Research And Implementation Of Paper Machine Cross Directional Basis Weight Control System

With the development of paper industry, continuous improvement of thespeed of the paper machine, increasing width of the paper, at the same time,consumers’ requirement on paper quality have become much higher. Whilemanufactures take control on the cost, traditional and single longitudinal controlcannot guarantee the quality of products of paper machine at all. Basis weight,moisture and thinness of paper are basic indicators of the characterization ofpaper products. Among them, basis weight is the most important physicalquantity. Basis weight control can be divided into machine direction control andcross direction control, but pure machine direction control can not meet therequirements of product quality. Adding paper machine cross directional basisweight control system is an inexorable trend. Advanced cross directional basisweight control system can better guarantee the machine high-speed and widen.In this paper, combine with “BQCS-I style paperboard production linequality control system, science and technology small and medium-sizedenterprise fund project(No：10C26216103241)”, the main contents and results ofthis paper are as follows:(1) To give an overview of the generation, development and recent progressof the machine cross directional basis weight control technology at home andabroad, and analyze advantages and disadvantages of regulatory mechanisms oflip plate with curve method and local density regulation method, proposingseveral problems and control puzzles, which need to be solved in developingdilution water cross directional basis weight process control system.(2) To study signal sampling and wavelet de-noising in cross directionalbasis weight. According to various communication ways by machine directionand cross direction systems in different factories, cross directional basis weightsignals can be collected by the communication of OPC from the machine direction control system. The collected signals mainly include practical crossdirection basis weight signals and random interference signals. In order to stoprandom interference signals from adjusting error, based on the principle ofwavelet de-noising, a function of self-optimizing threshold is proposed in thispaper to denoise the signals. The simulation result proves that self-optimizingthreshold is a better threshold method with better de-noising effect thanDonoho’s soft and hard threshold method, separately with SNR0.2577and6.0834higher and with RMS deviation0.0048and0.1263lower than the lattertwo.(3) To study the design of cross directional basis weight controller. Becausethe cross direction basis weight process is the most complex control process ofthe papermaking, the signal after de-noising can not achieve a better controleffect by the traditional PID control. Without the need of creating a preciseobject of mathematical models, and with better dynamic performance of thefuzzy control technique, especially fit for dealing with these problems, that is,difficult to establish the multi-variable of accurate mathematical model,time-varying and nonlinear uncertainty of the system. With the deviation of thecross direction basis weight and the change rate of the deviation as an input, PIDparameters are auto-tuned online by use of fuzzy inference methods. And asimulation study on tuning fuzzy PID control algorithm is done. The resultsshow that the effect of online self-tuning fuzzy PID controller is better thanconventional PID control.(4) To explore the specific methods of the application of the cross directionbasis weight control system in engineering. For precise positioning the crossdetection point and control point, the cross direction basis weight control systemis given. The system function blocks in the simatic software are used to achieveprecise positioning. And study the way to get the cross direction basis weightdata by the way of cross direction basis weight control system through OPCprotocol. For the control algorithm part, the global C script from the hostcomputer WINCC is given to finish the fuzzy inference. The parameters of thePID control algorithm of lower machine are online auto-tuned. The cross basisweight control systems are completed through the upper and lower machine. Thecross direction basis weight control system is successfully applied to a paper factory, the result shows that the cross direction basis weight control effect isimproved more than50%, and further illustrates the effectiveness of theimplementation.