Hardware-in-the-Loop estimation of kinetic energy loss in urban driving cycles using bond graph based unified modeling framework

Estimation of kinetic energy losses in urban driving cycle is a challenging task. Researchers uses mathematical or physical models while the industry uses test vehicle or high-fidelity test rigs. However, both have a common objective: reducing the gap between the model and the reality. Accurate estimation of energy losses is necessary to provide a better estimate of the amount of energy that can be harvested and to determine project feasibility. This paper presents an off-line Hardware-In-the-Loop (HIL) simulation wherein physical models and realtime data are used to estimate the kinetic energy losses in the urban driving cycle. A physical model of an drive-train of a passenger car is modeled using Bond Graph which also estimates the kinetic energy losses at the wheels. This model is further modified to include a modulated input with real-time urban driving cycle, thus forming a off-line HIL model. The real-time urban driving cycle is generated for representative road network using a passenger car fitted with a data acquisition unit. The physical model with modulated effort provides an estimate the amount of energy that can be gainfully harvested using Parallel Regenerative Braking System (PRBS). Using physical models and real data from actual vehicles in a low cost, general, off-line Hardware-in-the Loop (HIL) simulation, we show that designers can validate new automotive concepts even before prototypes become available while maintaining strict performance, quality requirements and time-to-market constraints.