CN108285071A - A kind of elevator Gernral Check-up method based on Bayesian network - Google Patents
A kind of elevator Gernral Check-up method based on Bayesian network Download PDFInfo
- Publication number
- CN108285071A CN108285071A CN201810072530.5A CN201810072530A CN108285071A CN 108285071 A CN108285071 A CN 108285071A CN 201810072530 A CN201810072530 A CN 201810072530A CN 108285071 A CN108285071 A CN 108285071A
- Authority
- CN
- China
- Prior art keywords
- elevator
- bayesian network
- data
- node
- network
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0037—Performance analysers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0006—Monitoring devices or performance analysers
- B66B5/0018—Devices monitoring the operating condition of the elevator system
- B66B5/0031—Devices monitoring the operating condition of the elevator system for safety reasons
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B5/00—Applications of checking, fault-correcting, or safety devices in elevators
- B66B5/0087—Devices facilitating maintenance, repair or inspection tasks
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
- Telephonic Communication Services (AREA)
Abstract
The elevator Gernral Check-up method based on Bayesian network that the invention discloses a kind of, this method carries out bayesian network structure and parameter learning using the different samples of elevator faults, and use Monte Carlo inference mechanism, it is established that a Bayesian network Elevator Fault Diagnosis model for meeting elevator operation mechanism.The diagnostic method predicted just for significant condition itself compared to other, Bayesian network Elevator Fault Diagnosis model considers the mutual restricting relation between complication system state variable, more scientific and precise, higher prediction rate can be realized by demonstrating this method by existing data sample, therefore, Gernral Check-up method based on Bayesian network is very effective in terms of elevator Gernral Check-up and prediction, is had broad application prospects.
Description
Technical field
The present invention relates to computer and elevator technology field, more particularly to a kind of elevator health based on Bayesian network is examined
Disconnected method.
Background technology
With the continuous development of China's science and technology, the traffic of domestic city becomes very convenient, and most cities start
Build subway.The normal order of subway station be unable to do without the operation of elevator, is just particularly important to the maintenance of elevator, subway station
Middle elevator Gernral Check-up is the important content of elevator maintenance.
Currently, the Bayesian network (Bayesian Net-work) based on probability inference has been developed in recent years, is
Solve the problems, such as uncertain, imperfection and a kind of technology proposed, complex device is uncertain and interconnectivity for solving for it
Caused failure has prodigious advantage, as a kind of graphic description method oriented to probabilistic relation, combines prior information, makes
The uncertain problem for solving to be generated due to system unlike signal and signal association in description with the correlation theory of probability, is led to
It crosses Bayes' theorem and calculates posterior probability, can be applied to the decision for relying on various control factor, Bayesian network has been at present
Through starting to be used for fault diagnosis field, but also in the unreferenced diagnosis to elevator faults.
Elevator system configuration is complicated, is made of seven big subsystems, and relationship is complicated between system unit, and shape is run according to it
State is difficult to establish corresponding mathematical model.In Elevator Fault Diagnosis field, there are many signal for characterizing elevator device malfunction,
And since failure predication is to the research process of related signal of interest before failure does not occur, diagnosis object is complicated, tests hand
Section has limitation and knowledge inaccurate so that the fault diagnosis of elevator includes mainly there are many uncertainties：Index signal
Between have many correlation and interconnectivity；Same failure may be caused by one or more abnormal signals or a certain exception
Signal may cause single or multiple failures simultaneously.Overwhelming majority document is the fault diagnosis around single system at present, and is taken
Certain scientific achievement was obtained, and domestic and foreign literature is few for the relevant report of elevator device failure predication at present.Therefore, it grinds
Study carefully the Gernral Check-up method based on Bayesian network is necessary in terms of elevator Gernral Check-up and prediction, has wide
Application prospect.
Invention content
The shortcomings that it is an object of the invention to overcome the prior art and deficiency, provide a kind of elevator based on Bayesian network
Gernral Check-up method realizes the screening to elevator faults feature, predicts the failure of elevator device and the failure rate of critical piece, right
The state of elevator operation carries out health evaluating.
The purpose of the present invention is realized by the following technical solution：A kind of elevator Gernral Check-up side based on Bayesian network
Method includes the following steps：
1, elevator easy break-down critical component and fault signature are screened, the Elevator Fault Diagnosis based on Bayesian network is established
Model；
2, it under the premise of giving a data sample set D, finds one and matches best net with training sample set D
Network structure；
3, Bayesian network parameters study is realized；
4, the complete Bayesian network of application training infers malfunctioning node hair by the real-time status of fault signature node
The probability of raw failure.
Preferably, prior probability derives from the probability data that forefathers are obtained by concrete practice, including comes from elevator explanation
The Research Literature for the data statistics and previous elevator that book, metro company are recorded using elevator；For data without or it is incomplete,
The data obtained in the case of not putting into practice, prior probability is from the long-term practical experience assessment of domain expert.
Preferably, include 2 kinds of nodes in Bayesian network model：Specific malfunctioning node and fault signature node；Failure
It is connected between characteristic node, is connected between simultaneous faults characteristic node and specific malfunctioning node.
Preferably, elevator device failure is tentatively established with matlab auxiliary tools on Full BNT-1.04 platforms to examine
Disconnected Bayesian network model figure.
Preferably, in step 2, Algorithm for Bayesian Networks Structure Learning uses K2 scoring algorithms, with P (G, D) as scoring
Function：
Wherein, P (G) is the prior probability of network structure G, XiFor network node, XiHaveEqual riA state,
I.e. Nodes XiCorresponding father node integrates as ∏ i, πiFor the configuration of ∏ i, πiPut in order is 1,2 ... qi, NijkIt is to meet X in data set Di=xi kAnd πiThe case where=j quantity.
Preferably, in step 3, Bayesian network parameters study is realized using EM algorithms, i.e., on the basis of sample data, is sought
Seek the probability distribution of each node of network；Using network topology structure and training sample set and priori, Bayesian network is determined
Conditional probability density at each node of network model is denoted as P (θ | D, G)；Solution converges to local nodes optimized parameterProcess：
It initializes firstThen configuration passes through two step of iteration E and M and finds maximum a posteriori probability it is assumed that and restraining optimal value.
Specifically, carrying out maximal possibility estimation to data, simulation best suits the parameter of structure, is as follows：
(1) E walks (expectation)
h,i,j,k,l,s∈N
Wherein E is desired value；D is training sample；Indicating the optimized parameter found, the codomain of Xi isqiIt is configuration πiPut in order 1 ..., qi；NijkIt is to meet variate-value X in data set Di=xi kAnd
πiThe condition frequency of=j；ylIt is the data amount check lost in D；ShIt is bayesian network structure selection
Assuming that.
(2) M walks (maximum estimated)
Maximal possibility estimation function：
MAP estimation：
Ni′jkIt is the abundant statistical factors of priori；NijkIt is the abundant statistical factors of sample data, i, j, k, h, q ∈ N.
Preferably, it to improve the operation efficiency of multinode complexity Bayesian network, is calculated using Bayesian network approximate resoning
Method：Monte carlo algorithm；The method only need to utilize a randomizer and shellfish without combining other supplementary structures such as tree
The conditional probability distribution table of this model of leaf determines the state of each elevator faults characteristic node, then generates a large amount of sample,
One counting of each variable save, while the number that the variable is in each state is recorded, generate calculating after all samples
Probability.
Compared with prior art, the present invention having the following advantages that and advantageous effect：
The present invention proposes a kind of novel elevator Gernral Check-up method.This method using elevator faults different samples into
Row bayesian network structure and parameter learning, and use Monte Carlo inference mechanism, it is established that one meets elevator operation mechanism
Bayesian network Elevator Fault Diagnosis model.It is tested using the data sample of variety classes failure, examines this method
There is higher accuracy in judging elevator faults, auxiliary is provided for the failure predication and security performance of current elevator device
Means.Compared to the diagnostic method that other are predicted just for significant condition itself, Bayesian network Elevator Fault Diagnosis model
The mutual restricting relation between complication system state variable, more scientific and precise are considered, is verified by existing data sample
This method can realize higher prediction rate, therefore, Gernral Check-up method based on Bayesian network in elevator Gernral Check-up and
It is very effective, has broad application prospects in terms of prediction.
Description of the drawings
Fig. 1 is the implementing procedure figure of embodiment method.
Fig. 2 is tractive driving Bayesian network model figure.
Fig. 3 is in the comparison for having complete data and rate of correct diagnosis under the conditions of incomplete data.
Fig. 4 is in the case where different data amount is trained to the accuracy of Elevator Fault Diagnosis.
Fig. 5 is the elevator Bayesian network training time under different data amount.
Specific implementation mode
Present invention will now be described in further detail with reference to the embodiments and the accompanying drawings, but embodiments of the present invention are unlimited
In this.
Embodiment
1, the fault signature of traction elevator is screened, including traction system, guidance system, door in elevator basic structure
System, car, weight balancing system, electric drive system, electric control system, eight big system of safety system.And it is right
The component that elevator safety operation plays an important role, including：Tractive driving, suspension arrangement, car frame and car, counterweight, door system,
Safeguard protection, electrical control, guide rail.Filter out that wherein easy break-down critical component and fault signature are as shown in table 1, and elevator is main
Fault signature is as shown in table 2：
1 elevator device failure critical component of table and its safety index
2 elevator major failure feature of table
2, in conjunction with Bayesian network principle, with the probability theory in Bayesian network：
Prior probability：Assessed by professional or consulted the obtained probability value of historical document.Prior probability is pressed
Two kinds are divided into according to source, the first is the probability data that objective data, in particular to forefathers are obtained by concrete practice, this
Data statistics and previous elevator of the priori data from elevator specification, metro company using elevator record in embodiment
Research Literature etc.；Second is subjective probability, is the data obtained in the case where data or is not put into practice entirely,
Mainly assessed with the long-term practical experience of domain expert.
Posterior probability：According to Bayesian formula, prior probability is modified, it is made more to tally with the actual situation.
Bayesian formula：
Elevator Fault Diagnosis method based on Bayesian network is the complete Bayesian network of application training, passes through failure
The real-time status of characteristic node infers the probability that malfunctioning node breaks down.
3, an Elevator Fault Diagnosis model based on Bayesian network is established：
Since elevator system configuration is extremely complex, complete Bayesian network model number of nodes is numerous, with tractive driving
For Bayesian network model, the Elevator Fault Diagnosis model based on Bayesian network is illustrated.
Include 2 kinds of nodes in Bayesian network model：(1) specific malfunctioning node；(2) fault signature node.Failure is special
It is connected between sign node, is connected between simultaneous faults characteristic node and specific malfunctioning node.The failure that model is included specifically is shown in
Table 3.The fault signature node and associated specific failure such as table 4, the value of characteristic node that model is tentatively chosen are divided into
Three grades：Slightly, critical and serious, difference value 0/1/2.Malfunctioning node is divided into normal and two kinds of failure, value 0/1.
Fig. 2 is the elevator traction system fault diagnosis tentatively established with matlab auxiliary tools on Full BNT-1.04 platforms
Bayesian network model figure.
The specific failure that 3 model of table includes
The fault signature nodename and associated malfunctioning node number that 4 model of table includes
4, with elevator device Bayes net algorithm, including：
Algorithm for Bayesian Networks Structure Learning, be exactly give a data sample set D under the premise of, find one with
Training sample set D matches best network structure.Algorithm for Bayesian Networks Structure Learning uses K2 scoring algorithms, is made with P (G, D)
For score function：
Wherein, P (G) is the prior probability of network structure G, XiFor network node, XiHaveEqual riA state,
I.e. Nodes XiCorresponding father node integrates as ∏ i, πiFor the configuration of ∏ i, πiPut in order is 1,2 ... qi, NijkIt is to meet X in data set Di=xi kAnd πiThe case where=j quantity.
Bayesian network parameters study is realized using EM algorithms, i.e., on the basis of sample data, seeks each node of network
Probability distribution.Using network topology structure and training sample set and priori, determine at each node of Bayesian network model
Conditional probability density, be denoted as P (θ | D, G).Solution converges to local nodes optimized parameterProcess.It initializes firstMatch
It sets, then passes through two step of iteration E and M and find maximum a posteriori probability it is assumed that and restraining optimal value.Maximum likelihood is carried out to data to estimate
Meter, simulation best suit the parameter of structure, are as follows：
(1) E walks (expectation)
h,i,j,k,l,s∈N
Wherein E is desired value；D is training sample；Indicating the optimized parameter found, the codomain of Xi isqiIt is configuration πiPut in order 1 ..., qi；NijkIt is to meet variate-value X in data set Di=xi kAnd
πiThe condition frequency of=j；ylIt is the data amount check lost in D；ShIt is bayesian network structure selection
Assuming that.
(2) M walks (maximum estimated)
Maximal possibility estimation function：
MAP estimation：
Ni′jkIt is the abundant statistical factors of priori；NijkIt is the abundant statistical factors of sample data, i, j, k, h, q ∈ N.
Increase since the Accurate Reasoning operation complexity of Bayesian network increases with the interstitial content in network and exponentially
Long, in conjunction with the structure of elevator device complexity and numerous malfunctions and fault signature node, network may exceed the reality of hardware
Existing range.To improve the operation efficiency of multinode complexity Bayesian network, using Bayesian network approximate resoning algorithm：Meng Teka
Lip river (Monte Carlo) algorithm.The method only need to utilize a randomizer without combining other supplementary structures such as tree
The state of each elevator faults characteristic node is determined with the conditional probability distribution table of Bayesian model, then generates a large amount of sample
This, one counting of each variable save, while the number that the variable is in each state is recorded, generate meter after all samples
Calculate probability.
The gloomy TE-evolution1 machine-roomless lifts fault diagnosis model example of the base of a fruit based on Bayesian network：
The fault data for the elevator faults feature and collection that this section is obtained according to above-mentioned analysis, establishes the Bayesian network of elevator
Network model carries out simulation study using Full BNT-1.04 platforms, and verifying the model has elevator device Analysis on Fault Diagnosis
Effect property.Fig. 2 systems are taken to carry out factoid.
1, Accuracy Verification is tested：
According to the gloomy elevator actual operating data of Guangzhou Underground company's vehicle pond station base of a fruit, the Bayesian network of known parameters is obtained simultaneously
As true value, 5000 groups of sample datas are generated using probability sampling method with the network, is divided into 10 experiments, tests number every time
According to being 500 groups, it is used as training data using first 480 groups, last 20 groups are used as test data.Table 5 briefly lists wherein 1 time experiment
Training and test data.Using the Structure learning K2 scoring algorithms and parameter learning EM algorithms of above-mentioned Bayesian network, use
480 groups of training datas train new Bayesian network.The characteristic node data of 20 groups of test datas are inputted into trained shellfish again
This network of leaf obtains the malfunctioning node prediction result of 20 groups of data, with the Q1-Q9 of original 20 groups of data totally 9 malfunctioning node shapes
State is compared.For taking Q1, if being predicted in 20 groups of data, correct data number is M, and total data number is N=20, then Q1 is correct
Rate is PQ1=M/N, counts the accuracy of 9 malfunctioning nodes such as Q1 to Q9 respectively.One of 9 specific malfunctioning nodes Q2 is chosen (to close
Brake force is inadequate after lock) for, study simultaneously statistical correction rate.The prediction result statistics of 10 experiment posterior nodal point Q2 is as shown in table 6.
To other malfunctioning nodes, experiment results are similar.
The 480 groups of training datas and 20 groups of test datas of the 1st experiment of table 5
The malfunctioning node Q2 prediction results of 6 complete data of table
| Number | 1 | 2 | 3 | 4 | 5 |
| Accuracy/% | 81.11 | 83.27 | 87.79 | 86.62 | 89.83 |
| Number | 6 | 7 | 8 | 9 | 10 |
| Accuracy/% | 88.04 | 90.11 | 86.70 | 82.73 | 84.29 |
It can be obtained by the data in table 6, the prediction average accuracy of 10 experiments is 86.749%.The average training of record
Time is 328s.Bayesian network after training is very high in complete data accuracy, and program operation speed is fast.
In the actual motion of elevator, the state of whole characteristic nodes can not monitor completely, therefore the data obtained
It is incomplete, in the practical operation situation of elevator, has about 10% data volume to be unable to monitor, therefore in 5000 groups of training
10% data are hidden on sample basis at random and carry out e-learning, hiding data are expressed as NAN (indicating shortage of data), equally
One of 9 specific malfunctioning nodes Q2 (brake force is inadequate after closing lock) is chosen, the prediction result statistics of 10 experiment posterior nodal point Q2 is such as
Shown in table 7.To other malfunctioning nodes, experiment results are similar.
The malfunctioning node Q2 prediction results of 7 partial data of table missing
| Number | 1 | 2 | 3 | 4 | 5 |
| Accuracy/% | 71.13 | 67.28 | 73.91 | 77.16 | 74.11 |
| Number | 6 | 7 | 8 | 9 | 10 |
| Accuracy/% | 72.86 | 66.66 | 69.53 | 72.33 | 75.37 |
It can be obtained by the data in table 7, the lower prediction average accuracy of 10 experiments is 72.034%.The average training of record
Time is 359s.
It draws the performance of fault diagnosis based on Bayesian network and compares figure such as Fig. 3：
As seen from Figure 3, compared with the perfect condition of complete data, in the incomplete reality of elevator faults characteristic
Under state, though the Bayesian network fault diagnosis model accuracy that the method for the present embodiment trains is declined slightly, accuracy
Still higher.
2, different data amount test experiments：
This experiment is to find the amount of training data for the Bayesian network for being suitble to be studied herein.By known network according to general
Rate sampling generates 10000 groups of data, is divided into 10 experiments, and each experimental data is 1000 groups, using first 980 groups as training number
According to last 20 groups are used as test data.Experimental procedure is identical as the malfunctioning node prediction steps that upper partial data lacks.
10% data are hidden on the basis of 980 groups of training samples at random and carry out e-learning, equally choose one of 9 specific malfunctioning nodes Q2
(brake force is inadequate after closing lock), the prediction result statistics for being repeated 10 times experiment posterior nodal point Q2 are as shown in table 8.When record is averagely trained
Between be 873s.To other malfunctioning nodes, experiment results are similar.
The precision of prediction table of 8 1000 groups of data of table
| Number | 1 | 2 | 3 | 4 | 5 |
| Accuracy/% | 84.46 | 82.29 | 88.31 | 86.64 | 79.33 |
| Number | 6 | 7 | 8 | 9 | 10 |
| Accuracy/% | 84.69 | 82.24 | 84.11 | 82.33 | 80.12 |
15000 groups of data are generated according to probability sampling by known network again, repeat above-mentioned experiment, the prediction result system of node Q2
Meter is as shown in table 9.Record average workout times are 1736s.
The precision of prediction table of 9 1500 groups of data of table
| Number | 1 | 2 | 3 | 4 | 5 |
| Accuracy/% | 7 6.66 | 81.17 | 83.3 | 79.34 | 76.86 |
| Number | 6 | 7 | 8 | 9 | 10 |
| Accuracy/% | 84.42 | 80.11 | 83.35 | 81.19 | 79.94 |
Line chart is drawn according to 7-table of table 9, the training sample for observing different number is correct to Bayesian network fault diagnosis
The influence of rate.
Fig. 4 and Fig. 5 in such as illustrating can be seen that since the algorithm used in this paper e-learnings includes iteration
Etc. processes, with the increase of data volume, although Bayesian network accuracy is increased slightly, the training time in increasing substantially,
It is not The more the better it is therefore seen that for training the data volume appropriateness needed for Bayesian network.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, it is other it is any without departing from the spirit and principles of the present invention made by changes, modifications, substitutions, combinations, simplifications,
Equivalent substitute mode is should be, is included within the scope of the present invention.
Claims (8)
1. a kind of elevator Gernral Check-up method based on Bayesian network, which is characterized in that include the following steps：
S1, screening elevator easy break-down critical component and fault signature, establish the Elevator Fault Diagnosis mould based on Bayesian network
Type；
S2, give a data sample set D under the premise of, find one match best network with training sample set D
Structure；
S3, Bayesian network parameters study is realized；
The complete Bayesian network of S4, application training infers that event occurs for malfunctioning node by the real-time status of fault signature node
The probability of barrier.
2. the elevator Gernral Check-up method according to claim 1 based on Bayesian network, which is characterized in that prior probability
The probability data obtained by concrete practice from forefathers, including come from elevator specification, metro company is recorded using elevator
Data statistics and previous elevator Research Literature；It is obtained in the case of for or not put into practice entirely in data
Data, prior probability is from the long-term practical experience assessment of domain expert.
3. the elevator Gernral Check-up method according to claim 1 based on Bayesian network, which is characterized in that in Bayes
Include 2 kinds of nodes in network model：Specific malfunctioning node and fault signature node；It is connected between fault signature node, while therefore
It is connected between barrier characteristic node and specific malfunctioning node.
4. the elevator Gernral Check-up method according to claim 1 based on Bayesian network, which is characterized in that in Full
The Bayesian network model figure of elevator device fault diagnosis is tentatively established on BNT-1.04 platforms with matlab auxiliary tools.
5. the elevator Gernral Check-up method according to claim 1 based on Bayesian network, which is characterized in that in step 2,
Algorithm for Bayesian Networks Structure Learning uses K2 scoring algorithms, and score function is used as with P (G, D)：
Wherein, P (G) is the prior probability of network structure G, XiFor network node, XiHaveEqual riA state, i.e., Nodes XiCorresponding father node integrates as ∏ i, πiFor the configuration of ∏ i, πiPut in order is 1,2 ... qi, NijkIt is to meet X in data set Di=xi kAnd πiThe case where=j quantity.
6. the elevator Gernral Check-up method according to claim 1 based on Bayesian network, which is characterized in that in step 3,
Bayesian network parameters study is realized using EM algorithms, i.e., on the basis of sample data, seeks the probability distribution of each node of network；
Using network topology structure and training sample set and priori, the conditional probability at each node of Bayesian network model is determined
Density is denoted as P (θ | D, G)；Solution converges to local nodes optimized parameterProcess：It initializes firstConfiguration, then passes through
Two step of iteration E and M finds maximum a posteriori probability it is assumed that and restraining optimal value.
7. the elevator Gernral Check-up method according to claim 6 based on Bayesian network, which is characterized in that data into
Row maximal possibility estimation, simulation best suit the parameter of structure, are as follows：
(1) E is walked：
Wherein E is desired value；D is training sample；Indicating the optimized parameter found, the codomain of Xi is
qiIt is configuration πiPut in order 1 ..., qi；NijkIt is to meet variate-value X in data set Di=xi kAnd πiThe condition of=j occurs
Number；ylIt is the data amount check lost in D；ShIt is that bayesian network structure selection is assumed；
(2) M is walked：
Maximal possibility estimation function：
MAP estimation：
N′ijkIt is the abundant statistical factors of priori；NijkIt is the abundant statistical factors of sample data, i, j, k, h, q ∈ N.
8. the elevator Gernral Check-up method according to claim 1 based on Bayesian network, which is characterized in that more to improve
The operation efficiency of node complexity Bayesian network, using Bayesian network approximate resoning algorithm：Monte carlo algorithm；The method without
Other supplementary structures such as tree need to be combined, need to only utilize the conditional probability distribution table of a randomizer and Bayesian model
It determines the state of each elevator faults characteristic node, then generates a large amount of sample, one counting of each variable save, simultaneously
The number that the variable is in each state is recorded, all samples is generated and calculates probability later.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810072530.5A CN108285071A (en) | 2018-01-25 | 2018-01-25 | A kind of elevator Gernral Check-up method based on Bayesian network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810072530.5A CN108285071A (en) | 2018-01-25 | 2018-01-25 | A kind of elevator Gernral Check-up method based on Bayesian network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108285071A true CN108285071A (en) | 2018-07-17 |
Family
ID=62835614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810072530.5A Pending CN108285071A (en) | 2018-01-25 | 2018-01-25 | A kind of elevator Gernral Check-up method based on Bayesian network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108285071A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109032872A (en) * | 2018-08-13 | 2018-12-18 | 广州供电局有限公司 | Equipment fault diagnosis method and system based on bayesian network |
| CN109270461A (en) * | 2018-10-15 | 2019-01-25 | 四川长虹电器股份有限公司 | Fault detection method based on Bayesian network |
| CN109614647A (en) * | 2018-11-09 | 2019-04-12 | 南京航空航天大学 | A kind of high-speed rail trailer system failure prediction method based on Bayesian network |
| CN109656228A (en) * | 2018-12-04 | 2019-04-19 | 江苏大学 | A kind of subway signal system onboard equipment fault automatic diagnosis method |
| CN109688604A (en) * | 2018-12-06 | 2019-04-26 | 北京航空航天大学 | A kind of high dynamic mobile ad-hoc network survivability evaluation method based on Bayesian network |
| CN109784724A (en) * | 2019-01-15 | 2019-05-21 | 常州大学 | A kind of subsea production system equipment fault diagnosis method based on Bayesian network |
| CN110097727A (en) * | 2019-04-30 | 2019-08-06 | 暨南大学 | Forest Fire Alarm method and system based on fuzzy Bayesian network |
| CN110262380A (en) * | 2019-07-05 | 2019-09-20 | 华南理工大学 | The hot-forming accuracy control system of microarray and method based on experience technique probability |
| CN110503217A (en) * | 2019-08-29 | 2019-11-26 | 新誉轨道交通科技有限公司 | A kind of slow leakage failure prediction technique of air conditioner coolant, device, equipment and system |
| CN111563603A (en) * | 2020-04-21 | 2020-08-21 | 西人马（厦门）科技有限公司 | Elevator health state evaluation method and device and storage medium |
| CN112906775A (en) * | 2021-02-05 | 2021-06-04 | 深圳市芯聚智科技有限公司 | Equipment fault prediction method and system |
| CN112938678A (en) * | 2021-01-29 | 2021-06-11 | 广东卓梅尼技术股份有限公司 | Diagnosis algorithm for elevator vibration fault |
| CN113407907A (en) * | 2021-06-04 | 2021-09-17 | 电子科技大学 | Hierarchical system structure function learning method fusing incomplete monitoring sequence |
| CN114004091A (en) * | 2021-11-03 | 2022-02-01 | 兰州理工大学 | CEEMDAN-BNs-based wind variable pitch system fault diagnosis method |
| CN114647992A (en) * | 2022-03-10 | 2022-06-21 | 北京航空航天大学 | Solid-liquid rocket engine fault diagnosis method based on improved Bayesian algorithm |
| CN117952598A (en) * | 2024-03-26 | 2024-04-30 | 广东电网有限责任公司 | Energy system safety evaluation analysis method and device based on fault rate approximation |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102293656A (en) * | 2011-05-25 | 2011-12-28 | 四川大学华西医院 | Emotional stability evaluation system based on magnetic resonance imaging and evaluation method thereof |
-
2018
- 2018-01-25 CN CN201810072530.5A patent/CN108285071A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102293656A (en) * | 2011-05-25 | 2011-12-28 | 四川大学华西医院 | Emotional stability evaluation system based on magnetic resonance imaging and evaluation method thereof |
Non-Patent Citations (3)
| Title |
|---|
| 冯鑫等: "基于贝叶斯模型的电梯安全评估方法研究", 《起重运输机械》 * |
| 钟文奇: "基于贝叶斯网络的列车故障诊断研究", 《中国优秀硕士学位论文全文数据库_信息科技辑》 * |
| 陈新亿: "基于KL距离的贝叶斯网络结构学习算法研究", 《中国优秀硕士学位论文全文数据库_信息科技辑》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109032872A (en) * | 2018-08-13 | 2018-12-18 | 广州供电局有限公司 | Equipment fault diagnosis method and system based on bayesian network |
| CN109032872B (en) * | 2018-08-13 | 2021-08-10 | 广东电网有限责任公司广州供电局 | Bayesian network-based equipment fault diagnosis method and system |
| CN109270461A (en) * | 2018-10-15 | 2019-01-25 | 四川长虹电器股份有限公司 | Fault detection method based on Bayesian network |
| CN109614647A (en) * | 2018-11-09 | 2019-04-12 | 南京航空航天大学 | A kind of high-speed rail trailer system failure prediction method based on Bayesian network |
| CN109614647B (en) * | 2018-11-09 | 2023-05-26 | 南京航空航天大学 | Bayesian network-based high-speed rail traction system fault prediction method |
| CN109656228A (en) * | 2018-12-04 | 2019-04-19 | 江苏大学 | A kind of subway signal system onboard equipment fault automatic diagnosis method |
| CN109688604A (en) * | 2018-12-06 | 2019-04-26 | 北京航空航天大学 | A kind of high dynamic mobile ad-hoc network survivability evaluation method based on Bayesian network |
| CN109688604B (en) * | 2018-12-06 | 2021-09-21 | 北京航空航天大学 | Bayesian network-based high-dynamic mobile ad hoc network survivability evaluation method |
| CN109784724A (en) * | 2019-01-15 | 2019-05-21 | 常州大学 | A kind of subsea production system equipment fault diagnosis method based on Bayesian network |
| CN110097727A (en) * | 2019-04-30 | 2019-08-06 | 暨南大学 | Forest Fire Alarm method and system based on fuzzy Bayesian network |
| CN110262380A (en) * | 2019-07-05 | 2019-09-20 | 华南理工大学 | The hot-forming accuracy control system of microarray and method based on experience technique probability |
| CN110503217A (en) * | 2019-08-29 | 2019-11-26 | 新誉轨道交通科技有限公司 | A kind of slow leakage failure prediction technique of air conditioner coolant, device, equipment and system |
| CN111563603A (en) * | 2020-04-21 | 2020-08-21 | 西人马（厦门）科技有限公司 | Elevator health state evaluation method and device and storage medium |
| CN112938678A (en) * | 2021-01-29 | 2021-06-11 | 广东卓梅尼技术股份有限公司 | Diagnosis algorithm for elevator vibration fault |
| CN112906775A (en) * | 2021-02-05 | 2021-06-04 | 深圳市芯聚智科技有限公司 | Equipment fault prediction method and system |
| CN112906775B (en) * | 2021-02-05 | 2023-12-01 | 深圳市芯聚智科技有限公司 | Equipment fault prediction method and system |
| CN113407907B (en) * | 2021-06-04 | 2022-04-12 | 电子科技大学 | Hierarchical system structure function learning method fusing incomplete monitoring sequence |
| CN113407907A (en) * | 2021-06-04 | 2021-09-17 | 电子科技大学 | Hierarchical system structure function learning method fusing incomplete monitoring sequence |
| CN114004091A (en) * | 2021-11-03 | 2022-02-01 | 兰州理工大学 | CEEMDAN-BNs-based wind variable pitch system fault diagnosis method |
| CN114647992A (en) * | 2022-03-10 | 2022-06-21 | 北京航空航天大学 | Solid-liquid rocket engine fault diagnosis method based on improved Bayesian algorithm |
| CN117952598A (en) * | 2024-03-26 | 2024-04-30 | 广东电网有限责任公司 | Energy system safety evaluation analysis method and device based on fault rate approximation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108285071A (en) | A kind of elevator Gernral Check-up method based on Bayesian network | |
| CN110262463B (en) | Rail transit platform door fault diagnosis system based on deep learning | |
| CN105279365B (en) | For the method for the sample for learning abnormality detection | |
| US9630637B2 (en) | Complex network-based high speed train system safety evaluation method | |
| CN101216998B (en) | An urban traffic flow information amalgamation method of evidence theory based on fuzzy rough sets | |
| CN110843870B (en) | Method for maintaining fixed capacity of high-speed railway network graph under abnormal event | |
| CN103984994B (en) | Method for predicting urban rail transit passenger flow peak duration | |
| CN106338406A (en) | On-line monitoring and fault early-warning system and method for traction electric transmission system of train | |
| CN110007652A (en) | A kind of Hydropower Unit degradation trend interval prediction method and system | |
| CN110161343A (en) | A kind of non-intrusion type real-time dynamic monitoring method of intelligence train exterior power receiving device | |
| CN109597401A (en) | A kind of equipment fault diagnosis method based on data-driven | |
| CN112529329A (en) | Infectious disease prediction method based on BP algorithm and SEIR model | |
| CN104765965A (en) | GIS fault diagnosis and reliability analysis method based on fuzzy Petri | |
| CN103760901A (en) | Rail transit fault identification method based on association rule classifier | |
| CN106529787A (en) | Subway energy management system based on face detection technology | |
| CN103425884B (en) | A kind of fuzzy prediction method of subway transfer channel traffic state | |
| CN102663264A (en) | Semi-supervised synergistic evaluation method for static parameter of health monitoring of bridge structure | |
| CN109002625A (en) | A kind of feeder bus sytem network design method based on robust optimization | |
| CN113610188B (en) | Bow net contact force non-section abnormality identification method and device | |
| CN103678924A (en) | Method for determining failure mode risk priority numbers of key component of rail transit vehicle | |
| CN108334902A (en) | A kind of track train equipment room smog fireproof monitoring method based on deep learning | |
| CN109543814A (en) | A kind of each equipment fault prediction technique of subway signal system | |
| Ma et al. | Short-term traffic flow prediction based on genetic artificial neural network and exponential smoothing | |
| CN110221173A (en) | A kind of power distribution network intelligent diagnosing method based on big data driving | |
| CN106202668B (en) | Complex equipment quality risk appraisal procedure based on quality problems data and reverse conduction neural network |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180717 |
|
| WD01 | Invention patent application deemed withdrawn after publication |