Vehicle fuel consumption prediction based on the data record obtained from an engine control unit

Šarkan, Branislav; Semanová, Štefánia; Harantová, Veronika; Stopka, Ondřej; Chovancová, Mária; Szala, Mirosław · 2019 · OpenAlex-citations

DOI: 10.1051/matecconf/201925206009

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the feasibility of predicting vehicle fuel consumption using data recorded from an engine control unit (ECU), specifically focusing on throttle position and engine speed. The research is motivated by the need for accurate, real-time methods to assess driving efficiency and driver behavior, which can support systems that warn drivers of inefficient driving styles. While fuel consumption is traditionally measured via laboratory tests or drive cycles, analyzing electronic parameters from the fuel mixture control system offers a complementary approach. The authors aim to determine the accuracy of such predictions and evaluate their potential utility for assessing economic driving practices. The experimental methodology involved laboratory measurements conducted at the University of Zilina using a MAHA MSR 1050 roller dynamometer. A Kia Ceed 1.6 CVVT equipped with a spark-ignition engine and multi-point indirect injection system served as the test vehicle. The study simulated the first 600 seconds of the Worldwide Harmonised Light Vehicles Test Procedure (WLTP) driving cycle to ensure consistent climatic conditions. Data acquisition was performed using the dynamometer’s control computer and universal OBD diagnostics via TouchScan software, recording engine speed, fuel consumption, throttle position, and vehicle speed at a frequency of approximately 3 Hz. This resulted in 1,800 data points for analysis. A flow meter with a variance of 0.5% was used to establish a baseline for actual fuel consumption, which averaged 8.35 l/100 km over the test distance. The results demonstrate strong correlations between the selected ECU parameters and fuel consumption. Analysis of engine speed revealed a significant power regression relationship, indicating that engine speed is a primary factor in fuel consumption. Similarly, throttle position showed a significant quadratic polynomial relationship with fuel consumption, particularly when idle engine speed values were excluded. The study further developed a 3D regression model combining both engine speed and throttle position. This linear polynomial model, derived using Microsoft Excel and Solver to minimize deviation from measured values, provided the most comprehensive prediction. When compared to the actual fuel consumption recorded by the flow meter, the 3D model achieved a prediction accuracy of 0.228%. The significance of this research lies in its validation of ECU-derived parameters as a reliable tool for estimating fuel consumption. The high accuracy of the combined model suggests that monitoring throttle position and engine speed can effectively assess the economic efficiency of a driver’s style. This approach supports the development of onboard systems capable of providing real-time feedback to drivers, potentially improving fuel economy and reducing emissions. The findings confirm that laboratory-based analysis of these electronic signals can serve as a precise complement to traditional fuel measurement methods, offering a viable pathway for integrating fuel efficiency monitoring into common vehicle systems.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• exposure measurement