Control measured variables

To keep the plant at its middle unstable steady state can be achieved by stabilizing the unstable steady state with a simple feedback control loop. For the sake of simplicity, we use a SISO (single input single output) proportional feedback control, in which the dense-phase temperature of the reactor is the controlled measured variable, while the manipulated variable can be any of the input variables of the system Yfa, FCD, etc. We use Yfa as the manipulated variable here. The set-point of the proportional controller is the dense-phase reactor temperature at the desired middle steady state in this case. Our simple SISO control law is... 

In petrochemical plants, fans are most commonly used ia air-cooled heat exchangers that can be described as overgrown automobile radiators (see HeaT-EXCHANGEtechnology). Process fluid ia the finned tubes is cooled usually by two fans, either forced draft (fans below the bundle) or iaduced draft (fans above the bundles). Normally, one fan is a fixed pitch and one is variable pitch to control the process outlet temperature within a closely controlled set poiat. A temperature iadicating controller (TIC) measures the outlet fluid temperature and controls the variable pitch fan to maintain the set poiat temperature to within a few degrees. 

Process measurements encompass the apphcation of the principles of metrology to the process in question. The objective is to obtain values for the current conditions within the process and make this information available in a form usable by either the control system, process operators, or any other entity that needs to know The term measured variable or process variable designates the process condition that is being determined. 

For regulatory control, repeatability is of major interest. The basic-objective of regulatory control is to maintain uniform process operation. Suppose that on two different occasions, it is desired that the temperature in a vessel be 80°C. The regulatoiy control system takes appropriate actions to bring the measured variable to 80°C. The difference between the process conditions at these two times is determined by the repeatability of the measurement device. 

Modern control systems permit the measurement device, the control unit, and the final actuator to be physically separated by several hundred meters, if necessary. This requires the transmission of the measured variable from the measurement device to the control unit, and the transmission of the controller output from the control unit to the final ac tuator. 

In each case, transmission of a single value in only one direction is required. Such requirements can be met by analog signal transmission. A span is defined for the value to be transmitted, and the value is basically transmitted as a percent of this span. For the measured variable, the logical span is the measurement span. For the controller output, the logical span is the range of the final actuator (e.g., valve fully closed to valve fully open). 

Analog alarms can be defined on measured variables, calculated variables, controller outputs, and the like. For analog alarms, the following possibilities exist ... 

Coupling digital controls with networking technology permits information to be passed from level-to-level within a corporation at high rates of speed. This technology is capable of presenting the measured variable brom a flow transmitter installed in a plant in a remote location anywhere in the world to the company headquarters in less than a second. 

This is a control algorithm that attempts to eliminate the offset (caused by proportional control) between the measurement and the setpoint of the controlled process variable. This control mode remembers how long the measurement has been off the setpoint. 

Instrument—A device that measures or controls a variable. 

Compensating control The process of au tomatically adjusting the control point of a controller to compensate for changes in a second measured variable. 

Measured variable A variable that is measured, and may be controlled. 

Practical activities should embody as best as possible the scientifie proeesses that have been preseribed by the American Association for the Advancement of Science observation, elassification, numerieal relations, measurements, time-spaee relations, eommunieation (oral, pictorial, written), deriving of conclusions, prediction ( what would happen if. .hypothesis making, production of operational definitions, identifieation and control of variables, experiment and explanation of experimental data. Different theoretical perspectives should be used with the aim to optimize the positive eognitive and affeetive outcomes. The use, sometimes together, sometimes separately, of different perspeetives can act complimentarily and can lead to positive results (Niaz, 1993 Tsaparhs, 1997). 

The manufacturing and quality control departments face higher costs because they have to eliminate process and measurement variability, even if they are already operating at the technological limit. They will have to add people to their staffs to mn all of the investigations and handle the additional paperwork (because malicious intent is suspected, peers and supervisors have to sign off at every step to confirm that each SOP was strictly adhered to whether the SOPs made sense, scientifically speaking, or were installed to satisfy formalistic requirements is of no interest here). 

Process automation implies the real time acquisition and control of process variables such as temperature, agitation, material delivery, or quality control measurements. As far as the MARS system is concerned, a real time process is just like any instrument. The acquisition module merely requires more interactive monitoring, alarms, and control. This can be accomplished by means of a real time multi-tasking data acquisition module. 

Bakshi and Stephanopoulos (1994b) have applied the above procedure to a fed-batch fermentation process. The problem involved 41 sets of batch records on 24 measured variables. Of these variables only very few were found by the decision tree to be relevant, and yield rules such as the following for guiding the diagnosis or control of a fermentor. 

Equilibrium relationships (e.g., Henry s law, relative volatilities, etc.). Controller equations with an input variable dependent on a measured variable. 

The components of the basic feedback control loop, combining the process and the controller can be best understood using a generalised block diagram (Fig. 2.29). The information on the measured variable, temperature, taken from the system is used to manipulate the flow rate of the cooling water in order to keep the temperature at the desired constant value, or setpoint. This is illustrated by the simulation example TEMPCONT, Sec. 5.7.1. 

In the basic conventional feedback control strategy the value of the measured variable is compared with that for the desired value of that variable and if a difference exists, a controller output is generated to eliminate the error. 

In control situations with more then one measured variable but only one manipulated variable, it is advantageous to use control loops for each measured variable in a master-slave relationship. In this, the output of the primary controller is usually used as a set point for the slave or secondary loop. 

Feedback control can never be perfect as it only reacts to the disturbances which are already measured in the system output. The feed-forward method tries to eliminate this drawback by an alternative approach. Instead of using the process output, the measured variable is taken as the measured inlet disturbances and its effect on the process is anticipated via the use of a model. The action is taken on the manipulated variable using the model to relate the measured variable at the inlet, the manipulated variable and the process output. The success of this control strategy depends largely on the accuracy of the model prediction, which is often imperfect as models can rarely predict the... 

An adaptive control system can automatically modify its behaviour according to the changes in the system dynamics and disturbances. They are applied especially to systems with non-linear and unsteady characteristics. There are a number of actual adaptive control systems. Programmed or scheduled adaptive control uses an auxiliary measured variable to identify different process phases for which the control parameters can be either programmed or scheduled. The "best" values of these parameters for each process state must be known a priori. Sometimes adaptive controllers are used to optimise two or more process outputs, by measuring the outputs and fitting the data with empirical functions. 

We use a simple liquid level controller to illustrate the concept of a classic feedback control system.1 In this example (Fig. 5.1), we monitor the liquid level in a vessel and use the information to adjust the opening of an effluent valve to keep the liquid level at some user-specified value (the set point or reference). In this case, the liquid level is both the measured variable and the controlled variable—they are the same in a single-input single-output (SISO) system. In this respect, the controlled variable is also the output variable of the SISO system. A system refers to the process which we need to control plus the controller and accompanying accessories such as sensors and actuators.2... 

To reduce derivative kick (the sudden jolt in response to set point changes), the derivative action can be based on the rate of change of the measured (controlled) variable instead of the rate of change of the error. One possible implementation of this idea is in Fig. 5.3. This way, the derivative control action ignores changes in the reference and just tries to keep the measured variable constant.2... 

Figure 5.3. Implementation of derivative control on the measured variable.

The first item on the agenda is process identification. We either derive the transfer functions of the process based on scientific or engineering principles, or we simply do a step input experiment and fit the data to a model. Either way, we need to decide what is the controlled variable, which is also the measured variable. We then need to decide which should be the manipulated variable. All remaining variables are delegated to become disturbances. 

We can use a block diagram to describe Fig. 10.1. Cascade control adds an inner control loop with secondary controller function Gc2 (Fig. 10.2a). This implementation of cascade control requires two controllers and two measured variables (fuel... 

A control mechanism is introduced that makes changes to the process in order to cancel out the negative impact of disturbances. In order to achieve this, instruments must be installed to measure the operational performance of the plant. These measured variables could include... 

To be able to control implies that there is some means of manipulating a variable. The element that makes the change in the variable is called the final control element. This is a valve, switch, or other item that is activated by the controller in order to maintain the measured variable at some desired value or within some set limits. For instance, the level in Figure 7-1 is controlled by opening and closing a valve that changes the flow rate out of the tank. The valve is the final control element. 

A controller is a device that generates an output signal based on the input signal it receives. The input signal is actually an error signal, which is the difference between the measured variable and the desired value, or setpoint. 

Figure 12 shows the input to output, characteristic waveform for a two position controller that switches from its "OFF" state to its "ON" state when the measured variable increases above the setpoint. Conversely, it switches from its "ON" state to its "OFF" state when the measured variable decreases below the setpoint. This device provides an output determined by whether the error signal is above or below the setpoint. The magnitude of the error signal is above or below the setpoint. The magnitude of the error signal past that point is of no concern to the controller. 

With proportional control, the final control element has a definite position for each value of the measured variable. In other words, the output has a linear relationship with the input. Proportional band is the change in input required to produce a full range of change in the output due to the proportional control action. Or simply, it is the percent change of the input signal required to change the output signal from 0% to 100%. 

Steam is admitted to the heat exchanger to raise the temperature of the cold water supply. The temperature detector monitors the hot water outlet and produces a 3 to 15 psi output signal that represents a controlled variable range of 100° to 300°F. The controller compares the measured variable signal with the setpoint and sends a 3 to 15 psi output to the final control element, which is a 3-in control valve. 

The combined action of the controller and control valve for different changes in the measured variable is shown in Figure 17. 

Initially, the measured variable value is equal to 100°F. The controller has been set so that this value of measured variable corresponds to a 100% output, or 15 psi, which in turn, corresponds to a "full open" control valve position. 

At time tb the measured variable increases by 100°F, or 50%, of the measured variable span. This 50% controller input change causes a 100% controller output change due to the controller s proportional band of 50%. The direction of the controller output change is decreasing because the controller is reverse-acting. The 100% decrease corresponds to a decrease in output for 15 psi to 3 psi, which causes the control valve to go from fully open to fully shut. 

At time tj, the measured variable decreases by 50°F, or 25%, of the measured variable span. The 25% controller input decrease causes a 50% controller output increase. This results in a controller output increase from 3 psi to 9 psi, and the control valve goes from fully shut to 50% open. 

If the measured variable drops below the setpoint, a positive error is developed, and the control valve opens further. If the measured variable goes above the setpoint, a negative error is developed, and the control valve throttles down (opening is reduced). The 50% proportional band causes full stroke of the valve between a +50°F error and a -50°F error. 

If a large difference exists between the setpoint and the measured variable, a large error results. This causes the final control element to change position rapidly. If, however, only a small difference exists, the small error signal causes the final control element to change position slowly. 

If the measured variable decreases from its initial value of 50 gpm to a new value of 45 gpm, as seen in Figure 21, a positive error of 5% is produced and applied to the input of the integral controller. The controller has a constant of 0.1 seconds 1, so the controller output rate of change is 0.5% per second. 

The controller acts to return the process to the setpoints. This is accomplished by the repositioning of the control valve. As the controller causes the control valve to reposition, the measured variable moves closer to the setpoint, and a new error signal is produced. The cycle repeats itself until no error exists. 

---