Data-driven Sensor Placement for Predictive Applications: A Correlation-Assisted Attribution Framework (CAAF)

TL;DR

This paper introduces CAAF, a novel framework combining clustering and feature attribution to improve sensor placement in complex systems, addressing correlation issues in input data.

Contribution

The paper presents a correlation-assisted attribution framework that enhances feature attribution for optimal sensor placement in correlated, real-world dynamical systems.

Findings

CAAF outperforms traditional methods in complex, nonlinear systems.

Clustering reduces redundancy and improves generalizability of sensor placement.

Effective in applications like structural health monitoring and turbulent flow estimation.

Abstract

Optimal sensor placement (OSP) is critical for efficient, accurate monitoring, control, and inference in complex physical systems. We propose a machine-learning-based feature attribution (FA) framework to identify OSP for target predictions. FA quantifies input contributions to a model output; however, it struggles with highly correlated input data often encountered in practical applications for OSP. To address this, we propose a Correlation-Assisted Attribution Framework (CAAF), which introduces a clustering step on the candidate sensor locations before performing FA to reduce redundancy and enhance generalizability. We first illustrate the core principles of the proposed framework through a series of validation cases, then demonstrate its effectiveness in realistic dynamical systems such as structural health monitoring, airfoil lift prediction, and wall-normal velocity estimation for turbulent channel flow. The results show that the CAAF outperforms alternative approaches that typically struggle due to the presence of nonlinear dynamics, chaotic behavior, and multi-scale interactions, and enables the effective application of FA for identifying OSP in real-world environments.