基于中国典型城市交通流特征与多参量解耦观测的客车混合动力系统运行优化方法

As the serious urban transport energy consumption and pollutant emissions are increasingly developed nowadays, plug-in hybrid electric bus (PHEB) has become one of the key technologies of new energy automotive industry. Recently, the energy saving performance of new energy bus running in many cities in China is far from the theoretical optimal level that is up to 50%. The reasons caused by the situation mentioned above could be summarized as follows. Firstly, a huge randomness characteristics which is generated by the varying rate of urban traffic information with the different time and zones, makes that conventional economic optimum external conditions might not achieve via the conventional driving mode. Secondly, it is hard to decouple the observed parameters which might extremely influent the energy efficiency as the single-shaft parallel configuration which is widely applied in the PHEB has a strong connection with the vehicle during the process of running. Thirdly, an uncertainty of dynamic state has arisen because of the repeated switches between the process and mode of the running system. Finally, it is extremely difficult to achieve the energy balance and optimization between the multi-adaptive vehicle operating conditions. To solve the problems mentioned above, several research interests are proposed as follows: 1)To build the method which is used to extract and quantify the statistical feature according to the urban traffic information with different time and zones; 2)To study dynamic decoupling mechanism and establish online observing methods of complex hybrid vehicle multi-parameter electromechanical system which are included of vehicle mass, equivalent slope, driving intent, SOC and so on; 3) To explore the operating mode of PHEB using the single-shaft parallel configuration, meanwhile explicit the optimizing mechanism and matching criterion; 4) To develop the realizing control method which could be utilized to excite the energy-saving potential of PHEB configuration. In summary, this research work will provide the theoretical basis and analyzing methods in the smooth and efficient running of PHEB.

应对日益严峻的城市交通能耗与排放的插电式混合动力客车（PHEB）是当前新能源汽车产业的前沿核心技术之一，但当前PHEB机电耦合系统节能减排性能远未达到其理论最优水平。原因有：城市交通随时间、区段具有强随机性，常规驾驶难以满足经济性最优的外部条件； PHEB普遍采用的同轴并联构型机制与运行工况深度耦合，解耦困难；混合动力系统瞬态模式切换带来的状态不确定性极度影响了系统控制平顺性；驾驶工况的自适应与多能源优化目标之间极难平衡。故项目将开展以下研究：1）建立针对PHEB交通流分时分区统计特征提取与量化方法；2）研究复杂混合动力机电系统多参量（质量、坡度、加速意图、SOC等）动态解耦机理与在线观测方法；3）探索同轴并联构型的PHEB运行模式与不确定工况间的优化匹配机制与判据；4）探索发掘PHEB构型节能潜能的可实现性控制方法。研究工作将为PHEB高效平顺运行优化控制提供理论支撑和分析方法。

为了实现插电式混合动力客车（PHEB）的节能减排性能，本项目主要从以下几个方面展开研究：1） 应对混合动力汽车在自适应巡航控制（ACC）和能量优化策略（EMS）方面，通过寻求最优解的方法将ACC和EMS进行融合控制并与串联控制方法进行比较，丰富了混合动力汽车在应对不同工况下的能量优化方法体系；2）通过探索不同城市交通工况差异，提出了一种基于粒子群优化算法（PSO）的非线性模型预测控制（NMPC）策略来实时获取更优的PHEB燃油经济性，并利用实际客车数据和行驶工况进行了仿真验证，与传统采用CD-CS方案相比，使用此方法，混合动力汽车的燃油消耗降低了10%以上，为PHEB的节能减排提供了坚实的理论基础；3） 在对混合动力客车机电耦合系统的研究中发现可通过建立汽车动力学稳定模型来采取预先防范措施，避免事故发生，同时有效提升燃油率，而车辆动力学关键状态观测问题研究是解决汽车极限工况失稳问题的基础，故推导出 变结构PID观测器，确保闭环系统的渐近稳定性并将该 综合问题被转化为LMI求解问题，通过求解LMI，所提出的观测器可以确保其性能。通过选择双移线操纵（double lane change, DLC）作为测试工况并通过离线实车试验，验证出纵向轮胎力观测值与实际值吻合。 在试验车蛇形试验中观测的纵向轮胎力的百分误差分别为8.06％，5.53％，8.79％，7.29％；进一步优化了混合动力汽车的机电耦合性能；4） 在对PHEB电池SOC性能研究中，提出了一种变结构的迭代扩展卡尔曼滤波算法，并通过仿真将变结构方法与常规扩展卡尔曼滤波方法进行了比较，通过比较发现变结构方法不论放电倍率大小，整体SOC的估算精度都维持在2%-8%，优于扩展卡尔曼滤波估算精度，而且在高倍率放电情况下变结构方法的收敛速度明显优于扩展卡尔曼滤波，此结论对于实际电池成组系统对SOC估算提供了应用前景。