Process Control Flashcards
This should be something that is readily measured
Controlled Variables
- reliable and accurate
- adequate sensitivity
- measured in real time
- something that responds with as little delay as possible
- something that is affected by changes to other process variables
Variable that can be adjusted by the operator or control system. This should be something that is readily manipulated. This should have a direct and predictable effect on the process
Manipulated variables
Inputs that cannot be controlled by an operator or control system
Disturbances
There can be both measurable and immeasurable disturbances
The control system has to adjust for these
Control approach
- Control variables before they affect the rest of the process
- Measure the controlled variable
- Use only one manipulated variable for each control variable
- No more that one control action on each stream
- Avoid interactions
- Position measuring devices close to where the disturbance occurs
Control design methodology
- Understand the process
- Identify the operating parameters
- Identify the hazardous scenarios
- Identify what you can measure
- Identify where you can measure
- Select measurement methods
- Select control method
This is not a full list: fail safe, redundancy etc. also
One controlled variable, one manipulated variable.
We change one variable to control one other,
Single input single output (SISO)
The output has no effect on the control system: no feedback and cannot self correct. Relies on good understanding of the process. Not really control. Can only use one independent variable and other variables have to remain constant.
Open loop “control”
Also known as feedback control. The main method used in process control. The effects of the disturbances on the controlled variable are used to adjust the manipulated variable and so correct for the disturbances.
Closed loop control
Sends a signal to the final control element to maintain tank level by adjusting supply flowrate based on the error
Controller
produces a measured value for use in control system
measurement device
unpredictable changes in systems
load changes
Closed loop control
- The condition of a system is monitored by a sensor which provides this information to the controller
- The comparator in the controller deterines the value of the error signal (difference between actual and target value)
- Based on the value of the error signal, the controller transmits a signal to the final control element to change the value of the manipulated variable
Control system acts to compensate for disturbances
Difference between actual and target value
error signal
Disadvantages of closed loop control
- feedback system cannot take a corrective action until after a disturbance has upset the process
- controlled output is continually changing until set point and measured point are in agreement
- Predicative control
- Useful where final controlled variable cannot be measured
- Measure process disturbances: apply a compensating control action & can provide greater stability
- Knowledge of process is required: understanding of the effect of independent variables on dependent variables
Feed forward control
shows how information flows around the control loop
the control block diagram
Uses a digital output (DO) 24Vdc signal from the control unit as a START or STOP signal to a manipulated variable.
Based on whether controlled variable is above or below set point
ON/OFF control
Relationship controller output (or manipulated value) and the error signal is a continual mathematical function
Continuous control action
Discontinuous (ON/OFF) Control Disadvantages
- inefficient
- physically wearing to valves and switches
- noisy
Manipulated variable is able to be continually and incrementally adjusted based on the deviation of the controlled variable from the desired value
Analogue output signal from control unit (AO) 4-20mA over a range
Continuous control
Calculates an error between the setpoint and measured variable and produces an action based on combinations of the Proportion, Integral and Derivative of this error
Continuous control
Control can be proportional only (P). Or combinations of proportional and:
- Integral (PI)
- Derivative (PD)
- Both integral and derivative (PID)
PID control
Simplest form of feedback control. Output action, J, is in proportional to the deviation from the measured value.
The bigger the error, the larger the corrective action
Proportional control
J = J0 + Kcε (t)
Jo = controller output when error ε=0 (steady state signal or bias)
Kc = proportional gain (or sensitivity)
prone to “offset”. If there is a change in load, the corrective action will cause a new steady state value that differs from the original. It can be stable but it is not accurate
Proportional only
Ratio of change of the output variable (responding variable) to the change of the input variable.
Defines the sensitivity of the output variable to a given change in the input variable.
The value for proportional gain Kc affects the speed of the response.
Gain
Very high gain
means a relatively small error signal can drive the final control element to the ends of its range
- the process acts very like an ON/OFF controller
Low gain
means an error signal may only produce a slight adjustment to the final control element.
- too low gain means the process takes a long time to get near the set point
- final control element may not be using full range
Amount of input change that is required to cause a 100% change in output
- the band of controller output over which the final control element will move from one extreme to another
Proportional band
Low gain
Wide proportional band
High gain
Narrow proportional band
Output action is proportional to the time integral of the error i.e. it is proportional to both the time and magnitude of the deviation
Integral control
Output action is proportional to the time derivative of the error i.e. the slope or rate of change of the error over time
Derivative control
- Integral term eliminates offset
- integral effectively measures the average error over time
- So for the case of P only, where we see a constant offset an integral of this would show an increase over time
- The integral element therefore means that the new steady state can be achieved without continuous error or offset
Proportional integral (PI)
- Derivative action provides a signal proportional to the rate of change of the error
- Derivative action acts against overshoots and oscillation
- Derivative action allows a higher proportional gain to be used
Proportional Derivative (PD)
Effect of derivative terms
- If derivative action is set too low, the control system will not respond quickly enough to load changes
- If the rate is set too high the system will become unstable and instead of progressively reducing the errors in the controlled variable, it may actually compound them
- Problematic in noisy systems
Needed to eliminate offset
Integral
- The full effect of a corrective action applied by a controller is not immediately shown by the controlled variable
- Almost inevitably there is some delay or lag (process lag)
Lag
Uses information about the presence of an error (digital)
ON/OFF control mode
uses information about the magnitude of the error
proportional mode
uses information about the average error over a time period
integral mode
uses information about the rate of error change
derivative mode
