Optimizing engine efficiency: An artificial neural network approach for fuel consumption reduction through engine remapping Available to Purchase

The automotive sector is diligently working towards reducing fuel consumption and carbon emissions by enhancing engine efficiency. Engine remapping, the adjustment of electronic engine settings for better performance and fuel efficiency, is a significant technique in this context. However, manual engine remapping is complex and time-consuming, often not yielding the expected efficiency improvements. This study aims to harness ANN simulation for engine remapping to improve fuel efficiency, focusing on predicting ignition timing mapping via ANN modeling to achieve more efficient fuel consumption. Utilizing the TRAINGDA feed-forward backpropagation training method, an ANN model was developed. Tests were conducted on a chassis dynamometer to simulate highway driving conditions. Initial vehicle data was recorded, followed by fuel consumption tests at different speeds. This data served as input for the ANN program to predict optimal ignition timing values, which were then integrated into the ECU for further testing and evaluation. The results showed that the ANN-based method effectively reduced fuel consumption at speeds between 10 km/h to 40 km/h, with a retardation of ignition timing by 2˚ improving fuel efficiency compared to the standard map. However, at 50 km/h, the standard ignition timing data yielded optimal fuel consumption. A robust correlation was found between the predicted values from the ANN model and the experimental measurements, highlighting a significant relationship between ignition timing and vehicle speed. The study concludes that ANN simulation in engine remapping is a promising strategy to optimize fuel efficiency and improve overall engine performance. The findings underscore the potential of ANN modeling in reducing fuel consumption, suggesting avenues for further research in this domain.