RESOURCE LIBRARY

Using modeFRONTIER to integrate GT-Valve train and GT-Power models for valve event optimization Cummins Engine, a leader in the manufacturing of diesel and natural gas-powered engines for a wide range of transportation and equipment purposes, has created a new power module ready to take on the stringent US - EPA regulations. It is significantly more compact and cost-effective than medium-speed engines at the same horsepower. It took 150 engineers to design it, and modeFRONTIER helped the High Horsepower group find the optimal valve timing, hence reducing fuel consumption. ## Challenge When designing piston engines, timing when opening and closing inlet and exhaust valves is a crucial parameter impacting the fuel consumption / power output ratio. Typically, delaying the Exhaust Valve Closing (EVC) and anticipating the Intake Valve Opening reduces Exhaust Gas Residuals, resulting in lower fuel consumption. Among the complex models composing the 16-cylinder engine, Cummins designers used GT-Suite (Valve Train and Power modules) to simulate valve event performance and dynamics. For optimal engine performance, valve timing and lift profile need to be perfected for given breathing configurations defined by engine speed, and valve and port geometry and performance. ## Solution Finding the optimal valve timing configuration required a two-step process; to start, a first workflow was created in mode FRONTIER and used to automate the calibration process. Valve train parameters were automatically adjusted with modeFRONTIER to calibrate the GT model and match measured push tube load. The second phase consisted in a second workflow, which was used to investigate the design space; initially with response surfaces and subsequently with the direct optimization algorithms NSGA and Hybrid - to find the best values for 12 output parameters measuring the exhaust and intake cam timing angles, the volumetric efficiency and the Brake Specific Fuel Consumption (BSFC)1. ## Benefits During both project phases, modeFRONTIER proved highly reliable for reducing design cycle time and improving the performance of the valve train system. From the outset “it helped drastically reduce the time taken for calibrating GT models” said Ambikapathy Naganathan (Structural and Dynamics Analysis Engineer at Cummins). “modeFRONTIER has an excellent capability for integrating with multiple GT models and post processing tools.” Continued Eng. Naganathan: “in fact it helped us link those GT models more efficiently and complement the in-house optimization tool, while at the same time maintaining concurrent use by different analysts in different locations.”

The optimization platform helped the BMW team create a reliable transient cooling system model When it comes to reducing fuel consumption, one effective strategy is to improve the performance of an engine’s transient cooling system. The Diesel Engine Development Department at BMW used modeFRONTIER to optimize engine parameters related to the cooling system and to support the validation steps for the newly updated thermal model, in this way determining an accurate framework for comparing and optimizing different cooling packages. Challenge In order to identify better configurations for the transient cooling system of the six cylinder/225kW diesel engine, the team of BMW engineers developed the air side and coolant circuit model using Kuli, supported also by 3D-CFD simulations. To validate the model, measurements were taken of two different driving cycles and load step on the test bench. After comparing the coolant heat balance for oil and water circuits in the vehicle and on the test bench, results showed significant differences, even for very similar engine operating points. In particular, the wind speed impacting the crankcase, the oil pan and the cylinder head beside the actual thermal conditions under the vehicle hood are difficult to precisely determine on the test bench, affecting the reliability of related transient cooling simulations. To increase confidence in the reliability of such simulations, the engine model was refined to reduce measurement divergence observed during validation tests. ## Solution The coolant circuit and the air path models represented in the engine model included two main groups of key parameters requiring enhancement: five heat transfer coefficients and four heat capacities. modeFRONTIER allowed the engineers to set up an effective optimization workflow that was capable of automatically interacting with the Kuli engine model and detecting the optimal configuration for the nine parameters. Günther Pessl, Head of Simulation at BMW says that “the easy-to-build integration between the two software enabled faster identification of the best heat transfer coefficients and thermal inertia in the engine analyzed.” When transferring the test bench model to the real vehicle model, some parameters indicated fluctuations during the validation cycles, especially oil temperatures which showed the biggest deviation during the hill climb. “Thanks to the optimization loop performed on the hill climb cycle, the engine parameters were refined, resulting in a significant improvement in accuracy measured on the Miramas BMW test track” says Günther Pessl. ## Benefits modeFRONTIER helped the BMW team create a reliable transient cooling system model, compliant with new testing guidelines and accurate enough to be reused for comparing and optimizing different cooling packages. The software contributed to a significant reduction in oil temperature and coolant side model errors and improved related temperature trends. Additionally, the automation capabilities of modeFRONTIER together with its powerful optimization algorithms, enabled automatic fine tuning of the parameters that supported and shortened the model validation steps.

Improvements in weight, aero-dynamics and brake distribution led to a lap time reduction ranging from 0.19 to 0.50 seconds Polestar Racing has been using modeFRONTIER in different steps of the development of the “Polestar Performance” concept. The software is used both as a tool for driving the aerodynamics, multibody and structural analysis process, and to improve the real lap time simulation. In 2012 Touring Car Team Association (TTA) championship, Polestar Racing won the first prize for the Drivers, Teams and Manufacturers categories with their innovative car design. ## Challenge The 2012 season of the Racing Elite League run by TTA introduced new rules leaving little room for engineering design changes. Therefore, the Polestar Racing vehicle model had to be modified without considering the chassis design parameters, which were previously the core of the optimization analysis. Design simulation acquired an even greater importance as the best combination of the Front-to-Rear weight, aerodynamics and brake distribution in less time (track testing was limited to three days) became the new target. ## Solution The simulation tools used - Adams.Car by MSC Software and the Lap Time Simulation (LTS) in-house code - were easily integrated into the modeFRONTIER optimization workflow, to optimally tune the Front-toRear weight distribution parameters. “Considering the limited amount of time we have had for testing in this championship” – says Per Blomberg, Manager of Chassis Development – “this kind of interrelated parameters study would not have been possible without the capability of modeFRONTIER to extend the simulation analysis ”. Improvements in weight, aerodynamics and brake distribution led to a lap time reduction ranging from 0.19 to 0.50 seconds. ## Benefits In only a few hours of simulation time modeFRONTIER piloted the evaluation of parameter combinations that might have never been tested, leading to enhanced configurations. The optimization framework initially set up for the TTA championship can be promptly adapted to the typical circuits of the STCC championship. Repeating the study considering different conditions is now “something that we can do even overnight on a weekend race , once we have completed the trial session and gathered some data about the track model”. Similarly, it is rather easy to run a new optimization cycle even with modified conditions, such as the driver, tires, and so forth: once the relevant parameters in the LTS vehicle model (e.g. engine or track model) are set, modeFRONTIER performs a robust sensitivity verification of the output (lap time) in relation to different vehicle parameters.