Real time plasma disruptions detection in JET implemented with the ITMS platform using FPGA based IDAQ

摘要

The use of FPGAs in data acquisition cards for processing purposes allows an efficient real time pattern recognition algorithm implementation. Using 13 JETs database waveforms an algorithm for detecting incoming plasma disruptions has been implemented. This algorithm is written in MATLAB using floating point representation. In this work we show the methodology used to implement the real time version of the algorithm using Intelligent Data Acquisition Cards (IDAQ), DAQ devices with field programmable gate array (FPGA) for local processing. This methodology is based on the translation of the MATLAB code to LabVIEW and the final coding of specific pieces of code in LabVIEW for FPGA in fixed point format. The whole system for evaluating the real time disruption detection (RTDD) has been implemented using the Intelligent Test and Measurement System (ITMS) platform. ITMS offers distributed data acquisition, distribution and real time processing capabilities with advanced, but easy to use, software tools that simplify application development and system setup. The RTDD implementation uses a standard PXI/PXIe architecture. Two 8 channel analog output cards play JETs database signals, two 8 channel DAQ with FPGA acquire signals and computes a feature vector based in FFT analysis. Finally the vector acquired is used by the system CPU to execute a pattern recognition algorithm to estimate an incoming disruption.