Managing energy transition alongside environmental protection by making use of AI-led butanol powered SI engine optimization in compliance with SDGs

Muhammad Ali Ijaz Malik; Muhammad Usman; Muhammad Waqas Rafique; Sohaib Raza; Muhammad Wajid Saleem; Naseem Abbas; Uzair Sajjad; Khalid Hamid; Mohammad Rezaul Karim; Md Abul Kalam

doi:10.1016/j.heliyon.2024.e29698

Managing energy transition alongside environmental protection by making use of AI-led butanol powered SI engine optimization in compliance with SDGs

Heliyon. 2024 Apr 18;10(9):e29698. doi: 10.1016/j.heliyon.2024.e29698. eCollection 2024 May 15.

Affiliations

¹ School of Civil and Environmental Engineering, FEIT, University of Technology Sydney, NSW, 2007, Australia.
² Mechanical Engineering Department, Main Campus, University of Engineering and Technology Lahore, Pakistan.
³ Department of Mechanical and Energy Engineering, De Montfort University Dubai, United Arab Emirates.
⁴ Department of Mechanical Engineering, Sejong University, Gwangjin-gu, Seoul, 05006, South Korea.
⁵ Department of Energy and Refrigerating Air-Conditioning Engineering, National Taipei University of Technology, Taipei, 10608, Taiwan.
⁶ Department of Energy and Process Engineering, Norwegian University of Science and Technology, 7491, Trondheim, Norway.
⁷ Department of Mechanical Engineering, College of Engineering, King Saud University, Riyadh, 11421, Saudi Arabia.

Abstract

Enormous consumption of fossil fuel resources has risked energy accessibility in the upcoming years. The price fluctuation and depletion rate of fossil fuels instigate the urgent need for searching their reliable substitute. The current study tries to address these issues by presenting butanol as a replacement for gasoline in SI engines at various speeds and loading conditions. The emission and performance parameters were ascertained for eight distinct butanol-gasoline fuel blends. The oxygenated butanol substantially increases engine efficiency and boosts power with lower fuel consumption. The carbon emissions were also observed to be lower in comparison with gasoline. Furthermore, the Artificial Intelligence (AI) approach was used in predicting engine performance running on the butanol blends. The correlation coefficients for the data training, validation, and testing were found to be 0.99986, 0.99942, and 0.99872, respectively. It was confirmed that the ANN predicted results were in accordance with the established statistical criteria. ANN was paired with Response Surface Methodology (RSM) technique to comprehend the influence of the sole design parameters along with their statistical interactions controlling the responses. Similarly, the R² value of responses in case of RSM were close to unity and mean relative errors (MRE) were confined under specified range. A comparative study between ANN and RSM models unveiled that the ANN model should be preferred. Therefore, a joint utilization of the RSM and ANN can be more effective for reliable statistical interactions and predictions.

Keywords: Artificial intelligence; Butanol-gasoline blends; Engine performance; Optimization; Statistical approach.