Towards Fast Numerical Studies for Maintenance and Lifecycle Problem Analysis: New Simulation Methods on Example of an Industrial Study Case

Abstract

Companies exploiting industrial installations are currently submitted to various constraints in the context of maintenance and lifecycle problem analysis: time and cost of production stops, efficiency of maintenance solutions, production safety criteria, etc. Hence, the reduction of the time of projects may lead to important research perspectives in terms of fast numerical prototyping and solution assessment methods. In this paper, we present a typical industrial problem of fast assessment of mechanical state of a production installation. We focused on the study of a cracked component of a reheater as well as we proposed a new design for this component to improve its mechanical behavior under an exceptional overload. In order to reduce the operational study time, the Extended Finite Element Method (X-FEM) has been used to model non-linear crack phenomenon without any human modification of a complex 3D model on the level of geometric model and mesh. An alternative design of a damageable element has been proposed to avoid all risk of possible cracking. This design task requires a mechanical analysis of locally modified structure under a cyclic loading. We also present a new prototyping method based on direct modification of meshes and allowing reducing the time of numerical study.

Direct Modification of Semantically-Enriched Finite Element Meshes

“…Numerical crack introduction and propagation schemes were augmented in an elegant manner with a X-FEM (eXtended Finite Element) method 4,5,6 . The use of special enrichment functions as well as a discontinuous function along the sides of the crack allows one to achieve "virtually" a complete crack analysis (on mechanical computation level only using so-called "level sets") without any geometric mesh modification 4 . These works aim at predicting crack behaviour and not to insert a specific crack shape on the mesh without using CAD model.…”

Methodology for automatic recovering of 3D partitions from unstitched faces of non-manifold CAD models

“…The reducing of CAD modelling time is particularly relevant for fast studies applied to maintenance/life cycle problems where it is critical to provide quickly the optimised solution and to ensure its effectiveness [4,5]. In the context of industrial maintenance, the product is already designed, and companies currently practice the re-use of existing non-manifold CAD models via exchange data formats to accelerate the study/improvement of the product during its exploitation.…”