Combustion stability analysis during engine stop and restart in a hybrid powertrain

A cycle-resolved analysis system was designed with the specified measurement instruments to investigate the characteristics of combustion stability in a mild gasoline hybrid powertrain. A Fast Response Flame Ionization Detector (FFID), cylinder pressure transducer and engine torque transducer were used to observe both the engine-out THC emissions and engine performance during a brief moment of engine restart. This research aimed to improve combustion stability and was performed by varying the battery State Of Charge (SOC), injection duration and ignition timing. The results indicate that engine combustion tends to be more stable with longer fuel injection durations and advanced ignition timing, while the effect of the battery SOC is negligible. Also, peculiar differences in the catalyst conversion efficiency at the front and rear of the catalyst during engine restart and deceleration were revealed, with the degree of HC oxidation being the suspected cause. This study not only analyzed the engine control and engine-out total hydrocarbon (THC) emission characteristics, but also implemented control strategies that allowed for combustion stability during engine stop and restart operation.     

基于决策树的异常电池分类方法

提出了一种基于决策树算法的异常电池精准定位和分类方法，帮助人们在电池维修或更换时能够迅速确定故障单体的位置和类型，从而实施准确的维修更换方法，提高故障处理的效率。搭建电池仿真模型获取异常电池的充放电循环数据；以电压数据为基础，训练用于异常电池分类的决策树算法，使用试验数据和云端实车数据对构建的模型进行验证。验证结果表明，该方法能够准确判断异常电池单体在电池组中的位置和异常类型。在不同验证数据上，该方法的分类准确率高达98%以上，能够有效筛选出动力电池组中的异常电池。该结果说明了提出的决策树算法在动力电池异常分类中的有效性和准确性。