Optimizing Engine Efficiency: An Artificial Neural Network Approach for Fuel Consumption Reduction through Engine Remapping
Agung Nugroho*, Randy Cahya Kurnianto, Tabah Priangkoso

Wahid Hasyim University

Abstract

In the pursuit of reducing fuel consumption and carbon emissions, the automotive industry has been exploring ways to enhance engine efficiency. One popular technique is engine remapping, which involves modifying electronic engine settings to optimize performance and fuel efficiency. However, manual engine remapping can be complex, time-consuming, and may not always yield the desired efficiency improvements. As a result, the use of Artificial Neural Network (ANN) simulation has emerged as a promising alternative for optimizing engine remapping. This paper presents a study on the use of ANN simulation in engine remapping to achieve more efficient fuel consumption. The research aims to optimize fuel efficiency by predicting ignition timing mapping using ANN modeling. The study utilizes the TRAINGDA feed-forward backpropagation training method to develop an ANN model and achieve a 10% increase in mileage compared to standard data. The research builds upon previous studies that have demonstrated the effectiveness of ANN in improving fuel efficiency and engine performance. The methodology involves conducting tests on a chassis dynamometer to simulate highway driving conditions. The initial vehicle data is recorded, and fuel consumption tests are performed at various speeds. The fuel consumption results are then used as input data for the ANN program, which predicts optimal ignition timing values. The resulting ignition timing map is incorporated into the engine control unit (ECU) for further testing and evaluation. The study^s results indicate that the ANN method effectively reduces fuel consumption at speeds ranging from 10 km/h to 40 km/h. By retarding the ignition timing by 2&#730-, the fuel efficiency is improved compared to the standard map. However, at a speed of 50 km/h, the standard ignition timing data is found to yield optimal fuel consumption. The analysis demonstrates a strong correlation between predicted values from the ANN model and experimental measurements, as well as a significant relationship between ignition timing and vehicle speed. In conclusion, the use of ANN simulation in engine remapping offers a promising approach to optimize fuel efficiency and improve overall engine performance. The study highlights the potential benefits of ANN modeling in achieving fuel consumption reduction and suggests avenues for further research in this field.

Keywords: Engine efficiency, Fuel consumption, Engine remapping, Artificial Neural Network (ANN), Ignition timing, Chassis dynamometer.

Topic: Artificial Intelligence and Data Science