FPGA based finite-set model predictive current control for small PMSM drives with efficient resource streaming

This paper discusses the experimental implementation of a finite control set model predictive control (FCS-MPC) approach for small permanent magnet synchronous machines (PMSM) on a field programmable gate array (FPGA). The small drives investigated in this paper have electrical time constants in the range of microseconds. The main challenge of using FCS-MPC is the high calculation effort. By respecting each switching possibility, the calculation effort increases exponentially with the time horizon. Therefore, methods for limiting the calculations in order to reduce the switching possibilities (e.g. branch and bound) are useful. However, this paper presents an exhaustive search approach which respects every calculation possibility, while sharing the available resources on the FPGA. The presented implementation is characterized by using the HDL Coder from MathWorks with an automated workflow for implementing a long horizon FCS-MPC with an efficient resource streaming and sharing at the example of a PMSM. The used rapid control prototyping approach avoids manual coding and facilitates the implementation of the algorithm. The benefit of the presented implementation is the use of an application-oriented calculation platform instead of a high-performance solution, which is not useful for the transfer into industrial applications.