Back to Search
Start Over
Driving Style Recognition Model Based on NEV High-Frequency Big Data and Joint Distribution Feature Parameters
- Source :
- World Electric Vehicle Journal, Volume 12, Issue 3, World Electric Vehicle Journal, Vol 12, Iss 142, p 142 (2021)
- Publication Year :
- 2021
- Publisher :
- MDPI AG, 2021.
-
Abstract
- With the promotion and financial subsidies of the new energy vehicle (NEV), the NEV industry of China has developed rapidly in recent years. However, compared with traditional fuel vehicles, the technological maturity of the NEV is still insufficient, and there are still many problems that need to be solved in the R&amp<br />D and operation stages. Among them, energy consumption and driving range are particularly concerning, and are closely related to the driving style of the driver. Therefore, the accurate identification of the driving style can provide support for the research of energy consumption. Based on the NEV high-frequency big data collected by the vehicle-mounted terminal, we extract the feature parameter set that can reflect the precise spatiotemporal changes in driving behavior, use the principal component analysis method (PCA) to optimize the feature parameter set, realize the automatic driving style classification using a K-means algorithm, and build a driving style recognition model through a neural network algorithm. The result of this paper shows that the model can automatically classify driving styles based on the actual driving data of NEV users, and that the recognition accuracy can reach 96.8%. The research on driving style recognition in this paper has a certain reference value for the development and upgrade of NEV products and the improvement of safety.
- Subjects :
- neural network
Computer science
Big data
Machine learning
computer.software_genre
high-frequency bid data
new energy vehicle
Set (abstract data type)
joint distribution characteristics
Feature (machine learning)
Driving range
TA1001-1280
Artificial neural network
driving style
business.industry
k-means clustering
Energy consumption
K-means algorithm
TK1-9971
Transportation engineering
Identification (information)
Automotive Engineering
Electrical engineering. Electronics. Nuclear engineering
Artificial intelligence
business
computer
Subjects
Details
- ISSN :
- 20326653
- Volume :
- 12
- Database :
- OpenAIRE
- Journal :
- World Electric Vehicle Journal
- Accession number :
- edsair.doi.dedup.....b9e5cd621e72c93ac0b7e6402235e6af