Evaluation And Optimization Of Energy Consumption Based On Driving Patterns For Plug-in Hybrid Electric Vehicle

The electrification is now an inevitable trend of the vehicle powertrain. Plug-inhybrid electric vehicle (PHEV), as a typical electric vehicle, has gained more and morepopularity in recent years. Yet, the configuraitons vary a lot among the PHEV products,e.g, Toyota, Ford and General Motor respectively choose10,20and40miles as the allelectric ranges for their PHEVs. Which configuration fits the Chinese market best? Toanswer this question, this paper provides solutions to three particular challenges facedby PHEV.Based on the fact that a PHEV has two operation stages (charge depleting stageand charge sustaining stage), the energy consumption evaluation method is studied toprovide a solution to the challenge of weighting the fuel economy of the two stages. Bythe comparison of the energy consumption evaluation methods between China and US,the current weighting method of the two stages used in China is not appropriate. Theutlity factor weighting method is then applied in this paper. The data of the utility factorcomes from the Beijing driving pattern database established by this study. A newmethod of calculating the utility factor is proposed. The new method requires the inputof daily driving range distribution function, instead of the original data. Thus, theevaluation result can be closer to the energy consumption in the real situation.Based on the fact that a PHEV can be charged from the grid, the energymanagement strategy (EMS) is studied to provide a solution to the challenge of optimalutility of the grid electricity. For the cases with known driving range, a realtimeimplementable EMS called approximate Pontryagin’s minimum principle (A-PMP)strategy is proposed. The core of this strategy is a piecewise approximation to the fuelrate curve of the engine. The A-PMP saves the fuel by6.96%compared to theconventional “all electric, charge sustaining”(AECS) strategy. For the cases withunknown driving range, the range adaptive optimal control (RADOC) strategy isproposed with the objective of minimizing the average fuel consumption. The historicaldriving range data is utilized to generate the utility factor of the driver. Then, the utilityfactor weighted fuel consumption, which suggests the average fuel consumption amongthese trips, becomes the objective of the proposed RADOC strategy. It is able to reduce the average fuel consumption by0.10%-4.07%compared to theAECS strategy.Based on the fact that a PHEV has a larger battery energy, the total cost ofownership (TCO) is studied to provide a solution to the challenge of optimizing thebattery size. A TCO model is established to evaluate the purchase and use cost of aPHEV owner. The model is a cash flow based model, integrating the powertrainsimulation, the optimal EMS, the driving pattern database, and the battery degradationtest data. The analysis based on the Beijing real scenario suggests that the optimalbattery size for PHEVs used in Beijing is6kW·h-8kW·h.