Sensor placement on slender structural systems

Abstract

There have been many interesting technical ideas and parallels in thinking that have resulted from the technical discussions that I have had with Professor Jerome J. Connor since 1986. This brief article provides a context and reflects upon his mentorship, which helped shape the thinking of a then young professor from another University, and on the impact that Professor Connor has had on his students and others beyond the halls of MIT's Building 1. The optimization topic presented here provides an example of that influence. Extracting the desired response information from an instrumented slender structure, while minimizing the number of sensors, is a challenging problem requiring well-defined objectives that can be used in an optimization process. In this study, a methodology that builds upon a Genetic Algorithm optimization procedure is used to investigate sensor placement needed to recover specific vibration modes. Data recorded from an experiment investigating the flow-induced vibration of a smooth horizontally towed cylinder is used to explore the optimization process and subtleties associated with its application subject to single or multiple objectives and gaps in sensor data due to several possible constraints. The use of the Paterno Front Method and the difficulty in accurately capturing higher modes are addressed.