A Physics-based Data-driven Platform for Damage Accumulation by Multiple Parallel Damage Scenarios

Damage accumulation in polymeric composites is a menace to airframes components in aerospace and naval applications , due to constant vibration, long storage time in extreme condition, rapid shift of environment during operation, aging due to accumulated damage is  relevant. Often in composite systems, degradation-induced failure occurs due to damage accumulated from discrete sources such as thermal damage, oxidation, hydrolysis, and moisture that are combined with progressive damage mechanisms such as mechanical damages induced by vibration. Since most damage mechanisms that lead to degradation act in parallel, estimation of the lifetime of a polymeric material is not possible without a realistic representation of the history and thus, it becomes a significant reliability issue.  We propose a systematic approach for coupling of different damage models, the project aim to develop physics-based data-driven predictive tools that can use large data gathered on the mission/storage condition to describe damage accumulation in composites that are concurrently exposed to a combination of mechanical and  environmental stressors. In the  proposed framework, our goal is to predict the accumulation of the mechanical damages  that are induced by  (i) deformation  and (ii) vibration with environmental damages that are induced by  (iii) water attack by hydrolysis/hygrothermal aging , and (iv)   oxidation by isothermal aging.  In view of the complexity of the mapping space, no model exists that can even consider two of those phenomena concurrently and that’s the main advantage of our proposed technology. Using the modular platform developed by MSU, this project provides software to describe and couples the effects of decay due to continuous transient environmental/mechanical loads on composites.  The proposed hybrid framework is a more rigorous, history-dependent, extensible based on location and environment, and implementable (based on continuum mechanics) approach and portends a transformational change in damage predictive approaches. the software  can provide component specific prediction by using using real-time data provided by airframes sensors throughout the missions and storage and also environmental history of storage location. The final models will be coupled into the platform and  two predictive tools will be developed based on that to describe damage accumulation and provide virtual test results for the composites. The model will be implemented and commercialized as Ansys User-programmable Feature. To show the feasibility of such platform, in Phase I we develop compatible models of the aforementioned four phenomena and implement them in the platform to validate them against experimental data. Model used in the platform can be updated/modified or replaced for any compound, and only onetime training is needed once the platform is assembled.