Energy saving neural network (NN) controlled flow-coupled hydraulic hybrid tricycle (HHT)
The paper proposes an energy saving, management and high level control of the HHT. The main components include gasoline engine. A multi-purpose shaft with varied thickness and two areas of external splines, couples both the rotational elements of the pumpmotor and the HHTs front axle. The pump-motor is attached to the front wheel hub and the reservoir (accumulator) in the central part of the HHT. During braking, the pump forces the hydraulic fluid out of a low pressure tank into the double high pressure. Conversely, the fluid is forced back to the low pressure tank via the pump/motor which applies torque to the wheels during acceleration. The regenerative braking energy which is saved to the accumulator is as much as 75-80％ and is fed back to the wheels during the next system operation. This proposed system is expected to improve the fuel economy and emission to about 30 - 45％. A multi-lawyer feed forward NN with back propagation training controls the system active power flow to the load and makes the outlet flow from pump constant regardless of the speed of the wheel. Experimental results proved its effectiveness in energy saving.
Energy Saving Pump/motor Neural network
Celestine Okoye Charles C Achebe Jihai Jiang
Department of Mechanical Engineering, Nnamdi Azikiwe University, Awka, Anambra State/Nigeria Department of Mechatronics Engineering, Harbin Institute of Technology, P. R China
国际会议
杭州
英文
920-923
2009-04-08（万方平台首次上网日期，不代表论文的发表时间）