Cognitive calculation-based wind power fault operation and maintenance management method

A technology for operation and maintenance management and failure, applied in computing, computer parts, data processing applications, etc., to solve problems such as no practical and effective solutions have been proposed

Pending Publication Date: 2021-01-01
XIANGTAN UNIV
0 Cites 0 Cited by

AI-Extracted Technical Summary

Problems solved by technology

[0003] The literature "Human reliability analysis of power system operation and its database system research" proposes human reliability analysis methods in three scenarios: time-related, process-related, and emergency-related, which can quantify the probability of human error; Human Factors Based on Hidden Markov Equipment Forced Outage Rate Model" also uses Hidden Markov to identify human factors in equipment strong wave outage in the above three scenarios; the literature "The impact of human errors on the reliability of protection systems "Based on the condition-based maintenance environment, the human error rate prediction technology (THERP), the...
View more

Abstract

The invention discloses a cognitive calculation-based wind power fault operation and maintenance management method. The method comprises the following steps: performing feature analysis on a turn-to-turn short circuit fault of a stator of a wind turbine generator; establishing a wind power fault operation and maintenance cognitive calculation model based on a Bayesian network according to a feature analysis result; collecting behavior actions and emotional state signals of operation and maintenance personnel in real time; and sending the acquired behavior action and emotion state signals intoa wind power fault operation and maintenance cognitive calculation model, and iterating the Bayesian network by using a Markov chain Monte Carlo method to obtain a cognitive result of the operation and maintenance personnel. According to the invention, emotion factors are considered in operation and maintenance personnel capability evaluation so as to cope with extremely severe conditions of fan maintenance, and by referring to the wind power fault operation and maintenance cognitive calculation model, power enterprises can more efficiently dispatch personnel, and the operation and maintenancecost is practically reduced.

Application Domain

Data processing applicationsCharacter and pattern recognition +2

Technology Topic

StatorBayesian network +10

Image

  • Cognitive calculation-based wind power fault operation and maintenance management method
  • Cognitive calculation-based wind power fault operation and maintenance management method
  • Cognitive calculation-based wind power fault operation and maintenance management method

Examples

  • Experimental program(1)

Example Embodiment

[0108]Example
[0109]Using the Markov Chain Monte Carlo method (MCMC) iterative Bayesian network, adjust the parameters c and d, and estimate the mean and variance of the parameters c and d in the direct dependence and joint relationship.
[0110]In the wind power fault operation and maintenance cognitive computing model: GSK, SK and PC; GSK, PK and BO; PK, SK and R&R, the three groups of nodes have a joint relationship. Taking GSK, SK and PC as examples, the parameter c And d are shown in Table 1:
[0111]Table 1 The mean and variance of the coefficient c and intercept d in the joint relationship
[0112]
[0113]among them, Is the mean and variance of the correlation coefficient c between the parent variables GSK, SK and the child variable PC; Is the mean and variance of the intercept d.
[0114]The other nodes are directly dependent. Take Cognition and PR; Cognition and PK as examples. The parameters c and d are shown in Table 2:
[0115]Table 2 The mean and variance of the coefficient c and intercept d in the direct dependence relationship
[0116]
[0117]among them, Is the mean and variance of the coefficient c directly dependent on the child node X and the parent node Cognition; Is the mean and variance of the intercept d.
[0118]The Metropolis sampler is used to construct a stable Markov chain, and a total of 6000 iterations of WinBUGS are used to achieve convergence and obtain the posterior distribution of the model. According to the Yerkes-Dodson curve, set the heart rate, body temperature, blood pressure level 2 as normal, and the levels of other observable variables are reduced from 1 to 3; the evaluation level of the knowledge ability nodes of GSK, PK, and SK is from 1 to 4 Decrease in turn, the rating of the PR emotional stress node is 2-3-1/4 from high to low. Emotional stress is not conducive to the improvement of operation and maintenance efficiency in the case of extremely high or extremely low; Cognition is the final cognitive evaluation As a result, the rating level is lowered from 1 to 4.
[0119]Figure 4In order to simulate the cognitive evaluation results of operation and maintenance personnel [1], [2], [3], it can be seen from the figure that operation and maintenance personnel [1] have a poor grasp of various knowledge, and GSK, PK, and SK are 3 and 4 levels The probability is large, the PR pressure is very high, and the probability of being level 4 is relatively large, so the cognitive ability is relatively poor; operation and maintenance personnel [2] have a good grasp of various knowledge, and the probability of GSK, PK, and SK being level 1 and 2 is higher. However, the PR pressure is high, and the probability of being grades 3 and 4 is high, so the cognitive ability is weak; operation and maintenance personnel [3] have a good grasp of various knowledge, and the probability of GSK, PK, and SK being grades 1 and 2 is higher. It is large, and the PK pressure is small, and the probability of being level 2 is high, so the cognitive ability is excellent.
[0120]Through the use of the MCMC algorithm, the WINDRIVE model demonstrates the use of probability distribution to judge the fault cognition ability of operation and maintenance personnel, and express the relationship between the ability and action of operation and maintenance personnel. It can be seen from the results of WinBUGS that the model can calculate the cognition of operation and maintenance personnel according to actual needs and specific background. In addition to common knowledge and operation mastery, the judgment factors also include extreme conditions for wind power operation and maintenance. The emotional state pressure level, PR node is very important for wind power operation and maintenance, which can greatly affect the operation level of operation and maintenance personnel.

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products