CN113505898A - Equipment predictive maintenance method based on AI technology - Google Patents
Equipment predictive maintenance method based on AI technology Download PDFInfo
- Publication number
- CN113505898A CN113505898A CN202110650951.3A CN202110650951A CN113505898A CN 113505898 A CN113505898 A CN 113505898A CN 202110650951 A CN202110650951 A CN 202110650951A CN 113505898 A CN113505898 A CN 113505898A
- Authority
- CN
- China
- Prior art keywords
- data
- equipment
- predictive maintenance
- mean
- processing
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses an AI technology-based equipment predictive maintenance method, which comprises the following steps: data acquisition, data processing, feature extraction, model training and safety early warning; wherein, the data acquisition step: the method comprises the following steps that a vibration sensor is arranged on running equipment, and the vibration sensor can synchronously and continuously acquire vibration data including data in a normal state and data in an abnormal state; and (3) data processing: processing the vibration data signals through a digital signal processing algorithm, wherein the processing comprises denoising and outlier processing; the method can give an alarm in the early stage of abnormal operation of the equipment, and eliminates the safety risk of the equipment in the bud state; accurately identifying the fault risk in the early stage of the fault, sending out early warning prompt, having sufficient time for a client to prepare spare parts and repair resources, and uniformly arranging maintenance and repair by arranging in idle production or centralized maintenance; the early failure can be identified in several weeks and months in advance, and sufficient time lead is strived for by customers to prepare spare parts for purchase and arrange for maintenance.
Description
Technical Field
The invention relates to a device safety maintenance prediction technology, in particular to a device predictive maintenance method based on an AI technology.
Background
At present, most industrial enterprises adopt reparative maintenance and preventive maintenance for production equipment. Wherein, the repairability maintenance is that the maintenance is carried out after the equipment breaks down, and the maintenance to the equipment can cause the influence to the production plan, if the urgent maintenance needs spare parts to be prepared in advance, more human cost is needed, and the maintenance also can spend more expenses in addition, increases the manufacturing cost of enterprise.
For preventive maintenance, scheduled periodic equipment maintenance and spare and accessory part replacement generally comprise several modes of maintenance, periodic inspection, periodic function detection, periodic overhaul, periodic replacement and the like, the periodic maintenance needs to perform shutdown integral detection and maintenance on equipment, the time consumption is long, and the efficiency is low; on the other hand, most of them rely on empirical values, which brings about a new risk of failure.
Disclosure of Invention
The objective of the present invention is to provide a predictive maintenance method for equipment based on AI technology to solve the above problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
an AI technology based equipment predictive maintenance method comprises the following steps: data acquisition, data processing, feature extraction, model training and safety early warning;
wherein, the data acquisition step: the method comprises the following steps that a vibration sensor is arranged on running equipment, and the vibration sensor can synchronously and continuously acquire vibration data including data in a normal state and data in an abnormal state;
and (3) data processing: processing the vibration data signals by a digital signal processing algorithm, including denoising and outlier processing, and performing dimension reduction processing on the vibration data by adopting an Auto-Encoder unsupervised learning algorithm;
a characteristic extraction step: respectively carrying out short-time Fourier transform on various data signals acquired by the vibration sensor after data processing, specifically, respectively carrying out windowing operation on each data signal to obtain a spectrogram, then carrying out cepstrum analysis on the spectrogram to obtain a cepstrum, and drawing to obtain a time-frequency graph by taking a horizontal axis as time and a vertical axis as frequency;
model training: constructing a predictive maintenance model based on a CNN convolutional neural network, setting a threshold value, respectively inputting each time-frequency graph in the step of feature extraction into the predictive maintenance model to obtain a three-dimensional feature group, inputting the three-dimensional feature group into the predictive maintenance model to obtain a prediction result, and training the predictive maintenance model;
a safety early warning step: when the equipment runs, the vibration sensor collects data of the equipment, the predictive maintenance model compares the models, and when the data exceed a threshold value, the predictive maintenance cloud platform sends out an alarm message to remind a manager to process the occurrence situation.
As a preferred embodiment of the present invention: the AutoEncoder unsupervised learning algorithm comprises an encode process and a decode process, and is provided with three layers of networks, namely an input layer, a hidden layer and an output layer, wherein the hidden layer positioned in the middle is a plurality of BP neural networks; and (3) carrying out forward conduction calculation on the neurons of each layer, and calculating by using a forward conduction formula:
a2＝σ(Z2)＝σ(a1*W+b2) (1)
in the formula, the superscript number represents the number of layers, the asterisk represents convolution, b represents a bias term, and sigma represents an activation function to obtain the activation value of each layer; the residual error between the final output layer and each layer of neurons is found by using a back propagation algorithm:
in the formula, the weight parameter W and the bias values b and J are cost functions, and the W and the b are continuously updated by using a gradient descent method, so that the output is closer to the input;
and finally, acquiring distance values from the center of all observation points based on the MAD by adopting a central distance calculation method of the median absolute deviation:
firstly, calculating median mean (X) of all observation points;
then, calculating the absolute deviation value abs (X-mean (X)) of each observation point and the median;
secondly, calculating the median of the absolute deviation value abs (X-mean (X)), namely MAD ═ mean (abs (X-mean (X));
finally, the absolute deviation value abs (X-mean (X))/MAD is divided by mean (abs (X-mean (X))) to obtain a set of range values abs (X-mean (X))/MAD from the center for all the viewpoints based on MAD.
During feature extraction, windowing and framing are carried out to convert a time domain signal into a frequency domain signal, so that the attenuation relation of energy at a certain moment along with time can be well reflected; the resonant band test and fault judgment during starting and stopping of the equipment can adopt a Hamming window calculation mode, and the calculation formula of the Hamming window is as follows:
performing cepstrum analysis on the spectrogram, converting by adopting Mel frequency cepstrum coefficients, and separating direct current signal components and sinusoidal signal components by adopting DCT (discrete cosine transformation) through the cepstrum analysis, wherein the formula is a cepstrum coefficient MFCC calculation formula as follows:
wherein L is the number of filters, and the MFCC range is [ min, max ]]The MFCC is normalized, the normalization coefficient pixel is calculated and mapped to the gray value of 0-255,
where pixels range from 0,255]；
Through the calculation, vibration signal frequency spectrums in different time periods are compared, comprehensive change analysis is carried out, and resonance band testing and fault judgment are carried out when equipment is started and stopped.
As a still further preferable embodiment of the present invention: during feature extraction, Hilbert change can be adopted to carry out envelope demodulation on frequency, for example, a bearing outer ring fault is taken as an example, the fault frequency of local mechanical damage is modulated to a high-frequency section due to impact, effective characteristic frequency components of the fault frequency are difficult to find only through spectrum analysis, and the envelope demodulation can realize separation of a low-frequency modulation signal from a carrier signal, so that the impact frequency of the fault frequency can be obtained, and a corresponding fault can be judged.
As a still further preferable embodiment of the present invention: during feature extraction, a calculation order analysis method is adopted for analyzing the rotating equipment, the order analysis is frequency spectrum analysis of a corner region sampling signal, the key point is to realize equal angle sampling of a vibration signal, namely sampling is carried out at a certain angle, the number of sampling points of each rotation is always the same no matter the rotating speed, and the sampling frequency is required to be correspondingly adjusted according to the rotating speed change of a reference shaft in order to guarantee equal angle sampling.
As a still further preferable embodiment of the present invention: under the condition that the equipment is unstable in work, the waterfall graph is used for analyzing, comparing vibration signal frequency spectrums in different time periods, and performing comprehensive change analysis for resonance band testing and fault judgment when the equipment is started and stopped.
Compared with the prior art, the invention has the beneficial effects that: 1. an alarm is given out in the early stage of abnormal operation of the equipment, so that the safety risk of the equipment is eliminated in the bud state;
2. the method can accurately identify the fault risk in the early stage of the fault, send out early warning prompt, allow the customer sufficient time to prepare spare parts and maintenance resources, and uniformly arrange maintenance and repair in the idle production or centralized maintenance;
3. the early failure can be identified in several weeks and months in advance, and sufficient time lead is strived for by customers to prepare spare parts for purchase and arrange for maintenance.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a main flow chart of the feature extraction step of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
In an embodiment of the present invention, an AI technology-based equipment predictive maintenance method includes the following steps: data acquisition, data processing, feature extraction, model training and safety early warning.
Wherein, the data acquisition step: the vibration sensor is arranged on the running equipment and can synchronously and continuously acquire vibration data, including data in a normal state and data in an abnormal state (the acquired data are time domain data).
And (3) data processing: and processing the vibration data signals by a digital signal processing algorithm, including denoising and outlier processing, and performing dimension reduction processing on the vibration data by adopting an Auto-Encoder unsupervised learning algorithm.
A characteristic extraction step: the method comprises the steps of respectively carrying out short-time Fourier transform on various data signals collected by a vibration sensor after data processing, specifically, carrying out windowing operation on each data signal to obtain a spectrogram, then carrying out cepstrum analysis on the spectrogram to obtain a cepstrum, and drawing to obtain a time-frequency graph by taking a horizontal axis as time and a vertical axis as frequency.
Specifically, the AutoEncoder comprises an encode process and a decode process, and is provided with three layers of networks, namely an input layer, a hidden layer and an output layer, wherein the hidden layer positioned in the middle is a plurality of BP neural networks; and (3) carrying out forward conduction calculation on the neurons of each layer, and calculating by using a forward conduction formula:
a2＝σ(Z2)＝σ(a1*W+b2)(1)
in the formula, the superscript number represents the number of layers, the asterisk represents convolution, b represents a bias term, and sigma represents an activation function to obtain the activation value of each layer; the residual error between the final output layer and each layer of neurons is found by using a back propagation algorithm:
in the formula, the weight parameter W and the bias values b and J are cost functions, and the W and the b are continuously updated by using a gradient descent method, so that the output is closer to the input;
and finally, acquiring distance values from the center of all observation points based on the MAD by adopting a central distance calculation method of the median absolute deviation:
firstly, calculating median mean (X) of all observation points;
then, calculating the absolute deviation value abs (X-mean (X)) of each observation point and the median;
secondly, calculating the median of the absolute deviation value abs (X-mean (X)), namely MAD ═ mean (abs (X-mean (X));
finally, the absolute deviation value abs (X-mean (X))/MAD is divided by mean (abs (X-mean (X))) to obtain a set of range values abs (X-mean (X))/MAD from the center for all the viewpoints based on MAD.
During feature extraction, windowing and framing are carried out to convert a time domain signal into a frequency domain signal, so that the attenuation relation of energy at a certain moment along with time can be well reflected; the resonant band test and fault judgment during starting and stopping of the equipment can adopt a Hamming window calculation mode, and the calculation formula of the Hamming window is as follows:
performing cepstrum analysis on the spectrogram, converting by adopting Mel frequency cepstrum coefficients, and separating direct current signal components and sinusoidal signal components by adopting DCT (discrete cosine transformation) through the cepstrum analysis, wherein the formula is a cepstrum coefficient MFCC calculation formula as follows:
wherein L is the number of filters, and the MFCC range is [ min, max ]]The MFCC is normalized, the normalization coefficient pixel is calculated and mapped to the gray value of 0-255,
where pixels range from 0,255]；
Through the calculation, vibration signal frequency spectrums in different time periods are compared, comprehensive change analysis is carried out, and resonance band testing and fault judgment are carried out when equipment is started and stopped.
Furthermore, in addition to extracting frequency characteristics, Hilbert change can be adopted to carry out envelope demodulation on frequency, for example, a bearing outer ring fault is taken as an example, the fault frequency of local mechanical damage is modulated to a high-frequency section due to impact, effective characteristic frequency components of the fault frequency are difficult to find only through spectrum analysis, and the envelope demodulation can realize separation of a low-frequency modulation signal from a carrier signal, so that the impact frequency of the fault frequency can be obtained, and a corresponding fault can be judged; for the rotating equipment, a computational order analysis method can be adopted for analysis, the order analysis is frequency spectrum analysis of a corner domain sampling signal, the key is to realize equal angle sampling of a vibration signal, namely sampling is carried out at a certain angle, the number of sampling points of each turn is always the same no matter the rotating speed, and the sampling frequency is required to be correspondingly adjusted according to the rotating speed change of a reference shaft in order to ensure equal angle sampling; meanwhile, under the condition that the equipment is unstable in work, the device can be analyzed through a waterfall graph, the frequency spectrums of vibration signals in different time periods are compared, comprehensive change analysis is carried out, and the device is commonly used for resonance band testing and fault judgment when the equipment is started and stopped.
Model training: constructing a predictive maintenance model based on the CNN convolutional neural network, setting a threshold value, respectively inputting each time-frequency graph in the step of feature extraction into the predictive maintenance model to obtain a three-dimensional feature group, then inputting the three-dimensional feature group into the predictive maintenance model to obtain a prediction result, and training the predictive maintenance model.
A safety early warning step: when the equipment runs, the vibration sensor collects data of the equipment, the predictive maintenance model compares the models, and when the data exceed a threshold value, the predictive maintenance cloud platform sends out an alarm message to remind a manager to process the occurrence situation.
The method can give an alarm in the early stage of abnormal operation of the equipment, and eliminates the safety risk of the equipment in the bud state; the method can accurately identify the fault risk in the early stage of the fault, send out early warning prompt, allow the customer sufficient time to prepare spare parts and maintenance resources, and uniformly arrange maintenance and repair in the idle production or centralized maintenance; the early failure can be identified in several weeks and months in advance, and sufficient time lead is strived for by customers to prepare spare parts for purchase and arrange for maintenance.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (7)
1. An AI technology based equipment predictive maintenance method, characterized by comprising the following steps: data acquisition, data processing, feature extraction, model training and safety early warning;
wherein, the data acquisition step: installing a vibration sensor on the running equipment, wherein the vibration sensor synchronously and continuously acquires vibration data comprising data in a normal state and data in an abnormal state;
and (3) data processing: processing the vibration data signals by a digital signal processing algorithm, including denoising and outlier processing, and performing dimension reduction processing on the vibration data by adopting an Auto-Encoder unsupervised learning algorithm;
a characteristic extraction step: respectively carrying out short-time Fourier transform on various data signals acquired by the vibration sensor after data processing, specifically, respectively carrying out windowing operation on each data signal to obtain a spectrogram, then carrying out cepstrum analysis on the spectrogram to obtain a cepstrum, and drawing to obtain a time-frequency graph by taking a horizontal axis as time and a vertical axis as frequency;
model training: constructing a predictive maintenance model based on a CNN convolutional neural network, setting a threshold value, respectively inputting each time-frequency graph in the step of feature extraction into the predictive maintenance model to obtain a three-dimensional feature group, inputting the three-dimensional feature group into the predictive maintenance model to obtain a prediction result, and training the predictive maintenance model;
a safety early warning step: when the equipment runs, the vibration sensor collects data of the equipment, the predictive maintenance model compares the models, and when the data exceed a threshold value, the predictive maintenance cloud platform sends out an alarm message to remind a manager to process the occurrence situation.
2. The AI-technology-based equipment predictive maintenance method according to claim 1, wherein the AutoEncoder unsupervised learning algorithm in the data processing step includes an encode process and a decode process, the algorithm has three layers of networks, namely an input layer, a hidden layer and an output layer, the hidden layer in between is a plurality of BP neural networks; and (3) carrying out forward conduction calculation on the neurons of each layer, and calculating by using a forward conduction formula:
a2＝σ(Z2)＝σ(a1*W+b2)
in the formula, the superscript number represents the number of layers, the asterisk represents convolution, b represents a bias term, and sigma represents an activation function to obtain the activation value of each layer; the residual error between the final output layer and each layer of neurons is found by using a back propagation algorithm:
in the formula, the weight parameter W and the bias values b and J are cost functions, and the W and the b are continuously updated by a gradient descent method, so that the output is closer to the input.
3. The AI-technology-based equipment predictive maintenance method according to claim 1, wherein in the data processing step, the distance values from the center of all the observation points based on the MAD are obtained by a median-of-absolute-deviation center distance calculation method:
firstly, calculating median mean (X) of all observation points;
then, calculating the absolute deviation value abs (X-mean (X)) of each observation point and the median;
secondly, calculating the median of the absolute deviation value abs (X-mean (X)), namely MAD ═ mean (abs (X-mean (X));
finally, the absolute deviation value abs (X-mean (X))/MAD is divided by mean (abs (X-mean (X))) to obtain a set of range values abs (X-mean (X))/MAD from the center for all the viewpoints based on MAD.
4. The AI-technology-based equipment predictive maintenance method according to claim 1, wherein in the feature extraction step, a short-time fourier transform is performed on the data signal acquired by the vibration sensor, specifically, a windowing operation is performed on the data signal to obtain a spectrogram, a hamming window is used in the windowing operation, and a calculation formula of the hamming window is as follows:
performing cepstrum analysis on the spectrogram, converting by adopting Mel frequency cepstrum coefficients, and separating direct current signal components and sinusoidal signal components by adopting DCT (discrete cosine transformation) through the cepstrum analysis, wherein the formula is a cepstrum coefficient MFCC calculation formula as follows:
wherein L is the number of filters, and the MFCC range is [ min, max ]]The MFCC is normalized, the normalization coefficient pixel is calculated and mapped to the gray value of 0-255,
where pixels range from 0,255]。
5. The AI-technology-based equipment predictive maintenance method of claim 4, wherein the extraction of features uses Hilbert transform for envelope demodulation of frequency.
6. The AI-technology-based equipment predictive maintenance method as in claim 4, wherein during feature extraction, for the rotating equipment, a computational order analysis is used for analysis, wherein the order analysis is a spectrum analysis of the angular domain sampling signals, and the sampling is performed at intervals of a set fixed angle value, and the number of sampling points per revolution is always the same no matter the rotation speed, thereby ensuring that the equal angle sampling adjusts the sampling frequency according to the rotation speed variation of the reference axis.
7. The AI-technology-based equipment predictive maintenance method as in claim 4, wherein in case of unstable equipment operation, analysis is performed by waterfall plot, comparing vibration signal spectra of different time periods, and performing comprehensive variation analysis for resonance band test and fault judgment when equipment is started or stopped.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110650951.3A CN113505898A (en) | 2021-06-10 | 2021-06-10 | Equipment predictive maintenance method based on AI technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110650951.3A CN113505898A (en) | 2021-06-10 | 2021-06-10 | Equipment predictive maintenance method based on AI technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113505898A true CN113505898A (en) | 2021-10-15 |
Family
ID=78010294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110650951.3A Withdrawn CN113505898A (en) | 2021-06-10 | 2021-06-10 | Equipment predictive maintenance method based on AI technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113505898A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115169505A (en) * | 2022-09-06 | 2022-10-11 | 杭州浅水数字技术有限公司 | Early warning method and early warning system for mechanical fault of special equipment moving part |
-
2021
- 2021-06-10 CN CN202110650951.3A patent/CN113505898A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115169505A (en) * | 2022-09-06 | 2022-10-11 | 杭州浅水数字技术有限公司 | Early warning method and early warning system for mechanical fault of special equipment moving part |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kumar et al. | Role of signal processing, modeling and decision making in the diagnosis of rolling element bearing defect: a review | |
| US10580228B2 (en) | Fault detection system and method for vehicle system prognosis | |
| US11193816B2 (en) | Health monitor method for an equipment and system thereof | |
| WO2019080367A1 (en) | Method for evaluating health status of mechanical device | |
| JP7199608B2 (en) | Methods and apparatus for inspecting wind turbine blades, and equipment and storage media therefor | |
| CN106649755B (en) | Threshold value self-adaptive setting abnormity detection method for multi-dimensional real-time power transformation equipment data | |
| CN110133500B (en) | Motor online monitoring and fault precursor diagnosis system and method based on multi-layer architecture | |
| CN113551765A (en) | Sound spectrum analysis and diagnosis method for equipment fault | |
| CN113505898A (en) | Equipment predictive maintenance method based on AI technology | |
| CN113567162A (en) | Fan fault intelligent diagnosis device and method based on acoustic sensor | |
| KR102545672B1 (en) | Method and apparatus for machine fault diagnosis | |
| CN114565006A (en) | Wind driven generator blade damage detection method and system based on deep learning | |
| Soualhi et al. | PHM survey: implementation of signal processing methods for monitoring bearings and gearboxes | |
| CN114120974A (en) | Fan blade fault diagnosis method based on deep learning | |
| CN116717437A (en) | Wind turbine generator system fault monitoring method and system | |
| CN115165274A (en) | Self-adaptive intelligent monitoring device and method for vibration state of engineering mechanical equipment | |
| Lv et al. | A new feature extraction technique for early degeneration detection of rolling bearings | |
| CN112162197A (en) | Online diagnosis method for stator and rotor center offset fault of vertical unit | |
| Cao et al. | Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator | |
| CN117217730A (en) | Power equipment fault identification method, device, equipment, medium and product | |
| CN113435228A (en) | Motor bearing service life prediction and analysis method based on vibration signal modeling | |
| CN115959549A (en) | Escalator fault diagnosis method based on digital twinning | |
| CN115539277A (en) | Fault early warning system and method based on hydroelectric machine voiceprint recognition | |
| CN113221292A (en) | Predictive maintenance model and maintenance method for wave-activated generator | |
| Yang et al. | A condition classification system for reciprocating compressors |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211015 |