Study On The Soft Measurement For Moisture Content Of Acrylonitrile Polymer

Since there is a large delay and poor control quality in the moisture control of the ploy dryer output materials in our factory, and the control system is not a real-time closed loop one, the production and quality of the dryer was restricted. In order to solve this problem, the development of a real time soft measurement and control system for the moisture of the output materials, is discussed in detail in this thesis.While building the mathematics model of the system, we first determine the least square parameter estimation method to be the theoretical fundament. Because the least square parameter estimation method is only adapted to the linear system, however, the dryer system involved in the thesis is a non-linear system, the dryer system is linearized at the steady point of the system. When the data are collected, first they are all filtered, then some exchanged unit and the error data are picked out, so that the error is eliminated to some degree. In order to eliminate the effect of the system lag, the different time data-collecting manner is adapted. To improve the accuracy of the parameters to be estimated, some process variables are combined and the dimension of the matrix equation is reduced after the analysis about the process theory mechanism.When the mathematical model is running on line, all the data to be calculated are also filtered, unit-exchanged and detected to eliminate the error. To eliminate the effect of the system lag, the data-delay manner is adapted. At the implement process, some calculation programs are written. To improve the accuracy of the soft measurement result, a timing self-organizing calibration method is adapted once a day.In development of the control system, an intermissive PID control scheme is adapted. The result of the soft measurement is considered as the measure value, and compared with the SP value, the output signal determined by the PID law adjusts the speed of the supply pump to control the moisture of the output materials in real time. To eliminate the effect of the system lag, the time compensation manner,as an intermissive control manner, is used. After the controller works for T1 span time, its output state will be held. And after T2 span time, the controller works again and going to next cycle period. As the automatic control of the system is based on the manual adjustment, the increment of output is combined with the manual adjust output.The real-time soft measurement and control system of the output materials moisture has been realized, as such, the measurement result is more accurate and the control effect has been much improved. Finally, in order that the accuracy of the soft measurement and the control effect may be improved further, the future directions for the study of the system are brought out in this thesis.