High-Efficiency Driving Cycle Generation Using a Markov Chain Evolution Algorithm
- 17 December 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 68 (2) , 1288-1301
Abstract
At present, driving cycles generated by Markov chain (MC) based methods have been used extensively in the evaluation of vehicle fuel and emissions, for designing the size of power components, and the optimization of energy management strategies. To study the adaptability of vehicles to driving cycles, it is necessary to generate representative driving cycles with a large number of vehicle and road parameters to reflect the actual driving environment. However, under this condition, the time efficiency of the MC simulation is very low, which makes it difficult to generate driving cycles with the required representativeness. Combined with random sampling, the evolution mechanism of a genetic algorithm (GA) can achieve this goal more efficiently than conventional random sampling. Therefore, we propose a novel MC evolution (MCE) algorithm that improves the efficiency of the MC by combining stochastic sampling with evolution. Genetic strategies are designed to satisfy the state transition of driving cycles, and the new GA is employed to optimize the stochastic sequences of the MC. Three-parameter (velocity, acceleration, and grade) driving cycles with different lengths are generated using the MCE from collected highway driving data. The validity and feasibility of the proposed method are evaluated by statistical analysis. A comparison of the MCE with other methods, including the MC stochastic method and a traditional GA, reveals that the proposed method yields driving cycles with remarkably improved representativeness and efficiency.Keywords
Funding Information
- Research on vehicle driving stability region based on driving torque and steering angle bifurcation (51475199)
This publication has 23 references indexed in Scilit:
- Development of simulated driving cycles for light, medium, and heavy duty trucks: Case of the Toronto Waterfront AreaTransportation Research Part D: Transport and Environment, 2015
- Using Large Driving Record Samples and a Stochastic Approach for Real-World Driving Cycle Construction: Winnipeg Driving CycleTransportation Science, 2014
- Comparison of the regulated air pollutant emission characteristics of real-world driving cycle and ECE cycle for motorcyclesAtmospheric Environment, 2014
- A robust, data-driven methodology for real-world driving cycle developmentTransportation Research Part D: Transport and Environment, 2012
- Synthesis of real-world driving cycles using stochastic process and statistical methodologyInternational Journal of Vehicle Design, 2011
- Computational Statistics in MATLAB®WIREs Computational Statistics, 2010
- Characterization of vehicle driving patterns and development of driving cycles in Chinese citiesTransportation Research Part D: Transport and Environment, 2008
- Fuel Use and Emissions Comparisons for Alternative Routes, Time of Day, Road Grade, and Vehicles Based on In-Use MeasurementsEnvironmental Science & Technology, 2008
- Development of a practical driving cycle construction methodology: A case study in Hong KongTransportation Research Part D: Transport and Environment, 2007
- Estimating Regional Air Quality Vehicle Emission Inventories: Constructing Robust Driving CyclesTransportation Science, 2003