Novel evaluation method of acoustic emission data based on statistical fiber bundle cells

In general, the aim of acoustic emission-assisted tensile tests of composite materials is to identify and characterize the damage and failure modes of the specimens. This paper presents a fiber-bundle-cells-based statistical model, which provides a possible solution to the problem of characterizing the mechanical and failure behavior of the material. The model, based on the results of mechanical tests and acoustic emission measurements, decomposes the measured acoustic emission event number and tensile force-load time processes into components corresponding to the different damage modes. The acoustic emission events belonging to different failure modes are described by inhomogeneous Poisson point processes, while failures are modeled with the breakage of fibers as elementary parts of the sample. Hence, damage modes can be characterized with the number fraction, and the tensile strength and signal energy distributions of the components. Moreover, the variation of the number fraction of the intact or damaged fibers as a function of the load time can be calculated and depicted as well. As reliability function or a kind of damage map, it reveals the mechanical load-bearing ability of the material tested. The applicability of the model is demonstrated by compact tension testing and the comparison of short and long glass fiber-reinforced VERTON PP sheets and injection molded wood fiber-reinforced PP composites.