A Method of NC Small Sample Reliability Modeling Based on Improved Bootstrap and Fault Correlation Optimization

By combining the improved Bootstrap method and the FP-Growth algorithm, the problem of insufficient accuracy in small-sample reliability modeling of CNC systems was solved, high-precision fault correlation analysis was achieved, and the reliability assessment effect of CNC systems was improved.

CN122174191APending Publication Date: 2026-06-09BEIJING UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING UNIV OF TECH
Filing Date
2026-03-05
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

The reliability modeling accuracy of CNC systems is insufficient under small sample conditions. Traditional methods are difficult to effectively uncover fault correlation patterns, and irrelevant interference information is easily introduced when supplementing large sample data.

Method used

An improved Bootstrap method is used to expand the sample size, and the FP-Growth algorithm is combined to mine frequent itemsets between modules, calculate the fault correlation factor Ξ±, correct the Weibull distribution parameters, and construct a reliability model of multi-module coupling.

Benefits of technology

It significantly improves the accuracy of reliability assessment under small sample conditions, maintains the statistical characteristics and distribution continuity of the data, and enhances the stability and accuracy of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122174191A_ABST
    Figure CN122174191A_ABST
Patent Text Reader

Abstract

The application discloses a numerical control small sample reliability modeling method based on improved Bootstrap and fault correlation optimization, first extracts fault data of the same numerical control system, and selects correlation fault data from the fault data, and constructs the correlation fault data as a benchmark data set, and constructs the correlation fault data of the fault data of the same numerical control system as a small sample data set; then, the small sample data set is expanded based on an improved Bootstrap self-help method, sample expansion and statistical characteristics are maintained; frequent item sets and confidence information between system modules in the benchmark data set are mined based on an FP-Growth algorithm, a fault correlation factor is calculated, a Weibull distribution likelihood function is introduced to correct shape parameters and scale parameters, and a Weibull distribution correction model containing module coupling information is formed; and a reliability function and an average failure-free time are calculated based on the correction parameters. The method can significantly reduce model errors, improve modeling stability and prediction accuracy, and is suitable for high-precision reliability evaluation and modeling optimization of the numerical control system.
Need to check novelty before this filing date? Find Prior Art