High-Fidelity Real-Time Digital Hardware Emulation of Power Electronic Motor Drives for Automotive Applications

为了提高能源效率、动力性能和满足减排标准，人们正在积极研究混合动力电动汽车（HEV）。混合动力汽车的能量管理系统是一个多物理场的子系统紧密集成的系统，其设计和优化非常复杂。实时硬件在环（HIL）仿真正在成为HEV控制系统快速原型设计和评估的重要工具，通过实时仿真，可以在闭环中对实际控制系统进行测试，从而减少对原型车的依赖，并快速跟踪设计过程。精确建模的HEV子系统是需要一个现实的实时HIL仿真，然而，现有的实时数字仿真器仍然受到建模和计算瓶颈，实现详细的仿真。该方案的目的是在并行可重构现场可编程门阵列（FPGA）上开发应用于混合动力汽车的电力电子电机驱动器的高保真实时模型。将为推进电机、电力电子转换器和储能装置开发详细的模型。将开发改进的参数估计和控制方法，以将详细的子系统模型集成到HEV HIL仿真中。将开发新的并行数值方法，在FPGA上实时仿真这些模型。所开发的模型、方法和HIL仿真结果将使用实验和行业标准器件级离线仿真工具进行充分验证。

Research into hybrid electric vehicles (HEVs) is being actively pursued for enhanced energy efficiency, power performance, and to meet emission reduction standards. The energy management system of the HEV is very complex to design and optimize because several subsystems in different physical fields are tightly integrated. Real-time hardware-in-the-loop (HIL) simulation is becoming an essential tool for rapid prototyping and evaluation of HEV control systems, whereby the actual control system can be tested against real-time simulations of the HEV subsystems in closed loop, thus reducing dependence on prototype vehicles and fast tracking the design process. Precise modeling of HEV subsystems is required for a realistic real-time HIL simulation; however, existing real-time digital simulators are still limited by modeling and computational bottlenecks for achieving a detailed simulation. This proposal aims to develop high-fidelity real-time models for power electronic motor drives applied in HEVs on parallel reconfigurable field programmable gate arrays (FPGAs). Detailed models will be developed for the propulsion motor, power electronic converters, and energy storage devices. Improved parameter estimation and control methods will be developed to integrate the detailed subsystem models into the HEV HIL simulation. Novel parallel numerical methods will be developed to emulate these models on FPGAs in real-time. The developed models, methods, and HIL simulation results will be fully validated using experiments and industry standard device-level off-line simulation tools.