Quantification, localization, and reconstruction of impact force on interval composite structures

Impact force identification has been intensively studied owing to its profound effects on health monitoring and safety assessment. This study investigates a novel uncertainty-oriented impact force identification framework for composite structures considering signal noises, material dispersion, and fabrication errors. The quantification of multi-point impact forces is achieved by the principal component analysis of the modal forces, which are deduced by the Newmark difference scheme using acceleration measurements. The localization and reconstruction of impact forces are obtained by a series of interactive iterations containing time-domain inversion of impact forces and correlation assessment of simulated/measured responses. Particularly, the partition optimization and polynomial fitting strategies are presented to alleviate the computation burden of impact localization and an optimum sensor allocation method based on two derived condition numbers and an original convergence criterion is developed to improve the identification precision of force reconstruction. Moreover, interval modeling and the first-order Talor series expansion are introduced for uncertainty quantification and propagation. Eventually, two numerical examples are discussed to demonstrate the validity and feasibility of the developed approach. The results suggest its promising applications in complicated engineering structures and various loading conditions