Saturday, November 29, 2008
A process is shown in Fig. 8-1 with a manipulated input M, a load input L, and a controlled output C, which could be flow, pressure, liquid level, temperature, composition, or any other inventory, environmental, or quality variable that is to be held at a desired value identified as the set point R. The load may be a single variable or aggregate of variables acting independently or manipulated for other purposes, affecting the controlled variable much as the manipulated variable does. Changes in load may occur randomly as caused by changes in weather, diurnally with ambient temperature, manually when operators change production rate, stepwise when equipment is switched in or out of service, or cyclically as the result of oscillations in other control loops.
Variations in load will drive the controlled variable away from set point, requiring a corresponding change in the manipulated variable to bring it back. The manipulated variable must also change to move the controlled variable from one set point to another.
An open-loop system positions the manipulated variable either manually or on a programmed basis, without using any process measurements. This operation is acceptable for well-defined processes without disturbances. An automanual transfer switch is provided to allow manual adjustment of the manipulated variable in case the process or the control system is not performing satisfactorily.
A closed-loop system uses the measurement of one or more process variables to move the manipulated variable to achieve control. Closedloop systems may include feedforward, feedback, or both. Feedback Control In a feedback control loop, the controlled variable is compared to the set point R, with the difference, deviation, or error e acted upon by the controller to move m in such a way as to minimize the error. This action is specifically negative feedback, in that an increase in deviation moves m so as to decrease the deviation.
(Positive feedback would cause the deviation to expand rather than diminish and therefore does not regulate.) The action of the controller is selectable to allow use on process gains of both signs. The controller has tuning parameters related to proportional, integral, derivative, lag, deadtime, and sampling functions. A negative feedback loop will oscillate if the controller gain is too high, but if it is too low, control will be ineffective.
The controller parameters must be properly related to the process parameters to ensure closed-loop stability while still providing effective control. This is accomplished first by the proper selection of control modes to satisfy the requirements of the process, and second by the appropriate tuning of those modes.
Feedforward Control A feedforward system uses measurements of disturbance variables to position the manipulated variable in such a way as to minimize any resulting deviation. The disturbance variables could be either measured loads or the set point, the former being more common. The feedforward gain must be set precisely to offset the deviation of the controlled variable from the set point.
Feedforward control is usually combined with feedback control to eliminate any offset resulting from inaccurate measurements and calculations and unmeasured load components. The feedback controller can either bias or multiply the feedforward calculation.
Computer Control Computers have been used to replace analog PID controllers, either by setting set points of lower level controllers in supervisory control, or by driving valves directly in direct digital control. Single-station digital controllers perform PID control
in one or two loops, including computing functions such as mathematical operations, characterization, lags, and deadtime, with digital logic and alarms. Distributed control systems provide all these functions, with the digital processor shared among many control loops; separate processors may be used for displays, communications, file servers, and the like. A host computer may be added to perform highlevel operations such as scheduling, optimization, and multivariable control. More details on computer control are provided later in this section.
Tuesday, November 25, 2008
Friday, November 21, 2008
Algoritma is usually used for describing the process of management flow that should be done in orderly but sometime should be done in the same period time or in the same time, according to the process flow order.
Saturday, November 15, 2008
An integral part of the industrial resolution and the ensuing manufacturing innovations was the necessity to many different production parameters. For example, to manufacture plastic it may be necessary to maintain the temperature of a reason vessel at a constant value. If the process were not controlled, the temperature of the vessel might vary radically and create a poor-quality product or even a dangerous situation. Procedures have been developed to provide such regulation. Regulation was accomplished manually at first, through measurement, evaluation, and adjustment of the variable. Later, automatic systems were developed that could measure, evaluate, and adjust without direct human interaction. Automatic regulation of this type makes use of the feedback of the value of a variable in order to effect necessary adjustments of the process involving that variable.
The complete assembly of the three elements, measurement, evaluation, and adjustment, constitutes what is called a process control loop, where the word loop convey the idea of feedback of adjustments to the process following measurements in the process. Most industrial operations involve many variables to be regulated and thus many such process control loops. Sometimes the loop variables interact with each other so that adjustments for one variable affect a second variable, and so on. The overall process-control system is the assemblage of all these loops.
Since the late 1970s, microprocessor-based process control systems have come into wide use. Depending on the user’s need, such systems use an array of microprocessors to control from one to hundreds of variables. Further, these microprocessors are “cross-coupled,” or connected, with each other in such a way that complex adjustments can be made across a series of variables based on information traded between microprocessors.
For more detail about this algoritma may can view on the drawing of every process that available on this blog.
