Prediction of journey characteristics for the intelligent control of a hybrid electric vehicle
UNSPECIFIED (1998) Prediction of journey characteristics for the intelligent control of a hybrid electric vehicle. In: IFAC Workshop on Intelligent Components for Vehicles, SEVILLE, SPAIN, MAR 23-24, 1998. Published in: INTELLIGENT COMPONENTS FOR VEHICLES pp. 141-146.Full text not available from this repository.
A hybrid electric vehicle is one which utilises both an internal combustion engine and electric motor for propulsion. This combination of power sources make its optimal control difficult. To enable optimised control of a hybrid electric powertrain it is desirable to have a priori knowledge of the characteristics for a given journey. Unfortunately this information is only known upon the completion of the journey and therefore it has to be intelligently estimated. This paper presents work that explores the possibility of prediction of Journey characteristics upon Journey departure. Usage data was collected from a number of vehicles with different usage characteristics over a period of one month each. The journey distance and duration were derived from this data. After ascertaining that there were predictable patterns inherent within the usage data, a fuzzy modelling approach was used to automatically generate rules for the prediction of journey distance and duration from departure time only. One conclusion drawn rs that It is possible to predict the journey characteristics for some users whereas others prove more problematic. Copyright (C) 1998 IFAC.
|Item Type:||Conference Item (UNSPECIFIED)|
|Subjects:||T Technology > TL Motor vehicles. Aeronautics. Astronautics
H Social Sciences > HE Transportation and Communications
|Journal or Publication Title:||INTELLIGENT COMPONENTS FOR VEHICLES|
|Publisher:||PERGAMON PRESS LTD|
|Number of Pages:||6|
|Page Range:||pp. 141-146|
|Title of Event:||IFAC Workshop on Intelligent Components for Vehicles|
|Location of Event:||SEVILLE, SPAIN|
|Date(s) of Event:||MAR 23-24, 1998|
Actions (login required)