Neural Network Based DC/DC PID Buck Control of Real Driving Behavior Shandong Jiaotong University, China

Format:

Book

Conference/Event

Author/Creator:

Gao, Yanfei, author.

Contributor:

Han, Guangde

Li, Aijuan

Wu, Qingsong

Yi, Fengyan

Zhang, Kesong

Conference Name:

3rd International Forum on Connected Automated Vehicle Highway System through the China Highway & Transportation Society (2020-10-29 : Jinan, China)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2020

Summary:

As an important part of intelligent transportation system, car following behavior is a very important phenomenon in the process of motorcade driving. Vehicle following theory can reflect the driving behavior of single lane or mixed lane, and the best comfortable reaction of rear drivers caused by the speed change of the front vehicle in the driving team. However, in the process of vehicle intelligent control, it is also necessary to consider the working characteristics of vehicle parts subsystem. In order to achieve driver comfort and ensure that the vehicle is in the best functional state, the 48V Light-duty Hybrid System with low system cost, significant fuel saving effect and little change to the existing vehicle structure has great potential of energy saving and emission reduction with practical significance. The bidirectional DC / DC converter, as the core component of 48V hybrid electric vehicle battery management, ensures the safe operation of the whole vehicle system, and the importance of its performance not can be ignored. In this paper, a neural network based DC / DC PID buck control of 48V micro hybrid electric vehicle under intelligent transportation conditions is proposed. This method can make the DC / DC load meet the wide range of no-load to full load, ensure the voltage stability, do not damage the device, and make the hybrid low-voltage electrical system run normally. According to the dynamic model of DC/DC, the relationship database between duty cycle and output voltage is obtained, and the complex input-output model between duty cycle and output voltage is identified by a neural network. Then a PID feedback is used to reduce the influence of neural network prediction error. The results show that the effect of PID control based on neural network is better than that of pure PID control

Notes:

Vendor supplied data

Publisher Number:

2020-01-5185

Access Restriction:

Restricted for use by site license