CN112098755A - Fault early detection method and system based on parallel sampling - Google Patents
Fault early detection method and system based on parallel sampling Download PDFInfo
- Publication number
- CN112098755A CN112098755A CN202010956356.8A CN202010956356A CN112098755A CN 112098755 A CN112098755 A CN 112098755A CN 202010956356 A CN202010956356 A CN 202010956356A CN 112098755 A CN112098755 A CN 112098755A
- Authority
- CN
- China
- Prior art keywords
- parallel sampling
- signal
- sampling
- fault detection
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention belongs to the technical field of signal processing, and discloses a parallel sampling-based early fault detection method and a parallel sampling-based early fault detection system, wherein the parallel sampling-based early fault detection method comprises the following steps: sampling the nth data sequence, selecting continuous m points, and constructing a data array; performing spectrum analysis on the data array; weighting and summing; and recording the failure and updating. According to the parallel sampling-based early fault detection method, the weak signals are enhanced by using random noise modulation, so that early fault detection is facilitated; the stochastic resonance algorithm which has low complexity and enhances the signal by noise is used as a core algorithm to meet the requirement of online detection; noise has an enhancing effect on the signal; the joint production of historical data and instant data is fully considered, and the output signal is more stable.
Description
Technical Field
The invention belongs to the technical field of signal processing, and particularly relates to a fault early detection method and a fault early detection system based on parallel sampling.
Background
Currently, weak signal detection is an important study in signal processing. How to eliminate noise interference has been a research focus of weak signal detection. However, most studies attempt to analyze the statistical properties of weak signals and noise in the aspects of information theory, electronics, and physical methods, and to construct filters to extract the weak signals. However, when the signal and noise bands overlap, the filtering method based on the noise cancellation idea will be adversely affected: firstly, weak signals with low signal-to-noise ratio are difficult to detect; the second is that detection inevitably causes signal damage or information loss.
Stochastic resonance based methods (SR) are an efficient new approach. SR means that noise energy will be converted into signal energy. When the input noise is matched to a particular non-linear system, to some extent the SNR of the output signal will increase and we can clearly detect the signal, which is essentially noise modulated. Stochastic Resonance (SR) theory was originally proposed by BENZI in 1981: in some non-linear systems, there is a non-zero value of noise, and the positive effect of the noise is exploited to enhance the weak signal and provide an optimal signal-to-noise ratio (SNR) output.
On the other hand, the automation degree of modern large-scale production is higher and higher, the structure of modern equipment is more and more complex, the functions are more and more perfect, and the relation among all parts in the equipment is more and more compact. For a dynamic system, a fault, an unstable working state or a performance disorder occurs inside the system, which not only causes huge economic loss, but also causes casualties and serious social influence. Obviously, most of the research methods adopt a complex off-line algorithm aiming at machine operation faults in a strong interference environment, and the real-time performance of fault monitoring is seriously influenced; the collected data is only for a certain time of data length, and the historical data is poor in application.
Through the analysis, in the detection of the fault signal in the early operation stage of the machine, the problems and the defects in the prior art are as follows:
(1) the conventional detection method fails aiming at the early machine operation fault in the strong interference environment;
(2) most of the existing methods adopt a relatively complex algorithm aiming at machine operation faults in a strong interference environment, and the real-time performance of fault monitoring is seriously influenced;
(3) the collected data is only for a certain time of data length, and the historical data is poor in application.
The difficulty in solving the above problems and defects is:
the invention effectively enhances weak signals by using noise by means of a stochastic resonance technology, and the complexity is not increased; compared with a plurality of methods, the non-parameter-based estimation technology has the advantages that the calculation amount is less and the adaptability is stronger as can be seen from the expression of signal processing.
The significance of solving the problems and the defects is as follows:
considering that a fault signal is weak and greatly influenced by noise in the early operation of a machine, the detection performance of the conventional technology is poor or the technical complexity is high, the influence of the noise on the signal enhancement function, historical data and the current monitoring data on the detection result is fully utilized, the early operation fault can be timely found, and the method has great economic and social values on safe production and product quality.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a parallel sampling-based early fault detection method.
The invention is realized in such a way that a parallel sampling-based early fault detection method comprises the following steps:
step one, sampling the nth data sequence, selecting continuous m points and constructing a data array.
And step two, performing spectrum analysis on the data array.
And step three, weighting and summing.
And step four, recording and updating the operation faults of the machine.
Further, in step one, the sampling sequence is composed of a plurality of harmonics, and the actual signal harmonics are only a small part. The random signal is represented as:
x(n)＝∑exp(jωn)+w(n)；
the fourier transform of a random signal can be given by the following equation:
wherein the content of the first and second substances,
a Fourier transform representing the signal; the fourier transform, which represents noise, consists of many harmonics, varying with different sample sequences. When there is a frequency coincidence between the two, the energy of the noise is diffused into the signal.When noise matches some signal characteristics, the signal frequency characteristics will be enhanced and clearly displayed, but the signal and noise frequency matches with randomness.
Under the influence of the driving force s (t), a one-dimensional nonlinear system u (x) with a potential function, the motion of the brownian particles is given by the following equation:
a bistable system is a classical nonlinear SR-generating system, and a simple symmetric bistable state has the following equation:
for a>0 and b>0, the potential is bistable. From a simple algebra point of view, there is an unstable state xu0, and two stable states
This can be obtained from the following equation:
in bistable systems, when the system parameters a and b, in particular b, are small, the influence caused by sinusoidal signals and white noise: a noise modulated sinusoidal signal may be output at the system.
Further, in step two, the signal output spectrum is given by the following formula:
where M represents the data length, i.e. the next M historical analyses. Can usually make
Further, the method for early fault detection based on parallel sampling further comprises the following steps:
and detecting signals by using FFT (fast Fourier transform), and judging the running state of the machine by parallel sampling and fusion of historical data and instant data.
Another object of the present invention is to provide a parallel sampling based early fault detection system, which includes:
the data array construction module is used for sampling the nth data sequence, selecting continuous m points and constructing a data array;
the spectrum analysis and processing module is used for carrying out spectrum analysis and weighted summation on the data array;
and the fault record updating module is used for recording and updating faults.
Further, the parallel sampling based early fault detection system further comprises:
and the machine running state analysis module is used for detecting signals through FFT (fast Fourier transform), and judging the machine running state through parallel sampling and fusion of historical data and instant data.
It is a further object of the invention to provide a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
sampling the nth data sequence, selecting continuous m points, and constructing a data array;
performing spectrum analysis on the data array;
weighting and summing;
and recording the failure and updating.
It is a further object of the present invention to provide a computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface for implementing said parallel sampling based early detection of faults method when executed on an electronic device.
It is another object of the present invention to provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the parallel sampling-based early failure detection method.
The invention also aims to provide a motor fault detector for implementing the parallel sampling-based early fault detection method.
By combining all the technical schemes, the invention has the advantages and positive effects that: according to the parallel sampling-based early fault detection method, the weak signals are enhanced by using random noise modulation, so that early fault detection is facilitated; the stochastic resonance algorithm which has low complexity and enhances the signal by noise is used as a core algorithm to meet the requirement of online detection; noise has an enhancing effect on the signal; the joint production of historical data and instant data is fully considered, and the output signal is more stable.
Experiments show that through comparison of the graph 4 and the graph 5, a fault signal in the early operation stage of the machine is weak, the machine is submerged in a strong interference background signal, the traditional detection method fails, and the method provided by the invention has a good detection effect.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained from the drawings without creative efforts.
Fig. 1 is a flowchart of a parallel sampling-based early fault detection method according to an embodiment of the present invention.
FIG. 2 is a schematic diagram of an improved SR algorithm provided by an embodiment of the present invention.
Fig. 3 is a schematic diagram of an FFT spectrum of a noiseless signal according to an embodiment of the present invention.
FIG. 4 shows an FFT spectrum (noise variance σ) of a noisy signal according to an embodiment of the present invention2Or 5) schematic.
FIG. 5 is a graph of a parallel-sampled stochastic resonance FFT spectrum (noise variance σ) provided by an embodiment of the present invention2Or 5) schematic.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In view of the problems in the prior art, the present invention provides a method for early detection of a fault based on parallel sampling, which is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the method for early detection of a fault based on parallel sampling according to an embodiment of the present invention includes the following steps:
s101, sampling the nth data sequence, selecting continuous m points and constructing a data array.
And S102, carrying out spectrum analysis on the data array.
And S103, weighting and summing.
And S104, recording and updating the operation faults of the machine.
The invention provides a parallel sampling-based early fault detection system, which comprises:
the data array construction module is used for sampling the nth data sequence, selecting continuous m points and constructing a data array;
the spectrum analysis and processing module is used for carrying out spectrum analysis and weighted summation on the data array;
and the fault record updating module is used for recording and updating faults.
And the machine running state analysis module is used for detecting signals through FFT (fast Fourier transform), and judging the machine running state through parallel sampling and fusion of historical data and instant data.
The present invention will be further described with reference to the following examples.
Example 1
In order to diagnose a fault in the system, it needs to be detected first. When a system fails, the type, the position and the reason of the failure need to be diagnosed, and finally, failure diagnosis and noise signal processing in the recovery of a solution for realizing the failure are provided, so that the method has important significance.
Based on the thought, the invention provides a novel weak signal detection method under the conditions of noise modulation of noise stochastic resonance and multi-sensor parallel sampling.
Noise modulation
For a noise contaminated signal, all practical signal detection methods are limited by the noise power or signal-to-noise ratio. Equation (2) shows that the sample sequence is composed of multiple harmonics, and the actual signal harmonics are only a small fraction. If the actual signal does not represent a signal salient feature of the sequence, it cannot be distinguished. All detection methods suffer from this property limitation, which, unlike other methods, improves the salient features of the frequency domain.
In general, the random signal is written as:
x(n)＝∑exp(jωn)+w(n) (1)
its fourier transform can be given by the following equation:
in the formula (2), the first and second groups,
a Fourier transform representing the signal; the fourier transform, which represents noise, consists of many harmonics, varying with different sample sequences. When there is a frequency coincidence between the two, the energy of the noise is diffused into the signal. When the noise matches some signal characteristics, the signal frequency characteristics will be emphasized and clearly displayed. But the signal is frequency matched to the noise with randomness.
(II) stochastic resonance:
under the influence of the driving force s (t), a one-dimensional nonlinear system u (x) with a potential function, the motion of the brownian particles is given by the following equation:
a bistable system is a classical nonlinear SR-generating system, and a simple symmetric bistable state has the following equation:
for a>0 and b>0, the potential is bistable. From a simple algebra point of view, there is an unstable state x u0, and two stable states
From equation (4), one can derive:
the invention makes it possible to conclude that in bistable systems, when the system parameters a and b, in particular b, are small, the influence caused by sinusoidal signals and white noise: a noise modulated sinusoidal signal may be output at the system.
(III) outputting and judging frequency spectrum
And detecting signals by using FFT (fast Fourier transform), and judging the running state of the machine by parallel sampling and fusion of historical data and instant data.
Example 2
For mechanical operation vibration signals containing noise interference, a plurality of fault detection methods exist, but most of the fault detection methods are not suitable for online real-time detection. The FFT algorithm has high calculation speed and strong adaptability, and is a preferred algorithm for industrial online detection. However, when the industrial machine operates in a strong random interference environment, the stability of the FFT spectrum is poor. In order to solve the problem of fault detection in a strong interference environment, an SR algorithm-based fault signal FFT detection method is designed, as shown in fig. 2.
Step 1: sampling the nth data sequence, selecting continuous m points, and constructing a data array;
step 2: performing spectrum analysis on the data array;
and 3, step 3: weighting and summing;
and 4, step 4: and recording the failure and updating.
In the algorithm, the weight of the SR analysis coefficient not only plays a role in learning new frequency components, but also weakens the interference of burst random noise. The signal output spectrum is given by (6).
In (6), M represents the data length, i.e., the next M number of history analyses. Can usually make
Simulation analysis: according to the proposed method, the following experiments are designed, obtaining a fault signal:
S＝0.3*sin(2*pi*0.01*t)+0.5*sin(2*pi*0.03*t)+w(n)
the above signals were simulated with a data window of 4096, a noise-free FFT spectrum as shown in fig. 3, a noise-added FFT spectrum as shown in fig. 4, and a parallel-sampled stochastic resonance FFT spectrum as shown in fig. 5.
Obviously, as can be seen from fig. 4 and 5, the performance of the stochastic resonance weighting algorithm for detecting early machine fault signals is significant, and the improved SR algorithm has stronger weak signal capability and is suitable for early machine operation fault diagnosis.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When used in whole or in part, can be implemented in a computer program product that includes one or more computer instructions. When loaded or executed on a computer, cause the flow or functions according to embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.)). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. The parallel sampling-based early fault detection method is characterized by comprising the following steps of:
sampling the nth data sequence, selecting continuous m points, and constructing a data array;
performing spectrum analysis on the data array;
weighting and summing;
and recording and updating the operation faults of the machine.
2. The parallel sampling based early fault detection method of claim 1, wherein the random signal of the sampling sequence is represented as:
x(n)＝∑exp(jωn)+w(n)；
the fourier transform of the random signal is the following equation:
wherein the content of the first and second substances,
a Fourier transform representing the signal; a fourier transform representing noise, consisting of many harmonics, varying with different sampling sequences; when the two have frequency coincidence, the energy of the noise is diffused into the signal; when noise matches some signal characteristics, the signal frequency characteristics are enhanced and clearly displayed;
under the driving force s (t), a one-dimensional nonlinear system u (x) with a potential function, the motion of the brownian particles is as follows:
the symmetric bistable state potential is given by the following equation:
for a>0 and b>0, the potential is bistable; the method comprises the following steps: unstable state is xu0, two stable states
The following equation:
in bistable systems, when the system parameter b is small, the influence caused by sinusoidal signals and white noise: the noise modulated sinusoidal signal outputs a bistable system.
3. The method for early detection of a fault based on parallel sampling according to claim 1, wherein the signal output spectrum is as follows:
wherein M represents the data length, and M adjacent historical analyses; order to
4. The parallel sampling based early fault detection method of claim 1, wherein the parallel sampling based early fault detection method further comprises:
and detecting signals by using FFT (fast Fourier transform), and judging the running state of the machine by parallel sampling and fusion of historical data and instant data.
5. An early fault detection system based on parallel sampling, characterized in that the early fault detection system based on parallel sampling comprises:
the data array construction module is used for sampling the nth data sequence, selecting continuous m points and constructing a data array;
the spectrum analysis and processing module is used for carrying out spectrum analysis and weighted summation on the data array;
and the fault record updating module is used for recording and updating faults.
6. The parallel sampling based early fault detection system of claim 5,
the parallel sampling based early fault detection system further comprises:
and the machine running state analysis module is used for detecting signals through FFT (fast Fourier transform), and judging the machine running state through parallel sampling and fusion of historical data and instant data.
7. A computer device, characterized in that the computer device comprises a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to carry out the steps of:
sampling the nth data sequence, selecting continuous m points, and constructing a data array;
performing spectrum analysis on the data array;
weighting and summing;
and recording the failure and updating.
8. A computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface for implementing a method for parallel sampling based early detection of faults as claimed in any one of claims 1 to 4 when executed on an electronic device.
9. A computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform a method for early detection of a fault based on parallel sampling according to any one of claims 1 to 4.
10. A motor fault detector for implementing the parallel sampling based early fault detection method according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010956356.8A CN112098755A (en) | 2020-09-11 | 2020-09-11 | Fault early detection method and system based on parallel sampling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010956356.8A CN112098755A (en) | 2020-09-11 | 2020-09-11 | Fault early detection method and system based on parallel sampling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112098755A true CN112098755A (en) | 2020-12-18 |
Family
ID=73750962
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010956356.8A Pending CN112098755A (en) | 2020-09-11 | 2020-09-11 | Fault early detection method and system based on parallel sampling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112098755A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2208981A1 (en) * | 2009-01-14 | 2010-07-21 | Lucent Technologies Inc. | Monitoring of rotating machines |
| CN103969067A (en) * | 2014-04-01 | 2014-08-06 | 中北大学 | Generalized-morphology-based automatic filling system fault diagnosis method |
| CN104483127A (en) * | 2014-10-22 | 2015-04-01 | 徐州隆安光电科技有限公司 | Method for extracting weak fault characteristic information of planetary gear |
| CN110196173A (en) * | 2019-05-05 | 2019-09-03 | 成都大学 | A kind of strong noise environment machine operation troubles online test method based on self study |
| CN111458092A (en) * | 2020-03-02 | 2020-07-28 | 河北工业大学 | Industrial robot early weak fault signal screening method |
-
2020
- 2020-09-11 CN CN202010956356.8A patent/CN112098755A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2208981A1 (en) * | 2009-01-14 | 2010-07-21 | Lucent Technologies Inc. | Monitoring of rotating machines |
| CN103969067A (en) * | 2014-04-01 | 2014-08-06 | 中北大学 | Generalized-morphology-based automatic filling system fault diagnosis method |
| CN104483127A (en) * | 2014-10-22 | 2015-04-01 | 徐州隆安光电科技有限公司 | Method for extracting weak fault characteristic information of planetary gear |
| CN110196173A (en) * | 2019-05-05 | 2019-09-03 | 成都大学 | A kind of strong noise environment machine operation troubles online test method based on self study |
| CN111458092A (en) * | 2020-03-02 | 2020-07-28 | 河北工业大学 | Industrial robot early weak fault signal screening method |
Non-Patent Citations (1)
| Title |
|---|
| 何大海等: "基于随机共振原理的微弱信号检测与应用", 《机电工程》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lv et al. | High-order synchroextracting transform for characterizing signals with strong AM-FM features and its application in mechanical fault diagnosis | |
| Xu et al. | Applying morphology to improve Canny operator's image segmentation method | |
| CN111881594B (en) | Non-stationary signal state monitoring method and system for nuclear power equipment | |
| CN104251934A (en) | Harmonic analysis method and apparatus, and method and apparatus for determining clutter in harmonic wave | |
| Chen et al. | Fault diagnosis method of rotating machinery based on stacked denoising autoencoder | |
| CN111291918B (en) | Rotating machine degradation trend prediction method based on stationary subspace exogenous vector autoregression | |
| Huijie et al. | Structured sparsity assisted online convolution sparse coding and its application on weak signature detection | |
| CN115165274A (en) | Self-adaptive intelligent monitoring device and method for vibration state of engineering mechanical equipment | |
| Deveci et al. | Transfer learning enabled bearing fault detection methods based on image representations of single-dimensional signals | |
| CN112098755A (en) | Fault early detection method and system based on parallel sampling | |
| Ren et al. | Research on fault feature extraction of hydropower units based on adaptive stochastic resonance and fourier decomposition method | |
| Wang et al. | Fast Spectral Correlation Based on Sparse Representation Self‐Learning Dictionary and Its Application in Fault Diagnosis of Rotating Machinery | |
| Yang et al. | Research on Fault Feature Extraction Method Based on FDM‐RobustICA and MOMEDA | |
| LU502454B1 (en) | A Bearing Fault Identification Method and System Based on EEMD Sparse Decomposition | |
| Ghosh | Comparative DNN model analysis for detection of various types of optical noise | |
| CN116318925A (en) | Multi-CNN fusion intrusion detection method, system, medium, equipment and terminal | |
| CN116226770A (en) | Time sequence data anomaly detection method and device | |
| Wu et al. | A novel denoising method for non‐linear and non‐stationary signals | |
| CN111343165B (en) | Network intrusion detection method and system based on BIRCH and SMOTE | |
| Zhong et al. | Fault feature extraction method of gear based on optimized minimum entropy deconvolution and accugram | |
| CN114415018A (en) | Self-learning grating interference spectrum analysis technology for motor fault early warning | |
| Zhang et al. | Application of morphological filter in pulse noise removing of vibration signal | |
| Li et al. | Fault feature extraction method based on EWT-SMF and MF-DFA for valve fault of reciprocating compressor | |
| Zhu et al. | A Novel Based on the Log‐Squared Envelope Harmonic Noise Ratio Fault Transient Feature Extraction Method | |
| Tan et al. | Graph constrained empirical wavelet transform and its application in bearing fault diagnosis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |