2018-01-0417 Published 0 3 Apr 2018 © 2018 SAE International. All Rights Reserved.System Design Model for Parallel Hybrid Powertrains using Design of Experiments Christian Schmidt AVL LIST GmbH Mario Hirz Graz University of TechnologyLaurent Allouchery and Stefan Lichtenegger AVL LIST GmbH Citation: Schmidt, C., Hirz, M., Allouchery, L., and Lichtenegger, S. “System Design Model for Parallel Hybrid Powertrains using Design of Experiments,” SAE Technical Paper 2018-01-0417, 2018, doi:10.4271/2018-01-0417. Abstract The paper focuses on an optimization methodology, which uses Design of Experiments (DoE) methods to define component parameters of parallel hybrid powertrains such as number of gears, transmission spread, gear ratios, progression factor, electric motor power, electric motor nominal speed, battery voltage and cell capacity. Target is to find the optimal configuration based on specific customer targets (e.g. fuel consumption, performance targets). In the method developed here, the hybrid drive train configuration and the combustion engine are considered as fixed components. The introduced methodology is able to reduce develop - ment time and to increase output quality of the early system definition phase. The output parameters are used as a first hint for subsequently performed detailed component development. The methodology integrates existing software tools like AVL CRUISE [ 5] and AVL CAMEO [1 ].The new approach of the present methodology includes specific integrated transmission and electric motor models in the vehicle simulation loop. This enables a calculation of appropriate efficiency maps, based on the defined parameters in every simulation loop. Characteristic parameters of the involved drive train components are delivered by integrated databases and in addition, a battery cell table of available products on the market is linked to the system. Restrictions for design space are technically and physi - cally based, as well as driven by benchmark data. Links between the parameters based on their dependencies to the targets, like maximum vehicle speed, improve the information quality output of the DoE model. An integrated plausibility check selects the valid combi - nations of drivetrain components based on vehicle targets. In a multidisciplinary optimization process, this procedure avoids results, which include optimum values for one param - eter (e.g. fuel consumption) but would lead to disadvantages in other criteria (e.g. performance). Introduction As global exhaust emission regulations are becoming increasingly strengthened, the focus of research and development in the automobile industry has strongly shifted towards alternative propulsion systems, e.g. hybrid technology. Furthermore, hybrid drives are considered as key technology on route to electric mobility. Many automobile manufacturers already offer hybrid models within their product line or have at least announced to do so in the near future. At AVL List GmbH [ 2], a method for optimizing a certain setting of hybrid drive systems has been developed [3]. A key element of this method is a DoE-model. Various test runs have shown that this model still has room for improvement - especially the low utilization of experimental plans is considered rather insufficient.Development Tool Chain A certain methodology needs capable tools, which allow to run several simulation loops stable and automatically. The present approach includes an integration of well-established simulation tools. AVL CAMEO [ 1] provides all necessary functionality to drive the overall methodology. CAMEO is a Design of Experiment managing software, mainly used as powertrain calibration tool. This powerful tool gives a window onto handling the complete calibration and sizing optimiza - tion process. CAMEO test and measure functionality commu - nicates with subsystem automation, controllers etc. to send parameters and receive measurements. iProcedures [ 3] is a self-adapting DoE techno