A convolutional neural network deep learning method for model class selection

TL;DR

This paper introduces a deep convolutional neural network method for model class selection that operates solely on response signals, enabling effective classification without full system identification, with applications in structural health monitoring.

Contribution

The paper presents a novel CNN-based approach for model class selection using response signals, enhanced by a physics-based Kalman filter, applicable to linear and nonlinear systems.

Findings

Accurately classifies model types with response signals alone.

Effective on both linear and nonlinear dynamic systems.

Applicable to 3D building finite element models.

Abstract

The response-only model class selection capability of a novel deep convolutional neural network method is examined herein in a simple, yet effective, manner. Specifically, the responses from a unique degree of freedom along with their class information train and validate a one-dimensional convolutional neural network. In doing so, the network selects the model class of new and unlabeled signals without the need of the system input information, or full system identification. An optional physics-based algorithm enhancement is also examined using the Kalman filter to fuse the system response signals using the kinematics constraints of the acceleration and displacement data. Importantly, the method is shown to select the model class in slight signal variations attributed to the damping behavior or hysteresis behavior on both linear and nonlinear dynamic systems, as well as on a 3D building finite element model, providing a powerful tool for structural health monitoring applications.