Computational analysis of spray pre-treatment in automotive applications 2020-01-0479
The automotive coating industry consists of several processes targeting the reliability and longevity of the manufactured Body-In-White (BIW) with process optimization playing a key role. Pre-treatment of BIW is one of the important aspects and this involves processes in the paint shop and body-in-white shop. The relevance of cleaning every part of the BIW is well known in the industry, and we will focus on the spray wash processes. While the industry currently relies on experiences from previous designs and experimental observations from model studies, this drastically slows down process optimization for new car models. Recent developments in Computer Aided Engineering (CAE) industry has shown capability to perform reliable studies using computer models that speeds up processes. The current study focuses on the Computational Fluid Dynamic (CFD) evaluation of spray washing of a BIW using a meshless method known as Smoothed Particle Hydrodynamics (SPH).
The study specifically discusses simulation of a washing process, where a car BIW is moving through pre-treatment line where, specifically arranged set of nozzles are spraying water at a constant flow rate. The complexity of this problem is too much for a conventional grid-based solver, as it involves a solid geometry moving through a huge empty space with fluids being sprayed from the domain boundaries. This could end up with high cost of conducting simulation studies, especially considering the massive 3-dimensional grid. As a Lagrangian based fully meshless method, SPH is well suited for simulating fluid dynamic problems involving large fluid deformations and free-surface flows. The solver is based on a Predictive-Corrective Incompressible (PCISPH) formulation of SPH, which obtains instantaneous physical properties of the fluids and their impact on the solids. The mass-based domain discretisation ensures only lesser computational cost in domain partially filled with fluids. Additionally, algorithms implemented on Graphics Processing Unit (GPU) makes the simulations faster and increases the scope for scalability.
Muraleekrishnan Menon, Samiullah Baig, Kevin Verma