In this paper, a Modelica library of electrified powertrain components is presented and its applications discussed. This library is used to construct digital twins of electrified powertrains during product development. These digital twins provide value by reducing development time and cost, while once the product is in-service, they enable improved condition monitoring. The library includes a multi-fidelity and multi-scale battery and power electronics sub-library, an Electrical Drive Unit (EDU) sub-library modelling different types of electrical machines, and an electrified propulsion system sub-library of template models that leverage the battery, power electronics and EDU components found in the other sublibraries. Finally, an example of applying the proposed library to electrified vehicle development is presented.

<div class="section abstract"><div class="htmlview paragraph">Stringent automotive legislation is driving requirements for increasingly complex battery pack solutions. The key challenges for battery pack development drive cost and performance optimisation, growth in architecture solutions, monitoring and safety through lifetime, and faster-to-market expectations.</div><div class="htmlview paragraph">The battery Virtual Product Development (VPD) toolchain addresses these challenges and provides a solution to reduce the battery pack development time, cost and risk. The battery VPD toolchain is built upon scalable, validated sub-models of the battery pack that capture the interactions between the various domains; mechanical, electrical, thermal and hydraulic. The model fidelity can be selected at each stage of the design process allowing the right amount of detail, and available data, to be incorporated. The toolchain is coupled with vehicle simulation tools to rapidly assess performance of the complete electrified powertrain.</div><div class="htmlview paragraph">The aim of this study is to demonstrate an agile approach to battery pack concept development using VPD, enabled by Design of Experiments (DoE) and optimisation. Key battery parameters such as cell type, electrical configuration, thermal heat path design and cooling strategy are chosen as the design variables of a multi-staged DoE. The DoE test matrices of these parameters are generated and imported into the battery VPD toolchain with the vehicle simulation model to perform the energy efficiency and performance simulations.</div><div class="htmlview paragraph">Finally, the simulation results are analysed to create surrogate models which can be used to predict powertrain attributes and optimise the battery pack design. The ability to front-load virtual battery pack concepts with vehicle simulations allow for wholistic performance assessment, ensuring that vehicle attribute targets such as pure electric range of WLTP, acceleration and maximum speed are met and reducing concept development time by up to 50% compared to the baseline approach.</div></div>