Design and Implementation of HDRT Real-time Simulation System

doi:10.16182/j.issn1004731x.joss.25-0932

Abstract

Abstract:

In view of the real-time simulation requirements of large-scale complex systems such as power electronics, a hidden dragon real-time(HDRT) simulation system was developed. The strict time constraints of simulation tasks were guaranteed based on a resource-dedicated real-time scheme, supporting fixed-step and multi-rate simulations from the second level to the hundred-nanosecond level. A hybrid CPU-field programmable gate array(FPGA) architecture was adopted to accelerate computation. The system could utilize multiple simulators for parallel simulation and realized microsecond-level real-time data interaction between simulators through a dedicated PCIe switch. A single simulator could be expanded through I/O interface equipment, supporting a maximum input or output of more than 1 500 channels of real-time analog signals or more than 3 000 channels of real-time digital signals. Test results indicate that the synchronization time of a single I/O interface equipment in the HDRT simulation system is only 40 ns, indicating the characteristic of fast synchronization among I/O interface equipment; the maximum jitters of the actual simulation steps of the master and slave simulation tasks do not exceed 180 ns and 400 ns, respectively; based on tests of actual simulation models, the results demonstrate that the HDRT simulation system possesses the real-time simulation capability for large-scale complex systems.

Key words: real-time system, hardware-in-the-loop simulation, parallel computation, synchronization, system design

CLC Number:

TP399

Zheng Huiji, Wang Guangsen, Liu Qing, Wang Kang, Wang Zhiwei, Zhang Zhenyu, Wang Shuo, Liu Zhu. Design and Implementation of HDRT Real-time Simulation System[J]. Journal of System Simulation, 2026, 38(8): 2364-2378.

Figures/Tables 22

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Table 1

Fig. 7

Fig. 8

Table 2

Fig. 9

Fig. 10

Fig. 11

Table 3

Fig. 12

Table 4

Fig. 13

Table 5

Fig. 14

Fig. 15

Fig. 16

Fig. 17

References 24

[1]	刘伟, 夏敏学, 梁晨鹏. 高压直挂大容量电池储能系统实时仿真及控制研究[J]. 电力系统保护与控制, 2024, 52(14): 154-166.
	Liu Wei, Xia Minxue, Liang Chenpeng. Real-time Simulation and Control of a Large Capacity Bettery Energy Storage System Directly Connecting to a High-voltage Grid Without a Transformer[J]. Power System Protection and Control, 2024, 52(14): 154-166.
[2]	洪泽华, 潘洪涛, 陆志沣, 等. 半实物仿真系统可信度评估方法对比研究[J]. 电子技术, 2015, 44(9): 1-5.
	Hong Zehua, Pan Hongtao, Lu Zhifeng, et al. The Comparative Study for the Credibility Evaluation Method of Hardware-in-the-loop System[J]. Electronic Technology, 2015, 44(9): 1-5.
[3]	孙瑞松. 基于RT-LAB的微电网运行控制硬件在环仿真[J]. 电气工程学报, 2019, 14(4): 85-92.
	Sun Ruisong. Hardware in the Loop Simulation on the Operation Control of Microgrid Based on RT-LAB[J]. Journal of Electrical Engineering, 2019, 14(4): 85-92.
[4]	孙冰. 一种基于RTDS的小电流接地选线装置的分析与改进[D]. 西安: 西安石油大学, 2024.
	Sun Bing. Analysis and Improvement of a Small Current Grounding Line Selection Device Based on RTDS[D]. Xi'an: Xi'an Shiyou University, 2024.
[5]	闫松. 混合动力汽车动态能量管理策略及硬件在环仿真研究[D]. 青岛: 青岛大学, 2020.
[6]	张淼. 基于麒麟操作系统的实时仿真平台关键技术研究[D]. 长沙: 国防科技大学, 2017.
	Zhang Miao. Research on Key Technologies of Real-time Simulation Platform Based on Kylin Operating System[D]. Changsha: National University of Defense Technology, 2017.
[7]	张晓龙. 基于Linux/RTAI的嵌入式系统半物理仿真研究[D]. 武汉: 华中科技大学, 2014.
	Zhang Xiaolong. Study on Hardware-in-loop Simulation Based on Linux/RTAI Embedded System[D]. Wuhan: Huazhong University of Science and Technology, 2014.
[8]	冉印. 面向Linux的实时仿真平台关键技术研究[D]. 武汉: 华中科技大学, 2022.
	Ran Yin. Research on Key Technologies of Real-time Simulation Platform for Linux[D]. Wuhan: Huazhong University of Science and Technology, 2022.
[9]	王钦盛. 基于FPGA的微电网电力电子设备实时仿真技术研究[D]. 哈尔滨: 哈尔滨工业大学, 2024.
	Wang Qinsheng. Research on Real-time Simulation Technology of Microgrid Power Electronic Equipment Based on FPGA[D]. Harbin: Harbin Institute of Technology, 2024.
[10]	Wu Zhuanhao, Kaushik Anirudh, Patel Hiren. High Performance and Predictable Shared Last-level Cache for Safety-critical Systems[J]. ACM Transactions on Embedded Computing Systems, 2024, 23(6): 97.
[11]	马可. 面向实时操作系统的实时性分析方法研究[D]. 成都: 电子科技大学, 2022.
	Ma Ke. Research on Real-time Analysis Method for Real-time Operating System[D]. Chengdu: University of Electronic Science and Technology of China, 2022.
[12]	刘小龙, 徐丹, 陈丽丽, 等. 一种基于Simulink系统分区并行计算的Speedgoat HIL测试系统研究[C]//2024年中国汽车工程学会年会论文集. 北京: 机械工业出版社, 2024: 219-222.
	Liu Xiaolong, Xu Dan, Chen Lili, et al. Research on Speedgoat HIL Testing System Based on System Partitioning for Concurrent Execution of Simulink[C]//SAECCE2024-IEE023. Beijing: China Machine Press, 2024: 219-222.
[13]	李旭, 刘金华, 潘敏. 基于dSPACE仿真系统的永磁同步直线电机驱动器的研制[J]. 电气时代, 2024(12): 123-127.
[14]	王效亮, 张芳, 曾宪科, 等. 基于NI实时控制器的六自由度平台测控系统设计与实现[J]. 计算机测量与控制, 2019, 27(2): 24-28, 33.
	Wang Xiaoliang, Zhang Fang, Zeng Xianke, et al. Design and Implementation of a Six-degree Platform Control System Based on NI Real Time Controller[J]. Computer Measurement & Control, 2019, 27(2): 24-28, 33.
[15]	任攀, 张舒涵, 胡娟. 基于Mworks的洁净室压差运行调节仿真研究[J]. 上海节能, 2025(4): 599-606.
	Ren Pan, Zhang Shuhan, Hu Juan. Simulation Analysis of Differential Pressure Operating Ad-justment in Clean Room Based on Mworks[J]. Shanghai Energy Conservation, 2025(4): 599-606.
[16]	邓晨, 陈功, 敖厚军, 等. 拦截弹复合控制半实物仿真系统设计与实现[J]. 北京航空航天大学学报, 2025, 51(12): 4178-4187.
	Deng Chen, Chen Gong, Ao Houjun, et al. Design and Implementation of a Hardware-in-the-loop Simulation System for Interceptor Composite Control[J]. Journal of Beijing University of Aeronautics and Astronautics, 2025, 51(12): 4178-4187.
[17]	Reghenzani Federico, Massari Giuseppe, Fornaciari William. The Real-time Linux Kernel: A Survey on PREEMPT_RT[J]. ACM Computing Surveys, 2019, 52(1): 18.
[18]	Madden M M. Challenges Using Linux as a Real-time Operating System[C]//AIAA Scitech 2019 Forum. Reston: AIAA, 2019: AIAA 2019-0502.
[19]	Chou C H, Bhuyan L N, Wong D. μDPM: Dynamic Power Management for the Microsecond Era[C]//2019 IEEE International Symposium on High Performance Computer Architecture (HPCA). Piscataway: IEEE, 2019: 120-132.
[20]	Deng Zhaomeng, Zhang Ziqi, Li Ding, et al. Interference-free Operating System: A 6 Years' Experience in Mitigating Cross-core Interference in Linux[C]//2024 IEEE Real-Time Systems Symposium (RTSS). Piscataway: IEEE, 2024: 308-321.
[21]	The Kernel Development Community. Reducing OS Jitter Due to Per-cpu Kthreads[EB/OL]. [2025-04-09]. .
[22]	Sohal P, Bechtel M, Mancuso R, et al. A Closer Look at Intel Resource Director Technology (RDT)[C]//Proceedings of the 30th International Conference on Real-Time Networks and Systems. New York: Association for Computing Machinery, 2022: 127-139.
[23]	Vaidyanathan K, Panda D K. Benefits of I/O Acceleration Technology (I/OAT) in Clusters[C]//2007 IEEE International Symposium on Performance Analysis of Systems & Software. Piscataway: IEEE, 2007: 220-229.
[24]	Ménard Sylvain. RT-LAB Version 11.1 User Guide[EB/OL]. [2025-04-09]. .

目录	命令	描述
内核编译选项	CONFIG_NO_HZ=y CONFIG_NO_HZ_COMMON=y CONFIG_NO_HZ_FULL=y	支持nohz_full模式，结合内核启动参数nohz_full=core_list使用，关闭指定实时CPU核心空闲和单个任务运行时的tick中断
	CONFIG_CPU_ISOLATION=y	支持CPU隔离，结合内核启动参数isolcpus=core_list使用，将指定CPU核心从调度算法中移除
	CONFIG_PREEMPT=y	使能抢占式内核(PREEMPT_RT)
内核启动参数	isolcpus=core_list	将指定CPU核心从调度算法中移除
	nohz_full=core_list	关闭指定实时CPU核心空闲和单个任务运行时的tick中断
	tsc=reliable	避免TSC时钟校准功能导致的tcik中断
	mce=off	避免机器检查异常功能导致的tick中断
用户层工具	echo 1>/proc/irq/*/smp_affinity	设置IRQ亲和性到指定核心
	taskset-pc 0/proc/*	将系统非实时进程迁移到非实时CPU核心
	echo performance>/sys/devices/system/cpu/cpu*/cpufreq/scaling_governor	设置CPU频率调节器为性能模式，避免造成性能抖动
	echo-1>/proc/sys/kernel/sched_rt_runtime_us	避免限制实时核心的CPU时间片导致tick中断
	echo 0>/sys/devices/system/cpu//online echo 1>/sys/devices/system/cpu//online	重启实时CPU，通过热插拔CPU，强制将该CPU上的部分定时器软中断等迁移到非实时CPU上^[21]

采样工具	显示位置及格式	采样方式	采样频率限制因素
上位机管理软件	上位机文件或图像	HDTarget实时任务缓存记录，非实时任务通过UDP上传上位机	网络和图像渲染速度
HDWriteFile模块	HDTarget文件	HDTarget实时任务缓存记录，非实时任务写入本地文件	内存写入文件速度
Get_Signal模块	上位机Simulink信号	HDTarget实时任务缓存记录，上位机Simulink模块通过TCP读取	网络与Simulink数据处理速度

监视工具	(事件)数据来源	监视周期	监视内容	精度	中断定位
imon	/proc/interrupts和SMI计数器	imon命令启动后	硬中断和SMI中断个数	s	CPU编号、中断类型
pmu监视器	硬件性能计数器	仿真过程中	硬中断个数和其他性能事件	ms	CPU编号
内核监视器	独占模式丢失监视补丁	仿真开始前、仿真过程中、仿真结束后	含硬中断和信号内陷等广义中断	μs	CPU编号、中断类型、调用栈

HDIO数量	时长
1	40
2	96
3	144

HDTarget	子系统	仿真步长/μs	最大仿真步长抖动/ns
Ⅰ	1(从)	75	249
	2(从)	100	271
	3(从)	100	279
	4(从)	100	281
	5(从)	25	179
	6(主)	25	172
Ⅱ	1(从)	25	269
	2(从)	25	356
	3(从)	25	378
	4(主)	25	169
Ⅲ	1(从)	25	377
	2(从)	100	250
	3(从)	25	347
	4(从)	25	318
	5(主)	25	170