OPTIMAL FBG SENSORS LAYOUT FOR IN-FLIGHT ROTOR BLADE SHAPE AND LOADS ESTIMATION

This paper presents a virtual sensing framework for the real-time reconstruction of sectional loads in helicopter rotor blades, obtained through an intermediate reconstruction of the blade deformed shape from strain data from fiber Bragg grating sensors, processed through a modal-based approach combined with the structural dynamic equations, both derived from a blade FEM model. A bi-objective genetic algorithm is employed to optimize the number and spatial distribution of sensors, as well as the number of retained eigenmodes. The algorithm simultaneously maximizes the Fisher information factor and minimizes the condition number of the reconstruction matrix. The optimization is carried out for both non-rotating and rotating blade configurations, thereby isolating the influence of rotation-induced variations in the blade eigenmodes on the resulting sensor layouts. Although the optimal sensor plants differ substantially between the two cases, shape and load reconstruction performed under static and rotating load conditions show that the non-rotating sensor configuration can be directly employed even under rotating conditions, without requiring a dedicated optimization.