Application Research On The Intelligent Controller For Liquid Nitrogen Filling Based On The Grey Prediction Theory

Liquid nitrogen is a kind of excellent low temperature coolant. Low-temperature Dewar vessel is usually used to store liquid nitrogen to refrigerate some objects in practice. It has become one of the important equipment in low-temperature refrigeration system.The liquid nitrogen auto-filling control system can identify the liquid nitrogen storage in Dewar vessel according to the gathered signals of liquid nitrogen level. It can achieve to control the liquid nitrogen level in Dewar vessel by controlling the solenoid valve in the delivery pipeline. Because there are characteristics as time-varying parameters and serious disturbance by load in the process of filling liquid nitrogen, the controller using conventional control algorithm can not satisfy quality of controlling requirements. An intelligent control-algorithm of auto-filling liquid nitrogen based on the grey prediction theory is offered in the paper, which has been used to design the intelligent controller of filling liquid nitrogen.Grey prediction is an important part of grey system theory. Grey system theory gives emphasis to investigate the issues that have little sample, poor information and can’t be solved by statistics or fuzzy mathematics. According to the overlay of the information, the practical orderliness through the sequence-building has been found. It doesn’t need the accurate model; its point is modeling based on few numbers. Grey prediction needn’t many samples and doesn’t request sample distributed regulation. It also has advantages of less computation and stronger adaptability.The primary contents of prediction are firstly introduced in the paper. The pretreatment of original data sequence is analyzed according to the primary procedure of prediction. Then it introduces the mechanism and steps of constructing grey prediction model. Control algorithm of filling liquid nitrogen to Dewar vessel based on grey prediction theory is designed, which has been used to increase the stability of system and filling efficiency. The controller has been used successfully in the refrigeration system of Large Sky Area Multi-Object Fiber Spectroscopy Telescope (LAMOST), achieved satisfactory control effect.