- 10:36 AM
636h

Integrated Process and Product Analysis for Automotive Paint Spray: a Multiscale Approach

Jia Li1, Jie Xiao2, Yinlun Huang1, and Helen H. Lou3. (1) Wayne State University, 5200 Anthony Wayne Dr. Apt#1103, Detroit, MI 48202, (2) Department of Chemical Engineering and Materials Science, Wayne State University, 5050 Anthony Wayne Drive, Detroit, MI 48202, (3) Department of Chemical Engineering, Lamar University, P.O.Box 10053, Beaumont, TX 77710

Paint spray is a key operational step in automotive surface coating, as this step largely determines the eventual coating quality, process efficiency, as well as environmental quality. In production, paint spray is controlled through process parameter adjustment based on the available macroscopic information in the length scale of 10-2 ~ 101 m and the time scale of 100 ~ 102 sec. It is recognized that even if such information is fully utilized, certain product quality issues may still not be addressable; the information at finer scales may be more crucial. However, the type of information is usually not accessible in operation, and its real value is yet to be understood.

In this paper, a multiscale system modeling method is introduced to characterize simultaneously product and process dynamic behavior. The characterization is established based on (i) systematic analysis of the multiscale aspects of paint spray, (ii) integrated process-product modeling for identification of key system parameters at multiple scales, and (iii) model-based applications for system estimation and improvement.

In systematic analysis, a graphic method is introduced to state the desired targets with their corresponding variables and mechanisms in a 2-D length and time scale map. This leads to the development of an integrated model that can describe the dynamics of paint-spray operation (e.g. booth air flow and electric field) and paint-film quality development (e.g. paint particle flow, deposition, and wet film leveling) in depth and breadth. Furthermore, an information integration approach is presented that facilitate the use of information at the length scale of 101 m to 10-6 m, and the time scale of 102 sec to 10-5 sec. It gives rise to various in-depth system understandings that are not conceivable using mono-scale modeling methods. Model-based simulation reveals various novel opportunities for improvement of product design and quality, energy and material efficiencies, and environmental performance. Computational scale-up of product and process designs are also investigated and applications in automotive paint spray are also presented.