Exploration of FPGA‐based electromagnetic transient real‐time simulation system design using high‐level synthesis. Issue 16 (7th January 2019)

• Record Type: Journal Article
• Authors: Li, Qiao; Xiang, Yinxing; Mu, Qing; Zhang, Xing; Li, Xiongfei; He, Guanghui
• Is Part Of: Journal of engineering. Volume 2019:Issue 16(2019)
• Page Start: 1217
• Page End: 1220
• Publication Date: 2019-01-07
• Subjects: hardware description languages -- field programmable gate arrays -- circuit simulation -- transient analysis -- floating point arithmetic -- logic design; design process -- high‐level language -- HDL -- Xilinx Virtex7‐690 T FPGA -- offline simulation tools -- FPGA‐based EMT simulation system -- high‐level synthesis -- field‐programmable gate array‐based simulation system -- real‐time electromagnetic transient simulation -- time‐step simulation -- HLS technology -- FPGA‐based electromagnetic transient real‐time simulation system design -- handcrafted hardware description language -- pipeline optimisation -- loop unroll optimisation -- switches -- voltage sources -- current sources -- transmission lines -- double‐precision floating‐point calculation; Engineering -- Periodicals; Engineering; Electronic journals; Periodicals; 620.005
• DOI: 10.1049/joe.2018.8879
• Languages: English
• ISSNs: 2051-3305

Abstract

Abstract : Field‐programmable gate array (FPGA)‐based simulation system is applied in real‐time electromagnetic transient (EMT) simulation for small time‐step simulation. However, the FPGA‐based simulation system requires handcrafted hardware description language (HDL) and significant design efforts. This paper explores FPGA‐based EMT real‐time simulation system design using high‐level synthesis (HLS) technology to accelerate the design process and improve the flexibility, because high‐level language can be transferred into HDL using HLS. Furthermore, loop unroll and pipeline optimisation directives are analysed and applied for HLS to improve the computation speed of the components. The system can simulate various components, including switches, R‐L‐C, voltage sources, current sources, and transmission lines. Especially, the system adopting double‐precision floating‐point calculation can support the network with 74 nodes and achieve 2 us for each small time step on Xilinx Virtex7‐690 T FPGA. Comparison of simulation results from offline simulation tools with those from the proposed simulation system validates the correctness of the FPGA‐based EMT simulation system using HLS technology.