Energy Management Control Systems

INTRODUCTION

Currently, almost all new buildings have some control systems to manage the operation of various building equipment, including the heating, ventilating, and air- conditioning (HVAC) systems. More elaborate control systems can simultaneously operate several pieces of mechanical and electrical equipment dispersed throughout the facility. In particular, these energy management control systems can be used to reduce and limit the energy demand of the entire facility. In the last decade, most of the advances in HVAC equipment are due to modern electronic controls, which are now cheap, flexible, and reliable.

The development of energy management and control systems (EMCS) is mostly attributed to the introduction of computerized building automation systems. In fact, energy management represents one of several tasks performed by an integrated building automation system (IBAS). Among other tasks of the IBAS include fire safety, vertical transportation control, and security regulation. Advanced IBAS include logic for interaction among lighting, HVAC, and security systems. For instance, if an automated occupancy sensor detects the presence of people in specific spaces during late hours (during night or weekends), the information can be used to adjust indoor temperature (for comfort) and to reinstate elevator service (to ensure that people can leave the building). Moreover, EMCS can provide facility operators with recommendations on maintenance needs (such as lighting fixture replacement) and alarms for equipment failures (such as motors when they burn out).

The use of energy-efficient equipment does not always guarantee energy savings. Indeed, good management of the operation of this equipment is a significant factor in reducing whole building energy use. Generally, building energy loads are continuously changing with time due to fluctuations in weather and changes in equipment use and occupancy. Thus, effective energy management requires knowledge of the facility loads. Two approaches are typically applied:

• 1. Load Tracking. The operation of equipment is modulated to respond to the actual needs in the facility. For example, the compressor in a centrifugal chiller may change speed to match the cooling demand. The actual needs of a facility can be determined by continuous monitoring. For example, the load on the chiller can be estimated if the chiller water flow and chilled water supply and return are monitored.
• 2. Load Anticipation. In some applications, the needs of a facility have to be predicted to be able to modulate the operation of the equipment adequately. For instance, in cooling plants with a thermal energy storage (TES) system, it is beneficial to anticipate future cooling loads to be able to decide when and how much to charge and discharge the storage tank. Load prediction can be achieved by analyzing the historical pattern variations of the loads.

Using monitored data and other parameters characterizing the building, energy control systems enable operators and managers to operate HVAC and lighting systems efficiently to maintain the comfort level. The following sections discuss building energy control systems and some of their applications.

BASIC CONTROL PRINCIPLES

Control Modes

Control systems are used to match equipment operation to load requirements by changing system variables. A typical control system includes four elements, as briefly described below:

• 1. The controlled variable is the characteristic of the system to be controlled (for instance, the indoor temperature is often the controlled variable in HVAC systems).
• 2. Sensors that measure the controlled variable (for instance, a thermocouple can be used to measure indoor temperatures).
• 3. Controllers that determine the needed actions to achieve the proper setting for the controlled variable [for instance, the damper position of the variable- air-volume (VAV) box terminal can be modulated to increase the air supply in order to increase the indoor temperature of the zone if it falls below a set-point].
• 4. Actuators are the controlled devices that need to be activated in order to complete the actions set by the controllers (to vary the air supplied by a VAV box, the position of the damper is changed by an actuator through direct linkages to the damper blades).

Generally, two categories of control systems can be distinguished: closed-loop systems and open-loop systems. In a closed-loop system (also known as a feedback control system), the sensors are directly affected by (and thus sense) the actions of the actuators. A typical control of a heating coil is an example of a closed-loop system. However, in an open-loop system (also called a feedforward control system), the sensors do not directly sense the actions of the controllers. The use of a timer to set the temperature of the heating coils would be an example of an open-loop system inasmuch as the time may not have a direct connection with the thermal load on the heating coils.

Figure 10.1 shows the various components and terms discussed above as well as an equivalent control diagram for a closed-loop control system for a heating coil.

Each control system can use different control modes to achieve the required objectives of the control actions. Four control modes are commonly used in operating HVAC systems:

FIGURE 10.1 Typical representations for a heating coil control system: (a) basic closed- loop control for a heating coil: (b) equivalent control diagram for the heating coil.

1. Two-Position. This control mode allows only two values (on-off or open- closed) for the controlled variable and is best suited for slow-reacting systems. Figure 10.2(a) shows the effect of two-position control on the time variation of the controlled variable (such as the air temperature due to the on-off valve position in a heating coil). In order to avoid rapid cycling, a control differential can be used. Due to the inherent time lag in the sensor response and to the thermal mass of the HVAC system, the controlled variable fluctuates with an operating range (called operating differential) with higher amplitude than the control differential. Thus, the operating differential is always higher than the control differential, as illustrated in Figure 10.2(b).

Examples of two-position controls are domestic hot water heating, residential space temperature control, and HVAC system electric preheat element.

2. Proportional. This mode has a linear relationship between the incoming sensor signal and the controller’s output. The relationship is established within an operating range for the sensor signal.

The set-point of a proportional controller is the sensor input that results in the controller output being at the midpoint of its range. Mathematically, the controller output и is given by the following equation for a proportional control:

FIGURE 10.2 Effect of two-position control on the time variation of a controlled variable: (a) two-position action when no control differential is used (rapid cycles); (b) two-position action with a control differential.

The offset or error e is the difference between the set-point and the value of the controlled variable. The proportionality constant K_p is called the proportional gain constant. The controller bias u₀ is the value of the controller output when no error exists.

As depicted by Eq. (10.1), the proportional control is not capable of reducing the error because an error is required to produce any controller action. Therefore, the controlled variable fluctuates within a throttling range as depicted in Figure 10.3.

FIGURE 10.3 Proportional control effect on the time variation of a controlled variable.

It should be noted that when the gain constant is very large, an unstable system can be obtained. Example 10.1 shows how the proportional gain constant can be determined.

Example 10.1

A hot water heating coil has a set-point of 35°C with a throttling range of 10°C. The heat output of the coil varies from 0 kW to 50 kW. Assuming that a proportional controller is used to maintain the air temperature set-point, determine the proportional gain for the controller and the relationship between the output air temperature and the heat rate provided by the coil. Assume steady-state operation.

SOLUTION

Using Eq. (10.1), the relationship between the heat rate Q, and the error in the air temperature at the coil outlet can be put in the form:

i. When the heat rate Q = Q_m,„ = 0 kW, the coil outlet air temperature is T=T = 35°C - 5°C = 30°C

ii. When the heat rate Q = Q_ma(= 50 kW, the coil outlet air temperature is T_air = Т_тгх = 35°C + 5°C = 40°C.

The proportional gain K_p can be determined as follows: or

Similarly, the constant Q₀ can be determined from or

Therefore, the relationship between the heat rate output and the air temperature for the heating coil is:

Thus, as long as the heat rate is different from Q₀ = 25 kW, the quantity (T_set._poinl- T_air), which is the error in the proportional control equation, cannot be equal to zero.

Generally, proportional controllers are used with slow stable systems that have small offset.

3. Integral. This control mode is typically incorporated with a proportional control mode to provide an automatic means to reset the set-point in order to eliminate the offset. The combination of proportional and integral actions is called “proportional-plus-integral” or simply PI control. Mathematically, the PI control can be expressed as follows:

where K, is the integral gain constant (also known as the reset rate) and has the effect of adding a correction to the controller output whenever an error exists. For HVAC systems, a typical KJK_{ ratio is less than 60 minutes.

The PI control can be applied to fast-acting systems that require large proportional bands for stability. Typical applications include mixed-air controls, heating or cooling coil controls, and chiller-discharge controls.

4. Derivative. This control action is used to speed up the response of the system in case of sudden changes. The derivative control mode is included in a combination of proportional-plus-integral-plus derivative (PID) control modes for fast-acting systems that tend to be unstable, such as duct static- pressure controls. The mathematical model for the PID control is given by Eq. (10.3):

where K_tl is the derivative gain constant. The derivative term generates a corrective action proportional to the time rate of change of the error. The ratio KJK_p is typically less than 15 minutes for most HVAC applications. If the system has a uniform offset, the derivative term has little effect. The use of PID controls is typically less common than the PI controls for HVAC systems inasmuch as no rapid control responses are needed.

To illustrate the action of P. PI, and PID control modes, Figure 10.4 compares the response of the system to an input step change. As expected, the proportional control results in an offset and the controlled variable does not reach the set-point. The correction term due to the PI control slowly forces the controlled variable to reach the set-point value. Finally, the derivative term of the PID control provides a faster action to allow the controlled variable to attain the set-point.

In addition to these conventional control modes, other intelligent controllers have been investigated in various engineering fields, including HVAC equipment controls as discussed in the following section.

FIGURE 10.4 Comparison of the reaction of three control modes to an input step change: (a) proportional (P) control action; (b) proportional-plus-integral (PI) control action time; (c) proportional-plus-integral-plus-derivative (PID) control action time.