Fuel Economy Improvement of Urban Buses with Development of an Eco-Drive Scoring Algorithm Using Machine Learning

Eco-drive is a widely used concept. It can improve fuel economy for different driving behaviors such as vehicle acceleration or accelerator pedal operation, deceleration or coasting while slowing down, and gear shift timing difference. The feasibility of improving the fuel economy of urban buses by applying eco-drive was verified by analyzing data from drivers who achieved high fuel efficiencies in urban buses with a high frequency of acceleration/deceleration and frequent operation. The items that were monitored for eco-drive were: rapid take-off/acceleration/deceleration, accelerator pedal gradient, coasting rate, shift indicator violation, average engine speed, over speed, and gear shifting under low-end engine speed. The monitoring method for each monitored item was set up, and an index was produced using driving data. A fuel economy prediction model was created using machine learning to determine the contribution of each index to the fuel economy. Furthermore, the contribution of each monitoring item was analyzed using the prediction model explainer. Accordingly, points (defined as the eco-drive score) were allocated for each monitoring item. It was verified that this score can represent the eco-drive characteristics based on the relationship between the score and fuel economy. In addition, it resulted in an average annual fuel economy improvement of 12.1%.

Modelling and experimental validation of the performance of a digital displacement® hydraulic hybrid truck

The development, modelling and testing of a novel, fuel-efficient hydraulic hybrid light truck is reported. The vehicle used a Digital Displacement® pump/motor and a foam-filled hydraulic accumulator in parallel with the existing drivetrain to recover energy from vehicle braking and use this during acceleration. The pump/motor was also used to reduce gear-shift times. The paper describes the development of a mathematical vehicle model and the validation of this model against an extensive testing regime. In testing, the system improved the fuel economy of the vehicle by 23.5% over the JE05 midtown drive cycle. The validated mathematical model was then optimised and used to determine the maximum fuel economy improvement over the diesel baseline vehicle for two representative cycles (JE05 midtown and WLTP). It was found that the hybrid system can improve the fuel economy by 24%–43%, depending on the drive cycle. When this was combined with engine stop-start, the system improved the fuel economy of the vehicle by 29%–95%, depending on the drive cycle.

A Receding-Horizon Framework for Co-Optimizing the Velocity and Power-Split of Automated Plug-In Hybrid Electric Vehicles

This paper presents a control framework to co-optimize the velocity and power-split operation of a plug-in hybrid vehicle (PHEV) online in the presence of traffic constraints. The principal challenge in its online implementation lies in the conflict between the long control horizon required for global optimality and limits in available computational power. To resolve the conflict between the length of horizon and its computation complexity, we propose a receding-horizon strategy where co-states are used to approximate the future cost, helping to shorten the prediction horizon. In particular, we update the co-state using a nominal trajectory and the temporal-difference (TD) error based on co-state dynamics. Our simulation results demonstrate a 12% fuel economy improvement over the sequential/layered control strategy for a given driving scenario. Moreover, its real-time practicality is evidenced by a computation time per model predictive controller (MPC) step on average of around 80 ms within a 10 s prediction horizon.

Combustion optimization for fuel economy improvement of a dedicated range-extender engine

In this study, the combustion system of a dedicated range-extender engine was optimized based on a production engine for fuel economy improvement with the use of enhanced tumble flow, higher compression ratio, Atkinson cycle and exhaust gas recirculation (EGR). First, the shape of the intake port was optimized to improve in-cylinder tumble and turbulence for combustion enhancement. The computational fluid dynamics (CFD) results showed that compared to the original intake port, the peak tumble ratio during the compression stroke of the new port is improved by 74.0%, and the turbulent kinetic energy at the spark timing is increased by 33.0%, and the results were verified through the flow test bench experiment. The dyno experiment showed that, with the new intake port, the engine brake specific fuel consumption (BSFC) was improved for all test conditions. Then, the late intake valve closing (IVC) and a higher compression ratio were used in combination to adopt the Atkinson cycle. The IVC timing was set to 642° ATDC based on the preset power target. And the compression ratio was set to 12 to balance knock tendency and BSFC improvement. Finally, the cooled EGR was optimized to further suppress the knocking tendency to improve fuel consumption. The results showed that, with the cooling Strategy 2, the attainable maximum EGR ratio at 2400 rpm full load and 70 Nm conditions was increased, the spark timing could be significantly advanced, and the BSFC was improved. The improvement of BSFC is between 6 g/kW·h and 13 g/kW·h for the load range from 40 Nm to the full load. After the optimization, the minimum BSFC of the range-extender engine reaches 233 g/kW·h, while it is around 242 g/kW·h for the base engine. The operation area where fuel consumption is lower than 240 g/kW·h becomes much wider.