Search Results

In this paper a rule-based energy management for parallel hybrid electric vehicles (HEVs) is presented, which is based on the principles describing the optimal control behavior. Therefore we first show the general relations that can be used to describe the optimal limit of electric driving as well as the optimal torque split among the two propulsion systems. Subsequently these relations are employed to derive maps, which represent the optimal behavior depending on several input parameters. These maps are then used as inputs for the rules in the proposed energy management. This not only makes it possible to automatically calibrate the rule-based controller but also gives the optimal control in every driving situation. Given it is not fuel-efficient to turn the internal combustion engine (ICE) on or off for short intervals, it is further shown how this approach allows to adjust the established limit for electric driving by additional rules.

Hybrid-electric vehicles provide additional functionality compared to conventional vehicles. So-called ‘hybrid’ software functions are required to coordinate the conventional powertrain control and these additional control functions. A key factor to reduce the fuel consumption lies in optimal control of the entire interconnected powertrain. This paper aims to provide a framework for efficient interface definition, connection and coordination of control units for hybrid electric vehicles. Such a framework supports an efficient development of control unit architectures and the distribution of software functions. The generic approach necessitates modular software functions. It defines the distribution of these functions in control units optimized with respect to reuse, interfaces and compatibility with different powertrain topologies and electrification variants, especially also considering compatibility with a conventional powertrain and its electric hybridization.

The overall efficiency of hybrid electric vehicles largely depends on the design and application of its energy management system (EMS). Despite the load coordination when operating the system in a hybrid mode, the EMS accounts for state changes between the different driving modes. Whether a transition between pure electric driving and internal combustion engine (ICE) powered driving is beneficial depends, among others, on the respective operation point, the route ahead as well as on the energetic expense for the engine start itself. The latter results from a complex interaction of the powertrain components and has a tremendous impact on the efficiency and quality of EMSs. Optimization based methods such as dynamic programming serve as benchmark for the design process of rule based control strategies. In case no energetic expenses are assigned to a state change, the resulting EMS suffers from being sub-optimal regarding the fuel consumption.

In order to achieve high customer satisfaction and to avoid high warranty costs caused by component failures of the power-train of hybrid electric vehicles (HEV), car manufacturers have to optimize the dimensioning of these elements. Hence, it is obligatory for them to gain knowledge about the different types of vehicle usage being predominant all over the world. Therefore, in this paper we present a Data Mining system that employs a Random Forest (RF) based dissimilarity measure in the dimensionality reduction technique t-Distributed Stochastic Neighbor Embedding (t-SNE) to automatically identify and visualize different types of vehicle usage by applying these methods to aggregated logged on-board data, i.e., load spectrum data. This kind of data is calculated and recorded directly on the control units of the vehicles and consists of aggregated numerical data, like the histogram of the velocity signal or the traveled distance of a vehicle.

The reduction of fuel consumption as well as the rising demands of customers regarding a vehicle’s driving dynamic and the legislator’s continually rising demands are a current issue in vehicle development. Hybrid vehicles offer a possibility to rise to this challenge. Realistic driving cycles are of utmost importance for the calibration of a hybrid vehicle’s operational strategy. Deriving replacement speed cycles from extensive customer data sets seems to be an approach for solving these problems. The contribution at hand describes the derivation of replacement cycles by using stochastic models, probabilistic (weighted) drawings and a combinatorial optimisation. The novelty value is that the characteristic influences of all drivers are being considered in the generation due to the stochastic modelling.