Every vehicle manufacturer in the transportation and equipment industries seeks reduction in development time and prototype costs. Boeing and Dassault Systèmes pioneered full vehicle modeling and design with the development of the 777 and 787 aircraft. The success of these programs demonstrates the feasibility of the approach for very complex vehicles and systems. In the long term, I believe the transportation and equipment industries will similarly migrate toward the dominant use of these techniques in vehicle development.
The topic of simulation can generate a wide variety of responses depending on who is speaking and their experience with the virtual development environment. Newcomers to the industry may rely heavily on simulation tools to defend the choices they make, while many veterans may dismiss simulation as foolhardy and dangerous until real hardware can be evaluated. The truth lies somewhere in between, and it depends entirely on the quality of the information captured in the models.
Modeling simple physics (e.g., acceleration of a vehicle) is straightforward, while accurately capturing the chemical kinetics and fuel-air mixing dynamics involved in the combustion of a hydrocarbon fuel inside an engine is still an ongoing topic of research.
For this reason, IAV believes in applying models only where confidence in the results is high while integrating physical hardware where that confidence is missing or where calculation speed is critical.
An example of this philosophy is a recent project completed by IAV where our experts married the simulation of vehicle components and behavior to an engine running on a dynamometer. Although the integration of simple vehicle models with engine dynamometer testing has been proven in the past, our engineers sought to expand this technique by incorporating more complex mechanical and hydraulic systems.
Working with an OEM partner, our experts identified a candidate machine that uses a hydraulic pump and valves to manipulate multibody booms and auxiliary systems. The challenge was to accurately simulate these systems in real time to recreate a torque demand on the engine dynamometer that matched recorded signals, especially engine speed, from the vehicle in the field. This required fast, accurate modeling software, detailed knowledge of the vehicle system, and an engine dynamometer with very low inertia.
Our engineers used LMS's AMESim to build the mechanical and hydraulic model of the vehicle and incorporate the load-sensing, compensation, and power management functions in cooperation with the OEM partner. The model was then validated with data recorded on the vehicle during typical drive cycles and maneuvers. Once the accuracy of the model was acceptable, the team then refined the model to run in real time. The final step in the setup was interfacing the model output data to the engine dynamometer controller to force the engine to recreate the conditions seen on the vehicle.
The result was a faithful re-creation of the vehicle in the engine test cell. This allows IAV to investigate alternate engine calibrations, control strategies, and hydraulic components to gauge their impact on the full vehicle system performance, all without building an expensive prototype vehicle. Other systems are easily integrated into the development platform and allow for expanded concept study.
For instance, aftertreatment systems are also highly complex and their performance is dependent upon the chemical composition of an engine’s exhaust gas. IAV employs tools like AxiSuite, a catalyst and substrate modeling software package, to investigate new concepts within this environment.
Data generated on a synthetic gas test bench is used to calibrate diesel oxidation catalysts, diesel particulate filter, and selective catalytic reduction catalysts to a sufficient level of accuracy over a broad range of inlet gas mixtures. These models can then be connected to the virtual vehicle environment and run in concert with the engine in the test cell. Real exhaust gas measurements are fed to the models to provide feedback on the feasibility of catalyst coatings, layout configurations, and component sizes. A manufacturer can gain confidence in proposed concepts without relying on prototype parts.
The key to bringing these tools together is an easy-to-use integration platform. Like many others, IAV relies upon Mathworks’ Matlab and Simulink software packages to act as the common hub for data communication. IAV’s VeLoDyn tool was developed in this environment and allows for the simple integration of commercial tools and home-brewed models. Because of the near universal acceptance of Matlab/Simulink, direct communication to a number of electronics hardware systems is guaranteed. This ensures that the virtual environment can be applied to common tools like HiL and SiL test benches from a variety of manufacturers.
By harnessing common tools, applying expert knowledge to advanced models, and employing physical hardware when model complexity reduces confidence, IAV expects to realize efficiencies of time and money in the traditional vehicle development process. The continuous improvement in processor power and the advances in basic science ensure that current complex systems will continue to make the jump into the virtual environment. The transportation and equipment industries are certain to benefit from simulation activity.
I expect to see continued growth in the area of simulation, as well as other areas, for IAV North America. Overall, we have quadrupled our sales over the past four years and tripled our workforce. As revenue continues to improve and services expand, the need for engineering and administrative talent is critical. This growth has influenced our decision to open a second office in Auburn Hills, MI., which reinforces our commitment to the importance of expanding and continuously adapting to the needs of customers.
Andy Ridgway, President, IAV Automotive Engineering Inc., wrote this article for SAE Off-Highway Engineering.