Optimal Placement of Sensors to Detect Delamination in Composite Beams

The paper describes an approach for the optimal placement of sensors in composite beam structures for online detection of damage. The ability to identify damage is based on establishing a mapping between the charactgeristics of specific damage mechanisms (location and extent) such as delamination, fiber breakage, and matrix cracking, and strain measurements at the selected sensor locations; a trained neural network is proposed as a tool to generate this mapping. The design problem considered in the present paper was to place the least number of sensors in the structure so that the ability of the neural network to predict the extent and location of damage is not compromised. The optimization problem involved a mix of discrete and integer variables, and a genetic algorithm was used as the search tool.