摘要
在保持当前动车组辅助变流器电力一次架构不变的前提下,优化动车组辅助变流器硬件拓扑及数据拓扑,使其数据系统可以支持运行超限学习机及多列神经网络的智能化控制程序,实现该新型拓扑对双机联合驱动及多机联合驱动的拓扑扩增过程。在仿真环境下,使用新拓扑并运行对应人工智能程序,在相同运行图的情况下,瞬时最大电流下降7.8%,工作电流标准差缩小69.2%,车辆平均驱动功率下降5.6%。优化后动车组辅助的硬件拓扑及数据拓扑有明显优势。
On the premise of keeping the power primary structure of the current EMU auxiliary converter unchanged,in order to optimize the hardware topology and data topology of the auxiliary converter of the EMU,the paper uses the data system to support the intelligent operation of the over-limit learning machine and multi-column neural network.The integrated control program is used to realize the topology amplification process of the new topology for dual-machine combined driving and multi-machine combined driving.In the simulation environment,using the new topology and running the corresponding artificial intelligence program,under the same operation diagram,the instantaneous maximum current decreases by 7.8%,the standard deviation of the working current decreases by 69.2%,and the average driving power of the vehicle decreases by 5.6%.The optimized hardware topology and data topology assisted by the EMU have obvious advantages.
作者
唐子辉
蒋学君
孙晓丽
Tang Zihui;Jiang Xuejun;Sun Xiaoli(CRRC Yongji Electric Co., Ltd., Shaanxi Xi'an,710016,China)
出处
《机械设计与制造工程》
2022年第2期130-134,共5页
Machine Design and Manufacturing Engineering
关键词
动车组
辅助变流器
拓扑设计
人工智能
拓扑扩增
EMU
auxiliary converter
topology design
artificial intelligence
topology amplification