Electromagnetic Interference Analysis And FlexRay Network Design For Dual-Motor Hybrid Power-Train System

The study in this paper is sponsored by the state "863" high-tech program (No.2008AAllA140) "Development of New Vehicle Technologies for the FAW-JieFang Hybrid Electric Bus" and the jointed action project among Jilin province, FAW, and Jilin University " Pre-research on the Key Technologies for the FAW Second Generation Hybrid Electric BUS ".This paper is targeted at electromagnetic interference (EMI) in the HEV development process. Firstly, the EMI which is generated during the operating process of dual-motor hybrid power-train system was analyzed. In order to eliminate the impact of EMI, the CAN bus was substituted by a new generation in-vehicle network FlexRay which became the backbone network. To enhance the time performance of FlexRay network, several optimizing methods were proposed. There are five parts of contents were proposed in this paper.1. The mechanism of EMI formed in the working process was analyzed, and concluded the theoretical value. The analyzing results showed that the key factors which leaded to EMI were the electricity generating process of rectifying circuit in the motor driven system and the process which motor was driven by a three-phase inverter. The study of EMI was divided into two parts:common-mode and differential-mode, the theoretical value was obtained by establishing the equivalent circuit model, and it was verified that the models could calculate the EMI produced by the system accurately by comparing the calculated and experimental results. Therefore, the theory basis of using FlexRay instead of CAN as the in-vehicle backbone network was obtained.2. The FlexRay network of dual-motor hybrid power-train system was established. FlexRay protocol was studied in detail on the media access control, distributed clock synchronization and protocol operation control. On this basis, the network development tool chain was used for FlexRay network design according to V model development process, including overall planning and network structure design, the definition of cluster and node parameters, network database was established and verified in a CANoe simulation environment. Hardware design of the main control chip, transceiver and its peripheral circuit of the node were carried out. The functions of initialization settings, communication controller configuration, wakeup and startup of the node and sending and receiving subroutines were implemented during the software design of the FlexRay mode.3. The optimization algorithm of FlexRay static segment was studied, so that the bandwidth utilization rate was significantly increased. The optimization design of FlexRay static segment could be divided into two steps:signal packaging and scheduling algorithm. A signal package method based on the maximum bandwidth utilization was designed and the static messages for transmission were obtained through integer linear programming (ILP) method. Meanwhile, a Fixed Height Conveyor (FHC) scheduling model which was based on the classical packing problem was designed, the ILP according to the constraint conditions in FlexRay protocol was established, the optimized scheduling table for static segment was obtained. In the simulation experiments, the signal in the network database was repackaged according to the optimized packing algorithm, by comparing with the original packing method and the classic packing algorithm, the improvement of the bandwidth utilization was concluded. Through the further expansion of the signal collection, more intuitive results were obtained, which verified that the static scheduling algorithm in this paper could improve the bandwidth utilization of FlexRay static segment.4. FlexRay dynamic segment configuration was optimized, which improved the time performance of the dynamic segment. On the basis of analyzing the transmission characteristics of dynamic message in detail, the optimization design of FlexRay dynamic segment could be divided into two parts, that is, scheduling optimization for a given message set in dynamic segment and to determine the optimal dynamic message length. In the process of scheduling optimization, schedulable analysis had to be carried out for each possible scheduling solution, in other words, which meant to judge whether each message in the collection met the conditions that the worst-case response time (WCRT) was less than a message deadline or not. So to establish a WCRT model of dynamic segment message should go first, then to obtain the key elements of optimal solution through ILP. Heuristic search strategy was used in the scheduling algorithm and the optimal solution of dynamic segment length was called in scheduling algorithm to reduce the complexity of the algorithm, the simulation results showed that the WCRT of dynamic message had obvious decreased after the optimum design.5. The FlexRay network developed in this paper was verified by hardware-in-the-loop simulation and bench experiments. In the simulation experiments, hardware-in-the-loop simulation system was established by CANoe.FlexRay, the network authentication and simulation were completed by FlexRay hardware interface VN7600. In the bench experiments, the differences of electromagnetic compatibility between the FlexRay network developed in this paper and the original CAN bus in the bench was compared, it was verified that FlexRay could be used as in-vehicle backbone network to reduce the influence of EMI in a serious EMI environment such as HEV.