Abstract
Abstract
To enable maximized use of electrical energy during the control of a hybrid electric power train, it is desirable to have a priori knowledge of the characteristics for a given journey. Unfortunately this information is only known on the completion of the journey and therefore at best can only be intelligently estimated. This paper presents work that explores the possibility of prediction of journey distance and duration on journey departure. Usage data were collected from a number of vehicles with different usage characteristics over a period of 1 month each. A preliminary visual analysis of the data demonstrated that there were predictable patterns inherent within the usage data which allowed simple rules to be developed heuristically. A fuzzy modelling approach was then used to generate automatically rules for the prediction of journey distance and duration from departure time only. This approach showed that there were three classifications of vehicle user among those logged: one with reasonable success, while the others required further information to enable effective prediction.
Get full access to this article
View all access options for this article.