Research And Application Based On Fieldbus Temperature Control System Of Reaction Kettle

Unsaturated polyester resin (UPR) as an important species in thedevelopment of the plastics industry in modern times, there is a wide rangeof applications in the field of industry, agriculture, construction, transportand defense industry. The reaction kettle temperature control system is animportant component in the production process. With the rapid developmentof the economy, people have higher requirements on the quality of chemicalproducts, therefore the reaction kettle research get more attention.Temperature control is the most important in the control of the reactionkettle, it directly affects the chemical reactions in the reaction kettle andplays a decisive role in the quality of the products. At present the mostproduction lines of temperature control system use conventional PID control,However, the UPR production have nonlinear, large time-delay,non-stationary characteristics. Although conventional PID is simple and easyto operate, but in the case of complex conditions is difficult to achieve thedesired control effect.For this reason automatic intelligent control has been experts andscholars generally concerned, Neural network not only has the ability toapproximate any nonlinear function, also the structure and the learningalgorithm are simple and clear, through learning system performanceachieve the best combination of PID control.The article takes Liaoyang a new UPR project control system design ascontrolled object, according to the actual situation, learning the reactionkettle working principle and product technology characteristics, described thedesign process of the project temperature control system. The systemselected Germany’s Siemens S7-300PLC as data acquisition and controlsystem, via PROFIBUS-DP fieldbus technology to achieve PLC master andslave between communications. In reaction kettles (esterification reactor,esterification polycondensation reactor) temperature heating stage useintegral separation PID control, in the constant temperature stage use BPneural network PID control. So that when the devices automatically run theprocess temperature curve process control systems play neural network PID control and PID control their own advantages. It is fast response, overshootsmall, without the static error. The actual operation situation is more ideal, itcan be applied in engineering practice.