CN109255395B - Service life prediction method of ball screw pair - Google Patents
Service life prediction method of ball screw pair Download PDFInfo
- Publication number
- CN109255395B CN109255395B CN201811243790.0A CN201811243790A CN109255395B CN 109255395 B CN109255395 B CN 109255395B CN 201811243790 A CN201811243790 A CN 201811243790A CN 109255395 B CN109255395 B CN 109255395B
- Authority
- CN
- China
- Prior art keywords
- value
- point
- features
- health
- service life
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000007477 logistic regression Methods 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims abstract description 17
- 230000015556 catabolic process Effects 0.000 claims abstract description 15
- 238000006731 degradation reaction Methods 0.000 claims abstract description 15
- 238000013507 mapping Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000001514 detection method Methods 0.000 claims abstract description 11
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 230000009467 reduction Effects 0.000 claims abstract description 7
- 230000001131 transforming effect Effects 0.000 claims abstract description 4
- 230000036541 health Effects 0.000 claims description 38
- 238000012549 training Methods 0.000 claims description 14
- 230000002123 temporal effect Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000003862 health status Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
- G06F18/21322—Rendering the within-class scatter matrix non-singular
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
- G06F18/21322—Rendering the within-class scatter matrix non-singular
- G06F18/21324—Rendering the within-class scatter matrix non-singular involving projections, e.g. Fisherface techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Pure & Applied Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Computation (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Complex Calculations (AREA)
Abstract
A service life prediction method of a ball screw pair belongs to the technical field of machinery; performing wavelet denoising processing on an original signal acquired by a vibration sensor; transforming the time domain signal to a frequency domain space through Fourier transform, and extracting time domain characteristics and frequency domain characteristics through characteristic extraction; performing dimensionality reduction processing on the extracted time domain features and frequency domain features through a Fisher criterion to screen out features which greatly contribute to lead screw degradation; mapping the features by adopting a logistic regression algorithm, wherein the mapping interval is [0,1 ]; carrying out critical point detection through a self-adaptive threshold detection algorithm; predicting the residual life of the screw rod through Gaussian process regression so as to obtain the residual life; the invention improves the stability and accuracy of prediction.
Description
Technical Field
The invention belongs to the technical field of machinery, and particularly relates to a service life prediction method of a ball screw pair.
Background
With the continuous progress of the artificial intelligence technology and the continuous development of the equipment intelligence level, the health management and fault prediction technology of the equipment becomes a new research hotspot, and the predictive maintenance technology of the equipment provides reliable parameters such as health indexes and residual life for the operation of the equipment, ensures the healthy operation of the equipment and prolongs the service life of the equipment. At present, a relatively systematic method is lacked in the technology related to the service life prediction of the screw pair. In the method for predicting the service life of the screw pair of the numerical control machine tool based on performance degradation, expected residual service life adopted by training data is obtained through theoretical calculation, so that a predicted value obtained through a trained network lacks reliability.
Disclosure of Invention
The invention overcomes the defects of the prior art and provides a service life prediction method of a ball screw pair, which extracts vibration signals on a vibration sensor of the screw pair, further extracts characteristics and selects important characteristics, maps the extracted characteristics into a health value through a logistic regression algorithm, predicts the trend of a health curve through Gaussian process regression, adds a detection algorithm of a self-adaptive threshold value on the basis, starts a service life prediction algorithm after a critical point is detected, and improves the stability and the accuracy of prediction.
The technical scheme of the invention is as follows:
a service life prediction method of a ball screw pair comprises the following steps:
step a, performing wavelet denoising processing on an original signal acquired by a vibration sensor;
b, transforming the time domain signal to a frequency domain space through Fourier transform, and extracting time domain characteristics and frequency domain characteristics through characteristic extraction;
step c, performing dimensionality reduction processing on the extracted time domain features and frequency domain features through a Fisher criterion to screen out features which greatly contribute to lead screw degradation;
d, mapping the features by adopting a logistic regression algorithm, wherein the mapping interval is [0,1 ];
e, detecting a critical point by using a self-adaptive threshold detection algorithm;
and f, predicting the residual service life of the screw rod through Gaussian process regression so as to obtain the residual service life.
Further, the extracted time domain feature and the extracted frequency domain feature include a root mean square value, a variance, a standard deviation, a maximum value, a minimum value, an average amplitude, a kurtosis factor, a waveform coefficient, a peak value factor, a pulse index, a root mean square value, a margin coefficient and a skewness.
Further, the characteristics contributing largely to the lead screw degradation include a root mean square value, a standard deviation, a maximum value, and a variance.
Further, the adaptive threshold detection algorithm comprises the steps of:
step e1, setting the health value sequence of the lead screw obtained by the logistic regression as { HV }; the health value sequence in the normal state is { HVNormal}, newly introduced health value is HVNew；
Step e2, continuously generating a health value in the operation process of the equipment, and recording the mean value of the points in front of the current point as mu and the variance of the points in front of the current point as sigma;
step e3, recording the difference value d between the current point and the mean value mu of the previous point, judging the size between the difference value d and the variance sigma, and defining the point as a critical point if 5 continuous points satisfy d > 3 sigma;
step e4, determine the first critical point as the initial critical point and determine this point as the initial point of life prediction.
Further, the gaussian process regression is a set of random variables, the combination of any finite-dimensional random variable in the set obeys joint normal distribution, the estimation of posterior distribution is completed through training data prior distribution, and a training data set is defined as:
wherein xnIn time series, ynSetting the initial degradation critical point as (x) for the health value obtained by logistic regression at the momentN,yN) Then will be
For training the model, if 200 points after the current point are predicted, the input is
The predicted output is
When the next point in time is generated, it will be
For training the model, if 200 points after the current point are predicted, the input is
The predicted output is
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a service life prediction method of a ball screw pair, which comprises the steps of extracting vibration signals on a vibration sensor of the screw pair, further extracting characteristics, selecting important characteristics, mapping the extracted characteristics into a health value through a logistic regression algorithm, and predicting the trend of a health curve through Gaussian process regression.
Drawings
FIG. 1 is a flow chart of the present invention;
fig. 2 is an effect diagram of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings.
A method for predicting the service life of a ball screw assembly, as shown in fig. 1, includes the following steps:
step a, performing wavelet denoising processing on an original signal acquired by a vibration sensor;
b, transforming the time domain signal to a frequency domain space through Fourier transform, and extracting time domain characteristics and frequency domain characteristics through characteristic extraction;
step c, performing dimensionality reduction processing on the extracted time domain features and frequency domain features through a Fisher criterion to screen out features which greatly contribute to lead screw degradation;
d, mapping the features by adopting a logistic regression algorithm, wherein the mapping interval is [0,1 ];
e, detecting a critical point by using a self-adaptive threshold detection algorithm;
and f, predicting the residual service life of the screw rod through Gaussian process regression so as to obtain the residual service life.
The data of the vibration sensor of the invention is used as input, and the obtained original signal can be interfered by other external signals, so that the obtained vibration signal of the screw pair needs to be preprocessed to remove the influence of interference noise. The signal denoising method adopted by the invention is wavelet denoising.
Further, the extracted time domain feature and the extracted frequency domain feature include a root mean square value, a variance, a standard deviation, a maximum value, a minimum value, an average amplitude, a kurtosis factor, a waveform coefficient, a peak value factor, a pulse index, a square root amplitude, a margin coefficient, and a signal obtained by deviating through the vibration sensor, and are time domain signals including time and an amplitude of the vibration signal at the moment. In order to fully consider the information of the signal, it is necessary to extract the time domain features and the frequency domain features of the signal. Since the original signal is a time domain signal, it needs to be fourier transformed to transform the time domain signal into a frequency domain space. The time domain features and the frequency domain features selected by feature extraction are 14, and the time domain features and the frequency domain features are root mean square values, variances, standard deviations, maximum values, minimum values, average amplitude values, kurtosis factors, waveform coefficients, peak values, peak value factors, pulse indexes, square root amplitude values, margin coefficients and skewness respectively.
Further, the characteristics contributing largely to the lead screw degradation include a root mean square value, a standard deviation, a maximum value, and a variance.
For the acquired time-frequency features extracted by the sensor, the dimension of the formed feature vector is high, and a certain coupling relation exists among the features, so that certain overlapping of the observed data can be caused, if all the features are used as sensitive features of the ball screw pair, the defects of large information processing quantity and the like can be caused, and the realization of a subsequent real-time prediction algorithm is not facilitated. Therefore, dimension reduction processing needs to be performed on the extracted features to select the optimal features for subsequent processing. The method adopts the Fisher criterion to perform dimensionality reduction processing on the extracted sensitive characteristics of the screw pair so as to screen out characteristics which greatly contribute to screw degeneration. The important features screened by the dimension reduction processing are respectively as follows: root mean square value, standard deviation, maximum value, and variance.
The Fisher criterion measures the difference between two feature values by using the mean and variance, and the calculation formula of the Fisher criterion is as follows:
wherein Jfisher(P, Q) is the Fisher criterion score value, P and Q represent the normal and failure states of the feature sample, mu and sigma respectively2Mean and variance of the feature samples are represented, respectively.
And c, extracting normal samples and failure samples in the full life cycle data of the screw, wherein each sample consists of the characteristics in the step b. And respectively calculating the mean value and the variance of each characteristic in the two parts of samples, sequentially substituting the mean value and the variance into a calculation formula of a Fisher criterion to obtain a score value of each characteristic, and screening out the characteristic which greatly contributes to the degradation of the screw rod according to the score value.
4 feature values are obtained by feature extraction and feature selection. Mapping is needed to describe the health state of the equipment operation, and the mapped value is defined as a health value and is used for describing the health state of the equipment. The method adopts a logistic regression algorithm to map the characteristic value of the screw pair obtained by characteristic extraction, and the mapping interval is [0,1 ]. Therefore, as the equipment continues to operate, the health values of the equipment are continuously updated, and the health values provide a basis for subsequent life prediction.
The service life prediction of the screw rod refers to the time from the current operation moment of the equipment to the fault occurrence, namely the specific remaining service time. Specifically, the time required for the device to reach the health value preset by the device from the current time is used. In order to improve the stability of prediction, the life prediction is divided into two parts, namely a critical point detection part and a life prediction part.
Because the equipment is stable in the early stage of operation and the time from the current state to failure is long, in order to improve the stability of prediction and avoid the occurrence of false alarm of prediction, the invention adopts a self-adaptive threshold detection algorithm and starts a service life prediction algorithm for the part behind a critical point to predict the residual service life of the screw rod. Wherein, the critical point detection part adopts a self-developed 3 sigma adaptive threshold value monitoring algorithm; and the life prediction part adopts Gaussian process regression to perform trend prediction on the curve obtained by health assessment so as to obtain the residual life of the equipment.
The logistic regression algorithm is to find an optimal logistic regression model to describe the relationship between the input feature vector and the output feature value by using healthy and invalid samples. The formula for calculating the logistic regression health value is as follows:
in the formula, alpha and beta are parameters of a logistic regression algorithm, (x)1,x2,...xn) The eigenvalues obtained from the Fisher criterion in step c.
Wherein h isθ(x) Closer to 1 indicates better health status, and closer to 0 indicates worse health status.
After obtaining the vibration signal, through feature extraction and feature selection, further selecting important feature values, and mapping the feature values to health values of [0,1] through a logistic regression algorithm. Because the equipment operates more stably at the initial stage of the operation of the equipment, the health value changes within a certain range, and when the change range exceeds a certain threshold value, the critical point is considered to be reached. The method adopts a 3 sigma criterion to solve the critical point of the regression curve of the lead screw. The 3 sigma criterion can objectively reflect the degradation condition of the screw rod, and the randomness of threshold selection is avoided. According to the case where the random variable follows a normal distribution, the probability that the measured value falls within (x-3 σ, x +3 σ) is 99.74%, the probability that the measured value falls outside this interval is 0.26%, and it is considered as a small probability event, and the performance is considered to deviate from the original state. However, for a component such as a lead screw, due to the introduction of uncertainty factors such as operating conditions, it is impossible to determine whether the performance state of the component has changed only when a measurement value is outside the interval, for example, in the operation of the lead screw, if there is an impact in the operating conditions, the measurement value may fall outside the interval, but after the impact, the lead screw returns to a steady operation state, and the state when there is an impact should not be determined as a critical point. In order to avoid the influence of uncertain factors such as impact, the invention considers that the performance state of the screw rod is changed when 5 continuous points fall outside the interval.
Further, the adaptive threshold detection algorithm comprises the steps of:
step e1, setting the health value sequence of the lead screw obtained by the logistic regression as { HV }; the health value sequence in the normal state is { HVNormal}, newly introduced health value is HVNew；
Step e2, continuously generating a health value in the operation process of the equipment, and recording the mean value of the points in front of the current point as mu and the variance of the points in front of the current point as sigma;
step e3, recording the difference value d between the current point and the mean value mu of the previous point, judging the size between the difference value d and the variance sigma, and defining the point as a critical point if 5 continuous points satisfy d > 3 sigma;
step e4, determine the first critical point as the initial critical point and determine this point as the initial point of life prediction.
Further, the gaussian process regression is a set of random variables, the combination of any finite-dimensional random variable in the set obeys joint normal distribution, the estimation of posterior distribution is completed through training data prior distribution, and a training data set is defined as:
wherein xnIn time series, ynSetting the initial degradation critical point as (x) for the health value obtained by logistic regression at the momentN,yN) Then will be
For training the model, if 200 points after the current point are predicted, the input is
The predicted output is
When the next point in time is generated, it will be
For training the model, if 200 points after the current point are predicted, the input is
The predicted output is
The effect graph is shown in fig. 2, wherein the scatter diagram is data used for training in the gaussian process regression, and the plus sign diagram is an actual degradation track in the gaussian process regression. The curve is a predicted track in the regression of the Gaussian process, and the interval is a confidence interval of the regression of the Gaussian process.
The service life of the screw pair can be predicted by acquiring the vibration signal of the screw pair and adopting the algorithm. The method can predict the residual service life of the screw pair in advance, and carry out health maintenance on the equipment at the optimal time point according to the predicted service life, and can reduce maintenance guarantee cost, improve the perfectness rate of the equipment and the completion rate of tasks, and increase the economic benefits of enterprises.
Claims (4)
1. A service life prediction method of a ball screw pair is characterized by comprising the following steps:
step a, performing wavelet denoising processing on an original signal acquired by a vibration sensor;
b, transforming the time domain signal to a frequency domain space through Fourier transform, and extracting time domain characteristics and frequency domain characteristics through characteristic extraction;
step c, performing dimensionality reduction processing on the extracted time domain features and frequency domain features through a Fisher criterion to screen out features which greatly contribute to lead screw degradation;
d, mapping the features by adopting a logistic regression algorithm, wherein the mapping interval is [0,1 ];
e, detecting a critical point by using a self-adaptive threshold detection algorithm; the method specifically comprises the following steps:
step e1, setting the health value sequence of the lead screw obtained by the logistic regression as { HV }; the health value sequence in the normal state is { HVNormal}, newly introduced health value is HVNew；
Step e2, continuously generating health values in the running process of the equipment, and keeping the mean value of the health values of the points in front of the health value of the current point as mu and the variance of the health value of the points in front of the health value of the current point as sigma;
step e3, recording the difference value d between the current point health value and the mean value mu of the previous point health value, judging the size between the difference value d and the variance sigma, and defining the point as a critical point if 5 continuous points satisfy d > 3 sigma;
step e4, determining the first critical point as the initial degradation critical point and determining the point as the initial point of life prediction;
and f, predicting the residual service life of the screw rod through Gaussian process regression so as to obtain the residual service life.
2. The method of predicting the lifespan of a ball screw assembly according to claim 1, wherein the extracted temporal and frequency domain features include a root mean square value, a variance, a standard deviation, a maximum value, a minimum value, a mean amplitude, a kurtosis factor, a form factor, a peak value, a peak factor, a pulse index, a square root amplitude, a margin factor, and a skewness.
3. The method for predicting the service life of a ball screw pair according to claim 2, wherein the features contributing greatly to the screw degradation include a root mean square value, a standard deviation, a maximum value and a variance.
4. The method for predicting the service life of the ball screw pair according to claim 3, wherein the Gaussian process regression is a set of random variables, the combination of any finite-dimension random variable in the set obeys joint normal distribution, the estimation of posterior distribution is completed through training data prior distribution, and the training data set is defined as follows:
wherein xnIn time series, ynSetting the initial degradation critical point as (x) for the health value obtained by logistic regression at the momentN,yN) Then will be
For training the model, if 200 points after the current point are predicted, the input is
The predicted output is
When the next point in time is generated, it will be
For training the model, if 200 points after the current point are predicted, the input is
The predicted output is
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811243790.0A CN109255395B (en) | 2018-10-24 | 2018-10-24 | Service life prediction method of ball screw pair |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811243790.0A CN109255395B (en) | 2018-10-24 | 2018-10-24 | Service life prediction method of ball screw pair |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109255395A CN109255395A (en) | 2019-01-22 |
| CN109255395B true CN109255395B (en) | 2021-11-02 |
Family
ID=65046687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811243790.0A Expired - Fee Related CN109255395B (en) | 2018-10-24 | 2018-10-24 | Service life prediction method of ball screw pair |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109255395B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111618655B (en) * | 2019-11-29 | 2021-12-28 | 东莞先知大数据有限公司 | Quantitative evaluation method for health degree of ball screw of numerical control machine tool |
| CN112053009B (en) * | 2020-09-30 | 2023-08-01 | 华人运通(江苏)技术有限公司 | Fault prediction method, device, system and storage medium |
| CN112621381B (en) * | 2020-12-25 | 2022-07-26 | 上海交通大学 | Intelligent health state evaluation method and device for machine tool feeding system |
| CN113553759B (en) * | 2021-06-17 | 2024-04-30 | 浙江工业大学之江学院 | Water pump residual service life prediction method, device, equipment and storage medium |
| CN115096359A (en) * | 2022-06-17 | 2022-09-23 | 北京航空航天大学 | Metal roof health monitoring system and method |
| CN117451348B (en) * | 2023-12-26 | 2024-05-28 | 宁德时代新能源科技股份有限公司 | Screw device detection method and device, electronic device and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101870075A (en) * | 2010-07-02 | 2010-10-27 | 西南交通大学 | Method for predicting service life of screw pair of numerical control machine on basis of performance degradation model |
| CN102682180A (en) * | 2011-03-07 | 2012-09-19 | 上海宝信软件股份有限公司 | Evaluation method for performance degradation of rotary mechanical equipment |
-
2018
- 2018-10-24 CN CN201811243790.0A patent/CN109255395B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101870075A (en) * | 2010-07-02 | 2010-10-27 | 西南交通大学 | Method for predicting service life of screw pair of numerical control machine on basis of performance degradation model |
| CN102682180A (en) * | 2011-03-07 | 2012-09-19 | 上海宝信软件股份有限公司 | Evaluation method for performance degradation of rotary mechanical equipment |
Non-Patent Citations (5)
| Title |
|---|
| Application of Gaussian Process Regression for bearing degradation assessment;Sheng Hong 等;《2012 6th International Conference on New Trends in Information Science, Service Science and Data Mining (ISSDM2012)》;20130613;全文 * |
| 基于人工智能的丝杠寿命预测技术;高宏力等;《西南交通大学学报》;20101015(第05期);全文 * |
| 基于小波包分析的滚珠丝杠状态监测;吴远昊等;《机械设计与制造》;20181008(第10期);全文 * |
| 基于深度学习理论的轴承状态识别研究;郭亮等;《振动与冲击》;20160628(第12期);全文 * |
| 钻头磨损检测与剩余寿命评估;敖银辉等;《机械工程学报》;20110105(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109255395A (en) | 2019-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255395B (en) | Service life prediction method of ball screw pair | |
| Wang et al. | Statistical modeling of bearing degradation signals | |
| KR101936283B1 (en) | Diagnostic and prognostics method for machine fault | |
| CN111459700B (en) | Equipment fault diagnosis method, diagnosis device, diagnosis equipment and storage medium | |
| CN114167838B (en) | Multi-scale health assessment and fault prediction method for servo system | |
| CN105071983A (en) | Abnormal load detection method for cloud calculation on-line business | |
| CN116415126A (en) | Method, device and computing equipment for anomaly detection of doctor blades of paper machine | |
| CN109255201B (en) | SOM-MQE-based ball screw pair health assessment method | |
| CN106405384A (en) | Simulation circuit health state evaluation method | |
| CN117370847B (en) | Deep learning-based disconnecting switch detection method and device | |
| CN114881087A (en) | Building robot bearing performance degradation assessment method | |
| CN115081673B (en) | Abnormality prediction method and device for oil and gas pipeline, electronic equipment and medium | |
| CN108333443B (en) | Method for alarming intermittent defects of power equipment | |
| CN117150244B (en) | Intelligent power distribution cabinet state monitoring method and system based on electrical parameter analysis | |
| CN118536082A (en) | Inverter fault prediction method and device based on artificial intelligence, terminal equipment and storage medium | |
| CN118036440A (en) | Fan foundation settlement early warning method, system, electronic equipment and storage medium | |
| CN113670611A (en) | Bearing early degradation evaluation method, system, medium and electronic equipment | |
| CN113341283A (en) | Transformer bushing partial discharge mode identification method | |
| CN111222203B (en) | Bearing service life model creation and prediction method thereof | |
| CN111258863B (en) | Data anomaly detection method, device, server and computer readable storage medium | |
| CN116415129A (en) | Method, device and computing equipment for assessing the health of a doctor blade of a paper machine | |
| CN112862019A (en) | Method for dynamically screening aperiodic anomaly | |
| CN110532698B (en) | Industrial equipment vibration characteristic value trend prediction method based on data model | |
| CN110297140B (en) | Fault prediction method and device of power distribution system | |
| KR20220122781A (en) | Abnormality detection system, abnormality detection method, and abnormality detection program stored in recording medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20211102 |