Multi-dimensional similarity statistical evaluation method for simulation and test results of lightning attachment points

Through the multi-dimensional similarity statistical evaluation method, the simulation and experimental data of lightning attachment points are constructed and processed, and the limitations and efficiency of similarity evaluation in the prior art are solved, and efficient and accurate multi-dimensional similarity evaluation of lightning attachment points is achieved.

CN119989541AActive Publication Date: 2025-05-13XIAN AIRBORNE ELECTROMAGNETIC TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202510453832.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-04-11
Publication Date
2025-05-13
Estimated Expiration
2045-04-11

AI Technical Summary

Technical Problem

The existing similarity evaluation technology for lightning attachment points and experimental results has problems such as single-dimensional analysis limitations, poor adaptability of small samples, sensitivity of outliers and low computational efficiency.

Method used

The multi-dimensional similarity statistical evaluation method is used to construct n-dimensional simulation attachment vectors and n-dimensional experimental attachment vectors, perform data preprocessing and outlier detection, calculate weighted multi-dimensional root mean square error, judge similarity, and adaptively select Hotelling's T² or permutation test for hypothesis test.

Benefits of technology

It improves the robustness and computing efficiency of multi-dimensional similarity evaluation of lightning attachment point simulation and experimental results, supports small sample data, and enhances the comprehensiveness and accuracy of the evaluation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119989541A_ABST
    Figure CN119989541A_ABST
Patent Text Reader

Abstract

The invention discloses a thunder and lightning attachment point simulation and test result multi-dimensional similarity statistical evaluation method. The method comprises the following steps: S1, constructing an evaluation vector: carrying out attachment area division on a simulation object and a test object, constructing an n-dimensional simulation attachment vector and an n-dimensional test attachment vector, and respectively assigning the n-dimensional simulation attachment vector and the n-dimensional test attachment vector; s2, performing abnormal value detection on data in the n-dimensional simulation attachment vector and the n-dimensional test attachment vector; s3, judging data similarity: judging whether the n-dimensional simulation attachment vector and the n-dimensional test attachment vector have similarity or not; s4, hypothesis testing: if the n-dimensional simulation attachment vector and the n-dimensional test attachment vector meet the similarity index, adaptively selecting Hotelline's Tor replacement testing, and if the n-dimensional simulation attachment vector and the n-dimensional test attachment vector meet the similarity index, adaptively selecting Hotelline's Tor replacement testing; and S5, generating a conclusion report. According to the thunder and lightning attachment point simulation and test result multi-dimensional similarity statistical evaluation method provided by the invention, the problems of instability in high-dimensional covariance estimation, sensitivity to abnormal values, low small sample inspection efficiency and the like are solved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Network traffic prediction method based on generative adversarial network

    CN112801411A

  • Aircraft lightning attachment point simulation evaluation method

    CN114757018A

  • Change detection method of pixel intensities using hypothesis testing with test statistics

    KR1019990084263A

  • Method and system for abnormal value detection in LTE network

    US20190261204A1