Piezoelectric transformer safety maintenance method and system

By preprocessing and in-depth analysis of the operating data of the piezoelectric transformer, using machine learning models to predict the optimal operating frequency and identify vibration modes, the problem of insufficient analysis in the existing technology is solved, and real-time monitoring and failure of the piezoelectric transformer is realized It is predicted that it has significantly improved its safety and stability.

CN118965055APending Publication Date: 2024-11-15ELECTRIC POWER RES INST OF GUANGXI POWER GRID CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202410837397.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-26
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The analysis of the vibration signals of piezoelectric transformers in the prior art is mainly limited to simple frequency domain or time domain analysis, and the failure to fully utilize modern signal processing and machine learning technologies for deep-level pattern recognition and fault prediction, resulting in long-term operation and complex environments. It is difficult to detect and deal with potential failures in a timely manner.

Method used

By obtaining the operating data of the piezoelectric transformer for preprocessing, the optimal operating frequency is predicted using improved adaptive frequency tuning algorithms and machine learning models, and the time-frequency domain analysis of the adjusted vibration signal is performed, the vibration mode is identified, and the driving frequency is dynamically adjusted. and operating parameters to avoid adverse states.

Benefits of technology

Real-time monitoring and intelligent analysis of the operating status of the piezoelectric transformer is realized, effectively avoiding resonance and other potential faults, and significantly improving its safety and stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965055A_ABST
    Figure CN118965055A_ABST
Patent Text Reader

Abstract

The invention discloses a piezoelectric transformer safety maintenance method and system, and relates to the technical field of piezoelectric transformer safety maintenance, and the method comprises the steps: obtaining the operation data of a piezoelectric transformer, and carrying out the preprocessing; adjusting the piezoelectric transformer by predicting the optimal working frequency under different conditions; the vibration signals of the adjusted piezoelectric transformer are analyzed and classified, and a vibration mode is recognized; and adjustment is performed according to a vibration mode identification result, and safe operation is maintained. According to the method, real-time monitoring and intelligent analysis of the operation state of the piezoelectric transformer can be achieved, resonance and other potential faults are effectively avoided through a self-adaptive adjustment mechanism, and the safety and stability of the piezoelectric transformer are remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field, and in particular to a piezoelectric transformer safety maintenance method and system. Background Art

[0002] With the rapid development of power electronics technology, piezoelectric transformers, as a new type of energy conversion device, are increasingly used in various electronic devices. Piezoelectric transformers use the inverse piezoelectric effect of piezoelectric materials to achieve voltage conversion and energy transmission through the mutual conversion between electrical energy and mechanical energy. They have significant advantages such as small size, high efficiency, and no electromagnetic interference. In recent years, with the advancement of microelectronics technology, nanomaterials, and intelligent control technology, the performance and application scope of piezoelectric transformers have been further expanded. However, despite remarkable achievements in theoretical research and application practice, existing piezoelectric transformers still face many problems during operation, especially the safety maintenance issues under long-term operation and complex environmental conditions need to be solved urgently.

[0003] At present, the safe maintenance of piezoelectric transformers mainly relies on traditional preventive maintenance and regular inspection methods. Although these methods can guarantee the normal operation of piezoelectric transformers to a certain extent, due to the lack of real-time monitoring and intelligent analysis of the operating status, potential faults and abnormalities are often unable to be discovered and handled in a timely manner. For example, when the load changes or the environmental conditions fluctuate, the piezoelectric transformer may experience problems such as resonance, detuning, and overheating, resulting in performance degradation or even damage. In addition, the existing technology for analyzing the vibration signal of piezoelectric transformers mostly stays at the simple frequency domain or time domain analysis level, and fails to make full use of modern signal processing and machine learning technology for in-depth pattern recognition and fault prediction. Therefore, there is an urgent need for a safe maintenance method for piezoelectric transformers based on real-time monitoring and intelligent analysis to improve their operating reliability and service life. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above problems or problems existing in the prior art, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the existing technology for analyzing the vibration signal of piezoelectric transformer mostly stays at the simple frequency domain or time domain analysis level, and fails to fully utilize modern signal processing and machine learning technology for in-depth pattern recognition and fault prediction.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a piezoelectric transformer safety maintenance method, which includes obtaining and preprocessing the operating data of the piezoelectric transformer;

[0008] Adjust piezoelectric transformers by predicting the optimal operating frequency under different conditions;

[0009] Analyze and classify the vibration signals of the adjusted piezoelectric transformer to identify the vibration mode;

[0010] Make adjustments based on the results of vibration pattern recognition to maintain safe operation.

[0011] As a preferred solution of the piezoelectric transformer safety maintenance method of the present invention, wherein: the obtaining of the operating data of the piezoelectric transformer and preprocessing includes:

[0012] Collecting operation data in real time through sensors, including but not limited to vibration frequency, load condition, voltage and current;

[0013] The collected operation data is processed, and the processing includes but is not limited to denoising and normalizing the data and extracting characteristic parameters.

[0014] As a preferred scheme of the piezoelectric transformer safety maintenance method described in the present invention, the prediction of the optimal operating frequency under different conditions includes: constructing a machine learning model through an improved adaptive frequency tuning algorithm; inputting the preprocessed characteristic parameters into the machine learning model; using the machine learning model to predict the optimal operating frequency under current conditions; based on the prediction results, dynamically adjusting the driving frequency of the piezoelectric transformer.

[0015] As a preferred solution of the piezoelectric transformer safety maintenance method of the present invention, wherein: the construction of the machine learning model includes:

[0016] The characteristic parameters in the running data are preprocessed and expressed as:

[0017]

[0018] Among them, x is the original feature parameter, x′ is the standardized feature parameter; μ is the mean of the feature parameter; σ is the standard deviation of the feature parameter;

[0019] Combined with the characteristic parameters, the adaptive frequency tuning algorithm is improved to dynamically adjust the driving frequency of the piezoelectric transformer, which is expressed as:

[0020] f new =f current +α·(tanh(β·(f target -f current )+o+p))

[0021]

[0022] Among them, f new is the new frequency; f current is the current frequency; f targetf is the target frequency; α is the adjustment coefficient; tanh is the hyperbolic tangent function, which is used to introduce nonlinear transformation; β and γ are adjustment parameters; δ is the interaction effect adjustment parameter; o is the weighted sum of the feature parameters, w i is the weight coefficient of the feature parameter, x′ i is the feature parameter after preprocessing, n is the number of feature parameters; p is the interaction effect between feature parameters; w ij is the weight coefficient of the interaction effect;

[0023] The optimal operating frequency is predicted by the improved neural network model, which is expressed as:

[0024]

[0025] g j (x′)=sin(a j ·x′ j )+cos(b j ·x′ j )+log(c j +(x′ j ) 2 )

[0026] Where y is the output of the machine learning model; f NN is the neural network model, x′ is the standardized feature parameter, θ is the parameter of the neural network; ∈ is the adjustment coefficient; v j is the weight coefficient, g j (x′) is the nonlinear transformation function, m is the number of nonlinear transformation functions; a j , b j and c j is the parameter of the nonlinear transformation; x′ j is the jth standardized feature parameter;

[0027] Combining the above improved adaptive frequency tuning algorithm and machine learning model, the final frequency prediction formula is expressed as:

[0028]

[0029] Among them, f optimal is the best operating frequency predicted under current conditions, f current is the current frequency.

[0030] As a preferred solution of the piezoelectric transformer safety maintenance method of the present invention, wherein: the vibration signal of the adjusted piezoelectric transformer is analyzed and classified, and the vibration mode is identified, including:

[0031] Use time domain and frequency domain analysis methods to extract characteristic parameters of vibration signals;

[0032] Based on the extracted feature parameters, a composite vibration pattern recognition model is constructed and trained using labeled vibration data.

[0033] Based on the identified vibration mode, the operating parameters of the piezoelectric transformer are adjusted to avoid unfavorable vibration conditions.

[0034] As a preferred solution of the piezoelectric transformer safety maintenance method of the present invention, wherein: the construction of the composite vibration mode recognition model includes:

[0035] Perform frequency domain analysis on the collected vibration signal, convert the time domain signal into frequency domain signal, identify the main frequency components, and express it as:

[0036]

[0037] Among them, X(f) is the frequency domain signal; x(n) is the time domain signal; N is the number of sampling points; f is the frequency; j is the imaginary unit;

[0038] Wavelet transform is performed to extract the time-frequency domain characteristics of the signal and capture the changes of the signal on different time scales, which can be expressed as:

[0039]

[0040] Among them, W ψ (a, b) are wavelet transform coefficients; x(t) is the time domain signal; is the complex conjugate of the wavelet basis function; a is the scale parameter; b is the translation parameter;

[0041] The high-dimensional feature data is reduced in dimension to extract the main feature components, which can be expressed as:

[0042] Z=XW

[0043] Among them, Z is the feature matrix after dimensionality reduction; X is the original data matrix; W is the feature vector matrix;

[0044] A comprehensive feature scoring formula is established to analyze and classify vibration modes, expressed as:

[0045]

[0046] Wherein, V(t) represents the vibration signal analysis result of the piezoelectric transformer at time t; x(t) is the real-time vibration signal of the piezoelectric transformer; eiωt It is the Fourier transform kernel function, which is used for frequency domain analysis; is the complex conjugate of the wavelet basis function, used for time-frequency domain analysis; z i is the characteristic component in principal component analysis; w k is the corresponding eigenvector; A and B are the upper limits of the scale and translation parameters of the wavelet transform, respectively; K is the number of principal components; I is the dimension of the input data.

[0047] As a preferred solution of the piezoelectric transformer safety maintenance method of the present invention, the adjustment according to the result of vibration mode recognition includes:

[0048] Based on the value range of the comprehensive feature scoring formula, the thresholds and features for identifying resonance phenomena, imbalance states and external interference are defined;

[0049] Identify resonance phenomena, imbalance conditions and external disturbances.

[0050] Another object of the present invention is to provide a system for safe maintenance of piezoelectric transformers, which can adjust the piezoelectric transformer by predicting the optimal operating frequency under different conditions; analyze and classify the vibration signals of the adjusted piezoelectric transformer, identify the vibration mode, and solve the current problem of inaccurate safe maintenance of piezoelectric transformers.

[0051] As a preferred solution of the system for safe maintenance of piezoelectric transformers of the present invention, it includes: a collection module, a prediction module, an analysis module and an identification module;

[0052] The acquisition module is used to obtain the operating data of the piezoelectric transformer and perform preprocessing;

[0053] The prediction module is used to adjust the piezoelectric transformer by predicting the optimal operating frequency under different conditions;

[0054] The analysis module is used to analyze and classify the vibration signal of the adjusted piezoelectric transformer and identify the vibration mode;

[0055] The recognition module is used to make adjustments based on the results of vibration pattern recognition to maintain safe operation.

[0056] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the piezoelectric transformer safety maintenance method according to any one of claims 1 to 7 when executing the computer program.

[0057] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the piezoelectric transformer safety maintenance method described in any one of claims 1 to 7 are implemented.

[0058] Beneficial effects of the present invention: The present invention obtains and pre-processes the operating data of the piezoelectric transformer, and uses advanced prediction algorithms to optimize the operating frequency of the piezoelectric transformer under different conditions, thereby achieving dynamic adjustment. By deeply analyzing and classifying the vibration signal of the adjusted piezoelectric transformer, different vibration modes are identified, and corresponding adjustments are made according to the pattern recognition results to ensure that the piezoelectric transformer maintains the best state under various operating conditions. Compared with the prior art, the present invention can not only realize real-time monitoring and intelligent analysis of the operating status of the piezoelectric transformer, but also effectively avoid resonance and other potential faults through an adaptive adjustment mechanism, significantly improving the safety and stability of the piezoelectric transformer. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0060] Figure 1 The figure is an overall flow chart of the safe maintenance method for piezoelectric transformers.

[0061] Figure 2 This is the overall flow chart of the piezoelectric transformer safety maintenance system.

[0062] Figure 3 It is a schematic diagram of the overall device of the computer. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0064] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0065] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0066] Example 1

[0067] Reference Figures 1 to 3 , as an embodiment of the present invention, provides a piezoelectric transformer safety maintenance method, comprising:

[0068] S1: Obtain the operating data of the piezoelectric transformer and perform preprocessing.

[0069] Furthermore, the obtaining and preprocessing of the operating data of the piezoelectric transformer includes:

[0070] Collect operation data in real time through sensors;

[0071] The collected operation data is processed, and the processing includes but is not limited to denoising and normalizing the data and extracting characteristic parameters.

[0072] S2: Adjust the piezoelectric transformer by predicting the optimal operating frequency under different conditions.

[0073] Furthermore, the prediction of the optimal operating frequency under different conditions includes:

[0074] Building a machine learning model with an improved adaptive frequency tuning algorithm;

[0075] Input the preprocessed feature parameters into the machine learning model;

[0076] Use machine learning models to predict the optimal operating frequency under current conditions;

[0077] Based on the prediction results, the driving frequency of the piezoelectric transformer is dynamically adjusted.

[0078] Furthermore, the construction of the machine learning model includes:

[0079] The characteristic parameters in the running data are preprocessed and expressed as:

[0080]

[0081] Among them, x is the original feature parameter, x′ is the standardized feature parameter; μ is the mean of the feature parameter; σ is the standard deviation of the feature parameter;

[0082] Combined with the characteristic parameters, the adaptive frequency tuning algorithm is improved to dynamically adjust the driving frequency of the piezoelectric transformer, which is expressed as:

[0083] f new =f current +α·(tanh(β·(f target -f current )+o+p))

[0084]

[0085] Among them, fnew is the new frequency; f current is the current frequency; f targetf is the target frequency; α is the adjustment coefficient; tanh is the hyperbolic tangent function, which is used to introduce nonlinear transformation; β and γ are adjustment parameters; δ is the interaction effect adjustment parameter; o is the weighted sum of the feature parameters, w i is the weight coefficient of the feature parameter, x′ i is the feature parameter after preprocessing, n is the number of feature parameters; p is the interaction effect between feature parameters; w ij is the weight coefficient of the interaction effect;

[0086] It should be noted that, when predicting the optimal operating frequency of a piezoelectric transformer, the characteristic parameters include vibration frequency, load conditions, voltage, and current. These characteristic parameters have the following interactive effects:

[0087] Interaction effect of voltage and current: Voltage and current jointly affect the operating frequency of the piezoelectric transformer. When the voltage changes, the trend of the current will also adjust accordingly, affecting the driving frequency.

[0088] Interaction effect of vibration frequency and load condition: The vibration frequency and load condition act together on the performance of the piezoelectric transformer. Under high load conditions, the effect of vibration frequency on frequency tuning may be more significant.

[0089] By introducing the interaction effects of characteristic parameters (such as the interaction between voltage and current, and the interaction between vibration frequency and load conditions) to capture the complex relationship between characteristic parameters, the accuracy and robustness of the model are improved. Especially in the case of variable data and the presence of noise, the prediction error can be effectively reduced, thereby achieving the effect of improving the accuracy and stability of frequency prediction, so that the piezoelectric transformer can always maintain the best performance under various working conditions.

[0090] The optimal operating frequency is predicted by the improved neural network model, which is expressed as:

[0091]

[0092] g j (x′)=sin(a j ·x′ j )+cos(b j ·x′ j )+log(c j +(x′ j ) 2 )

[0093] Where y is the output of the machine learning model; f NN is the neural network model, x′ is the standardized feature parameter, θ is the parameter of the neural network; ∈ is the adjustment coefficient; vj is the weight coefficient, g j (x′) is the nonlinear transformation function, m is the number of nonlinear transformation functions; a j , b j and c j is the parameter of the nonlinear transformation; x′ j is the jth standardized feature parameter;

[0094] Combining the above improved adaptive frequency tuning algorithm and machine learning model, the final frequency prediction formula is expressed as:

[0095]

[0096] Among them, f optimal is the best operating frequency predicted under current conditions, f current is the current frequency.

[0097] It should be noted that the frequency prediction formula realizes the accurate capture of the complex relationship between multi-dimensional features and the precise control of dynamically adjusting the driving frequency. It also improves the explanatory power and user operability of the model by comprehensively processing the characteristic parameters and their interaction effects. It shows higher transparency and controllability in practical applications, and achieves the effect of improving the working efficiency of piezoelectric transformers, reducing energy consumption and optimizing performance.

[0098] S3: Analyze and classify the vibration signal of the adjusted piezoelectric transformer to identify the vibration mode.

[0099] Furthermore, the analyzing and classifying the vibration signal of the adjusted piezoelectric transformer and identifying the vibration mode includes:

[0100] Use time domain and frequency domain analysis methods to extract characteristic parameters of vibration signals;

[0101] Based on the extracted feature parameters, a composite vibration pattern recognition model is constructed and trained using labeled vibration data.

[0102] Based on the identified vibration mode, the operating parameters of the piezoelectric transformer are adjusted to avoid unfavorable vibration conditions.

[0103] Furthermore, the construction of the composite vibration mode recognition model includes:

[0104] Perform frequency domain analysis on the collected vibration signal, convert the time domain signal into frequency domain signal, identify the main frequency components, and express it as:

[0105]

[0106] Among them, X(f) is the frequency domain signal; x(n) is the time domain signal; N is the number of sampling points; f is the frequency; j is the imaginary unit;

[0107] Wavelet transform is performed to extract the time-frequency domain characteristics of the signal and capture the changes of the signal on different time scales, which can be expressed as:

[0108]

[0109] Among them, W ψ (a, b) are wavelet transform coefficients; x(t) is the time domain signal; is the complex conjugate of the wavelet basis function; a is the scale parameter; b is the translation parameter;

[0110] The high-dimensional feature data is reduced in dimension to extract the main feature components, which can be expressed as:

[0111] Z=XW

[0112] Among them, Z is the feature matrix after dimensionality reduction; X is the original data matrix; W is the feature vector matrix;

[0113] A comprehensive feature scoring formula is established to analyze and classify vibration modes, expressed as:

[0114]

[0115] Wherein, V(t) represents the vibration signal analysis result of the piezoelectric transformer at time t; x(t) is the real-time vibration signal of the piezoelectric transformer; e iωt It is the Fourier transform kernel function, which is used for frequency domain analysis; is the complex conjugate of the wavelet basis function, used for time-frequency domain analysis; z i is the characteristic component in principal component analysis; w k is the corresponding eigenvector; A and B are the upper limits of the scale and translation parameters of the wavelet transform, respectively; K is the number of principal components; I is the dimension of the input data.

[0116] It should be noted that by extracting frequency domain features through Fourier transform, time-frequency domain features through wavelet transform, and main feature components through principal component analysis, a comprehensive feature score V(t) is constructed, which realizes multi-dimensional comprehensive analysis of vibration signals, captures more key features, avoids the limitations of a single analysis method, and can more comprehensively reflect the working status and vibration characteristics of the piezoelectric transformer, thereby improving the accuracy of fault identification and enabling the piezoelectric transformer to maintain stable operation in a complex and changing environment.

[0117] S4: Make adjustments based on the results of vibration pattern recognition to maintain safe operation.

[0118] Further, the adjustment according to the result of vibration pattern recognition includes:

[0119] Based on the value range of the comprehensive feature scoring formula, the thresholds and features for identifying resonance phenomena, imbalance states and external interference are defined;

[0120] Identify resonance phenomena, imbalance conditions and external disturbances.

[0121] Furthermore, the resonance phenomenon is identified: the resonance phenomenon is usually manifested as a significant increase in the amplitude of the vibration signal and a higher energy in a specific frequency range.

[0122] Identification threshold: If the value of V(t) in a certain frequency band is significantly higher than the average value of the normal working state, and the frequency is stable near the resonance frequency, it can be judged as a resonance phenomenon.

[0123] Formula threshold: Let V1 be the resonance threshold. When V(t)>V1 and the frequency is stable near the resonance frequency, it is identified as a resonance phenomenon.

[0124] Identify imbalance: An imbalance usually manifests itself as a periodic change in the vibration signal in a specific direction, which may be accompanied by multiple frequency components.

[0125] Identification threshold: If the vibration signal of V(t) in a specific direction fluctuates periodically and has multiple frequency components, and the energy of these components is higher than that in the normal working state, it can be judged as an unbalanced state.

[0126] Formula threshold: Let V2 be the unbalanced threshold. When V(t) shows periodic fluctuations and the energy of a specific frequency component exceeds V2, it is identified as an unbalanced state.

[0127] Identify external interference: External interference usually manifests itself as abnormal fluctuations or transient shocks in vibration signals, which have irregular characteristics in the time and frequency domains.

[0128] Identification threshold: If irregular transient fluctuations appear in V(t) and these fluctuations are inconsistent with the normal working state of the piezoelectric transformer, it can be judged as external interference.

[0129] Formula threshold: Let V3 be the interference threshold. When V(t) has irregular transient fluctuations and the energy of these fluctuations exceeds V3, it is identified as external interference.

[0130] It should be noted that by identifying different vibration modes such as resonance, imbalance and external interference through the comprehensive score V(t), it is possible to accurately classify and identify different types of vibration phenomena, provide targeted maintenance and adjustment measures, avoid unnecessary maintenance and downtime due to misjudgment, and improve operational efficiency, thereby significantly improving the sensitivity and accuracy of fault detection and ensuring safety and stability under different working conditions.

[0131] In summary, the present invention obtains and preprocesses the operating data of the piezoelectric transformer, and uses advanced prediction algorithms to optimize the operating frequency of the piezoelectric transformer under different conditions, thereby achieving dynamic adjustment. By deeply analyzing and classifying the vibration signal of the adjusted piezoelectric transformer, different vibration modes are identified, and corresponding adjustments are made according to the pattern recognition results to ensure that the piezoelectric transformer maintains the best state under various operating conditions. Compared with the prior art, the present invention can not only realize real-time monitoring and intelligent analysis of the operating status of the piezoelectric transformer, but also effectively avoid resonance and other potential faults through an adaptive adjustment mechanism, significantly improving the safety and stability of the piezoelectric transformer.

[0132] The above is a schematic scheme of a piezoelectric transformer safety maintenance method of this embodiment. It should be noted that the technical scheme of the piezoelectric transformer safety maintenance system and the technical scheme of the piezoelectric transformer safety maintenance method described above belong to the same concept, and the details of the technical scheme of the piezoelectric transformer safety maintenance system not described in detail in this embodiment can be referred to the description of the technical scheme of the piezoelectric transformer safety maintenance method described above.

[0133] This embodiment also provides a piezoelectric transformer safety maintenance system, including:

[0134] Acquisition module, prediction module, analysis module and recognition module;

[0135] The acquisition module is used to obtain the operating data of the piezoelectric transformer and perform preprocessing;

[0136] The prediction module is used to adjust the piezoelectric transformer by predicting the optimal operating frequency under different conditions;

[0137] The analysis module is used to analyze and classify the vibration signal of the adjusted piezoelectric transformer and identify the vibration mode;

[0138] The recognition module is used to make adjustments based on the results of vibration pattern recognition to maintain safe operation.

[0139] This embodiment further provides a computing device, which is applicable to the case of safe maintenance of a piezoelectric transformer, including:

[0140] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the piezoelectric transformer safety maintenance method proposed in the above embodiment.

[0141] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for safely maintaining a piezoelectric transformer as proposed in the above embodiment is implemented.

[0142] The storage medium proposed in this embodiment and the method for implementing safe maintenance of piezoelectric transformer proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0143] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0144] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A piezoelectric transformer safety maintenance method, characterized in that: include, Obtaining and preprocessing the operating data of the piezoelectric transformer; Adjust piezoelectric transformers by predicting the optimal operating frequency under different conditions; Analyze and classify the vibration signals of the adjusted piezoelectric transformer to identify the vibration mode; Make adjustments based on the results of vibration pattern recognition to maintain safe operation.

2. The piezoelectric transformer safety maintenance method according to claim 1, characterized in that: The step of obtaining the operating data of the piezoelectric transformer and performing preprocessing includes: Collect operation data in real time through sensors; The collected operation data is processed, and the processing includes but is not limited to denoising and normalizing the data and extracting characteristic parameters.

3. The piezoelectric transformer safety maintenance method according to claim 2, characterized in that: The prediction of the optimal operating frequency under different conditions includes, Building a machine learning model with an improved adaptive frequency tuning algorithm; Input the preprocessed feature parameters into the machine learning model; Use machine learning models to predict the optimal operating frequency under current conditions; Based on the prediction results, the driving frequency of the piezoelectric transformer is dynamically adjusted.

4. The piezoelectric transformer safety maintenance method according to claim 3, characterized in that: The construction of the machine learning model includes: The characteristic parameters in the running data are preprocessed and expressed as: Among them, x is the original feature parameter, x′ is the standardized feature parameter; μ is the mean of the feature parameter; σ is the standard deviation of the feature parameter; Combined with the characteristic parameters, the adaptive frequency tuning algorithm is improved to dynamically adjust the driving frequency of the piezoelectric transformer, which is expressed as: f new =f current +α·(tanh(β·(f target -f current )+o+p)) Among them, f new is the new frequency; f current is the current frequency; f targetf is the target frequency; α is the adjustment coefficient; tanh is the hyperbolic tangent function, which is used to introduce nonlinear transformation; β and γ are adjustment parameters; δ is the interaction effect adjustment parameter; o is the weighted sum of the feature parameters, w i is the weight coefficient of the feature parameter, x′ i is the feature parameter after preprocessing, n is the number of feature parameters; p is the interaction effect between feature parameters; w ij is the weight coefficient of the interaction effect; The optimal operating frequency is predicted by the improved neural network model, which is expressed as: g j (x′)=sin(a j ·x′ j )+cos(b j ·x′ j )+log(c j +(x′ j ) 2 ) Where y is the output of the machine learning model; f NN is the neural network model, x′ is the standardized feature parameter, θ is the parameter of the neural network; ∈ is the adjustment coefficient; v j is the weight coefficient, g j (x′) is the nonlinear transformation function, m is the number of nonlinear transformation functions; a j , b j and c j is the parameter of the nonlinear transformation; x′ j is the jth standardized feature parameter; Combining the above improved adaptive frequency tuning algorithm and machine learning model, the final frequency prediction formula is expressed as: Among them, f optimal is the best operating frequency predicted under current conditions, f current is the current frequency.

5. The piezoelectric transformer safety maintenance method according to claim 4, characterized in that: The analyzing and classifying the vibration signal of the adjusted piezoelectric transformer and identifying the vibration mode includes: Use time domain and frequency domain analysis methods to extract characteristic parameters of vibration signals; Based on the extracted feature parameters, a composite vibration pattern recognition model is constructed and trained using labeled vibration data. Based on the identified vibration mode, the operating parameters of the piezoelectric transformer are adjusted to avoid unfavorable vibration conditions.

6. The piezoelectric transformer safety maintenance method according to claim 5, characterized in that: The construction of the composite vibration mode recognition model comprises: Perform frequency domain analysis on the collected vibration signal, convert the time domain signal into frequency domain signal, identify the main frequency components, and express it as: Among them, X(f) is the frequency domain signal; x(n) is the time domain signal; N is the number of sampling points; f is the frequency; j is the imaginary unit; Wavelet transform is performed to extract the time-frequency domain characteristics of the signal and capture the changes of the signal on different time scales, which can be expressed as: Among them, W ψ (a, b) are wavelet transform coefficients; x(t) is the time domain signal; is the complex conjugate of the wavelet basis function; a is the scale parameter; b is the translation parameter; The high-dimensional feature data is reduced in dimension to extract the main feature components, which can be expressed as: Z=XW Among them, Z is the feature matrix after dimensionality reduction; X is the original data matrix; W is the feature vector matrix; A comprehensive feature scoring formula is established to analyze and classify vibration modes, expressed as: Wherein, V(t) represents the vibration signal analysis result of the piezoelectric transformer at time t; x(t) is the real-time vibration signal of the piezoelectric transformer; e iωt It is the Fourier transform kernel function, which is used for frequency domain analysis; is the complex conjugate of the wavelet basis function, used for time-frequency domain analysis; z i is the characteristic component in principal component analysis; w k is the corresponding eigenvector; A and B are the upper limits of the scale and translation parameters of the wavelet transform, respectively; K is the number of principal components; I is the dimension of the input data.

7. The piezoelectric transformer safety maintenance method according to claim 6, characterized in that: The adjustment according to the result of vibration pattern recognition includes: Based on the value range of the comprehensive feature scoring formula, the thresholds and features for identifying resonance phenomena, imbalance states and external interference are defined; Identify resonance phenomena, imbalance conditions and external disturbances.

8. A system using the piezoelectric transformer safety maintenance method according to any one of claims 1 to 7, characterized in that: It includes acquisition module, prediction module, analysis module and recognition module; The acquisition module is used to obtain the operating data of the piezoelectric transformer and perform preprocessing; The prediction module is used to adjust the piezoelectric transformer by predicting the optimal operating frequency under different conditions; The analysis module is used to analyze and classify the vibration signal of the adjusted piezoelectric transformer and identify the vibration mode; The recognition module is used to make adjustments based on the results of vibration pattern recognition to maintain safe operation.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the piezoelectric transformer safety maintenance method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the piezoelectric transformer safety maintenance method according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Distribution transformer equipment fault research and judgment method and system

    CN112729531A

  • Large transformer transportation monitoring method and system based on optical fiber sensing

    CN117553903A

  • Equipment vibration fault diagnosis method based on artificial intelligence

    CN118114186A

  • Power supply device with piezoelectric transformer and control method therefor

    JP2001054280A

  • Intrinsic adaptive and autonomic piezotransformer circuits

    US20140247633A1