Optimal sensor location and stress prediction on a plate using machine learning
Document Type
Conference Proceeding
Publication Date
1-19-2023
Publisher
American Institute of Aeronautics and Astronautics, Inc.
Abstract
The selection of the type and placement of sensors is crucial to successfully predicting the structural health of a system. This paper expands on previous work involving sensor placement selection for a simple beam to apply the same methodology to a more complex geometry. The sensor placement problem is solved using machine learning feature importance selection, in this case random forest regression. The features correspond to the signal measured by accelerometers or strain gauges, and the output corresponds to the maximum stress. Using a combined dataset combining multiple loading conditions, strain gauge and accelerometer placement can be determined for a 2D geometry. Once optimal sensor location has been determined, a neural network is trained to predict maximum stresses within the plate based on data recorded from these sensors.
Recommended Citation
Choppala, S., Kelmar, T. W., Chierichetti, M., Davoudi, F., & Huang, D. (2023). Optimal sensor location and stress prediction on a plate using machine learning. In AIAA SCITECH 2023 Forum. American Institute of Aeronautics and Astronautics. https://doi.org/10.2514/6.2023-0370
Comments
AIAA SCITECH 2023 Forum
23-27 January 2023
National Harbor, MD & Online