A Review of the Data-Driven Prediction Method of Vehicle Fuel Consumption

Abstract

Accurately and efficiently predicting the fuel consumption of vehicles is the key to improving their fuel economy. This paper provides a comprehensive review of data-driven fuel consumption prediction models. Firstly, by classifying and summarizing relevant data that affect fuel consumption, it was pointed out that commonly used data currently involve three aspects: vehicle performance, driving behavior, and driving environment. Then, from the model structure, the predictive energy and the characteristics of the traditional machine learning model (support vector machine, random forest), the neural network model (artificial neural network and deep neural network), and this paper point out that: (1) the prediction model of fuel consumption based on neural networks has a higher data processing ability, higher training speed, and stable prediction ability; (2) by combining the advantages of different models to build a hybrid model for fuel consumption prediction, the prediction accuracy of fuel consumption can be greatly improved; (3) when comparing the relevant indicts, both the neural network method and the hybrid model consistently exhibit a coefficient of determination above 0.90 and a root mean square error below 0.40. Finally, the summary and prospect analysis are given based on various models’ predictive performance and application status.

Strategies for Reducing Automobile Fuel Consumption

“…This review emphasizes the impact of driving behaviors, road conditions, and vehicle characteristics on fuel consumption. Zhao et al [14] present a review of data-driven methods for the forecasting of vehicle fuel consumption, in which they compare traditional machine learning methods with neural network methods. This study highlights the advantages of using hybrid predictive models and multivariate data fusion technology for enhanced prediction accuracy, underlining that it can be further increased by using big data and advanced computational methods.…”

Modeling Vehicle Fuel Consumption Using a Low-Cost OBD-II Interface

“…Zhao et al [29] conducted a study of methods to predict vehicle fuel consumption by means of algorithms and statistical models, including neural networks and traditional machine learning. According to the authors, the estimation model for fuel consumption based on neural networks had high accuracy only if the input data were sufficient.…”

Does Engine Oil Type Affect Fuel Consumption in Passenger Vehicles? A Two-Year Investigation

“…The fuel consumption of vehicles is assessed primarily through driving tests with specific driving cycles, e.g., NEDC [5]. The second method is to rely on a physical model built on the principle of vehicle dynamics [6]. Another way is to use the on-board diagnostic system (OBD), which is used to monitor the engine condition [7].…”

“…Considering individual engine systems, these losses can be divided into losses in the piston-ringscylinder assembly 45%, 30% in the bearings, 15% in the timing system, and 10% for pumping losses [23,28]. Identifying key elements where energy is lost allows for the introduction of new advanced technologies in engine design and its operation (e.g., reducing engine friction by means of surface texturing and coating [29,30], modifying the engine oil composition [31], downsizing [32,33], improving the efficiency of the engine's thermodynamic cycle [6]).…”