High-sensitivity vibration sensing method based on Beidou difference

By utilizing the differential positioning technology and target vibration recognition model of Beidou satellite navigation system, combined with the quartile method and XGBoost algorithm, the shortcomings of traditional vibration monitoring methods in high accuracy and high sensitivity are solved, and the precise detection of micro vibrations is achieved and Analysis, suitable for vibration monitoring in different environments.

CN119066534AInactive Publication Date: 2024-12-03SHENZHEN TEFA EAST SERVICE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411106982.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional vibration monitoring methods have shortcomings in high accuracy and high sensitivity, and are difficult to meet the monitoring needs of high-risk scenarios, especially in terms of environmental interference and cost.

Method used

Using differential positioning technology based on Beidou satellite navigation system, the relative position change information is obtained by installing Beidou differential receiver, data preprocessing and inputting the trained target vibration recognition model, and vibration perception is improved by combining quartile method and XGBoost algorithm. Monitoring sensitivity and accuracy.

Benefits of technology

It realizes accurate detection and analysis of tiny vibrations, improves the sensitivity and accuracy of vibration monitoring, is suitable for vibration monitoring in different environments, and has a wide range of application prospects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066534A_ABST
    Figure CN119066534A_ABST
Patent Text Reader

Abstract

The invention provides a high-sensitivity vibration sensing method based on Beidou difference, and relates to the technical field of vibration sensing, and the method comprises the steps: installing at least one pair of Beidou difference receivers on a monitoring target, and enabling the Beidou difference receivers to be used for receiving Beidou satellite signals; performing differential processing on the Beidou satellite signal by using a differential technology to obtain relative position change information between receivers; performing data preprocessing on the relative position change information to obtain preprocessed data; inputting the preprocessed data into a trained target vibration recognition model to obtain a recognition result; and based on a quartile method, performing vibration sensing on the monitoring target according to the identification result. The method improves the sensitivity and precision of vibration monitoring, is suitable for vibration monitoring in different environments, and has a wide application prospect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of vibration sensing technology, and in particular to a high-sensitivity vibration sensing method based on Beidou differential. Background Art

[0002] With the rapid development of the industrial and construction industries, the demand for high-precision and high-sensitivity vibration monitoring is increasing. Traditional vibration monitoring methods are limited by factors such as equipment accuracy, environmental interference and cost, and are difficult to meet the monitoring needs of certain high-risk scenarios.

[0003] Beidou Navigation Satellite System (BDS) is a global satellite navigation system independently developed by China. It is also the third mature satellite navigation system after GPS and GLONASS. Beidou Navigation Satellite System, GPS of the United States, GLONASS of Russia and GALILEO of the European Union are all suppliers recognized by the United Nations Satellite Navigation Committee.

[0004] The Beidou satellite navigation system consists of three parts: the space segment, the ground segment and the user segment. It can provide high-precision, high-reliability positioning, navigation and timing services to all types of users around the world around the clock, and has short message communication capabilities. It has preliminary regional navigation, positioning and timing capabilities, with positioning accuracy at the decimeter and centimeter levels, speed measurement accuracy of 0.2 meters per second, and timing accuracy of 10 nanoseconds.

[0005] The differential positioning technology provided by the Beidou Satellite Navigation System (BDS) has high precision and strong anti-interference ability. Based on this, a technology that can use the differential positioning technology provided by the Beidou Satellite Navigation System (BDS) to realize vibration monitoring is in urgent need of emergence. Summary of the invention

[0006] In order to overcome the deficiencies of the prior art, the object of the present invention is to provide a high-sensitivity vibration sensing method based on Beidou differential.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A high-sensitivity vibration sensing method based on Beidou differential, comprising:

[0009] Install at least one pair of Beidou differential receivers on the monitoring target to receive Beidou satellite signals;

[0010] Performing differential processing on the Beidou satellite signals using differential technology to obtain relative position change information between receivers;

[0011] Performing data preprocessing on the relative position change information to obtain preprocessed data;

[0012] Inputting the preprocessed data into a trained target vibration recognition model to obtain a recognition result;

[0013] Based on the quartile method, vibration perception is performed on the monitoring target according to the recognition result.

[0014] Preferably, the training method of the target vibration recognition model includes:

[0015] Obtaining a training data set consisting of relative position change information of historical monitoring targets;

[0016] Preprocessing the training data set, and performing data fusion on the preprocessed data to obtain vibration characteristic data;

[0017] Extracting the vibration characteristic data based on the Hilbert-Huang transform algorithm to obtain a vibration signal;

[0018] Determine vibration influencing variables based on structural mechanism analysis of the monitored target;

[0019] The vibration influencing variables are used as input parameters of the preset model, the vibration signals are used as output parameters of the preset model, and the preset model is trained and its accuracy is verified using the XGBoost algorithm to obtain a trained target vibration recognition model.

[0020] Preferably, based on the quartile method, vibration sensing of the monitoring target is performed according to the recognition result, including:

[0021] Comparing the recognition result with a preset vibration mode signal of the monitoring target to obtain a vibration residual curve;

[0022] The quartile method is used to perform vibration change analysis on the vibration residual curve, and whether the monitored target is in an abnormal vibration state is determined based on the analysis result. If so, the vibration state of the monitored target is determined to be an abnormal state; if not, the vibration state of the monitored target is determined to be a normal state.

[0023] Preferably, the training data set is preprocessed, and the preprocessed data is fused to obtain vibration characteristic data, including:

[0024] Determine the outliers of the training data set according to the 3σ method, and remove the training data set according to the determination result to obtain the removed data;

[0025] Filling the eliminated data according to Newton interpolation method to obtain the preprocessed data; the preprocessed data includes horizontal vibration data and vertical vibration data;

[0026] Extracting features from the horizontal vibration data and the vertical vibration data according to preset vibration influence factors to obtain horizontal vibration feature data and vertical vibration feature data;

[0027] The horizontal vibration characteristic data and the vertical vibration characteristic data are fused to obtain the vibration characteristic data.

[0028] Preferably, the calculation formula for fusing the horizontal vibration characteristic data and the vertical vibration characteristic data is:

[0029]

[0030] Among them, x i represents the i-th horizontal vibration characteristic data, y i represents the vertical vibration characteristic data of the i-th, z i Represents the i-th vibration characteristic data.

[0031] Preferably, data preprocessing is performed on the relative position change information to obtain preprocessed data, including:

[0032] Performing multi-scale decomposition on the relative position change information based on wavelet basis functions to obtain wavelet coefficients at different decomposition scales;

[0033] The denoising threshold is constructed based on the median of the wavelet coefficients at different decomposition scales;

[0034] Constructing a denoising function using the denoising threshold;

[0035] Processing the wavelet coefficients at different decomposition scales based on the denoising function to obtain denoised wavelet coefficients;

[0036] Performing inverse wavelet transform on the denoised wavelet coefficients to obtain the denoised preprocessed data.

[0037] Preferably, the denoising threshold is constructed based on the median of the wavelet coefficients at different decomposition scales, including:

[0038] The mean square error of the wavelet coefficients at the corresponding decomposition scale is calculated according to the median of the wavelet coefficients at different decomposition scales;

[0039] The denoising threshold is constructed according to the mean square error of the wavelet coefficients; the calculation formula of the denoising threshold is:

[0040]

[0041] Among them, λ j represents the denoising threshold, σ j represents the mean square error of the wavelet coefficients at the jth decomposition scale, median(ω0 ) represents the median value of the wavelet coefficients at the jth decomposition scale, k j Represents the length of the signal at the jth decomposition scale.

[0042] Preferably, the denoising function is:

[0043]

[0044] in, represents the kth filtered wavelet coefficient at the jth decomposition scale, ω j,k represents the kth original wavelet coefficient under the jth decomposition scale, a represents the preset parameter, and sign represents the sign function.

[0045] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0046] The present invention provides a highly sensitive vibration sensing method based on Beidou differential, comprising: installing at least one pair of Beidou differential receivers on a monitoring target to receive Beidou satellite signals; performing differential processing on the Beidou satellite signals using differential technology to obtain relative position change information between the receivers; performing data preprocessing on the relative position change information to obtain preprocessed data; inputting the preprocessed data into a trained target vibration recognition model to obtain a recognition result; and based on the quartile method, performing vibration sensing on the monitoring target according to the recognition result. The present invention improves the sensitivity and accuracy of vibration monitoring, is suitable for vibration monitoring in different environments, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. 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 paying creative labor.

[0048] Figure 1 A flow chart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The purpose of the present invention is to provide a high-sensitivity vibration sensing method based on Beidou differential, which can improve the sensitivity and accuracy of vibration monitoring, is suitable for vibration monitoring in different environments, and has broad application prospects.

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a high-sensitivity vibration sensing method based on Beidou differential, comprising:

[0053] Step 100: Install at least one pair of Beidou differential receivers on the monitoring target to receive Beidou satellite signals;

[0054] Step 200: performing differential processing on the Beidou satellite signals using differential technology to obtain relative position change information between receivers;

[0055] Step 300: performing data preprocessing on the relative position change information to obtain preprocessed data;

[0056] Step 400: inputting the preprocessed data into a trained target vibration recognition model to obtain a recognition result;

[0057] Step 500: Based on the quartile method, vibration sensing is performed on the monitoring target according to the recognition result.

[0058] Specifically, in this embodiment, Beidou differential positioning technology is used to achieve accurate detection and analysis of small vibrations. Exemplarily, the monitoring target in this embodiment can be a key structure such as a bridge, a large machine, or a building.

[0059] Optionally, in this embodiment, a suitable Beidou differential receiver is first selected to ensure that it can receive Beidou satellite signals and provide differential positioning functions; secondly, the receiver is installed and ensured that it can accurately receive satellite signals.

[0060] Preferably, the training method of the target vibration recognition model includes:

[0061] Obtaining a training data set consisting of relative position change information of historical monitoring targets;

[0062] Preprocessing the training data set, and performing data fusion on the preprocessed data to obtain vibration characteristic data;

[0063] Extracting the vibration characteristic data based on the Hilbert-Huang transform algorithm to obtain a vibration signal;

[0064] Determine vibration influencing variables based on structural mechanism analysis of the monitored target;

[0065] The vibration influencing variables are used as input parameters of the preset model, the vibration signals are used as output parameters of the preset model, and the preset model is trained and its accuracy is verified using the XGBoost algorithm to obtain a trained target vibration recognition model.

[0066] Specifically, this embodiment uses the Hilbert-Huang transform algorithm, which is divided into two parts: empirical mode decomposition (EMD decomposition) and Hilbert algorithm (HT algorithm). It is a frequency domain signal processing algorithm that can process nonlinear and unstable signals. The HHT algorithm first performs EMD decomposition on the signal to obtain the intrinsic mode functions IMFs, and then performs Hilbert transform on each IMF component to extract the special mode. At the same time, the HHT side spectrum characterizes the change of the signal amplitude with the frequency in the entire frequency band. If there is energy of a certain frequency in the signal, it means that there must be a vibration wave of this frequency.

[0067] Furthermore, this embodiment uses the XGBoost algorithm for data modeling. The XGBoost (extreme gradient boosting) algorithm is an improved algorithm of the Boosting algorithm and an integrated learning algorithm that can improve the performance of the classifier through transformation. When the amount of data is large, the Boosting algorithm has the disadvantages of many calculation iterations and long time consumption. When the training set itself is huge and the data is complex and diverse, it is necessary to perform thousands or even tens of thousands of iterations to obtain a good training model, which is time-consuming and laborious. The XGBoost algorithm can solve the above-mentioned drawbacks very well. Its biggest advantage is that it can automatically use the multi-threading of the CPU for parallel calculations and optimize the algorithm to improve accuracy.

[0068] Preferably, based on the quartile method, vibration sensing of the monitoring target is performed according to the recognition result, including:

[0069] Comparing the recognition result with a preset vibration mode signal of the monitoring target to obtain a vibration residual curve;

[0070] The quartile method is used to perform vibration change analysis on the vibration residual curve, and whether the monitored target is in an abnormal vibration state is determined based on the analysis result. If so, the vibration state of the monitored target is determined to be an abnormal state; if not, the vibration state of the monitored target is determined to be a normal state.

[0071] Specifically, the quartile used in this embodiment is a common method in statistics, which is used to describe the distribution of data. It divides a set of data into four parts, each of which contains an equal amount of data. Through the quartile, this embodiment can understand the central position of the data, the degree of dispersion, and the existence of outliers, thereby realizing the analysis of the vibration residual curve.

[0072] Preferably, the training data set is preprocessed, and the preprocessed data is fused to obtain vibration characteristic data, including:

[0073] Determine the outliers of the training data set according to the 3σ method, and remove the training data set according to the determination result to obtain the removed data;

[0074] Filling the eliminated data according to Newton interpolation method to obtain the preprocessed data; the preprocessed data includes horizontal vibration data and vertical vibration data;

[0075] Extracting features from the horizontal vibration data and the vertical vibration data according to preset vibration influence factors to obtain horizontal vibration feature data and vertical vibration feature data;

[0076] The horizontal vibration characteristic data and the vertical vibration characteristic data are fused to obtain the vibration characteristic data.

[0077] Preferably, the calculation formula for fusing the horizontal vibration characteristic data and the vertical vibration characteristic data is:

[0078]

[0079] Among them, x i represents the i-th horizontal vibration characteristic data, y i represents the vertical vibration characteristic data of the i-th, z i Represents the i-th vibration characteristic data.

[0080] Optionally, the vibration characteristic data in this embodiment may include but is not limited to data information such as frequency and amplitude.

[0081] Preferably, data preprocessing is performed on the relative position change information to obtain preprocessed data, including:

[0082] Performing multi-scale decomposition on the relative position change information based on wavelet basis functions to obtain wavelet coefficients at different decomposition scales;

[0083] The denoising threshold is constructed based on the median of the wavelet coefficients at different decomposition scales;

[0084] Constructing a denoising function using the denoising threshold;

[0085] Processing the wavelet coefficients at different decomposition scales based on the denoising function to obtain denoised wavelet coefficients;

[0086] Performing inverse wavelet transform on the denoised wavelet coefficients to obtain the denoised preprocessed data.

[0087] Preferably, the denoising threshold is constructed based on the median of the wavelet coefficients at different decomposition scales, including:

[0088] The mean square error of the wavelet coefficients at the corresponding decomposition scale is calculated according to the median of the wavelet coefficients at different decomposition scales;

[0089] The denoising threshold is constructed according to the mean square error of the wavelet coefficients; the calculation formula of the denoising threshold is:

[0090]

[0091] Among them, λ j represents the denoising threshold, σ j represents the mean square error of the wavelet coefficients at the jth decomposition scale, median(ω 0 ) represents the median value of the wavelet coefficients at the jth decomposition scale, k j Represents the length of the signal at the jth decomposition scale.

[0092] Preferably, the denoising function is:

[0093]

[0094] in, represents the kth filtered wavelet coefficient at the jth decomposition scale, ω j,k represents the kth original wavelet coefficient under the jth decomposition scale, a represents the preset parameter, and sign represents the sign function.

[0095] Specifically, under normal circumstances, when filtering, the noise of the traditional wavelet threshold function will also change with the decomposition scale, thereby affecting the denoising effect of the vibration data. However, the present invention uses the decomposition scale of the relative position change information to construct an adaptive denoising threshold and function, which can self-adjust the wavelet coefficients at different scales, thereby improving the denoising effect of the relative position change information.

[0096] Furthermore, the present embodiment can also establish a preset vibration pattern database for comparison with actual vibration characteristic data; and develop an algorithm or software to determine whether abnormal vibration exists.

[0097] Furthermore, in this embodiment, a data processing and vibration alarm process is also designed, through which the abnormal vibration information is recorded and analyzed, and the vibration alarm process can output a vibration alarm signal to notify relevant personnel or systems.

[0098] The beneficial effects of the present invention are as follows:

[0099] The present invention realizes real-time perception of vibration and effective recognition and analysis of complex vibration signals through differential technology, improves the sensitivity and accuracy of vibration monitoring, is suitable for vibration monitoring in different environments, and has broad application prospects.

[0100] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0101] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A high-sensitivity vibration sensing method based on Beidou differential, characterized in that: include: Install at least one pair of Beidou differential receivers on the monitoring target to receive Beidou satellite signals; Performing differential processing on the Beidou satellite signals using differential technology to obtain relative position change information between receivers; Performing data preprocessing on the relative position change information to obtain preprocessed data; Inputting the preprocessed data into a trained target vibration recognition model to obtain a recognition result; Based on the quartile method, vibration perception is performed on the monitoring target according to the recognition result.

2. The high-sensitivity vibration sensing method based on Beidou differential according to claim 1 is characterized in that: The training method of the target vibration recognition model includes: Obtaining a training data set consisting of relative position change information of historical monitoring targets; Preprocessing the training data set, and performing data fusion on the preprocessed data to obtain vibration characteristic data; Extracting the vibration characteristic data based on the Hilbert-Huang transform algorithm to obtain a vibration signal; Determine vibration influencing variables based on structural mechanism analysis of the monitored target; The vibration influencing variables are used as input parameters of the preset model, the vibration signals are used as output parameters of the preset model, and the preset model is trained and its accuracy is verified using the XGBoost algorithm to obtain a trained target vibration recognition model.

3. The high-sensitivity vibration sensing method based on Beidou differential according to claim 1 is characterized in that: Based on the quartile method, vibration sensing is performed on the monitoring target according to the recognition result, including: Comparing the recognition result with a preset vibration mode signal of the monitoring target to obtain a vibration residual curve; The quartile method is used to perform vibration change analysis on the vibration residual curve, and whether the monitored target is in an abnormal vibration state is determined based on the analysis result. If so, the vibration state of the monitored target is determined to be an abnormal state; if not, the vibration state of the monitored target is determined to be a normal state.

4. The high-sensitivity vibration sensing method based on Beidou differential according to claim 2 is characterized in that: The training data set is preprocessed, and the preprocessed data is fused to obtain vibration characteristic data, including: Determine the outliers of the training data set according to the 3σ method, and remove the training data set according to the determination result to obtain the removed data; Filling the eliminated data according to Newton interpolation method to obtain the preprocessed data; the preprocessed data includes horizontal vibration data and vertical vibration data; Extracting features from the horizontal vibration data and the vertical vibration data according to preset vibration influence factors to obtain horizontal vibration feature data and vertical vibration feature data; The horizontal vibration characteristic data and the vertical vibration characteristic data are fused to obtain the vibration characteristic data.

5. The high-sensitivity vibration sensing method based on Beidou differential according to claim 4 is characterized in that: The calculation formula for fusing the horizontal vibration characteristic data and the vertical vibration characteristic data is: Among them, x i represents the i-th horizontal vibration characteristic data, y i represents the vertical vibration characteristic data of the i-th, z i Represents the i-th vibration characteristic data.

6. The high-sensitivity vibration sensing method based on Beidou differential according to claim 1 is characterized in that: Performing data preprocessing on the relative position change information to obtain preprocessed data includes: Performing multi-scale decomposition on the relative position change information based on wavelet basis functions to obtain wavelet coefficients at different decomposition scales; The denoising threshold is constructed based on the median of the wavelet coefficients at different decomposition scales; Constructing a denoising function using the denoising threshold; Processing the wavelet coefficients at different decomposition scales based on the denoising function to obtain denoised wavelet coefficients; Performing inverse wavelet transform on the denoised wavelet coefficients to obtain the denoised preprocessed data.

7. The high-sensitivity vibration sensing method based on Beidou differential according to claim 6 is characterized in that: The denoising threshold is constructed based on the median of the wavelet coefficients at different decomposition scales, including: The mean square error of the wavelet coefficients at the corresponding decomposition scale is calculated according to the median of the wavelet coefficients at different decomposition scales; The denoising threshold is constructed according to the mean square error of the wavelet coefficients; the calculation formula of the denoising threshold is: Among them, λ j represents the denoising threshold, σ j represents the mean square error of the wavelet coefficients at the jth decomposition scale, median(ω0) represents the median of the wavelet coefficients at the jth decomposition scale, k j Represents the length of the signal at the jth decomposition scale.

8. The high-sensitivity vibration sensing method based on Beidou differential according to claim 7 is characterized in that: The denoising function is: in, represents the kth filtered wavelet coefficient at the jth decomposition scale, ω j,k represents the kth original wavelet coefficient under the jth decomposition scale, a represents the preset parameter, and sign represents the sign function.