A Supervised Tensor Dimension Reduction-Based Prognostics Model for Applications with Incomplete Imaging Data

TL;DR

This paper introduces a supervised tensor dimension reduction method for prognostics that effectively handles incomplete imaging data and leverages time-to-failure information to improve feature extraction for better prognostic accuracy.

Contribution

The proposed approach allows prognostics with incomplete tensor data and incorporates TTF supervision, which is a novel combination not present in existing models.

Findings

Handles incomplete tensor data effectively

Utilizes TTF for supervised feature extraction

Provides an optimization algorithm with closed-form solutions

Abstract

This paper proposes a supervised dimension reduction methodology for tensor data which has two advantages over most image-based prognostic models. First, the model does not require tensor data to be complete which expands its application to incomplete data. Second, it utilizes time-to-failure (TTF) to supervise the extraction of low-dimensional features which makes the extracted features more effective for the subsequent prognostic. Besides, an optimization algorithm is proposed for parameter estimation and closed-form solutions are derived under certain distributions.