[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN108304941A - A kind of failure prediction method based on machine learning - Google Patents

A kind of failure prediction method based on machine learning Download PDF

Info

Publication number
CN108304941A
CN108304941A CN201711362861.4A CN201711362861A CN108304941A CN 108304941 A CN108304941 A CN 108304941A CN 201711362861 A CN201711362861 A CN 201711362861A CN 108304941 A CN108304941 A CN 108304941A
Authority
CN
China
Prior art keywords
data
predicted
prediction model
subset
feature
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
Application number
CN201711362861.4A
Other languages
Chinese (zh)
Inventor
乔立中
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CHINA SOFTWARE AND TECHNOLOGY SERVICE Co Ltd
Original Assignee
CHINA SOFTWARE AND TECHNOLOGY SERVICE Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CHINA SOFTWARE AND TECHNOLOGY SERVICE Co Ltd filed Critical CHINA SOFTWARE AND TECHNOLOGY SERVICE Co Ltd
Priority to CN201711362861.4A priority Critical patent/CN108304941A/en
Publication of CN108304941A publication Critical patent/CN108304941A/en
Pending legal-status Critical Current

Links

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
    • G06N20/00—Machine learning
    • 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a kind of failure prediction methods based on machine learning.This method is：1) the setting operating index data for acquiring object to be predicted obtain the time series data of each setting operating index；Acquire the historical failure data of the object to be predicted；2) feature extraction is carried out respectively to the data of step 1) acquisition, the feature of extraction is input in machine learning system and is trained, obtain a basic failure predication model；3) real time data for collecting the setting operating index when object operation to be predicted, carries out feature extraction to it and inputs the basis failure predication model, predicts that the object to be predicted currently whether there is failure.The present invention improves equipment safety operation efficiency, shortens maintenance time, reduces maintenance cost, extends service life of equipment, the influence generated due to certain device fails is reduced or avoided.

Description

Fault prediction method based on machine learning
Technical Field
The invention belongs to the field of artificial intelligence machine learning, relates to a maintenance method, and particularly relates to a fault prediction method based on machine learning.
Background
Currently, in real life, the dependence on systems and machines is beyond the imagination of people. In daily travel, a vehicle needs to be driven, an elevator needs to be taken, a high-speed rail or an airplane needs to be carried, and in the manufacturing and production of enterprises, machines liberate workers, but faults occur to the machines or systems, some faults only bring inconvenience, and some faults are vital.
When the risk is high, regular maintenance of the system is required. Since the cost of failure is much higher than the cost on the surface. For example, high-speed rail is routinely inspected daily, automobiles are maintained every few months, and airplanes are maintained daily, which results in significant waste of resources, excessive, and even excessive, maintenance.
Predictive maintenance can predict failures, take action in advance, and even predict when failures will occur, which can greatly save overhead, bring high predictability, and enhance system availability. At the same time, predictive maintenance avoids both extremes, maximizing resource utilization. The method detects the abnormal and fault modes, gives early warning information, avoids or minimizes fault downtime, and optimizes periodic maintenance operation, thereby greatly improving maintenance efficiency and benefit.
Disclosure of Invention
In order to improve the safe operation efficiency of equipment, shorten the maintenance time, reduce the maintenance cost, prolong the service life of the equipment, reduce or avoid the influence caused by the fault of some equipment, and reasonably arrange the maintenance time plan in a certain range so as to reduce the loss caused by shutdown to the minimum, the invention provides a fault prediction method based on machine learning.
The technical scheme of the invention is as follows:
a fault prediction method based on machine learning comprises the following steps:
1) acquiring set operation index data of an object to be predicted to obtain time sequence data of each set operation index; collecting historical fault data of the object to be predicted;
2) respectively extracting the characteristics of the data acquired in the step 1), and inputting the extracted characteristics into a machine learning system for training to obtain a basic fault prediction model;
3) and collecting real-time data of the set operation indexes of the object to be predicted when the object to be predicted operates, extracting the characteristics of the real-time data, inputting the characteristics into the basic fault prediction model, and predicting whether the object to be predicted has faults at present.
Further, the machine learning system carries out denoising and feature engineering on the input features in sequence, trains to obtain a basic fault prediction model and carries out hyper-parameter optimization on the basic fault prediction model.
Further, the processing method of the feature engineering comprises the following steps: for each time series data, calculating a sliding window variance of the time series data, and using a plurality of sliding window variances of the time series data as a feature subset; then carrying out k-means clustering on each feature subset to obtain the feature subset with the highest prediction capability; and calculating the variance of the sliding window of the historical fault data to obtain a characteristic subset corresponding to the historical fault data.
Further, a method of calculating the sliding window variance: setting a diagnosis window with the width of h for time sequence data { x (t) }, namely, h data arranged according to the acquisition sequence are contained in the diagnosis window, and the data sequence at the time k of the diagnosis window is { xkX (k-j) } (j-h-1, h-2, …,1,0), its corresponding sliding window variance Is the sequence { xkMean of samples of { n }, nqThe number of singular points in the diagnostic window at time k.
Further, the method for performing k-means clustering on each feature subset to obtain the feature subset with the highest prediction capability comprises the following steps: for each feature subset, 1) selecting k objects from the feature subset data space as initial clustering centers; 2) calculating Euclidean distances between the data objects in the feature subset and each clustering center, and classifying the data objects in the feature subset to a class corresponding to the clustering center closest to the data objects according to a closest criterion; 3) updating a clustering center: taking the mean values corresponding to all the objects in each category as the clustering center of the category, and calculating the value of a target function; 4) judging whether the values of the clustering center and the target function are changed or not, if not, outputting a result, and if so, returning to 2); and repeating the steps until the characteristic subset corresponding to the characteristic subset with the most prediction capability is obtained.
Further, training by utilizing the feature subset with the highest prediction capability to obtain the basic fault prediction model, and performing hyper-parameter optimization processing on the basic fault prediction model; and evaluating and verifying the basic fault prediction model by utilizing the characteristic subset corresponding to the historical fault data, if the obtained conclusion is matched with the known fault, confirming the availability of the basic fault prediction model, and if not, correcting the basic fault prediction model.
Further, the optimized hyper-parameters include: the number of iterations, the distribution, the activation function, and the number of hidden layers.
Furthermore, the neural network Auto-Encoder is used for denoising the extracted features in sequence.
Further, the extracted features include a mean, a sliding window variance, a root mean square, a peak factor, a kurtosis coefficient, and a form factor of the data.
Further, the set operation index includes a device temperature, a heat quantity, a rotation speed, a displacement, a process parameter, and a vibration quantity.
The invention has the beneficial effects that:
the invention mainly aims to predict the time when the equipment is likely to have faults and then take relevant actions to prevent the faults, so as to monitor the future faults and schedule maintenance time in advance, and have better economic and social benefits. Not only can reduce cost, but also can achieve the following effects:
1. the maintenance frequency is reduced.
2. The time spent on certain maintained equipment is reduced, and the efficiency is improved.
3. And the maintenance cost is reduced.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
In order to facilitate the understanding of the technical contents of the present invention by those skilled in the art, the present invention will be further explained with reference to the accompanying drawings.
Data is the root of the machine learning solution problem, and if the data is not selected, the problem cannot be solved. Firstly, the data is divided into two parts, wherein one part mainly comprises basic technical parameters of equipment, operation index data and known data of past faults, and the specific parameters relate to equipment temperature, heat, rotating speed, displacement, process parameters, vibration quantity and the like. The other part is time sequence data of the parameters of equipment operation, the data is real-time data of the operation of a prediction object (such as equipment or a system) by additionally arranging a sensor in the object to be predicted, the two parts of data are subjected to preprocessing of feature extraction (mean value, variance, root mean square, peak factor, kurtosis coefficient and form factor), data cleaning and standardization in sequence, then a formed data set is input into a machine learning system, the steps of denoising, feature engineering and hyper-parameter optimization are sequentially carried out, an intelligent reasoning algorithm is carried out, a supervised and unsupervised algorithm is firstly adopted to construct a basic fault prediction model, and a reinforced learning algorithm is used for further optimizing the model. And then, evaluating and verifying the prediction model through an existing fault sample, if the obtained conclusion is matched with the known fault, confirming the usability of the model, if the obtained conclusion has deviation, correcting the prediction model, if the prediction conclusion has high deviation, the prediction model is so-called under-fitting, parameters such as the number of input features, the training step length is reduced, the number and the depth of decision trees are increased, if the inverse situation is that the over-fitting occurs, the weight of part of features is reduced or is 0 through increasing the input data quantity and then through a regularization method, and the complexity of the model is reduced.
FIG. 1 is a block diagram of the system framework of the present invention. The invention will be explained in detail below with reference to the drawings and the practical application. A training and evaluation framework diagram of predictive maintenance data as shown in fig. 1.
Firstly, raw data are input into a machine learning system after being preprocessed, and denoising, characteristic engineering and hyper-parameter optimization are sequentially carried out. Noise removal using a simple neural network Auto-Encoder, which uses the same numberThe data set is used as input and output of a network to train the model, and the number of parameters of the network is less than the dimensionality of the data set. This is very similar to principal component analysis, where data is represented as its several major dimensions. Because the dimensionality of noise is far higher than that of conventional data, the noise can be reduced in the process, and then the noise removing effect is improved through Auto-encoder optimization with three hidden layers. The data after de-noising processing is processed by feature engineering to obtain feature subsets, firstly, features as much as possible are selected, firstly, dimension reduction is carried out, namely, useful information is extracted from initial data, a data set in a high-dimensional space is mapped to low-dimensional space data through dimension reduction, meanwhile, information is lost as little as possible, features reserved after dimension reduction are selected, and finally, the feature subsets are stored. The processing steps of the characteristic engineering are as follows, and the variance of the sliding window is as follows: the time sequence of the data sampling is { x (t) }, a diagnosis window with the width of h is set, namely h data arranged according to the acquisition sequence are contained in the window, and the time sequence of the diagnosis window of the x variable at the moment k is called as { x (t) }, and the time sequence of the diagnosis window of the x variable at the moment k is { xkX (k-j) } (j-h-1, h-2, …,1,0), sample standard deviation thereof The estimate of the standard deviation of the x variable at time k is:σkthe characteristic state of the x variable at time k is characterized. Wherein,is the sequence { xkThe mean of samples at time k, assuming n is within the diagnostic windowqAnd (4) singular points.
k-mean value: 1) selecting k objects in each feature subset data space as initial clustering centers aiming at feature subsets obtained after feature engineering processing, wherein each object represents one clustering center; 2) for the data objects in the sample, according to Euclidean distances between the data objects and the clustering centers, classifying the data objects into the class corresponding to the clustering center (most similar) closest to the data objects according to the closest criterion; 3) updating a clustering center: taking the mean values corresponding to all the objects in each category as the clustering center of the category, and calculating the value of a target function; 4) and judging whether the values of the clustering center and the objective function are changed or not, if not, outputting the result, and if so, returning to the step 2). This is repeated until the feature subset with the best predictive power is obtained.
And (3) constructing a prediction base model based on the obtained feature subset by using an intelligent reasoning and deep neural network algorithm, and carrying out hyper-parameter optimization processing on the prediction base model to optimize the following hyper-parameters: iteration times, normal distribution, activation functions and the number of hidden layers. When the daily hyper-parameter optimization work is carried out, the hyper-parameter optimization method can be used for manually trying, and good parameters can be adjusted by using a random search method. Firstly setting an initial hyper-parameter value, then carrying out basic model training, feeding back the index of the model to a hyper-parameter tuning mechanism, tuning the hyper-parameter, continuing training the model, repeating the training in the way, and obtaining the current optimal hyper-parameter value after a plurality of times.
And finally, verifying the prediction system by using the optimal hyper-parameter, assisting an existing fault sample, further evaluating and correcting the prediction system, confirming the availability of the prediction system if the obtained conclusion is matched with the known fault, and correcting the prediction system if the obtained conclusion has a deviation. In order to avoid the problem that unbalanced class distribution occurs if the same class is over-represented in the data set, the invention also adopts the generated feature subset to evaluate the machine learning model and uses precision ratio and recall ratio as the measuring standard. For accuracy, recall ratio and precision ratio, the best value is close to 1, and the trained model has good performance.
The invention discloses a fault prediction system based on machine learning, which inputs preprocessed data into the machine learning system, sequentially carries out denoising, characteristic engineering and hyper-parameter optimization processing, obtains a fault prediction model through an intelligent reasoning algorithm, evaluates the model through a typical fault sample and a characteristic subset to improve the accuracy of the model prediction model, thereby constructing the fault prediction system, and finally inputs the acquired real-time operation data of equipment into the fault prediction system to realize the prediction of equipment faults.
The foregoing description of the preferred embodiments of the present invention has been included to describe the features of the invention in detail, and is not intended to limit the inventive concepts to the particular forms of the embodiments described, as other modifications and variations within the spirit of the inventive concepts will be protected by this patent. The subject matter of the present disclosure is defined by the claims, not the detailed description of the embodiments.

Claims (10)

1. A fault prediction method based on machine learning comprises the following steps:
1) acquiring set operation index data of an object to be predicted to obtain time sequence data of each set operation index; collecting historical fault data of the object to be predicted;
2) respectively extracting the characteristics of the data acquired in the step 1), and inputting the extracted characteristics into a machine learning system for training to obtain a basic fault prediction model;
3) and collecting real-time data of the set operation indexes of the object to be predicted when the object to be predicted operates, extracting the characteristics of the real-time data, inputting the characteristics into the basic fault prediction model, and predicting whether the object to be predicted has faults at present.
2. The method of claim 1, wherein the machine learning system performs denoising and feature engineering on the input features in sequence, trains to obtain a basic failure prediction model, and performs hyper-parameter optimization on the basic failure prediction model.
3. The method of claim 2, wherein the processing method of the feature engineering is: for each time series data, calculating a sliding window variance of the time series data, and using a plurality of sliding window variances of the time series data as a feature subset; then carrying out k-means clustering on each feature subset to obtain the feature subset with the highest prediction capability; and calculating the variance of the sliding window of the historical fault data to obtain a characteristic subset corresponding to the historical fault data.
4. The method of claim 3, wherein the method of calculating the sliding window variance: setting a diagnosis window with the width of h for time sequence data { x (t) }, namely, h data arranged according to the acquisition sequence are contained in the diagnosis window, and the data sequence at the time k of the diagnosis window is { xkX (k-j) } (j-h-1, h-2, …,1,0), its corresponding sliding window variancene＝h-nq-1，Is the sequence { xkMean of samples of { n }, nqThe number of singular points in the diagnostic window at time k.
5. The method of claim 3, wherein k-means clustering is performed on each feature subset to obtain the most predictive feature subset by: for each feature subset, 1) selecting k objects from the feature subset data space as initial clustering centers; 2) calculating Euclidean distances between the data objects in the feature subset and each clustering center, and classifying the data objects in the feature subset to a class corresponding to the clustering center closest to the data objects according to a closest criterion; 3) updating a clustering center: taking the mean values corresponding to all the objects in each category as the clustering center of the category, and calculating the value of a target function; 4) judging whether the values of the clustering center and the target function are changed or not, if not, outputting a result, and if so, returning to 2); and repeating the steps until the characteristic subset corresponding to the characteristic subset with the most prediction capability is obtained.
6. The method of claim 3, wherein the basic fault prediction model is obtained by training with the most predictive feature subset and is subjected to hyper-parameter optimization; and evaluating and verifying the basic fault prediction model by utilizing the characteristic subset corresponding to the historical fault data, if the obtained conclusion is matched with the known fault, confirming the availability of the basic fault prediction model, and if not, correcting the basic fault prediction model.
7. The method of claim 6, wherein the optimized hyper-parameters comprise: the number of iterations, the distribution, the activation function, and the number of hidden layers.
8. The method of claim 2, wherein the extracted features are denoised sequentially using a neural network Auto-Encoder.
9. The method of claim 1, wherein the extracted features include a mean, a sliding window variance, a root mean square, a peak factor, a kurtosis coefficient, and a form factor of the data.
10. The method of claim 1, wherein the set operating indicators include device temperature, heat, rotational speed, displacement, process parameters, and vibration magnitude.
CN201711362861.4A 2017-12-18 2017-12-18 A kind of failure prediction method based on machine learning Pending CN108304941A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711362861.4A CN108304941A (en) 2017-12-18 2017-12-18 A kind of failure prediction method based on machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711362861.4A CN108304941A (en) 2017-12-18 2017-12-18 A kind of failure prediction method based on machine learning

Publications (1)

Publication Number Publication Date
CN108304941A true CN108304941A (en) 2018-07-20

Family

ID=62870169

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711362861.4A Pending CN108304941A (en) 2017-12-18 2017-12-18 A kind of failure prediction method based on machine learning

Country Status (1)

Country Link
CN (1) CN108304941A (en)

Cited By (44)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109412892A (en) * 2018-10-23 2019-03-01 株洲中车时代电气股份有限公司 A kind of network communication quality assessment system and method
CN109472285A (en) * 2018-09-29 2019-03-15 北京中油瑞飞信息技术有限责任公司 Leakage recognition methods, device and computer equipment
CN109492752A (en) * 2018-11-05 2019-03-19 常熟理工学院 Using intelligent learning system and method for the data-signal in conjunction with knowledge elicitation
CN109547251A (en) * 2018-11-27 2019-03-29 广东电网有限责任公司 A kind of operation system failure and performance prediction method based on monitoring data
CN109582724A (en) * 2018-12-07 2019-04-05 厦门铅笔头信息科技有限公司 Distributed automated characterization engineering system framework
CN109583751A (en) * 2018-11-27 2019-04-05 中国科学院空间应用工程与技术中心 The failure decision-making technique of payload
CN109670551A (en) * 2018-12-20 2019-04-23 树根互联技术有限公司 The failure prediction method and device of engineering mechanical device
CN109685331A (en) * 2018-12-06 2019-04-26 中国软件与技术服务股份有限公司 A kind of high-speed rail bogie sensor fault diagnosis method based on machine learning
CN109737045A (en) * 2019-01-11 2019-05-10 南方电网科学研究院有限责任公司 Air compressor fault early warning method and system applied to chip production and related device
CN110033135A (en) * 2019-04-15 2019-07-19 北京交通大学 The train braking system failure prediction method that Multivariate Time Series feature is reinforced
CN110175701A (en) * 2019-04-29 2019-08-27 北京六捷科技有限公司 A kind of railway service Call failure prediction technique and device
CN110349289A (en) * 2019-06-21 2019-10-18 东软集团股份有限公司 Failure prediction method, device, storage medium and electronic equipment
CN110428109A (en) * 2019-08-07 2019-11-08 佳都新太科技股份有限公司 Subway shield door time between failures prediction model is established and prediction technique
CN110750784A (en) * 2019-09-24 2020-02-04 深圳市智莱科技股份有限公司 Safety prevention and control method and system for automatic vending equipment
CN110794227A (en) * 2018-08-02 2020-02-14 阿里巴巴集团控股有限公司 Fault detection method, system, device and storage medium
CN110995384A (en) * 2019-11-27 2020-04-10 杭州联汇科技股份有限公司 Broadcast master control fault trend prejudging method based on machine learning
CN111105063A (en) * 2018-10-26 2020-05-05 北京国双科技有限公司 Fault prediction method, fault prediction device, model construction method, fault prediction device, processor and readable storage medium
CN111221919A (en) * 2018-11-27 2020-06-02 波音公司 System and method for generating aircraft failure prediction classifier
CN111219942A (en) * 2018-11-26 2020-06-02 珠海格力电器股份有限公司 Refrigerator fault prediction method and device
CN111459692A (en) * 2019-01-18 2020-07-28 伊姆西Ip控股有限责任公司 Method, apparatus and computer program product for predicting drive failure
CN111724048A (en) * 2020-06-03 2020-09-29 浙江中烟工业有限责任公司 Characteristic extraction method for finished product library scheduling system performance data based on characteristic engineering
CN111829425A (en) * 2020-08-06 2020-10-27 厦门航空有限公司 Health monitoring method and system for civil aircraft leading edge position sensor
CN111860943A (en) * 2020-06-08 2020-10-30 清华大学 Power grid fault prediction method and system based on numerical meteorological data and machine learning
CN111985651A (en) * 2019-05-22 2020-11-24 中国移动通信集团福建有限公司 Operation and maintenance method and device for business system
CN112241511A (en) * 2019-07-19 2021-01-19 上海杰之能软件科技有限公司 Performance detection method and device of auxiliary electrical system of vehicle and terminal
CN112506687A (en) * 2020-11-24 2021-03-16 四川长虹电器股份有限公司 Fault diagnosis method based on multi-period segmented sliding window standard deviation
CN112534370A (en) * 2018-08-12 2021-03-19 斯凯孚人工智能有限公司 System and method for predicting industrial machine failure
CN112541597A (en) * 2019-09-04 2021-03-23 上海杰之能软件科技有限公司 Multi-equipment maintenance method and device, storage medium and terminal
CN112613619A (en) * 2020-11-30 2021-04-06 重庆电子工程职业学院 Safety operation management system for gas area in workshop
CN112947356A (en) * 2019-12-10 2021-06-11 佳能株式会社 Control method, control device, mechanical equipment and recording medium
CN112989501A (en) * 2021-05-10 2021-06-18 中国标准化研究院 Balance car safety evaluation method and device and terminal equipment
CN113383282A (en) * 2019-01-31 2021-09-10 应用材料公司 Correcting component failure in an ion implanted semiconductor manufacturing tool
CN113609187A (en) * 2021-07-23 2021-11-05 四川虹美智能科技有限公司 Fault diagnosis method and system based on big data
CN113656287A (en) * 2021-07-28 2021-11-16 北京宝兰德软件股份有限公司 Method and device for predicting software instance fault, electronic equipment and storage medium
CN113947248A (en) * 2021-10-21 2022-01-18 广东电网有限责任公司广州供电局 Method and related device for predicting risk of cable moisture aging tripping
CN114251295A (en) * 2020-09-23 2022-03-29 株式会社岛津制作所 Pump monitoring device, vacuum pump, pump monitoring method, and storage medium
CN114418810A (en) * 2021-12-08 2022-04-29 福州物联网开放实验室有限公司 Parallel maintenance management system for municipal facilities
CN114492517A (en) * 2022-01-10 2022-05-13 南方科技大学 Elevator detection method, elevator detection device, electronic device and storage medium
CN114781762A (en) * 2022-06-21 2022-07-22 四川观想科技股份有限公司 Equipment fault prediction method based on life consumption
CN114987773A (en) * 2022-05-20 2022-09-02 成都飞机工业（集团）有限责任公司 Method, device, equipment and medium for identifying performance abnormity of aircraft hydraulic system
US11520564B2 (en) 2021-01-20 2022-12-06 International Business Machines Corporation Intelligent recommendations for program code
CN116364263A (en) * 2022-12-15 2023-06-30 四川大学 CT equipment anomaly prediction method and system based on Internet of things data
CN116579762A (en) * 2023-04-14 2023-08-11 广州林旺空调工程有限公司 Intelligent operation and maintenance platform for cooling tower
CN116364263B (en) * 2022-12-15 2024-11-15 四川大学 CT equipment anomaly prediction method and system based on Internet of things data

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103257921A (en) * 2013-04-16 2013-08-21 西安电子科技大学 Improved random forest algorithm based system and method for software fault prediction
CN103745229A (en) * 2013-12-31 2014-04-23 北京泰乐德信息技术有限公司 Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine)
US20170249234A1 (en) * 2016-02-29 2017-08-31 B. G. Negev Technologies And Applications Ltd., At Ben-Gurion University Data-augmented software diagnosis method and a diagnoser therefor

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103257921A (en) * 2013-04-16 2013-08-21 西安电子科技大学 Improved random forest algorithm based system and method for software fault prediction
CN103745229A (en) * 2013-12-31 2014-04-23 北京泰乐德信息技术有限公司 Method and system of fault diagnosis of rail transit based on SVM (Support Vector Machine)
US20170249234A1 (en) * 2016-02-29 2017-08-31 B. G. Negev Technologies And Applications Ltd., At Ben-Gurion University Data-augmented software diagnosis method and a diagnoser therefor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
白小琴 等: ""流程工业仪表故障快速检测方法"", 《化工自动化及仪表》 *

Cited By (60)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110794227A (en) * 2018-08-02 2020-02-14 阿里巴巴集团控股有限公司 Fault detection method, system, device and storage medium
CN112534370A (en) * 2018-08-12 2021-03-19 斯凯孚人工智能有限公司 System and method for predicting industrial machine failure
CN109472285A (en) * 2018-09-29 2019-03-15 北京中油瑞飞信息技术有限责任公司 Leakage recognition methods, device and computer equipment
CN109412892A (en) * 2018-10-23 2019-03-01 株洲中车时代电气股份有限公司 A kind of network communication quality assessment system and method
CN109412892B (en) * 2018-10-23 2022-03-01 株洲中车时代电气股份有限公司 Network communication quality evaluation system and method
CN111105063A (en) * 2018-10-26 2020-05-05 北京国双科技有限公司 Fault prediction method, fault prediction device, model construction method, fault prediction device, processor and readable storage medium
CN109492752A (en) * 2018-11-05 2019-03-19 常熟理工学院 Using intelligent learning system and method for the data-signal in conjunction with knowledge elicitation
CN111219942A (en) * 2018-11-26 2020-06-02 珠海格力电器股份有限公司 Refrigerator fault prediction method and device
CN109547251A (en) * 2018-11-27 2019-03-29 广东电网有限责任公司 A kind of operation system failure and performance prediction method based on monitoring data
CN109547251B (en) * 2018-11-27 2022-05-10 广东电网有限责任公司 Service system fault and performance prediction method based on monitoring data
CN111221919A (en) * 2018-11-27 2020-06-02 波音公司 System and method for generating aircraft failure prediction classifier
CN109583751A (en) * 2018-11-27 2019-04-05 中国科学院空间应用工程与技术中心 The failure decision-making technique of payload
CN109685331A (en) * 2018-12-06 2019-04-26 中国软件与技术服务股份有限公司 A kind of high-speed rail bogie sensor fault diagnosis method based on machine learning
CN109582724A (en) * 2018-12-07 2019-04-05 厦门铅笔头信息科技有限公司 Distributed automated characterization engineering system framework
CN109582724B (en) * 2018-12-07 2022-04-08 厦门铅笔头信息科技有限公司 Distributed automatic feature engineering system architecture
CN109670551A (en) * 2018-12-20 2019-04-23 树根互联技术有限公司 The failure prediction method and device of engineering mechanical device
CN109737045A (en) * 2019-01-11 2019-05-10 南方电网科学研究院有限责任公司 Air compressor fault early warning method and system applied to chip production and related device
CN111459692A (en) * 2019-01-18 2020-07-28 伊姆西Ip控股有限责任公司 Method, apparatus and computer program product for predicting drive failure
CN111459692B (en) * 2019-01-18 2023-08-18 伊姆西Ip控股有限责任公司 Method, apparatus and computer program product for predicting drive failure
CN113383282A (en) * 2019-01-31 2021-09-10 应用材料公司 Correcting component failure in an ion implanted semiconductor manufacturing tool
CN110033135A (en) * 2019-04-15 2019-07-19 北京交通大学 The train braking system failure prediction method that Multivariate Time Series feature is reinforced
CN110175701A (en) * 2019-04-29 2019-08-27 北京六捷科技有限公司 A kind of railway service Call failure prediction technique and device
CN111985651A (en) * 2019-05-22 2020-11-24 中国移动通信集团福建有限公司 Operation and maintenance method and device for business system
CN110349289A (en) * 2019-06-21 2019-10-18 东软集团股份有限公司 Failure prediction method, device, storage medium and electronic equipment
CN112241511A (en) * 2019-07-19 2021-01-19 上海杰之能软件科技有限公司 Performance detection method and device of auxiliary electrical system of vehicle and terminal
CN112241511B (en) * 2019-07-19 2024-05-14 上海杰之能软件科技有限公司 Performance detection method and device for auxiliary electrical system of vehicle and terminal
CN110428109B (en) * 2019-08-07 2020-11-03 佳都新太科技股份有限公司 Subway shield door fault interval time prediction model establishing and predicting method
CN110428109A (en) * 2019-08-07 2019-11-08 佳都新太科技股份有限公司 Subway shield door time between failures prediction model is established and prediction technique
CN112541597A (en) * 2019-09-04 2021-03-23 上海杰之能软件科技有限公司 Multi-equipment maintenance method and device, storage medium and terminal
CN110750784B (en) * 2019-09-24 2023-10-03 深圳市智莱科技股份有限公司 Security prevention and control method and system for automatic vending equipment
CN110750784A (en) * 2019-09-24 2020-02-04 深圳市智莱科技股份有限公司 Safety prevention and control method and system for automatic vending equipment
CN110995384A (en) * 2019-11-27 2020-04-10 杭州联汇科技股份有限公司 Broadcast master control fault trend prejudging method based on machine learning
CN110995384B (en) * 2019-11-27 2021-08-10 杭州联汇科技股份有限公司 Broadcast master control fault trend prejudging method based on machine learning
CN112947356A (en) * 2019-12-10 2021-06-11 佳能株式会社 Control method, control device, mechanical equipment and recording medium
CN111724048A (en) * 2020-06-03 2020-09-29 浙江中烟工业有限责任公司 Characteristic extraction method for finished product library scheduling system performance data based on characteristic engineering
CN111860943A (en) * 2020-06-08 2020-10-30 清华大学 Power grid fault prediction method and system based on numerical meteorological data and machine learning
CN111829425A (en) * 2020-08-06 2020-10-27 厦门航空有限公司 Health monitoring method and system for civil aircraft leading edge position sensor
CN114251295B (en) * 2020-09-23 2024-06-11 株式会社岛津制作所 Pump monitoring device, vacuum pump, pump monitoring method, and storage medium
TWI826803B (en) * 2020-09-23 2023-12-21 日商島津製作所股份有限公司 Pump monitoring device, vacuum pump, pump monitoring method and pump monitoring program
CN114251295A (en) * 2020-09-23 2022-03-29 株式会社岛津制作所 Pump monitoring device, vacuum pump, pump monitoring method, and storage medium
CN112506687A (en) * 2020-11-24 2021-03-16 四川长虹电器股份有限公司 Fault diagnosis method based on multi-period segmented sliding window standard deviation
CN112613619A (en) * 2020-11-30 2021-04-06 重庆电子工程职业学院 Safety operation management system for gas area in workshop
CN112613619B (en) * 2020-11-30 2024-04-09 重庆电子工程职业学院 Safety operation management system for gas area in workshop
US11520564B2 (en) 2021-01-20 2022-12-06 International Business Machines Corporation Intelligent recommendations for program code
CN112989501B (en) * 2021-05-10 2022-02-01 中国标准化研究院 Balance car safety evaluation method and device and terminal equipment
CN112989501A (en) * 2021-05-10 2021-06-18 中国标准化研究院 Balance car safety evaluation method and device and terminal equipment
CN113609187A (en) * 2021-07-23 2021-11-05 四川虹美智能科技有限公司 Fault diagnosis method and system based on big data
CN113656287A (en) * 2021-07-28 2021-11-16 北京宝兰德软件股份有限公司 Method and device for predicting software instance fault, electronic equipment and storage medium
CN113656287B (en) * 2021-07-28 2024-06-04 北京宝兰德软件股份有限公司 Method and device for predicting software instance faults, electronic equipment and storage medium
CN113947248B (en) * 2021-10-21 2024-05-31 广东电网有限责任公司广州供电局 Risk prediction method and related device for cable damp aging tripping
CN113947248A (en) * 2021-10-21 2022-01-18 广东电网有限责任公司广州供电局 Method and related device for predicting risk of cable moisture aging tripping
CN114418810A (en) * 2021-12-08 2022-04-29 福州物联网开放实验室有限公司 Parallel maintenance management system for municipal facilities
CN114492517A (en) * 2022-01-10 2022-05-13 南方科技大学 Elevator detection method, elevator detection device, electronic device and storage medium
CN114492517B (en) * 2022-01-10 2022-11-25 南方科技大学 Elevator detection method, elevator detection device, electronic device and storage medium
CN114987773A (en) * 2022-05-20 2022-09-02 成都飞机工业（集团）有限责任公司 Method, device, equipment and medium for identifying performance abnormity of aircraft hydraulic system
CN114781762A (en) * 2022-06-21 2022-07-22 四川观想科技股份有限公司 Equipment fault prediction method based on life consumption
CN116364263A (en) * 2022-12-15 2023-06-30 四川大学 CT equipment anomaly prediction method and system based on Internet of things data
CN116364263B (en) * 2022-12-15 2024-11-15 四川大学 CT equipment anomaly prediction method and system based on Internet of things data
CN116579762B (en) * 2023-04-14 2023-10-20 广州林旺空调工程有限公司 Intelligent operation and maintenance platform for cooling tower
CN116579762A (en) * 2023-04-14 2023-08-11 广州林旺空调工程有限公司 Intelligent operation and maintenance platform for cooling tower

Similar Documents

Publication Publication Date Title
CN108304941A (en) A kind of failure prediction method based on machine learning
CN109102005B (en) Small sample deep learning method based on shallow model knowledge migration
Yan et al. Deep order-wavelet convolutional variational autoencoder for fault identification of rolling bearing under fluctuating speed conditions
Pandhare et al. Convolutional neural network based rolling-element bearing fault diagnosis for naturally occurring and progressing defects using time-frequency domain features
CN111353482A (en) LSTM-based fatigue factor recessive anomaly detection and fault diagnosis method
CN118154174B (en) Intelligent operation and maintenance cloud platform for industrial equipment
Wang et al. A sequence-to-sequence model with attention and monotonicity loss for tool wear monitoring and prediction
CN112784920B (en) Yun Bianduan coordinated rotating component reactance domain self-adaptive fault diagnosis method
CN110685868A (en) Wind turbine generator fault detection method and device based on improved gradient elevator
CN116010900A (en) Multi-scale feature fusion gearbox fault diagnosis method based on self-attention mechanism
CN114218872B (en) DBN-LSTM semi-supervised joint model-based residual service life prediction method
CN114580291B (en) LSTM-VAE-based dynamic equipment health condition assessment method
CN111459144A (en) Airplane flight control system fault prediction method based on deep cycle neural network
Li et al. Gear pitting fault diagnosis using raw acoustic emission signal based on deep learning
CN110737976A (en) mechanical equipment health assessment method based on multi-dimensional information fusion
Shi et al. Health index synthetization and remaining useful life estimation for turbofan engines based on run-to-failure datasets
CN114500004A (en) Anomaly detection method based on conditional diffusion probability generation model
Zheng et al. An unsupervised transfer learning method based on SOCNN and FBNN and its application on bearing fault diagnosis
CN110779988A (en) Bolt life prediction method based on deep learning
Huang et al. An adversarial learning approach for machine prognostic health management
CN115618267A (en) Unsupervised domain self-adaption and entropy optimization equipment sensing diagnosis method and system
Cao et al. Fault diagnosis of rolling bearing based on multiscale one-dimensional hybrid binary pattern
CN116910574A (en) Mechanical equipment diagnosis method and device based on meta-learning and time convolution network
CN117633688A (en) Large-scale power data anomaly detection method based on ridge regression-k-means clustering-LOF-LSTM fusion algorithm
Wang et al. An efficient anomaly detection for high-speed train braking system using broad learning system

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

Application publication date: 20180720

RJ01 Rejection of invention patent application after publication