DFT algorithm analysis in low-cost power quality measurement systems based on a DSP processor

Abstract

The application of power quality parameters measurement algorithms requires utilization of computational methods which transform the signal from the time- to the frequency-domain. Currently, the most frequently applied method is the fast Fourier transformation (FFT). Because of the amendment of the proper power quality standards, it is also necessary to consider of implementation discrete Fourier transformation (DFT) algorithms. Considering the number of computations, DFT is more complicated than FFT. This paper presents basic analog-to-digital conversion parameters, which are defined in the norms, the time duration analysis of DFT algorithms, considering described assumptions and comparing them to FFT. Codes created in C and assembler languages, as well as the Matlab/Simulink environment, were taken into account. The software including the aforementioned codes was implemented in a low-cost power quality measurement system based on the TMS320c6713 floating point DSP processor from Texas Instruments. In the software, a real time DSP/BIOS operation system was used. Multi thread programming in power quality measurement systems was presented, and an analysis of computational efficiencies of the systems based on the TMS320c6 713 DSP processors was performed.