Applet을 실행할 수 없는 경우에는 Java Run-time Environment(JRE)을 설치하십시오. http://java.com/download/
보안설정으로 차단된 응용프로그램과 같은 문제가 발생할 경우에는 Java 제어판의 보안 레벨 설정을 참조하시기 바랍니다.
예외사항 사이트에 https://www.cheric.org를 추가한 후 브라우저를 모두 닫고 다시 시작하면 Applet을 사용하실 수 있습니다.
PROCESS CONTROL APPLET SERIES IV
Feedforward control and Cascade Control
Using this applet, user can investigate the feedforward and cascade control strategies for a jacketed tank heater.
Feedforward Control and Cascade Control
The feedback control can drive the process to the desired operating condition based on the error between the set point and process output. But the action will be taken only after the error appears in the output measurement. If a known change (disturbance) is introduced to a process and the predictable effect of change will appear with time lag, one doesnot need to wait until the change affects the process condition. In this case, feedforward control will compensate the effect of certain change based on the process model when the change is detected. Thus, feedforward control can act quickly before the change affects the output. However, if the prediction of the effect of the change from the model is not accurate, the anticipatory action will be erroneous. Therefore, it is common to use feedforward control and feedback control together. The feedforward control can enhance the dynamics but it requires disturbance measurement and a model for the prediction. If there is a secondary measurement which detects the changes in primary output sooner, a cascade control can be applied to improve the process dynamics. It is particulary useful when the disturbances are associated with the manipulated variable or when the final control element exhibits nonliear behavior.
Objectives of This Applet
This applet is designed to illustrate the followings :
- - Effect of disturbance
- - Effect of nonlinear actuator
- - Feedforwad and feedback control
- - Cascade control
This applet consists of 5 parts. The horizontal thin panels on top and bottom are the title panel and status panel, respectively. The title panel is permanent one and the status panel is changing depending on the situation to give the explanation to users. In the Middle, there are three panels. The leftmost panel shows the process and instrumentation. The middle panel shows the response of each variable and controls the run of this applet. The rightmost panel shows the PID controller and its option. In this help, only these three panels will be described.
If one of the dots in the process inlet/outlet lines is pressed, the instrumentation will show. By clicking the instrumentation it will hide. Only for the feed stream, if the display is clicked, the adjustment panel will appear and the feed flow will be increased or decreased by clicking the triangle. If the green dot in the middle is pressed an arbitrary number can be type in the top of the Controller Panel.
The button on steam line notated as 'Normal P' will toggle the pressure of steam between normal and high. If the steam pressure is high, less steam is required for the same process condition. Cascade control can compensate this type of change effectively.
This panel shows the input, reactor and jacket temperatures and setpoint. The time axis is fixed as 10 times of dominant time constant and y axis is adjusted automatically. In the lower left corner, the thumbtack figure with "freeze" label will freeze or release the current plot. If it is freezed, next plots will be drawn on top of previous plots. In this case, the y axis scale is not adjusted automatically.
There are 5 preset-run options. After selecting one option, by pressing 'RUN' button, the simulation will be proceeded after 'RUN' button is chaged to 'PAUSE'. Also, two checkboxes are present for the feedforward control and cascade control. At the bottom, two listboxes will decide the speed of simulation and the difficulties of the process nature.
The setpoint (SP) and manipulated variable (MV) can directly be changed by dragging the indicators in the master controller faceplate. Drag them slowly because the cursor may loose tracking the indicator depending on the performance of your computer. The AUTO/MAN mode can be toggled by clicking the area. By clicking the pull tab on the side of the faceplate, the parameter plate will be toggled to be shown. In cascade mode, the master controller will not be switched to 'AUTO' unless the slave controller is in 'cascade' mode and all instrumentations appear. All other options are same as Applet I.
By clicking the tab for 'Slave', the faceplate for slave controller will show up. The AUTO/MAN mode can be toggled by clicking the area. In cascade simulation, if the slave and master controllers are both in automatic mode, the 'AUTO' in slave controller is changed to 'CAS'. To switch from 'CAS' to 'MAN', the master should be in 'MAN' model first.Also, to switch from 'MAN' to 'AUTO for master controller, the slave should be in 'AUTO' first.
The feedforward controller is in the form of lead-lag module. User has to speify the feedforward gain and time constants in the equation shown in the faceplate. The input method is same as explained in previous modules.
Explain the concept of FF & Cascade control (Java Applet will help you interactively.)
To be Completed by the instructor.