Study On The Critical Technique Of NIR Wheat Quality Analysis System

Focusing on the practical problems of NIR wheat quality analysis system, this paper redesigns light source control system of the spectrometer and studies on wheat quality analysis. The main research contents of this paper are as follows:(1) In this paper, the system structure of the self-made system is overall designed. We choose CCD detector ranging from900to1700nm. Design principle and feature of the spectrometer and key algorithms of NIRS are briefly introduced.(2) After analyzing deficiencies of the original source instrument control scheme, this paper measures a scheme that current is controlled by light intensity. A comprehensive comparison of two kinds of source control scheme is given. Light source control system based on feedback control theory is designed, including light intensity control system and temperature control system.Temperature control system of light intensity detector:in order to improve the performance of detector, this article designs the power control module, temperature control module, serial communication module and main control module. All modules collect temperature signals real-time and adjust chilling current by PID algorithm to achieve the refrigeration and temperature stable objective.Light intensity control system:based on feedback control theory, the paper designs a light source calibration with double close-loop control system. Light current is variable inner loop and light intensity is in outer loop. Light current is also control object in the circuit. The power control module, power control module, serial communication module, and main control module are designed. The circuit will collect light intensity and current signal, and used as feedback signal to adjust source current.(3) The experiment consists of temperature and light intensity test and contrast analysis experiment among the self-made instrument, CA-06grating spectrometer and MPA Fourier NIR spectroscopy. In this part, data processing algorithm is edited to convert voltage signals to spectral data and spectral graph. The experimental results show that, in temperature control system which guarantees working temperature at-10℃, response time is about500s, the system’s, error is±0.1℃. In light intensity control system, response time is about20s, and its current standard deviation is3.2032*10-4A for3.5hours working. The light source control system responses speedily and works stable. The performance of improved instrument is enhanced with better predictability, repeatability and stability, so it can be used for detecting and analyzing wheat quality. However, the performance of the self-made instrument is poorer than that of MPA spectrometer in precision and accuracy.