CONTROL SYSTEM CHARACTERISTICS
Automatic controls are used wherever a variable condition must be controlled. In HVAC systems, the most commonly controlled conditions are pressure, temperature, humidity, and rate of flow. Applications of automatic control systems range from simple residential temperature regulation to precision control of industrial processes.
CONTROLLED VARIABLES
Automatic control requires a system in which a controllable variable exists. An automatic control system controls the variable by manipulating a second variable. The second variable, called the manipulated variable, causes the necessary changes in the controlled variable.
In a room heated by air moving through a hot water coil, for example, the thermostat measures the temperature (controlled variable) of the room air (controlled medium) at a specified location. As the room cools, the thermostat operates a valve that regulates the flow (manipulated variable) of hot water (control agent) through the coil. In this way, the coil furnishes heat to warm the room air.
CONTROL LOOP
In an air conditioning system, the controlled variable is maintained by varying the output of the mechanical equipment by means of an automatic control loop. A control loop consists of an input sensing element, such as a temperature sensor; a controller that processes the input signal and produces an output signal; and a final control element, such as a valve, that operates according to the output signal.
The sensor can be separate from or part of the controller and is located in the controlled medium. The sensor measures the value of the controlled variable and sends the resulting signal to the controller. The controller receives the sensor signal, compares it to the desired value, or setpoint, and generates a correction signal to direct the operation of the controlled device. The controlled device varies the control agent to regulate the output of the control equipment that produces the desired condition.
HVAC applications use two types of control loops: open and closed. An open-loop system assumes a fixed relationship between a controlled condition and an external condition. An example of open-loop control would be the control of perimeter radiation heating based on an input from an outdoor air temperature sensor. A circulating pump and boiler are energized when an outdoor air temperature drops to a specified setting, and the water temperature or flow is proportionally controlled as a function of the outdoor temperature. An open-loop system does not take into account changing space conditions from internal heat gains, infiltration/exfiltration, solar gain, or other changing variables in the building. Open-loop control alone does not provide close control and may result in underheating or overheating. For this reason, open-loop systems are not common in residential or commercial applications.
A closed-loop system relies on measurement of the controlled variable to vary the controller output. Figure 19 shows a block diagram of a closed-loop system. An example of closed-loop control would be the temperature of discharge air in a duct determining the flow of hot water to the heating coils to maintain the discharge temperature at a controller setpoint.
In this example, the sensing element measures the discharge air temperature and sends a feedback signal to the controller. The controller compares the feedback signal to the setpoint. Based on the difference, or deviation, the controller issues a corrective signal to a valve, which regulates the flow of hot water to meet the process demand. Changes in the controlled variable thus reflect the demand. The sensing element continues to measure changes in the discharge air temperature and feeds the new condition back into the controller for continuous comparison and correction.
Automatic control systems use feedback to reduce the magnitude of the deviation and produce system stability as described above. A secondary input, such as the input from an outdoor air compensation sensor, can provide information about disturbances that affect the controlled variable. Using an input in addition to the controlled variable enables the controller to anticipate the effect of the disturbance and compensate for it, thus reducing the impact of disturbances on the controlled variable.