Abstract:Together with stringent emission standards, meeting fuel consumption, and performance targets have made engine calibration process complicated, lengthy, and costly. For this reason, it is crucial to develop an efficient methodology for the optimization of a diesel engine calibration. In this study, a novel automated fine tuning methodology using genetic algorithms and data driven models is proposed. Fuel consumption of a light commercial vehicle with 2.0-L diesel engine was optimized for a predefined Real Driving Emissions (RDE) cycle. During the optimization, tail pipe emission limits were maintained by taking the aftertreatment system efficiency into account. To this end, a representative engine control unit, internal combustion engine, and Selective Catalyst Reduction (SCR) system models were built by using empirical data and neural networks. Moreover, the effect of choosing different tailored optimization points for a predefined route was investigated by comparison of frequency and fuel consumption based weighting. The cost function, related to the optimization problem, was defined based on fuel consumption, performance, and mechanical limits of the engine, and in accordance with tailpipe emission limits defined by Euro 6d-Temp regulations. Compared to previous model-based calibration studies, tailpipe emissions, and regulation limits have been incorporated in this study in order to obtain more realistic calibrations. After testing different genetic algorithm configurations, the fuel efficiency over the predefined route was improved by 3.07% without violating any of the limits while the smoothness of the calibration maps was also preserved. This route specific calibration methodology is believed to have the potential to increase fuel economy in real world applications such as daily commuting routes or repetitive driving patterns for instance, operation vehicle use cases in factories, airports, and construction areas. As for further development, this can be achieved by running simulations and the optimization on cloud and updating the calibration remotely based on the selected route before the trip.

Vehicle Fuel Optimization Under Real-World Driving Conditions: An Explainable Artificial Intelligence Approach

Interpretable Machine Learning Models for Predicting and Explaining Vehicle Fuel Consumption Anomalies

Investigation of an efficient model-based fuel economy optimization methodology for diesel engines focusing on real world driving emission cycles

Random Forest Machine Learning Model for Predicting Combustion Feedback Information of a Natural Gas Spark Ignition Engine

Innovative composite machine learning approach for biodiesel production in public vehicles

Machine Learning for Real-Time Fuel Consumption Prediction and Driving Profile Classification Based on ECU Data

Determination of Energy Savings via Fuel Consumption Estimation with Machine Learning Methods and Rule-Based Control Methods Developed for Experimental Data of Hybrid Electric Vehicles

Fuel Consumption Prediction for a Passenger Ferry using Machine Learning and In-service Data: A Comparative Study

Real-time vehicular fuel consumption estimation using machine learning and on-board diagnostics data

Fuel Efficiency Modeling and Prediction for Automotive Vehicles: A Data-Driven Approach

Measuring fuel consumption in vehicle routing: new estimation models using supervised learning

A Comparative Study on Fuel Consumption Prediction Methods of Heavy-Duty Diesel Trucks Considering 21 Influencing Factors

An Artificial Neural Network Model to Predict Efficiency and Emissions of a Gasoline Engine

Efficient and Interpretable Traffic Destination Prediction using Explainable Boosting Machines

Model predictive eco-driving control for heavy-duty trucks using Branch and Bound optimization

A Case Study on Predicting Fuel Consumption in VLGC Ship Propulsion

Toward Sustainable Mobility: AI-Enabled Automated Refueling for Fuel Cell Electric Vehicles

Analyzing Vehicle Fuel Saving Opportunities through Intelligent Driver Feedback

Personalized Driving Behaviors and Fuel Economy Over Realistic Commute Traffic: Modeling, Correlation, and Prediction

Validation of Ecology and Energy Parameters of Diesel Exhausts Using Different Fuel Mixtures, Consisting of Hydrogenated Vegetable Oil and Diesel Fuels, Presented at Real Market: Approaches Using Artificial Neural Network for Large-Scale Predictions

Integrating Machine Learning for Predicting Internal Combustion Engine Performance and Segment-Based CO2 Emissions Across Urban and Rural Settings