CN112485001A - Fault diagnosis method for intelligent elevator - Google Patents
Fault diagnosis method for intelligent elevator Download PDFInfo
- Publication number
- CN112485001A CN112485001A CN202011280097.8A CN202011280097A CN112485001A CN 112485001 A CN112485001 A CN 112485001A CN 202011280097 A CN202011280097 A CN 202011280097A CN 112485001 A CN112485001 A CN 112485001A
- Authority
- CN
- China
- Prior art keywords
- cnn
- signal
- fault
- noise reduction
- time
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000003745 diagnosis Methods 0.000 title claims abstract description 24
- 230000009467 reduction Effects 0.000 claims abstract description 24
- 239000013598 vector Substances 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 9
- 230000001133 acceleration Effects 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims description 18
- 238000007781 pre-processing Methods 0.000 claims description 12
- 238000011176 pooling Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 abstract description 4
- 230000036541 health Effects 0.000 abstract description 2
- 238000013527 convolutional neural network Methods 0.000 description 35
- 238000004458 analytical method Methods 0.000 description 9
- 238000005096 rolling process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000000354 decomposition reaction Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000010183 spectrum analysis Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic 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
-
- 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
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Acoustics & Sound (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a fault diagnosis method for an intelligent elevator, which comprises the following steps: collecting acceleration and speed signal data of an elevator tractor and a door machine, and classifying a data set according to the running state of a motor to obtain a training sample; carrying out signal noise reduction pretreatment on original vibration signal data to construct a characteristic vector of a signal; constructing a CNN input image by using a PWVD method; and (4) learning input images of different fault categories by using the CNN, and finishing accurate identification of different fault states. The ability and the speed of CNN learning complex characteristics can be strengthened through the pretreatment mechanism, the model classification accuracy is improved, the robustness is enhanced, and the training time is reduced, so that the system health condition of the intelligent elevator can be monitored and analyzed in real time.
Description
Technical Field
The invention belongs to the field of mode classification, and relates to a fault diagnosis method for an intelligent elevator.
Background
The intelligent elevator is an important component in the intelligent city construction process and is an essential part in daily life, so the normal operation of the elevator is extremely important, wherein a traction machine responsible for the lifting of the elevator and a door motor responsible for the opening and closing of an elevator door are important components for ensuring the safe and effective work of the elevator, and the intelligent elevator has great significance for the fault diagnosis of the two motors. Meanwhile, the rolling bearing is a key component of rotating machinery such as a motor, and has the characteristics of high rotating speed, complex structure, easiness in operation failure and the like. The defects and damages of the rolling bearing directly affect the performance and the service life of the whole motor, so that the fault diagnosis of the rolling bearing has important practical significance.
Vibration signal analysis is commonly used for fault diagnosis of rolling bearings, including time, frequency and time-frequency analysis (TFA). The time domain analysis includes waveform analysis, correlation analysis, kurtosis analysis, and the like. The time domain analysis method is simple and easy to understand, but has higher limitation. For example, when the dimensional index is used, the stability of the operating environment of the equipment needs to be ensured, otherwise, the dimensional index becomes worthless. The parameter is concerned about the signal distribution condition and is slightly influenced by the amplitude and frequency of the signal, so that the adaptability is stronger, but the dimensionless index is limited and cannot be suitable for diagnosing each fault condition. The frequency domain analysis includes fourier transform, envelope demodulation, cepstrum analysis, and the like. The conditions required by these current more common spectrum analysis techniques should satisfy the condition that the signal is stationary, so these more common spectrum analysis techniques all have a serious disadvantage, and if the original vibration signal does not strictly comply with periodicity or stationarity, the spectrum analysis result loses its corresponding physical meaning, so that the final analysis result only contains information in the frequency domain but loses information in the time-frequency domain. However, most of the vibration signals of mechanical equipment in the industry are nonlinear and non-stationary, so that the final result is also affected if the non-stationary vibration signals are subjected to a frequency domain feature extraction method by fourier transform as a core. Due to vibration caused by faults, TFA can simultaneously reproduce time domain and frequency domain characteristics of signals, which is very effective for nonlinear and non-stationary signals, so TFA is widely applied to fault diagnosis of rolling bearings. The most common method for using TFA is wavelet transform, which also has the problem of difficulty in selecting basis functions. Meanwhile, algorithms related to the traditional machine learning field cannot meet the requirement on the accuracy of fault diagnosis gradually, for example, the SVM algorithm has the problems of difficulty in kernel function solving and large storage space requirement, and is far behind the deep learning algorithm in accuracy.
Disclosure of Invention
The invention aims to: after the noise reduction processing is carried out on an original vibration signal, a new method is adopted to construct a feature vector, the noise reduction signal is decomposed, redundant features are removed, a 2D contour map is constructed by combining a pseudo Wigner-Ville distribution (PWVD) technology and serves as the input of a Convolutional Neural Network (CNN), and the accuracy of fault diagnosis is improved.
The technical scheme of the invention is as follows: a fault diagnosis method for a smart elevator comprises the following steps:
s1: collecting acceleration and speed signal data of an elevator tractor and a door machine, and classifying a data set according to the running state of a motor to obtain a training sample;
s2: carrying out signal noise reduction pretreatment on original vibration signal data to construct a characteristic vector of a signal;
s3: constructing a CNN input image by using a PWVD method;
s4: and (4) learning input images of different fault categories by using the CNN, and finishing accurate identification of different fault states.
The further technical scheme is as follows: the preprocessing of signal noise reduction on the original vibration signal data in step S2 includes:
and (4) decomposing the signal data acquired in the step (S1) by adopting a wavelet packet technology, and performing noise reduction reconstruction on the original vibration signal data by taking the first M wavelet packets with large energy to minimize the mean square error between the reconstructed signal and the original signal.
The further technical scheme is as follows: the constructing the feature vector of the signal in step S2 includes:
further preprocessing the data after noise reduction and reconstruction, setting the noise reduction signal obtained in S2 as h (t), and performing the following processing on h (t): a (t) ═ h (t) - [ hmax(t)+hmin(t)]And repeating the above operations for a (t) until h (t) is decomposed into a series of stationary components ai(t) and reconstructing the input feature vector using the correlation coefficient selection component.
The further technical scheme is as follows: the constructing of the CNN input image using the PWVD method in step S3 includes:
and processing each selected component by using a PWVD method, superposing PWVD results of different components to obtain a 3D time-frequency distribution graph of the vibration signal, performing projection operation on the obtained 3D time-frequency distribution graph to obtain a 2D contour line time-frequency graph, and taking the 2D contour line time-frequency graph as an input image of the CNN.
The further technical scheme is as follows: in step S4, the learning of the input images of different failure categories by using CNN completes accurate identification of different failure states, including:
and inputting the 2D contour line time-frequency graph serving as an input sample into the CNN, and learning time-frequency distribution graphs of different fault types by using the CNN to finish accurate identification of different fault states.
The further technical scheme is as follows: the structure of the CNN is as follows: a multi-layer feature extraction network structure consisting of 3 convolutional layers, 3 pooling layers, a full-connection layer and a classification layer is adopted;
and the convolutional layers and the pooling layers are alternately arranged, the output of the convolutional layers is subjected to batch normalization through batch scaling and translation, then the normalized convolutional layers are transmitted to the activation layer for nonlinear processing, and finally the Softmax layer is added to classify the fault categories so as to complete the mapping from the characteristic domain to the fault category domain.
The invention has the advantages that:
1. the method is combined with a convolutional neural network in the field of image processing, the complexity of the network is reduced, the risk of overfitting is reduced, noise reduction preprocessing is performed on a vibration signal, a CNN input image is constructed by a PWVD method, the method is different from the method of directly inputting original signal data into the neural network, so that the accuracy fluctuation is large, and more time is needed for training a model;
2. the method comprises the steps of conducting noise reduction processing on collected original one-dimensional vibration signal data of a motor bearing through a wavelet packet decomposition technology, conducting decomposition processing on noise reduction signals, selecting components according to relevant coefficients, reconstructing input feature vectors, conducting PWVD processing on each selected component to obtain a 2D time-frequency distribution graph of the vibration signals, enabling the 2D time-frequency distribution graph to serve as input of a CNN (computer network), conducting fault diagnosis and classification, conducting self-learning on two-dimensional time-frequency features through the CNN, enabling a preprocessing mechanism to enhance the capability and speed of the CNN for learning complex features, avoiding uncertainty of feature selection in a traditional method, reducing complexity of a diagnosis process by using the CNN, improving accuracy and increasing model classification accuracy from 97.4% to 99.8%.
Drawings
The invention is further described with reference to the following figures and examples:
fig. 1 is a flowchart of a fault diagnosis method for an intelligent elevator provided by the present application;
FIG. 2 is a 3D time-frequency distribution diagram obtained by PWVD processing on the selected feature vectors provided in the present application;
fig. 3 is a 2D contour time-frequency diagram after performing a projection operation on the 3D time-frequency distribution diagram after PWVD processing.
Detailed Description
Example (b): since the introduction of deep learning, deep learning has been rapidly developed not only in terms of voice, images, and the like, but also widely used in the field of failure diagnosis. Compared with other deep learning algorithms, the Convolutional Neural Network (CNN) reduces the complexity of the network and the risk of overfitting, and therefore is applied to the field of mechanical fault diagnosis. According to the method, the 2D time frequency distribution graph is obtained after the original vibration signals are subjected to data preprocessing, the two-dimensional time frequency characteristics are self-learned through the CNN, the uncertainty of characteristic selection in the traditional method is avoided, the complexity of the diagnosis process is reduced by using the CNN, and meanwhile, the accuracy is improved.
With reference to fig. 1 to 3, the present application provides a fault diagnosis method for an intelligent elevator, which includes the following steps.
S1: and collecting acceleration and speed signal data of the elevator traction machine and the door machine, and classifying the data set according to the running state of the motor to obtain a training sample.
For example, in practical application, the original signal may be divided into four states, namely, Normal (NC), rolling element fault (BF), Inner ring fault (IF), and Outer ring fault (OF), and classified according to the four states as an original data set, that is, the original vibration signal.
S2: and carrying out signal noise reduction pretreatment on the original vibration signal data to construct a characteristic vector of the signal.
The preprocessing of signal noise reduction on the original vibration signal data in step S2 includes: decomposing the signal data collected in the step S1 by adopting a wavelet packet technology, and performing noise reduction reconstruction on the original vibration signal data by taking the first M wavelet packets with large energy to minimize the mean square error between the reconstructed signal and the original signal, namelyWhere h (t) and x (t) are the reconstructed signal and the original signal, respectively. Therefore, noise elimination processing can be realized, and compared with the original data, the amplitude is obviously reduced, the intensity is reduced, and the impact process is more obvious.
Constructing a feature vector of the signal in step S2, including: further preprocessing the data after noise reduction and reconstruction, setting the noise reduction signal (reconstruction signal) obtained in S2 as h (t), and performing the following processing on h (t): a (t) ═ h (t) - [ hmax(t)+hmin(t)]And repeating the above operations for a (t) until h (t) is decomposed into a series of stationary components ai(t) (i ═ 1,2, 3.) the sum, using the correlation coefficient culling components, reconstructs the input feature vector, removing redundant features in the dataset.
Illustratively, the components with the correlation coefficient exceeding 0.3 are considered to have correlation, and the components with the correlation coefficient smaller than 0.3 are removed.
S3: the CNN input image is constructed using the PWVD method.
Step S3 is specifically implemented as: and processing each selected component by using a PWVD method, and superposing PWVD results of different components to obtain a time-frequency distribution map of the vibration signal.
Further, a 3D time-frequency distribution graph of the vibration signal is obtained (as shown in fig. 2), and then the obtained 3D time-frequency distribution graph is subjected to a projection operation to obtain a 2D contour time-frequency graph (as shown in fig. 3), and the 2D contour time-frequency graph is used as an input image of the CNN.
On the time-frequency plane, the information represented in the 3D time-frequency distribution graph and the 2D contour time-frequency graph is the same, where the 2D contour time-frequency graph is used as input for the CNN.
S4: and (4) learning input images of different fault categories by using the CNN, and finishing accurate identification of different fault states.
Step S4 is specifically implemented as: and inputting the time-frequency distribution graph (2D contour line time-frequency graph) serving as an input sample into the CNN, and learning the time-frequency distribution graphs of different fault types by using the CNN to finish the accurate identification of different fault states.
Illustratively, after adjusting the model parameters through multiple experiments, the structure of the CNN is determined as follows: a multi-layer feature extraction network structure consisting of 3 convolutional layers, 3 pooling layers, a full-connection layer and a classification layer is adopted; and the convolutional layers and the pooling layers are alternately arranged, the output of the convolutional layers is subjected to batch normalization through batch scaling and translation, then the normalized convolutional layers are transmitted to the activation layer for nonlinear processing, and finally the Softmax layer is added to classify the fault categories so as to complete the mapping from the characteristic domain to the fault category domain.
The main parameters of the CNN network are as follows: the size of a convolution kernel of the first layer is 64 multiplied by 64, the sizes of other 2 layers are set to be 3 multiplied by 3, the number of convolution kernels is 16-32-64 in sequence, the sizes of 3 layers of pooling layers are all 2 multiplied by 2, the pooling mode is maximum pooling, the number of neurons of a full connecting layer is 100, an activation function ReLu, an optimizer Adam, and the number of training iterations is 20.
Table 1 shows the comparative experiment results of the fault diagnosis method for the smart elevator proposed in the present application:
TABLE 1
In summary, the fault diagnosis method for the intelligent elevator, provided by the application, combines a convolutional neural network in the field of image processing, reduces the complexity of the network, reduces the risk of overfitting, performs noise reduction preprocessing on a vibration signal, constructs a CNN input image through a PWVD method, is different from the problems that the accuracy fluctuation is large and more time is needed for training a model due to a method of directly inputting original signal data into the neural network, can enhance the capability and speed of learning complex features of the CNN through a preprocessing mechanism, improves the model classification accuracy, enhances the robustness, and is beneficial to monitoring and analyzing the system health condition of the intelligent elevator in real time when training is reduced;
in addition, the collected original one-dimensional vibration signal data of the motor bearing is subjected to noise reduction treatment through a wavelet packet decomposition technology, the noise reduction signal is subjected to decomposition treatment, components are selected according to related coefficients, input feature vectors are reconstructed, PWVD treatment is carried out on each selected component, a 2D time-frequency distribution graph of the vibration signal is obtained, the 2D time-frequency distribution graph is used as input of a CNN (computer network), fault diagnosis and classification are carried out, the CNN is used for self-learning of two-dimensional time-frequency features, a preprocessing mechanism can enhance the capability and speed of the CNN for learning complex features, uncertainty of feature selection in the traditional method is avoided, complexity of a diagnosis process is reduced by using the CNN, accuracy is improved, and model classification accuracy is improved from 97.4% to 99.8%.
The terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying a number of the indicated technical features. Thus, a defined feature of "first", "second", may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless otherwise specified.
The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disk, or the like.
The above description is only exemplary of the present application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (6)
1. A fault diagnosis method for a smart elevator is characterized by comprising the following steps:
s1: collecting acceleration and speed signal data of an elevator tractor and a door machine, and classifying a data set according to the running state of a motor to obtain a training sample;
s2: carrying out signal noise reduction pretreatment on original vibration signal data to construct a characteristic vector of a signal;
s3: constructing a CNN input image by using a PWVD method;
s4: and (4) learning input images of different fault categories by using the CNN, and finishing accurate identification of different fault states.
2. The method for diagnosing faults of an intelligent elevator according to claim 1, wherein the preprocessing of signal noise reduction on the original vibration signal data in step S2 includes:
and (4) decomposing the signal data acquired in the step (S1) by adopting a wavelet packet technology, and performing noise reduction reconstruction on the original vibration signal data by taking the first M wavelet packets with large energy to minimize the mean square error between the reconstructed signal and the original signal.
3. The method of claim 2, wherein the constructing the feature vector of the signal in step S2 comprises:
further preprocessing the data after noise reduction and reconstruction, setting the noise reduction signal obtained in S2 as h (t), and performing the following processing on h (t): a (t) ═ h (t) - [ hmax(t)+hmin(t)]And repeating the above operations for a (t) until h (t) is decomposed into a series of stationary components ai(t) and reconstructing the input feature vector using the correlation coefficient selection component.
4. The intelligent elevator-oriented fault diagnosis method according to claim 3, wherein the constructing a CNN input image using the PWVD method in step S3 includes:
and processing each selected component by using a PWVD method, superposing PWVD results of different components to obtain a 3D time-frequency distribution graph of the vibration signal, performing projection operation on the obtained 3D time-frequency distribution graph to obtain a 2D contour line time-frequency graph, and taking the 2D contour line time-frequency graph as an input image of the CNN.
5. The method of claim 4, wherein the step S4 of learning the input images of different failure categories by using CNN to complete the precise identification of different failure states includes:
and inputting the 2D contour line time-frequency graph serving as an input sample into the CNN, and learning time-frequency distribution graphs of different fault types by using the CNN to finish accurate identification of different fault states.
6. The method for diagnosing faults of an intelligent elevator according to any one of claims 1 to 5, wherein the CNN has a structure of: a multi-layer feature extraction network structure consisting of 3 convolutional layers, 3 pooling layers, a full-connection layer and a classification layer is adopted;
and the convolutional layers and the pooling layers are alternately arranged, the output of the convolutional layers is subjected to batch normalization through batch scaling and translation, then the normalized convolutional layers are transmitted to the activation layer for nonlinear processing, and finally the Softmax layer is added to classify the fault categories so as to complete the mapping from the characteristic domain to the fault category domain.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011280097.8A CN112485001A (en) | 2020-11-16 | 2020-11-16 | Fault diagnosis method for intelligent elevator |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011280097.8A CN112485001A (en) | 2020-11-16 | 2020-11-16 | Fault diagnosis method for intelligent elevator |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112485001A true CN112485001A (en) | 2021-03-12 |
Family
ID=74931122
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011280097.8A Pending CN112485001A (en) | 2020-11-16 | 2020-11-16 | Fault diagnosis method for intelligent elevator |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112485001A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117361256A (en) * | 2023-10-10 | 2024-01-09 | 广东全联富士电梯有限公司 | Elevator safety management method and system based on artificial intelligence |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110472563A (en) * | 2019-08-13 | 2019-11-19 | 浙江大学 | The vibrated major break down diagnostic method of vertical ladder based on WAVELET PACKET DECOMPOSITION and neural network |
CN110550518A (en) * | 2019-08-29 | 2019-12-10 | 电子科技大学 | Elevator operation abnormity detection method based on sparse denoising self-coding |
CN111397901A (en) * | 2019-03-12 | 2020-07-10 | 上海电机学院 | Rolling bearing fault diagnosis method based on wavelet and improved PSO-RBF neural network |
CN111783558A (en) * | 2020-06-11 | 2020-10-16 | 上海交通大学 | Satellite navigation interference signal type intelligent identification method and system |
-
2020
- 2020-11-16 CN CN202011280097.8A patent/CN112485001A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111397901A (en) * | 2019-03-12 | 2020-07-10 | 上海电机学院 | Rolling bearing fault diagnosis method based on wavelet and improved PSO-RBF neural network |
CN110472563A (en) * | 2019-08-13 | 2019-11-19 | 浙江大学 | The vibrated major break down diagnostic method of vertical ladder based on WAVELET PACKET DECOMPOSITION and neural network |
CN110550518A (en) * | 2019-08-29 | 2019-12-10 | 电子科技大学 | Elevator operation abnormity detection method based on sparse denoising self-coding |
CN111783558A (en) * | 2020-06-11 | 2020-10-16 | 上海交通大学 | Satellite navigation interference signal type intelligent identification method and system |
Non-Patent Citations (4)
Title |
---|
刘江等: "基于变模式分解降噪的滚动轴承故障诊断研究", 《机械设计与制造》 * |
褚东亮 著: "《旋转机械非平稳信号分析及故障诊断技术》", 30 November 2019 * |
郑海波等: "基于小波包变换的一种降噪算法", 《合肥工业大学学报(自然科学版)》 * |
黄鑫等: "基于深度卷积神经网络与WPT-PWVD的轴承故障智能诊断", 《振动与冲击》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117361256A (en) * | 2023-10-10 | 2024-01-09 | 广东全联富士电梯有限公司 | Elevator safety management method and system based on artificial intelligence |
CN117361256B (en) * | 2023-10-10 | 2024-03-12 | 广东全联富士电梯有限公司 | Elevator safety management method and system based on artificial intelligence |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106682688B (en) | Particle swarm optimization-based stacked noise reduction self-coding network bearing fault diagnosis method | |
Yang et al. | SuperGraph: Spatial-temporal graph-based feature extraction for rotating machinery diagnosis | |
Chen et al. | Dual-path mixed-domain residual threshold networks for bearing fault diagnosis | |
Al-Bugharbee et al. | A fault diagnosis methodology for rolling element bearings based on advanced signal pretreatment and autoregressive modelling | |
CN112906644B (en) | Mechanical fault intelligent diagnosis method based on deep migration learning | |
CN113176092B (en) | Motor bearing fault diagnosis method based on data fusion and improved experience wavelet transform | |
CN109389171B (en) | Medical image classification method based on multi-granularity convolution noise reduction automatic encoder technology | |
CN110991424A (en) | Fault diagnosis method based on minimum entropy deconvolution and stacking sparse self-encoder | |
CN111753891B (en) | Rolling bearing fault diagnosis method based on unsupervised feature learning | |
CN111397896A (en) | Fault diagnosis method and system for rotary machine and storage medium | |
CN115017945A (en) | Mechanical fault diagnosis method and system based on enhanced convolutional neural network | |
Zhang et al. | Bearing fault diagnosis under various operation conditions using synchrosqueezing transform and improved two-dimensional convolutional neural network | |
CN112633098A (en) | Fault diagnosis method and system for rotary machine and storage medium | |
CN114169377A (en) | G-MSCNN-based fault diagnosis method for rolling bearing in noisy environment | |
CN113705396A (en) | Motor fault diagnosis method, system and equipment | |
Al Tobi et al. | Using MLP‐GABP and SVM with wavelet packet transform‐based feature extraction for fault diagnosis of a centrifugal pump | |
Cao et al. | Fault diagnosis of rolling bearing based on multiscale one-dimensional hybrid binary pattern | |
Zhang et al. | Fault diagnosis based on optimized node entropy using lifting wavelet packet transform and genetic algorithms | |
CN112485001A (en) | Fault diagnosis method for intelligent elevator | |
CN117592543A (en) | Aeroengine rolling bearing fault diagnosis method based on self-supervision learning | |
CN113409213B (en) | Method and system for enhancing noise reduction of time-frequency diagram of fault signal of plunger pump | |
CN112345251B (en) | Mechanical intelligent fault diagnosis method based on signal resolution enhancement | |
CN113158769A (en) | CEEMDAN and FastICA-based electromechanical device bearing vibration signal denoising method | |
CN112884093B (en) | Rotary machine fault diagnosis method and equipment based on DSCRN model and storage medium | |
Chen et al. | Gear fault diagnosis based on SGMD noise reduction and CNN |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210312 |