Abstract
Despite long-haul trucks becoming more fuel-efficient, their share in CO2 production continues to rise due to increased transport demand and decreasing CO2 emissions in other sectors. Possible solutions to this issue include hybridization and/or increasing the transport capacity of long-haul truck combinations, both of which improve fuel efficiency. One attractive vehicle combination is the A-double High Capacity Vehicle (HCV), which consists of a tractor, two semitrailers, and a dolly. Since conventional semitrailers can be used in this combination with only minor adjustments, investment costs can be kept low. Furthermore, the vehicle combination can be easily split into conventional vehicles, improving flexibility.In the A-double configuration, the dolly is attached to the first semitrailer via a drawbar and connected to the last semitrailer via a fifth-wheel coupling. Currently, A-doubles (as well as other HCVs) are not allowed to drive cross-border on the European road network due to their increased vehicle length and regulatory issues. To recover braking energy and improve steering behavior, the A-double can be equipped with a steered and driven dolly.
Because maneuverability is significantly worse for the A-double compared to a Conventional Tractor-Semitrailer (CTS) combination, a path-following algorithm has been developed to make the second semitrailer follow the first semitrailer by controlling the steering angles of the dolly tires individually. Simulations with the TU/e Commercial Vehicle Library (CVL) show that the proposed path-following steering algorithm can reduce the tracking error to the centimeter range. The idea behind this algorithm is to predict the required steering angles based on zero side-slip of the dolly tires. Furthermore, the path-following algorithm improves high-speed stability, although stability remains worse than that of a conventional tractor-semitrailer combination.
If the dolly is equipped with an electric powertrain, it is possible to make the A-double behave similarly to a CTS combination at low velocity by minimizing drawbar forces through dolly propulsion. Additionally, a control algorithm is developed to generate a yaw moment to stabilize the vehicle combination at high speeds. Simulations with a CVL model show that high-speed stability can be improved but will still be worse than that of a CTS combination. Finally, constraints on motor power determine whether the vehicle can be sufficiently stabilized to prevent roll-over during an extreme single-lane change maneuver.
To determine the most profitable dolly design and minimum fuel consumption, scalable quasi-static models are created for the engine and electric powertrain. A forward simulation model is used to create feasible velocity setpoints and evaluate control strategies, whereas a backward model is used to determine the optimal control strategy for a given powertrain size. A nested optimization approach is employed, where the powertrain size is varied in the outer loop, and the control signal is optimized for each powertrain size using a Dynamic Programming (DP) algorithm.
Finally, a comparison is made with an Equivalent Consumption Minimization Strategy (ECMS), which shows that ECMS works almost optimally when the co-state can be accurately determined for the cycle. The optimal control strategy can be identified as an assist-only strategy for the standardized CO2-cycle used for calculating fuel consumption.
| Date of Award | 2017 |
|---|---|
| Original language | English |
| Sponsors | TNO Location Helmond |
| Supervisor | Henk Nijmeijer (Supervisor 1), Steven Wilkins (Supervisor 1), Igo J.M. Besselink (Supervisor 1) & Theo Hofman (Supervisor 1) |