Avionics fault diagnosis method based on multi-mode distillation and related equipment

By employing a two-stage architecture of dual-encoder teacher model and lightweight student model, along with a multi-level knowledge distillation strategy, the problem of insufficient cross-modal semantic alignment in multimodal avionics fault diagnosis is solved, achieving efficient knowledge transfer and diagnostic accuracy. This approach is suitable for avionics fault diagnosis equipment with limited resources.

CN121743091APending Publication Date: 2026-03-27XI AN JIAOTONG UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202511879189.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-12
Publication Date
2026-03-27

AI Technical Summary

Technical Problem

Existing technologies lack cross-modal semantic alignment in multimodal avionics fault diagnosis, and traditional knowledge distillation methods struggle to transfer cross-modal interactive knowledge, resulting in limited model performance when fusing visual and textual information.

Method used

A two-stage architecture of dual-encoder teacher model and lightweight student model is adopted. Combined with low-rank adapter and multi-level knowledge distillation strategy, the parameters are updated through low-rank decomposition to realize supervised adjustment of cross-modal knowledge and dynamic multimodal fusion.

Benefits of technology

While maintaining parameter efficiency, it significantly improves the cross-modal joint reasoning capability of the student model, ensuring the accuracy and reliability of avionics fault diagnosis, and is suitable for deployment of resource-constrained equipment.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of avionics fault diagnosis, and discloses an avionics fault diagnosis method based on multi-modal distillation and related equipment. Through a double-stage architecture design of a double-encoder teacher model and a lightweight student model, low-rank adapter injection and a multi-level knowledge distillation strategy are combined; the problems that an existing efficient parameter fine adjustment method is insufficient in cross-modal semantic alignment in a multi-modal diagnosis task, and cross-modal interaction knowledge is difficult to transmit in traditional knowledge distillation are effectively solved. Due to the introduction of a low-rank adapter, parameter efficiency is kept, and meanwhile, the model is assisted to accurately learn specific features of a multi-modal task; the multi-level knowledge distillation not only aligns model output, but also more fully migrates key knowledge such as cross-modal attention association and a multi-level feature fusion mode in the teacher model, and significantly improves the cross-modal joint reasoning ability of the student model.
Need to check novelty before this filing date? Find Prior Art