Application On Fuzzy PID Control In Temperature Control System Of Chromatograph

The chromatograph is an online analysis apparatus, widely used in the fields like chemical, petroleum and steel. It is intelligentized and network, which plays an important role in the quality control and improves production efficiency. Presently, the chromatograph used in many medium-sized is import production. The CX8800 chromatograph designed by Nanjing analysis apparatus factory and our university has double kilns, multiflows and it is also intelligentized, self-checked, roboticized. This project is a key technology one, belonging to Nanjing technology bureau in 2002, the design of whose intelligentized measurement and control system belongs to Jiangsu province. This project improves the intelligentized and homemade development of chromatograph.The three control objects of doule kilns and melting kiln demand high control precision, which should be digitlized and intelligentized. Moreover, the chromatograph demands that the temperature control should be in the range from 60 to 200 Celsius and the control precision should be in the range from -3 Celsius to 3 Celsius.In order to meet the blastproof demand, the electric box and analysis box of the CX8800 chromatograph both have plus pressure structure. Infulenced by different season and temperature, the temperature and pressure of the compressed air filled in the boxes are changed, which further greatly influent the temperature transfer among the kilns. The experiments have proved that this charactersitcis is nonlinear and time varying.This project involves hardware design of temperature measurement and control loop of PC/104 system and intelligentized PID control. In the hardware design, pt1000 is used to measure temperature, module amplifier, IC control are also used. The solid-state relay is used to control temperature to reduce the influence of high order wave of power. In order to decrease the influence of outside inferences, different anti-inference measures are designed to settle down different states such as power and ground inferences, electrostatic and electromagnetism induction, I/O channels inferences. Furthermore, in the software design process, the wavelet nonlinear threshold filter is applied to increase the signal noise ratio of sampling signal, which ensures the following high precision temperature control. In the design of high efficiency PID controller, the conventional fuzzy PID controller is discussed, whose structure and fuzzy control rules are improved to construct the coefficient-changing PID parameter selftuning fuzzy controller. The simulation presents it has higher control precision than BP nervous network PID controller and conventional PID fuzzy controller, and meets the design control demand.