YOUNG'S MODULUS ESTIMATION AND LOAD IDENTIFICATION USING VISION AND FEM BASED PARTICLE FILTERING
Marcin Tekieli, Marek Słoński · 2013
This paper presents an example of a vision-based application of particle filtering (PF) [1] and finite element method (FEM) to two identification problems for a laboratory aluminum frame, shown in Fig. 1. In the first problem, we have successfully estimated the elastic modulus of the frame material. In the second problem, for known elastic constants, we have managed to identify the position and the magnitude of a quasi-static concentrated load. In both problems, the solutions are based on the displacement field which is obtained with a digital camera, computer vision techniques and digital image correlation method (DIC), see Fig. 1. The main element of the system is the algorithm responsible for displacement measurements and material or load parameters estimation. The first phase of the algorithm is the full-field displacement measurement using PF for markers detection and DIC for markers tracking. The aluminum frame was loaded by application of a quasi-static concentrated force. This phase is described in more details in [2] and the results are shown in Fig. 2-A. In the second stage, FEM and PF are used together to determine the Young’smodulusEalum. A single particle represents a particular value of Ealum. We have tested few FEM models with different number of FEs (see an example on the right-hand side of Fig. 2-C). The tests have shown that, in case of the loading force applied in the middle of the beam, it is enough to discretize the beam and the columns with only two FEs, respectively. The number of FEs should be increased when the force is applied in different points i.e. model nodes.