Markov Chain-Based Approach for Drive Cycle Generation in Electric Motorcycles

Electric Motorcycles are increasingly seen as a key solution for sustainable mobility, necessitating accurate evaluation of their energy consumption and performance under realistic driving conditions. Traditional standardized drive cycles, such as the Worldwide Motorcycle Test Cycle (WMTC) and the Indian Drive Cycle (IDC), often fail to capture the variability and stochastic nature of real-world driving. This discrepancy can lead to inaccurate estimations of energy consumption, range, and powertrain efficiency, affecting the reliability of EV simulations and development. To address these limitations, this study proposes a Markov chain-based methodology for generating a stochastic drive cycle that better represents real-world driving conditions. The approach leverages real-world driving data collected from multiple electric two-wheelers, constructing a transition probability matrix (TPM) to model speed and acceleration variations probabilistically. The generated drive cycle undergoes statistical validation, ensuring its consistency with observed real-world data. Key comparisons with WMTC and IDC are conducted to evaluate the accuracy of the proposed cycle in reflecting real-world driving patterns. Results indicate that the Markov-based drive cycle closely matches real-world driving statistics. The proposed methodology enhances EV performance evaluation, improving battery range estimation, power management strategies, and regenerative braking assessments. By providing a more reliable and adaptable approach to drive cycle generation, this study contributes to the development of more efficient and accurate EV testing frameworks.