Picture Exp_7 Picture Exp_9 Picture Exp_8 Picture Exp_10 Picture Exp_11 Picture Exp_13 Picture Exp_12

Automobile Light Design: Thermal Simulation of Lighting Systems

Motivation

The development of head and rear lamps is totally conditioned by the outer style surfaces. Often this forces lamp designers to work with reduced spaces. Heat dissipation is very difficult in these conditions and the plastic material may degrade, losing its optical and esthetic properties. Simulation is a powerful tool to test virtually the thermal and mechanical performance of lighting systems. The simulation process consists of several stages. Each one requires a lot of time to generate the models and to obtain results. The new procedure based on the CloudFlow portal cuts off the time requested in the whole process, saving at the same time and cost in license fees.

In addition to the time and cost/project reduction, the collaboration with the partner CSUC (HPC) has given to an SME like BTECHC the possibility to run bigger models improving the simulation quality and increasing the number of simulation rounds per project. This aspect is crucial to check more strategies for heat dissipation, to decide the convenience of using thermal shields and, to select materials and surface treatments properly.

Goals of the experiment

  • Improving the simulation process to speed up the product innovation in lighting systems. This means:
    • Optimized design in terms of the material thermal performance.
    • Optimized design in terms of the best strategy of thermal dissipation.
  • Increasing the number of simulation rounds per project.
  • Lowering costs due to savings in commercial license fees.

The new simulation methodology has been validated by comparison with previous simulated results and experimental measurements.

Technical impact

Any CFD software to be used in thermal simulation of lighting systems is very demanding in computing resources. Robustness and reliability are strongly dependent on the number of cells used in models. The use of high performance computing is very convenient to run this kind of simulation. Although the computational cost provided by HPC is quite reasonable, the license fee of commercial CFD software depends on the number of cores to be used in the simulation and makes it unaffordable for an SME like BTECHC. The CloudFlow experiment has enabled BTECHC the chance of working with OpenFoam (CFD open source) after enhancing its capabilities of modeling properly the radiation phenomena. It has been estimated savings of 80 percent in license fees and 38 percent in computing costs.

In addition to the cost reduction, the cloudified process reduces by 75 percent the workflow execution time due to the use of supercomputing and the automatization of several stages in only one single workflow. This fact enables BTECHC increasing the number of simulation rounds per project from 3 to 10 which allows better product optimization.

Another important advantage for BTECHC is being able to run the workflow via the CloudFlow portal from any place with Internet connection as well as using the remote post processing through the portal by means of a powerful graphical node. Now the simulation results can be discussed directly with the customer without investing time reporting or creating videos to support explanations.

Electronics Design Automation Electronics Design Automation

Economic impact

Every year the automotive industry starts near 500 new developments of head and rear lamps. It is estimated that the funds devoted to outsourcing CAD and CAE activities (BTECHC core business) are approximately 360 million euros worldwide. BTECH began supporting the automotive lighting industry in 2009. One of the BTECHC strategies to win market share is offering services not covered by the competitors. Thermal simulation, due to its complexity, is one of these services. Until now, the simulation service had two main drawbacks: cost and number of loops per project. The CloudFlow experiment is a solution for BTECHC clients who claimed for time and cost reduction.

Summing up all cost/project involved in the project execution, the total cost difference between the old and new procedure is approximately 76 percent. This huge cost reduction should lead BTECH to a better position to win market share as provider of engineering services.

Taking the development cost reduction and the competitive advantage through better products and a faster time-to market into account, it is estimated that BTECHC can increase their revenues in simulation services by about 240,000 euros over the next 3 years. It is expected the creation of 2 new jobs allocated to the CAE&CFD department for this period of time. It is estimated that synergies with the CAD team would increase the revenue in CAD services in two times reaching 1.2 million euros.

Regarding the CSUC activities as HPC, after the CloudFlow portal deployment CSUC users have easier ways to follow up their jobs. Before the experiment, a pay-per-use payment modality was in use and the value of each simulation was accounted by the CPU time. After the CloudFlow experiment other resource which do not consume much CPU time (such remote visualization) are properly accounted using the wall clock.

Addressing a customer segment defined by the CAE simulation sector in the European industry, in a three-year horizon CSUC estimates to face a global market size of around 100 customers, with a potential share reaching 5 percent and leading to incomes of 50,000 euros. With this new resources available CSUC plan to create 1 new job to support the new users and develop the existing infrastructure.

Another important advantage for BTECHC is being able to run the workflow via the CloudFlow portal from any place with Internet connection as well as using the remote post processing through the portal by means of a powerful graphical node. Now the simulation results can be discussed directly with the customer without investing time reporting or creating videos to support explanations.