Asosiy kontentga oʻtish
Maqola

Machine Learning for Fuel Consumption Prediction and Driving Profile Classification Based on ECU Data Next-Generation Engineering

Deepthi PM. NithyaP. MounikaK. MouryaP.Mano VardhanReddyChristo AnanthFaculty of Artificial Intelligence and Digital Technologies, Samarkand State University,UzbekistanYusupov Ozod Rabbimovich
2026en
ABI

Annotatsiya

The classification of driving profiles and predicting real-time fuel consumption have gained significant interest over the recent years since it determines environmental sustainability besides efficiency of the vehicle. The project objectives are to use machine learning techniques to classify different driving behaviour and calculate fuel consumption with the help of the information received by the engine control unit (ECU). Currently, the use of such techniques as XGBoost, SVR (Support Vector Regression), or Ridge Regression are used. The proposed system will enhance the effectiveness in prediction and profile classification since it will involve additional algorithms like the Random Forest, Logistic Regression and Adaboost. The fuel consumption patterns are categorized into five patterns Sporty, Eco, Calm, Normal, and Aggressive which form the driving habits of the car. This does not only help in the better understanding of driving habits but also helps in the creation of adaptive fuel efficiency strategies. The project will not only enhance the performance of the vehicles but also improve the environment with the use of more advanced machine learning procedures in Eco-Engineering Approaches to Reduce Carbon and Methane Emissions.

Hali tarjima qilinmagan

Identifikatorlar

Iqtiboslar va manbalar

0 ta iqtibos0 ta foydalanilgan manba