Multi-perspective evaluation of fuel-efficient architecture for China's plug-in hybrid electric vehicles across levels, time, and driving cycles

Abstract

The architecture of plug-in hybrid electric vehicles (PHEVs) defines their technical framework, and the question of which PHEV architecture delivers superior performance has ignited a fervent debate. Previous studies evaluating different PHEV architectures through performance comparisons of multiple vehicle cases lacked generality due to parameter randomness. To investigate the optimal PHEV architecture for fuel economy in the Chinese market, this paper proposes a combined statistical and experimental approach to analyze 23 PHEV architectures. The statistical analysis utilizes fuel consumption data from 765 vehicles over a span of 12 years, while the experimental analysis involves dynamometer tests on three vehicles with different architectures under three driving cycles. The Monte Carlo method was utilized to calculate the sampling probability of the best fuel economy among various architectures. The results indicate that the rankings of fuel economy superiority vary across different perspectives, and no single architecture consistently outperformed others in all scenarios. The superiority or inferiority of architectures is reflected in the fuel consumption rankings in case comparisons and the statistics rankings of fuel consumption or the probability rankings of the lowest fuel consumption in the overall analysis. The series-parallel P1+P3 and P1+P2 architectures demonstrate better fuel economy performance from multiple perspectives. This analysis offers a comprehensive overview of the fuel economy across various PHEV architectures in China, serving as a valuable reference for selecting technical routes.