Refine
Year of publication
- 2019 (2)
Document Type
Language
- English (2)
Has Fulltext
- no (2)
Publication reviewed
- begutachtet (2)
Institute
Two-Dimensional ECMS for System Analysis of Hybrid Concepts featuring Two Electric Traction Motors
(2019)
This work is focusing on the extension of the well-known equivalent consumption minimization strategy (ECMS) for analysis of new hybrid electric vehicle (HEV) concepts featuring two traction motors. Given the contribution to advanced systems engineering, the control algorithm is derived under consideration of causality and practicality. Having more than one traction motor, a two-dimensional torque-split allowing full parallel operation of the HEV powertrain is required. Furthermore, this HEV enables series operation mode, and an example of implementation is given as well. Since we define the capability of comprehensive system analysis as a major attribute of the energy management, we abstain from restricting the solution space via heuristics e.g. mode selection in advance. This gives birth to affect the control behavior only by applying constraints directly to the control problem and thus enables analysis based on the respective (sub)optimal solutions. In this paper, a brief example of implementing so-called soft constraints for reducing the number of engine starts is given. The practicality of the proposed 2D-ECMS is proven within a simulation study and control behavior is analyzed. Additionally, different effects of variable constraints and parameter settings are investigated.
To find the optimal system-level design of hybrid electric vehicles (HEVs), component models are used in simulations to evaluate a large number of different designs within a high dimensional design space. As the electrical machine (EM) represents a key component of the HEV powertrain in terms of energy consumption, models require scalability and sufficient accuracy with manageable computational effort. This paper presents a novel approach for the development of scalable EM models based on Neural Networks (NN). The models are trained with data derived by a Finite Element Analysis (FEA) based scaling procedure and capable to represent the characteristics of a wide range of EM designs without the incorporation of further details. Once a model is trained, it can be directly used in system-level design optimization. The practicality of the model is proven within an exemplary simulation study and its goodness of fit to the training data is validated by a statistical analysis. This approach can help to reduce the computational effort of EM efficiency maps calculation, since only a small number of time-consuming FEA based scaling simulations must be performed prior to the optimization.