Circuits, Communications and Systems, Pacific-Asia Conference on (2009)
May 16, 2009 to May 17, 2009
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/PACCS.2009.12
An optimized algorithm using neural network is presented, and the characteristics of neural network are introduced. This algorithm has advantages in increasing precision, reducing calculating time, reducing additional memory and the deterioration of the harmonic. This paper introduces a new FPGA: ALTERA’S ACEX chip and EDA tool: MAX+PLUS II. For recent years, FPGA is increasingly developing very fast and has the advantages of high speed, large scale and easy-to-design software. It is convenient to use FPGA to design the circuit, simulate and verify the result, and it is also easy to debug the hardware using ISP (In System Programmable). So this algorithm is easily realized. If it is necessary to adopt the improved algorithm, the only work we should do is to modify the algorithm in software, then download file to chip using ISP, it has nothing to do with the hardware circuit. Therefore, it is fit to design product and verify the algorithm using FPGA. At the end of this paper, the experimental results are shown, and the advantages are verified by these results.
SVPWM; Neural network; FPGA; Switching
X. Honghua, L. Jianlin and Z. Chao, "Based FPGA Development of Optimized SVPWM Algorithm," 2009 Pacific-Asia Conference on Circuits, Communications and Systems (PACCS 2009)(PACCS), Chengdu, 2009, pp. 217-219.