A Fault Prediction Method for Power Equipment Based on Vibration Analysis

Abstract

A fault prediction method for power equipment based on vibration analysis according to the present invention includes: collecting vibration signal data of power equipment; preprocessing the vibration signal data to transform it into vibration analysis data; segmenting the vibration analysis data into multiple sub-segments according to a time series, where a sub-segment contains continuous vibration analysis data points, and extracting feature data from the vibration analysis data within the sub-segment; calculating a dynamic adjustment coefficient for the sub-segment according to the feature data within the sub-segment, using the dynamic adjustment coefficient of the sub-segment to perform dynamic smoothing processing on the vibration analysis data within the sub-segment, and calculating the deviation value between the predicted value and the actual value of the vibration analysis data within each sub-segment; presetting a deviation value threshold, comparing the deviation value with the deviation value threshold, and determining whether there is an abnormality in the vibration analysis data within the sub-segment. The present invention realizes early warning of power equipment faults, improves the operation safety and reliability of the power system, has higher accuracy, and provides strong support for the monitoring and management of power equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis and monitoring, and specifically to a method for predicting faults in power equipment based on vibration analysis. Background Art

[0002] With the continuous expansion of the scale and the increase in complexity of the power system, the stable operation of power equipment is crucial for ensuring the quality and safety of power supply. However, various faults may occur during the long-term operation of power equipment, which may not only lead to power supply interruption but also pose a serious threat to the safety of the power grid. To detect and prevent potential faults in a timely manner and improve the operation reliability of power equipment, it is particularly important to develop efficient and accurate fault prediction technologies. Therefore, the present invention aims to provide a method for predicting faults in power equipment based on vibration analysis to achieve effective detection and early warning of early faults in power equipment.

[0003] Traditional methods for detecting faults in power equipment mainly rely on regular manual inspections and simple on-line monitoring means, such as temperature monitoring, oil quality analysis, etc. Although these methods can reflect the operation status of the equipment to a certain extent, due to the lack of in-depth analysis of the internal status of the equipment, it is often difficult to detect early faults in a timely manner. In addition, manual inspections are time-consuming, laborious, costly, and there are certain subjectivity and limitations; while simple on-line monitoring means usually can only reflect the state changes on the surface of the equipment and cannot comprehensively evaluate the health status of the equipment.

[0004] In recent years, with the development of sensor technology and data analysis technology, fault diagnosis technologies based on vibration analysis have gradually become a research hotspot. In the prior art, there are already some methods for predicting faults using vibration signals. These methods usually collect the vibration signals generated during the operation of the equipment, analyze and process them to identify the abnormal status of the equipment. However, most of the existing vibration analysis methods focus on the extraction of static features and pay insufficient attention to the dynamic changes of vibration signals, resulting in poor accuracy and real-time performance of fault prediction. In addition, when dealing with vibration signals, the prior art often ignores the importance of different frequency components, making the robustness of the fault prediction model insufficient.

[0005] In view of the problems existing in the prior art, the present invention proposes a new method for predicting faults in power equipment based on vibration analysis. This method not only considers the static features of vibration signals but also introduces a dynamic adjustment coefficient to perform dynamic smoothing processing on vibration signals, improving the accuracy and stability of fault prediction. Summary of the Invention

[0006] In view of the above technical problems, the present application proposes a method for predicting faults in power equipment based on vibration analysis, including:

[0007] Collect vibration signal data of power equipment;

[0008] Preprocess the vibration signal data, and after filtering, noise reduction, and normalization processing, the vibration signal data is transformed into vibration analysis data;

[0009] Slice the vibration analysis data into multiple sub - segments according to the time series. Each sub - segment contains continuous vibration analysis data points, and extract feature data from the vibration analysis data within the sub - segment;

[0010] According to the feature data within the sub - segment, calculate the dynamic adjustment coefficient of the sub - segment, use the dynamic adjustment coefficient of the sub - segment to perform dynamic smoothing processing on the vibration analysis data within the sub - segment, and calculate the deviation value between the predicted value and the actual value of the vibration analysis data within each sub - segment;

[0011] Preset a deviation value threshold, compare the deviation value with the deviation value threshold, and determine whether there is an abnormality in the vibration analysis data within the sub - segment.

[0012] Preferably, the preprocessing of the vibration signal data is transformed into vibration analysis data through filtering, noise reduction, and normalization processing;

[0013] The filtering removes high - frequency noise and low - frequency drift in the vibration signal data through a band - pass filter, and retains the vibration signal frequency band;

[0014] The noise reduction reduces the influence of random noise further by using the wavelet transform method to denoise the filtered vibration signal data;

[0015] The normalization processing normalizes the denoised vibration signal data so that the mean of the data is 0 and the standard deviation is 1, generating vibration analysis data.

[0016] Preferably, the vibration analysis data is sliced into multiple sub - segments according to the time series, including the following steps:

[0017] Determine the sub - segment length and the step size where is the number of vibration analysis data points contained in each sub - segment, is the number of interval data points between adjacent sub - segments;

[0018] Starting from the starting point of the vibration analysis data, select sub - segments with a length of in sequence according to the step size to form multiple sub - segments;

[0019] The starting point position of the th sub - segment is calculated by the following formula:

[0020]

[0021] wherein, is the serial number of the sub - segment;

[0022] The end - point position of the th sub - segment is calculated by the following formula:

[0023]

[0024] wherein, is the position of the last data point of the th sub - segment.

[0025] Preferably, the extraction of characteristic data from the vibration analysis data within the sub - segment includes frequency - domain characteristics and time - domain characteristics;

[0026] The frequency - domain characteristics are obtained by performing a fast Fourier transform on the vibration analysis data within the sub - segment to obtain frequency - domain data :

[0027]

[0028] wherein, is the vibration analysis data of the th data point within the sub - segment, is the frequency, is the imaginary unit;

[0029] Calculate the frequency - domain energy of the vibration analysis data within the sub - segment :

[0030]

[0031] Calculate the main frequency of the vibration analysis data within the sub - segment through the frequency - domain data :

[0032]

[0033] The main frequency marks the characteristic vibration frequency of the vibration analysis signal.

[0034] Preferably, the time - domain characteristics include the following steps:

[0035] Calculate the mean value of the vibration analysis data within the sub - segment :

[0036]

[0037] wherein, is the vibration analysis data of the th data point within the sub - segment, is the sub - segment length, and are respectively the starting point and the ending point positions of the th sub-segment;

[0038] Calculate the standard deviation of the vibration analysis data within the sub-segment through the mean value of the vibration analysis data ;

[0039]

[0040] Peak value is calculated through the vibration data within the sub-segment, and the formula is as follows:

[0041]

[0042] The time-domain characteristic data is the mean value of the vibration analysis data , standard deviation and peak value , realizing the key characteristic values in the time domain characteristics.

[0043] Preferably, the dynamic adjustment coefficient of the sub-segment includes the following steps:

[0044] Calculate the comprehensive time-domain characteristic index of the sub-segment through the mean value of the vibration analysis data within the sub-segment , standard deviation and peak value ;

[0045]

[0046] Among them, , , are respectively preset weight coefficients;

[0047] Calculate the comprehensive frequency-domain characteristic index of the sub-segment through the main frequency and frequency-domain energy within the sub-segment;

[0048]

[0049] Among them, , are respectively preset weight coefficients;

[0050] Calculate the comprehensive characteristic index of the sub-segment according to the comprehensive time-domain characteristic index and the comprehensive frequency-domain characteristic index of the sub-segment;

[0051] ​​​​

[0052] Among them, is the comprehensive weight of the preset time-domain feature and frequency-domain feature;

[0053] According to the comprehensive feature index of the sub-segment calculate the dynamic adjustment coefficient of the sub-segment ;

[0054]

[0055] Among them, is the preset threshold.

[0056] Preferably, the vibration analysis data within the sub-segment is dynamically smoothed by the dynamic adjustment coefficient of the sub-segment. For each sub-segment the vibration analysis data within is dynamically smoothed, and the smoothed value of each data point is calculated , and the dynamic smoothing processing formula is:

[0057]

[0058] Among them, is the smoothed value of the previous sub-segment at the th data point.

[0059] Preferably, the deviation value is calculated by the predicted value and the actual value of the vibration analysis data within the sub-segment. The predicted value of the vibration analysis data within the sub-segment is:

[0060]

[0061] The deviation value is obtained by the difference between the predicted value and the actual value, and the formula is:

[0062]

[0063] Among them, is the deviation value of the th sub-segment at the th data point.

[0064] Preferably, the deviation value threshold includes the following steps:

[0065] Calculate the average deviation value of all sub-segments

[0066]

[0067] Among them, is the total number of sub-segments;

[0068] Calculate the standard deviation of the deviation values of all sub - segments ;

[0069]

[0070] By averaging the deviation values and the standard deviation of the deviation values a preset deviation value threshold :

[0071]

[0072] wherein, is a preset multiple coefficient.

[0073] Preferably, the vibration analysis data within the sub - segment is compared with the preset deviation value threshold through the deviation value to determine whether there is an abnormality;

[0074] If > , it is determined that the vibration data within the sub - segment is abnormal and the power equipment has a fault;

[0075] If ≤ , it is determined that the vibration data within the sub - segment is normal and the power equipment has no fault.

[0076] Compared with the prior art, the technical solution of the present application has the following technical effects:

[0077] The present invention solves the problem of noise interference through the pre - processing technology of vibration signals, obtains more accurate vibration analysis data. Use a band - pass filter to remove high - frequency noise and low - frequency drift in the vibration signal, and retain the main frequency band of the vibration signal; then use the wavelet transform method to denoise the filtered vibration signal to further reduce the influence of random noise, and perform normalization processing on the denoised vibration signal data to make the mean of the data 0 and the standard deviation 1, generating vibration analysis data. This series of pre - processing steps effectively eliminates noise interference, improves the quality of vibration signals, and thus provides a more accurate data basis for subsequent feature extraction and fault prediction.

[0078] The present invention solves the problem of data processing efficiency through the time - series segmentation technology of vibration analysis data, improves the real - time performance of fault prediction, segments the vibration analysis data into multiple sub - segments according to the time series, each sub - segment contains continuous vibration analysis data points, and determines appropriate sub - segment lengths and step sizes. In this way, a large amount of vibration data can be processed in segments, reducing the amount of data processed at one time and improving the data processing efficiency. At the same time, the division of sub - segments also helps to capture the local characteristics of vibration signals, improving the real - time performance and accuracy of fault prediction.

[0079] This invention solves the problem of adaptability to dynamic changes in vibration signals and improves the accuracy of fault prediction by calculating and applying dynamic adjustment coefficients. By calculating the comprehensive time-domain and frequency-domain characteristic indices of a subsegment, a comprehensive characteristic index for the subsegment is derived, and the dynamic adjustment coefficient for the subsegment is calculated based on the comprehensive characteristic indices. Using the dynamic adjustment coefficients to dynamically smooth the vibration analysis data within the subsegment can better adapt to the dynamic changes in the vibration signal and reduce misjudgments caused by sudden changes in the signal. This technical solution significantly improves the accuracy and robustness of fault prediction.

[0080] This invention addresses the sensitivity issue of fault detection by comparing deviation values with preset thresholds, enabling early warning of faults. By calculating the deviation between the predicted and actual values of vibration analysis data within each subsegment and comparing it with a preset deviation threshold, it determines whether the vibration data within that subsegment is abnormal. This method can promptly detect subtle vibration changes, thereby providing early warning of power equipment failures. Compared to traditional fault detection methods, the technical solution of the present invention has higher sensitivity and reliability, effectively preventing potential failure risks and ensuring the safe and stable operation of the power system.

[0081] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application so that it can be implemented in accordance with the contents of the specification, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following is a detailed description of the preferred embodiment of the present application in conjunction with the accompanying drawings.

[0082] Based on the detailed description of the specific embodiments of the present application in conjunction with the accompanying drawings below, those skilled in the art will become more aware of the above and other objects, advantages and features of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without inventive work. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0084] Figure 1 It is the system flow chart. Specific embodiments

[0085] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, rather than all of them. In the following description, specific details such as specific configurations and components are provided only to assist in a comprehensive understanding of the embodiments of this application. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described here without departing from the scope and spirit of this application. Additionally, for clarity and conciseness, descriptions of known functions and structures are omitted in the embodiments.

[0086] It should be understood that the "one embodiment" or "this embodiment" mentioned throughout the specification means that the specific features, structures, or characteristics related to the embodiment are included in at least one embodiment of this application. Therefore, the "one embodiment" or "this embodiment" that appears throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics can be combined in one or more embodiments in any suitable manner.

[0087] In addition, this application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0088] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, B exists alone, and both A and B exist simultaneously. The term " / and" in this article is a description of another association object relationship, indicating that there can be two relationships. For example, A / and B can represent: A exists alone, and both A and B exist. Additionally, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0089] The term "at least one" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, at least one of A and B can represent: A exists alone, both A and B exist simultaneously, and B exists alone.

[0090] It should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise", or any other variant thereof are intended to cover non-exclusive inclusion.

[0091] Embodiment 1

[0092] This embodiment mainly specifically describes a power equipment fault prediction method based on vibration analysis, as Figure 1 shown, including the following steps:

[0093] S1: Collect vibration signal data of power equipment; preprocess the vibration signal data, and after filtering, noise reduction, and normalization processing, the vibration signal data is transformed into vibration analysis data;

[0094] Utilize vibration sensors to monitor the vibration signal data of power equipment in real-time and continuously. Through the integration of an advanced data acquisition system, it is precisely configured to set the data acquisition interval in seconds. The high-frequency data acquisition frequency ensures the integrity and real-time nature of the collected vibration signal data, and can effectively capture the minute changes in the vibration of power equipment, providing reliable basic data for subsequent fault prediction.

[0095] Preprocess the vibration signal data

[0096] In order to improve the accuracy and reliability of subsequent analysis, preprocess the collected vibration signal data.

[0097] The preprocessing steps include filtering, noise reduction, and normalization processing, specifically as follows:

[0098] Filtering process: Remove high-frequency noise and low-frequency drift in the vibration signal data, and retain the main frequency band of the vibration signal, thereby reducing the interference of irrelevant signals. Use a band-pass filter to filter the vibration signal data. The band-pass filter can effectively remove high-frequency noise and low-frequency drift and retain the vibration signal within the intermediate frequency range.

[0099] For the filtering process, select the cut-off frequency according to the operating frequency range of the power equipment. If the operating frequency range of the power equipment is from 50 Hz to 1 kHz, select a band-pass filter with a passband range of 50 Hz to 1 kHz, and apply the selected band-pass filter to the original vibration signal data to remove high-frequency noise and low-frequency drift and retain the main vibration signal.

[0100] Noise reduction process: Reduce the influence of random noise and improve the signal-to-noise ratio of the signal. Use the wavelet transform method to perform noise reduction processing on the filtered vibration signal data. The wavelet transform is a multi-resolution analysis method that can effectively separate the signal and noise.

[0101] Using wavelet basis functions: According to the characteristics of the vibration signal, select appropriate wavelet basis functions, such as Db4 and Sym8; perform wavelet decomposition: Perform multi-scale wavelet decomposition on the filtered vibration signal data to obtain wavelet coefficients at different scales, threshold processing: Perform threshold processing on the wavelet coefficients to remove the noise components in the wavelet coefficients; for common threshold processing methods, there are hard threshold method and soft threshold method. Reconstruct the signal: According to the processed wavelet coefficients, reconstruct the vibration signal to obtain the vibration signal data after noise reduction.

[0102] The normalization processing normalizes the vibration signal data after noise reduction, making the mean of the data 0 and the standard deviation 1, and generating vibration analysis data.

[0103] The vibration analysis data generated by S1 provides high-quality basic data for subsequent feature extraction and fault prediction, ensuring the accuracy and reliability of fault prediction.

[0104] S2: Cut the vibration analysis data into multiple sub-segments according to the time series. Each sub-segment contains continuous vibration analysis data points, and extract the feature data from the vibration analysis data within the sub-segment;

[0105] Time series segmentation In the method for predicting faults in power equipment based on vibration analysis, a key step is to cut the vibration analysis data into multiple sub-segments according to the time series. This process is to better capture the characteristics of the vibration signal changing over time, so as to more accurately identify potential fault patterns. Specifically, this step includes the following aspects:

[0106] Determine the sub-segment length and step size: First, it is necessary to determine the number of vibration analysis data points included in each sub-segment (i.e., the sub-segment length) and the number of interval data points between adjacent sub-segments (i.e., the step size). For example, if 1000 vibration data points are collected per second, the sub-segment length can be selected as 1000 data points (equivalent to 1 second of data), and the step size can be set to 500 data points, which means moving half of the sub-segment length each time to form the next sub-segment. This choice depends on the requirements for time resolution in practical applications and the availability of computing resources.

[0107] Determination of sub-segments: After determining the sub-segment length and step size, starting from the starting point of the vibration analysis data, select the corresponding length of data points in sequence according to the set step size to form multiple sub-segments. For example, assuming we have a vibration signal containing 10000 data points, if the above sub-segment length and step size are used, 10 sub-segments can be formed, and each sub-segment contains 1000 data points.

[0108] Sub-segment position calculation: Starting from the starting point of the vibration analysis data, according to the step size select in sequence with a length of sub-segments are formed, resulting in multiple sub-segments;

[0109] The starting point position of the nth sub-segment is calculated by the following formula:

[0110]

[0111] where is the sequence number of the sub-segment;

[0112] The ending point position of the nth sub-segment is calculated by the following formula:

[0113]

[0114] where is the position of the last data point of the nth sub-segment;

[0115] After the vibration analysis data is segmented into multiple sub-segments, feature extraction is performed on the data within each sub-segment. This step aims to extract key information from a large amount of raw data that can reflect the equipment status. Feature extraction generally includes two main parts: time-domain features and frequency-domain features:

[0116] Time-domain features: Mainly focus on the statistical characteristics of the vibration signal on the time axis, such as mean, standard deviation, and peak value. These features can directly reflect the intensity and fluctuation of the vibration signal and are important bases for evaluating the health status of the equipment. For example, if the peak value of the vibration signal is found to be abnormally high in a certain sub-segment, it may mean that the equipment has suffered a large impact or vibration during this period.

[0117] Frequency-domain features: The time-domain signal is transformed into the frequency domain through the fast Fourier transform (FFT), and then features such as frequency-domain energy and main frequency are calculated. Frequency-domain features can reveal the frequency composition of the vibration signal and are particularly useful for diagnosing specific faults of rotating machinery. For example, faults in motor bearings often generate abnormally high energy at specific frequencies, and these problems can be located by analyzing the frequency-domain features.

[0118] For time-domain features, the following steps are included:

[0119] Calculate the mean of the vibration analysis data within the sub-segment :

[0120]

[0121] where is the vibration analysis data of the nth data point within the sub-segment, is the length of the sub-segment, and are respectively the starting point and the ending point positions of the th sub-segment;

[0122] Calculate the standard deviation of the vibration analysis data within the sub-segment through the mean value of the vibration analysis data ;

[0123]

[0124] The said peak value is calculated through the vibration data within the sub-segment, and the formula is as follows:

[0125]

[0126] The said time-domain characteristic data are the mean value, standard deviation and peak value of the vibration analysis data, realizing the key characteristic values in the time domain characteristics.

[0127] For the frequency-domain characteristics, the said frequency-domain characteristics are obtained by performing a fast Fourier transform on the vibration analysis data within the sub-segment to obtain frequency-domain data :

[0128]

[0129] wherein, is the vibration analysis data of the th data point within the sub-segment, is the frequency, is the imaginary unit;

[0130] Calculate the frequency-domain energy of the vibration analysis data within the sub-segment :

[0131]

[0132] Calculate the main frequency of the vibration analysis data within the sub-segment through the frequency-domain data :

[0133]

[0134] The said main frequency marks the characteristic vibration frequency of the vibration analysis signal.

[0135] Through the detailed analysis of the vibration data, the signs of equipment performance degradation can be detected earlier, and measures can be taken in time to avoid faults, thus significantly reducing the maintenance cost and downtime. By adopting a variety of characteristic combination methods, even when the signal is interfered to a certain extent, a high fault recognition rate can be maintained.​

[0136] S3: Calculate the dynamic adjustment coefficient of the sub-segment based on the characteristic data within the sub-segment, perform dynamic smoothing processing on the vibration analysis data within the sub-segment using the dynamic adjustment coefficient of the sub-segment, and calculate the deviation value between the predicted value and the actual value of the vibration analysis data within each sub-segment;

[0137] The dynamic adjustment coefficient of the sub-segment is calculated by the mean value , standard deviation and peak value of the vibration analysis data within the sub-segment to calculate the comprehensive time-domain characteristic index ;

[0138]

[0139] wherein, , , are respectively preset weight coefficients, where + + = 1.

[0140] Calculate the comprehensive frequency-domain characteristic index within the sub-segment through the main frequency and frequency-domain energy of the vibration analysis data within the sub-segment;

[0141]

[0142] wherein, , are respectively preset weight coefficients; where + = 1.

[0143] Calculate the comprehensive characteristic index of the sub-segment based on the comprehensive time-domain characteristic index and the comprehensive frequency-domain characteristic index of the sub-segment;

[0144]

[0145] wherein, is the comprehensive weight of the preset time-domain characteristics and frequency-domain characteristics;

[0146] Calculate the dynamic adjustment coefficient of the sub-segment based on the comprehensive characteristic index of the sub-segment;

[0147]

[0148] wherein, is the preset threshold.

[0149] Dynamically smooth the vibration analysis data within a sub - segment by means of the dynamic adjustment coefficient of the sub - segment, for each sub - segment Dynamically smooth the vibration analysis data within it, and calculate each data point Smoothing value , the dynamic smoothing formula is:

[0150]

[0151] where, is the data value of the previous sub - segment at time .

[0152] The deviation value is calculated through the predicted value and the actual value of the vibration analysis data within the sub - segment. The predicted value of the vibration analysis data within the sub - segment is:

[0153]

[0154] Deviation value is obtained by the difference between the predicted value and the actual value, and the formula is:

[0155]

[0156] where, is the th deviation value of the sub - segment at time .

[0157] Dynamic smoothing can effectively reduce the random noise in the data, make the signal smoother, and make it easier to identify the true fault characteristics. The smoothed data is more stable, which helps to improve the accuracy of the prediction model and enables more accurate monitoring of potential faults in power equipment.

[0158] S4: Preset a deviation value threshold, compare the deviation value with the deviation value threshold, and determine whether there is an abnormality in the vibration analysis data within the sub - segment;

[0159] Presetting the deviation value threshold and comparing the deviation value with the deviation value threshold are the keys to determining whether there is an abnormality in the vibration analysis data within the sub - segment; the deviation value threshold includes the following steps:

[0160] Calculate the average deviation value of all sub - segments

[0161]

[0162] where, is the total number of sub - segments;

[0163] Calculate the standard deviation of the deviation values of all sub - segments ;

[0164]

[0165] By the average deviation value and the standard deviation of the deviation value a preset deviation value threshold :

[0166]

[0167] wherein, is a preset multiple coefficient.

[0168] The vibration analysis data within the sub - segment is compared with the preset deviation value threshold through the deviation value to determine whether there is an abnormality;

[0169] If > , it is determined that there is an abnormality in the vibration data within the sub - segment, and the power equipment has a fault;

[0170] If ≤ , it is determined that there is no abnormality in the vibration data within the sub - segment, and the power equipment has no fault.

[0171] Meanwhile, for any abnormality in the vibration analysis data within a sub - segment, it is determined that the machine has a fault, further ensuring the safety of the power equipment.

[0172] Through the above implementation steps, this embodiment can more accurately determine whether there are potential faults in the power equipment, thus realizing early warning of power equipment faults and improving the operation safety and reliability of the power system. Compared with traditional methods, the technical solution of the present invention has higher accuracy and real - time performance, providing strong support for the health management of power equipment.

[0173] The above is only the preferred embodiment of the present invention, and it does not limit the protection scope of the present invention. For those skilled in the art, the present invention can have various changes and modifications; within the spirit and principle of the present invention, through conventional substitutions or those that can achieve the same function, without departing from the principle and spirit of the present invention, changes, modifications, substitutions, integrations, and parameter changes to these embodiments all fall within the protection scope of the present invention.

Claims

1. A power equipment fault prediction method based on vibration analysis, characterized in that, It includes the following steps: Collect vibration signal data of power equipment; Preprocess the vibration signal data. After filtering, noise reduction, and normalization processing, the vibration signal data is transformed into vibration analysis data; Slice the vibration analysis data into multiple sub - segments according to the time series. Each sub - segment contains continuous vibration analysis data points, and extract feature data from the vibration analysis data within the sub - segment; According to the feature data within the sub - segment, calculate the dynamic adjustment coefficient of the sub - segment, use the dynamic adjustment coefficient of the sub - segment to perform dynamic smoothing processing on the vibration analysis data within the sub - segment, and calculate the deviation value between the predicted value and the actual value of the vibration analysis data within each sub - segment; Preset a deviation value threshold, compare the deviation value with the deviation value threshold, and determine whether there is an abnormality in the vibration analysis data within the sub - segment.

2. The method for predicting faults of power equipment based on vibration analysis according to claim 1, wherein, The preprocessing of the vibration signal data is transformed into vibration analysis data through filtering, noise reduction, and normalization processing; The filtering removes high - frequency noise and low - frequency drift in the vibration signal data through a band - pass filter, and retains the vibration signal frequency band; The noise reduction further reduces the influence of random noise by using the wavelet transform method to denoise the filtered vibration signal data; The normalization processing normalizes the denoised vibration signal data so that the mean of the data is 0 and the standard deviation is 1, generating vibration analysis data.

3. The method for predicting faults of power equipment based on vibration analysis according to claim 2, wherein The vibration analysis data is sliced into multiple sub - segments according to the time series, including the following steps: Determine the sub-segment length and the step size , where is the number of vibration analysis data points included in each sub-segment, is the number of interval data points between adjacent sub-segments; Starting from the starting point of the vibration analysis data, in accordance with the step size select sub-segments with a length of in sequence to form multiple sub-segments; The starting point position of the nth sub-segment is calculated by the following formula: ； Among them, is the serial number of the sub - segment; The end point position of the nth sub-segment is calculated by the following formula: ； Among them, is the position of the last data point of the th sub-segment.

4. The method for predicting power equipment faults based on vibration analysis according to claim 3, characterized in that The extraction of feature data from the vibration analysis data within the sub - segment includes frequency - domain features and time - domain features; The frequency domain features are obtained by performing a fast Fourier transform on the vibration analysis data within the sub-segment to obtain frequency domain data : ； Among them, is the vibration analysis data of the th data point within the sub-segment, is the frequency, is the imaginary unit; Calculate the frequency-domain energy of the vibration analysis data within the sub-segment : ； Calculating the main frequency of vibration analysis data within a sub-segment from frequency domain data : ， The main frequency Mark the characteristic vibration frequency of the vibration analysis signal.

5. A method for predicting faults of power equipment based on vibration analysis according to claim 4, characterized in that, The time - domain features include the following steps: Calculate the mean value of the vibration analysis data within the sub-segment : ， Among them, is the vibration analysis data of the th data point within the sub-segment, is the length of the sub-segment, and are respectively the starting point and the ending point positions of the th sub-segment; By means of the mean value of vibration analysis data Calculate the standard deviation of vibration analysis data within the sub-segment ; ， Peak value It is calculated through the vibration data within the sub-segment, and the formula is as follows: ， The time-domain characteristic data is the mean value of vibration analysis data , standard deviation and peak value , realizing the key characteristic values in the time-domain characteristics.

6. A fault prediction method for power equipment based on vibration analysis according to claim 1, characterized in that, The dynamic adjustment coefficient of the sub - segment includes the following steps: By the mean value of the vibration analysis data within the sub-segment , standard deviation and peak value Calculate the comprehensive time-domain feature index of the sub-segment ; ， Among them, , , are respectively preset weight coefficients; By the main frequency of the vibration analysis data within the sub-segment and the frequency domain energy Calculate the comprehensive index of the frequency domain characteristics within the sub-segment ; ； Among them, and are respectively preset weight coefficients; Based on the comprehensive time-domain feature index of the sub-segment and the comprehensive frequency-domain feature index calculate the comprehensive feature index of the sub-segment ; ； Among them, is the comprehensive weight of the preset time-domain feature and frequency-domain feature; According to the comprehensive characteristic index of the sub-segment Calculate the dynamic adjustment coefficient of the sub-segment ; ； Among them, is a preset threshold value.

7. A method for predicting faults of power equipment based on vibration analysis according to claim 6, characterized in that, The vibration analysis data within the sub-segment is dynamically smoothed by the dynamic adjustment coefficient of the sub-segment, for each sub-segment the vibration analysis data within is dynamically smoothed, and the smoothed value of each data point is calculated , and the dynamic smoothing processing formula is: ； Among them, is the smoothed value of the previous sub-segment at the th data point.

8. A method for predicting faults of power equipment based on vibration analysis according to claim 1, characterized in that The deviation value is calculated by using the predicted value and the actual value of the vibration analysis data within the sub-segment, and the predicted value of the vibration analysis data within the sub-segment is as follows: ； The deviation value is obtained by the difference between the predicted value and the actual value, and the formula is: ； Among them, is the deviation value of the th sub-segment at the th data point.

9. A fault prediction method for power equipment based on vibration analysis according to claim 1, characterized in that The deviation value threshold includes the following steps: Calculate the average deviation value of all sub-segments ; ； Among them, is the total number of sub - segments; Calculate the standard deviation of the deviation values of all sub-segments ; ； By the average deviation value and the standard deviation of the deviation value a preset deviation value threshold : ； Among them, is a preset multiple coefficient.

10. A power equipment fault prediction method based on vibration analysis according to claim 8 or 9, characterized in that, The vibration analysis data within the sub-segment is compared with a preset deviation value threshold through a deviation value to determine whether there is an abnormality by comparison; ​ If > , it is determined that the vibration data within the sub-segment is abnormal and the power equipment has a fault; If ≤ , it is determined that the vibration data within the sub-segment is normal and the power equipment has no fault.

Images