Field programmable gate array implementation of wheel–rail contact laws

Abstract

This study presents the development of a field programmable gate array (FPGA)-based accelerator that is highly suited to the implementation of complex and computationally demanding control or real-time simulation tasks. The application studied in this article is for the real-time simulation of wheel–rail contact laws, which may be used for hardware-in-the-loop experimental studies of the latest active control technology for wheelset stabilisation and steering. The complex wheel–rail contact laws are implemented using a single FPGA chip that outperforms substantially modern general-purpose CPU or DSP in the aspects of processing time, configuration flexibility and cost. The algorithms are restructured to utilise the FPGA's parallel processing capability. Reusable IP core are used for the floating point operations. The scheduling of the operations is optimised to ensure effective and efficient use of the FPGA's resources.