Implementing an Adaptive Algorithm for Hybrid EVs: Recognising Driving Patterns with Artificial Intelligence

¹ Assistant Professor, Department of Electronics, Sanskriti University, Mathura, Uttar Pradesh, India .
² Assistant Professor, Maharishi School of Engineering & Technology, Maharishi University of Information Technology, Uttar Pradesh, India .
³ Assistant Professor, Electrical Engineering, Vivekananda Global University, Jaipur, India .
⁴ Professor, Department of Computer Science and Information Technology, Jain (Deemed to be University), Bangalore, India .

^* Corresponding Author :rishisikka.ec@sanskriti.edu.in
^** ccnitj@gmail.com
^*** mansingh.meena@vgu.ac.in
^**** mn.nachappa@jainuniversity.ac.in

Abstract

This review article delves into the enhancement of fuel efficiency in hybrid electric vehicles (HEVs) through the use of adaptive algorithms for precise driving pattern recognition. The review explores studies that delve into two distinct methodologies. Firstly, a method utilising a Learning Vector Quantisation neural network is highlighted, which analyses six standard driving cycles. By employing micro-trip extraction and Principal Component Analysis, this method ensures a comprehensive training sample, subsequently simplifying the model and reducing data convergence time. Simulations reveal a significant reduction in sampling duration whilst maintaining satisfactory accuracy, leading to an 8% improvement in fuel economy when paired with a parallel hybrid vehicle model. Additionally, the article examines the Neural Network Fuzzy Energy Management Strategy (NNF-EMS), designed to address the adaptability constraints of traditional energy management strategies. Through neural network learning and parameter analysis, the NNF-EMS showcases enhanced adaptability and practicality across diverse driving cycles, underscoring the potential of artificial intelligence in HEV algorithm development..