No abstract

No abstract

In this study we present a procedure for the design and implementation of a control strategy to optimize energy use within a light weight hydraulic hybrid passenger vehicle. The hydraulic hybrid utilizes a high pressure accumulator for energy storage which has superior power density than conventional battery technology. This makes fluid power attractive for urban driving applications in which there are frequent starts and stops and large startup power demands. A dynamic model of a series hydraulic hybrid powertrain is presented along with the design of a model predictive control based energy management strategy. Model predictive control was chosen for this study because it uses no future information about the drive cycle in its design. This increases the flexibility of the controller allowing it to be directly applied to a variety of drive cycles. Using the model predictive framework, a holistic view of the powertrain was taken in the design of the control strategy, and the impact of each actuator’s efficiency on overall efficiency was evaluated. A hardware-in-the-loop experiment using an electro-hydraulic powertrain testbed was then used to validate the dynamic model and control performance. Through a simulation study in which each actuator’s efficiency was given varying levels of priority in the objective function, it was found that overall system efficiency could be improved by allowing for small sacrifices in individual component performance. In fact, the conventional wisdom of using the additional degrees of freedom within a hybrid powertrain to optimize engine efficiency was found to yield the lowest overall powertrain efficiency. In this work we present a rigorous framework for the design of an energy management strategy. The design method improves the powertrain’s operational efficiency by finding the best balance between optimizing individual component efficiencies. Furthermore, since the design of the control strategy is built upon an analysis of individual components, it can be readily extended to other architectures employing different actuators.

No abstract

The sensitivity of energy management strategies (EMS) with respect to variations in drive cycle and system parameters is considered. The design of three strategies is presented: rule-based, stochastic dynamic programming (SDP), and model predictive control (MPC). Each strategy is applied to a series hydraulic hybrid powertrain and validated experimentally using a hardware-in-the-loop system. A full factorial design of experiments (DOE) is conducted to evaluate the performance of these controllers under different urban and highway drive cycles as well as with enforced modeling errors. Through this study, it is observed that each EMS design method represents a different level of tradeoff between optimality and robustness based on how much knowledge of the system is assumed. This tradeoff is quantified by analyzing the standard deviation of system specific fuel consumption (SSFC) and root mean square (RMS) tracking error over the different simulation cases. This insight can then be used to motivate the choice of which control strategy to use based on the application. For example, a city bus travels a repeated route and that knowledge can be leveraged in the EMS design to improve performance. Through this study, it is demonstrated that there is not one EMS design method which is best suited for all applications but rather the underlying assumptions of the system and drive cycle must be carefully considered so that the most appropriate design method is chosen.