网联混合动力汽车车速控制与能量管理的耦合机理及动态博弈控制方法研究

The concept of eco-driving strategy is an efficient way to improve the energy efficiency of connected hybrid electric vehicles (HEVs) further. Serving the vehicle body as an energy storage unit, the eco-driving strategy can improve the powertrain efficiency by controlling the kinetic energy of the vehicle body (or vehicle speed). In HEVs, the velocity planning couples with the energy management system. To explore the fuel saving potential of connected HEVs, the coordinated control strategy of vehicle velocity and energy management system is required. In this program, we will investigate the relationship between the powertrain energy consumption (including fuel and electricity) and the vehicle speed change in the past or future firstly, and then derive the coordinated control rule for both vehicle speed and energy management with optimal energy efficiency. In addition, the dynamic game control strategy under the dynamic traffic conditions, such as car following and intersection crossing, will be investigated to solve the multi-objective problem, which takes the equivalent fuel consumption, ride comfort, battery life and travel time into consideration. The objective of the program is to reveal the coupling mechanism between vehicle velocity planning and energy management, propose an enhanced equivalent consumption minimization strategy considering the change of vehicle kinematic energy, reveal the influence of road and traffic condition on the eco-driving of HEVs, and achieve the model predictive control of the hybrid dynamic system. As a result, the fuel saving potential of the connected HEVs can be fully explored.

经济型辅助驾驶是进一步提高网联混合动力汽车能量效率的有效手段，其本质是将车身作为储能元件，通过控制车身动能（车速）的变化达到提高动力总成效率的目的，这与混合动力能量管理策略相互耦合，必须协调二者以充分挖掘网联混合动力汽车的节油潜力。本课题拟建立多模混合动力汽车当前动力源能量损耗（油耗与电耗）与过去或未来车速变化的联系；探索具有油耗最优特性的车速与动力总成协调控制规律；研究跟车、路口通行等动态约束交通环境下，计及车辆综合油耗、行驶平顺性、电池寿命与通行时间的多目标动态博弈控制策略；旨在掌握混合动力汽车车速与能量管理控制的耦合机理，提出计及车辆动能变化的增强型等效燃油消耗最小化方法，揭示道路条件、交通流特征等因素对混合动力汽车经济型驾驶的影响机制，实现混杂动态系统的模型预测控制，充分挖掘混合动力汽车的节能潜力。

车辆能耗不仅取决于动力总成的控制，还受到周围复杂交通环境的约束（道路坡度、弯道曲率、信号灯、交通拥堵状况等）。智能网联汽车利用对部分未来行驶工况的预测或多车协同，通过合理的运动规划可以大幅度降低车辆能耗，提高交通效率。.本项目围绕智能网联汽车的运动规划与动力总成能量管理协同优化问题开展研究，建立网联汽车动力源能量损耗与过去或未来车速变化联系，提出了基于最小等效能耗的瞬时车速控制策略，以及基于网格缩放与迭代更新的迭代动态规划算法，实现满足实时性计算需求的经济性车速规划方法，解决了混合动力汽车速度规划与能量管理的控制精度与计算效率难以权衡的问题；设计基于长短期记忆网络的前车车速预测模型，提出计及前车运动状态估计的经济性车速优化方法与路口节能通行策略；考虑多路口信号灯相位配时的时空耦合关系，设计网联汽车高效通行与节能车速规划分层式控制架构，提高汽车在多交叉口的能量与通行效率；并将单车控制方法扩展到多车协同领域，解析车辆编队能效最优的几何位置关系，设计队列变间距控制机制，提出基于空间域的队列节能控制策略，实现安全节能的车辆编队控制；最后，面向高速匝道入口合流区的混合交通场景多车控制问题，提出基于控制势垒函数和控制李雅普诺夫函数的严格安全混合交通匝道协同合流策略，实现安全高效的匝道协同合流控制。相关研究成果可应用于搭载辅助驾驶或高级别自动驾驶系统的智能汽车上实现更优异的车辆行驶效率，对于提高道路车辆的能量效率及交通的通行效率具有重要作用与意义。