High-Accuracy Freight Vehicle Highway Driving Cycle Construction Based on Improved Elitist Genetic Algorithm

To enable the precise evaluation of vehicle energy consumption and emissions under highway driving conditions, and to address the limitations of conventional methods, such as insufficient accuracy and low convergence efficiency, this study proposes a high-precision freight vehicle driving cycle development method based on an improved elitist genetic algorithm (IEGA). The core innovation of the proposed algorithm is the replacement of the inefficient single-point mutation operator in conventional genetic algorithms (GA) with a novel segment reconstruction algorithm based on the Markov reachable domain. Furthermore, a two-stage elitist selection strategy is designed to establish a new and efficient GA framework. This framework facilitates global exploration through large-scale crossover while conducting intensive local optimization for elite individuals, thereby effectively balancing the algorithm’s exploration and exploitation capabilities. Experiment results show that the average deviation between the developed freight vehicle highway driving cycle and real-world data across 16 key characteristic parameters is merely 0.59%. Moreover, the fuel consumption estimation error based on this cycle is as low as 6.24%. This research demonstrates that the proposed method effectively overcomes the tendency of conventional algorithms to converge to local optima, significantly enhancing the accuracy and robustness of driving cycle development. Consequently, it provides a reliable and representative driving cycle for vehicle energy consumption prediction and emissions evaluation.