Neural Network Data Analysis in the Large Helical Device Thomson Scattering System

In Thomson scattering diagnostics systems, a combination of the lookup table and minimum χ2 methods has been widely used to determine electron temperature. The concept method is based on clearly defined mathematical statistics. However, calculation requires large amount time because all values have be calculated at temperatures included in table. Thus, this unsuitable for real-time data analysis required next generation fusion devices, e.g., International Thermonuclear Experimental Reactor France. To establish diagnostics, we developed neural network program helical device (LHD) (TS) system. First, systematically studied number nodes training cycles obtain satisfactory results, then applied them LHD TS was successfully reduced by approximately 1/50 - 1/100 time. addition, experimental error estimation performed according study.