; S240: defining a fuzzy control rule table, taking the fuzzy value set of the vibration deviation and the fuzzy value set of the vibration deviation change rate as inputs of the Mamdani fuzzy inference method, and using the Mamdani fuzzy inference method in combination with the rule activation strength and the fuzzy set truncation formula to infer the fuzzy control rule table to obtain an inferred control quantity fuzzy set, and merging the control quantity fuzzy sets of all rules using the maximum membership method based on the inferred control quantity fuzzy set to obtain a merged fuzzy set; S250: Divide the merged fuzzy set into M discrete points by an equal interval discretization method, and calculate the membership value of each discrete point by using a triangular membership function; S260: Set a selection threshold, compare the membership value corresponding to the discrete point with the threshold, and retain the discrete point when the membership value corresponding to the discrete point is greater than or equal to the threshold; otherwise, remove the discrete point; and calculate the preliminary control amount based on the retained discrete point using the discretization formula.

4. The method according to claim 3, characterized in that In S230, based on the obtained input variables, a triangular membership function is used to perform fuzzy conversion to obtain the converted membership value, which is specifically: The fuzzy formula of vibration deviation is: ； Where, Indicates the vibration deviation of the input variable Membership value of the set of fuzzy values belonging to the vibration deviation, Indicates vibration deviation, Indicates the maximum absolute value of vibration deviation; The fuzzy value set of vibration deviation is expressed as: ； Where, Indicates a large negative. Indicates a small negative. represents the center value, Indicates positive small, It means upright; The fuzzy formula of the vibration deviation change rate is: ； Where, Indicates the input variable vibration deviation change rate The membership value of the fuzzy value set belonging to the vibration deviation change rate, Indicates the vibration deviation change rate, Indicates the maximum absolute value of the vibration deviation change rate; The fuzzy value set of the vibration deviation change rate is expressed as: ； Where, Indicates negative speed, Indicates zero speed, Indicates positive speed; The rule activation formula in the Mamdani fuzzy inference method in S240 is: ； Where, represents the rule activation strength, Indicates taking the minimum value of the membership value; The truncation formula of the rule output fuzzy set is: ； Where, Indicates the activation intensity Truncated fuzzy set membership function, Represents the output variable obtained through Mamdani fuzzy inference, The original membership function of the fuzzy set representing the output of the fuzzy rule; Based on the inferred control quantity fuzzy set, the maximum membership method is used to merge the control quantity fuzzy sets of all rules to obtain the merged fuzzy set, which is specifically: ； Where, represents the merged fuzzy set, Indicates taking the maximum membership value, represents the fuzzy set output by the nth rule; The discretization formula used in S250 to calculate the preliminary control quantity is as follows: ； Where, represents the initial control quantity at time t, represents the weight of the discrete points, Represents the i-th discrete point The membership value of , S represents the total number of discrete points retained.

5. The method according to claim 4, characterized in that S400 includes: S410: Calculating a driving force signal according to the final control amount, and using a sliding window method to obtain the vibration data and driving force signal of the previous moment based on the normalized vibration data and driving force signal; S420: constructing a time series set at each time point based on the driving force signal at the previous moment and the corresponding vibration data at the previous moment, and accumulating the driving force signal and vibration data in the time series set to obtain an accumulated driving force signal and accumulated vibration data; S430: Construct a rectangular matrix based on the accumulated driving force signal and the accumulated vibration data, construct a target vector based on the accumulated vibration data, and perform a transposition operation on the rectangular matrix and the target vector to obtain a parameter vector Then, the matrix decomposition technology is further used to decompose the vibration data to obtain the coefficients , coefficient of driving force signal and static offset ; S440: Based on the accumulated driving force signal, accumulated vibration data, coefficients of the vibration data, coefficients of the driving force signal and the static offset, a differential equation is constructed to output the accumulated value at the future moment, and the accumulated value at the future moment is converted using the cumulative subtraction generation method to obtain instantaneous data.

6. The method according to claim 5, characterized in that In S410, the driving force signal is calculated according to the final control amount, specifically: ； Where, Indicates the driving force signal, represents the stiffness parameter, set through experiments, is the final control quantity; In S430, a rectangular matrix is constructed based on the accumulated driving force signal and the accumulated vibration data, specifically: ； Where, represents a rectangular matrix, represents the total number of data points in the time series, Indicates the accumulated vibration data, Indicates the cumulative driving force signal; In S430, a target vector is constructed based on the accumulated vibration data, specifically: ； Where, represents the target vector, Indicates the total length of the time series; In S430, the rectangular matrix and the target vector are transposed using a transposition operation to obtain a parameter vector, which is further decomposed using a matrix decomposition technique to obtain coefficients of the vibration data, coefficients of the driving force signal, and a static offset, specifically: ； ； Where, represents the transpose operation, represents the parameter vector, The coefficients representing the vibration data, represents the coefficient of the driving force signal, Indicates static offset; In S440 , a differential equation is constructed based on the accumulated driving force signal, accumulated vibration data, coefficients of the vibration data, coefficients of the driving force signal, and the static offset, and the accumulated value at the future moment is output. Specifically, it is: ； Where, Indicates the future moment The accumulated value of In S440, the accumulated value at the future moment is converted using the cumulative-subtraction generation method to obtain instantaneous data, specifically: ； Where, Represents instantaneous data, Represents the accumulated value at the current time t.

7. The method according to claim 6, characterized in that In S500, the control error is calculated based on the instantaneous data, and the control adjustment based on the control error includes: S510: Subtracting the instantaneous data from the normalized vibration data to obtain a control error; S520: Setting an optimization threshold, comparing the control error with the threshold, when the control error is greater than or equal to the threshold, re-optimizing the control amount through the final optimization model, and when the control error is less than the threshold, using the control error as the basis for vibration isolation control.

8. The method according to claim 7, characterized in that S500 uses a database to store adjustment results, including: Based on the obtained control error, a database is used for storage. During the storage process, the control error is recorded in time series, and the vibration data and time series set corresponding to the error are stored in association. The historical error data is classified and organized through the database, and trend analysis data is generated.

9. An intelligent vibration isolation control system based on vibration monitoring using the method according to any one of claims 1 to 8, characterized in that: It includes data preprocessing module, preliminary control quantity calculation module, model optimization module, integration and transformation module and adjustment and storage module; A data preprocessing module is used to collect vibration data, preprocess the vibration data, and obtain preprocessed data; The preliminary control quantity calculation module is used to calculate the target signal and vibration deviation based on the preprocessed data, analyze the vibration deviation change rate, perform fuzzy conversion on the vibration deviation and change rate and integrate the membership values, infer and merge the control quantity fuzzy sets through the Mamdani fuzzy inference method, divide the discrete points and calculate the membership values, screen the discrete points and calculate the preliminary control quantity; The model optimization module is used to build the optimization model and define the objective function, randomly initialize the membership value and generate the initial population, iteratively optimize the population position using the locust optimization algorithm until the objective function value converges, output the final optimization model, and use the preliminary control quantity as the input of the final optimization model to obtain the final control quantity; The integration and conversion module is used to calculate the driving force signal based on the final control quantity, further construct a differential equation based on the accumulation of the driving force signal, and output a predicted value. The predicted value is converted to obtain instantaneous data; The adjustment and storage module is used to calculate the control error based on the instantaneous data, perform control adjustment based on the control error, and use the database to store the adjustment results.

Images