Günter Prochart

Model predictive control of a battery emulator for testing of hybrid and electric powertrains

Thorough testing of the powertrain is a key aspect for the development of reliable hybrid and electric vehicles. Instead of a real traction battery, a battery emulator is used to supply the electric motor inverters of hybrid powertrains on a testbed. This approach avoids time-consuming preconditioning of batteries and allows automated testing of hybrid powertrains. An electric motor inverter acts as a constant power load towards the battery emulator. This degrades the dynamic performance and can even destabilize a supply with a conventional controller. The contributions are a model predictive controller design approach that includes a model of the system and the constant power load. A robustness concept is utilized in order to achieve stable operation. An algorithm for real-time execution of constrained model predictive control is proposed as well. Simulation results and experimental results are presented.

Battery impedance emulation for hybrid and electric powertrain testing

With battery emulators, development engineers can test electrical drives on testbeds without the need for costly battery prototypes. Furthermore, valuable time for battery pre-conditioning can be saved. However, high quality battery emulation is a challenge that requires a good battery model and accurate emulation of the model with a controllable power supply. A high-fidelity nonlinear battery model based on local model networks is used. This allows the extraction of local linear models for impedance emulation with a switch model DC-DC converter. Local linear battery models and a model of the converter are combined to a prediction model which is used for designing a model predictive controller. In contrast to existing impedance emulation schemes with cascaded feedback loops, the virtual impedance is included in the inner voltage control loop in order to achieve the highest possible bandwidth. Owing to constrained optimal control, this is achieved while respecting control variable constraints and inductor current safety limits. Experimental results show the effectiveness of the proposed battery impedance emulation scheme.

Predictive pulse pattern control for a synchronous multiphase buck converter

Finite control set model predictive control (FCS-MPC) has been proven to be a powerful control strategy in power electronics with very high dynamic performance compared to carrier-based pulse width modulation (PWM) approaches while avoiding unnecessarily high switching frequencies that would increase switching losses. However, the limited time resolution often leads to poor steady-state performance and can only be overcome by increasing the sampling rate, which would involve an infeasibly high computational demand. This paper presents a predictive pulse pattern control (PPPC) strategy for a synchronous multiphase buck converter (SMPB) that achieves both objectives yielding in high performance for transient and steady-state operating conditions with a feasible computational demand. At every time step, a pulse pattern with minimal switching effort is selected whose switching instants are obtained by minimizing an objective function yielding in high tracking accuracy. Simulation results demonstrate the potential of the proposed control strategy to outperform state-of-the art control approaches.

Datengetriebene Batteriemodelle zur Emulation des Impedanzverhaltens von Traktionsbatterien

ZusammenfassungIn dieser Arbeit wird der Einsatz von datengetriebenen Batteriemodellen und einer darauf aufbauenden modellprädiktiven Regelungsstrategie für die Emulation von Traktionsbatterien vorgestellt. Dieses Konzept ermöglicht effiziente und reproduzierbare Prüfstandsläufe von Antriebssträngen für Elektro- und Hybridfahrzeuge, indem das Verhalten der realen Batterie durch einen Batterieemulator nachgebildet wird. Somit können die verbliebenen Komponenten zeiteffizient und ohne Umrüstvorgänge unter verschiedensten Bedingungen (Ladezustand, Zelltyp, Konfiguration der Zellen etc.) getestet werden.Um das Verhalten der Traktionsbatterie auch bei hochdynamischer Anregung durch den Prüfling noch realitätsgetreu emulieren zu können, werden für dieses Konzept einerseits ein genaues mathematisches Batteriemodell, und andererseits eine hochdynamische Spannungsregelung benötigt. Ein rein datenbasierter Ansatz zur nichtlinearen Batteriemodellierung ermöglicht dabei die Anwendbarkeit für verschiedene Zellchemien. Der Reglerentwurf zur Impedanzemulation basiert auf einem modellprädiktiven Regelungskonzept und berücksichtigt dabei explizit die Dynamik des Batterieemulators, des Prüflings und die zu emulierende Batterieimpedanz.AbstractIn this paper, a data driven battery modelling approach and the associated model predictive control strategy for traction battery emulation is described. The proposed concept makes it possible to test powertrains of electric or hybrid vehicles under repeatable conditions. When the impedance behaviour of the real battery is reproduced by the battery emulator, the remaining components of the powertrain can be tested very efficiently and without hardware modifications under various conditions/settings (state of charge, cell type, cell configuration, etc.).In order to obtain a precise emulation of the battery impedance even when a highly dynamic excitation is applied by the unit under test, an accurate mathematical battery model and an effective voltage controller are required. A purely data driven modelling approach enables the application of the proposed concepts to different battery cell chemistries. The dynamic impedance emulation is based on a model predictive controller which is based on a model of the testbed setup including the virtual battery model.