Abstract: This paper mainly describes the applicability of fast fourier transform (FFT) deconvolution technique, which is used in assessing the residence time distribution (RTD) of reactors. And then use the ideal system models and these models after added noise to verify it , the results consistent with the analytical results very well. Finally use the twin-screw extruder as research object, apply FFT deconvolution technique to obtain local RTD curves of every location of extruder. Obtained Residence time distribution curves shows a single peak shape, both ends of the peak (i.e. error) is relatively a little large. Generally speaking, in the same extruder conditions, the RTD curves of pushing zone are narrower than the RTD curves of kneading zone, it means that, kneading zone have a good mix effect. The research of this paper indicate that, FFT deconvolution technique have a good application foreground in assessing RTD, but FFT deconvolution method is sensitive to the quality of original data, it still have some troubles in treating with real data system. If use effectively filtering method, it is expected to increase its adaptability to the data and expand its application.
Keywords: Residence time distribution; Deconvolution; Fast Fourier transform; Twin-screw extruder