Improving Comfort and Air Conditioner Performance by Optimizing Controllers under Actual Usage Conditions

Abstract

This study is to increase the efficiency of the air conditioning system under actual conditions of use. In order to improve the increase in energy consumption due to control without considering indoor cooling load fluctuations, we review it compared to conventional control methods in two respects. First, we examined the control method to reduce energy consumption by varying the evaporative pressure of the system according to the cooling load, and secondly, further reducing energy consumption by controlling the revolutions per minute (RPM) of the fan of the indoor unit at the same time as the pressure fluctuation under the cooling load. We found that changing the target pressure depending on the difference between the target temperature and room temperature can control temperature more efficiently and save energy than using fixed target pressure, and that the effect increases when applying the control of the air volume of the indoor air. Cleaning up the resulting values, the fan-load-control condition of the cooling-load-estimation control method showed an energy savings effect of 16.1–48.7% compared with the fixed-pressure control method and 1.2–37.7% compared with the cooling-load-estimation control method.

1. Introduction

As temperatures are rising owing to global warming and living standards are improving, the use of air conditioners is increasing rapidly. Accordingly, research on ways to improve indoor-space comfort and air conditioner efficiency in actual conditions should be followed by research on energy efficiency.

An inverter air conditioner can control the amount of circulating refrigerant by varying compressor driving frequency, thereby modulating cooling ability according to heat load and ultimately maintaining room temperature while reducing deviation.

Consequently, unnecessary energy loss can be reduced, and seasonal energy efficiency ratio, which is an energy efficiency index considering environmental temperature changes, is higher in inverter air conditioners than in conventional constant-speed-type air conditioners [1,2]. Riegger et al. [3] demonstrated how control errors can lead to reduced reliability and abnormal cycle problems along with decreased system capability and efficiency. Kinab et al. [4] examined the characteristics of compressor efficiency and power consumption and compared the effect of cooling load changes on the energy efficiency ratio. Luis et al. [5] summarized the energy regulations of major countries and emphasized the importance of system efficiency.

Control algorithms were developed continuously along with inverter hardware control technology. Hewitt et al. [6] modeled a system by examining the transient response characteristics of system overheating owing to disturbances using variable area expansion and electronic expansion valves. Moreover, they modeled the transient response caused by changes in the opening degree of an electronic expansion valve mathematically. Chung et al. [7] used compressor speed, compressor outlet superheat and pressure, condenser temperature, evaporator temperature, compressor inlet temperature and pressure, indoor temperature, outdoor temperature, and fan speed to configure an inverter cycle controller. The authors also considered the algorithm technology involved in optimal cycle control. Fallahsohi et al. [8] compared the proportional integral derivative (PID) and power factor correction (PFC) control methods with evaporator superheat and confirmed the energy savings effect of 2% through PFC control.

The main idea behind inverter air conditioners is variable capacity control based on the cooling load by varying the operating frequency of the compressor. Simultaneously, to match the optimum cycle, the superheat control logic through the electronic expansion valve is operated, and the refrigerant quality of each component is controlled appropriately. As an inverter control method, the fixed-pressure control method controls the refrigerant flow rate to the indoor unit to maintain the refrigerant evaporation temperature at a constant level regardless of the indoor cooling load condition. However, this control method increases the capacity of the indoor unit relative to the cooling indoor load because refrigerant at a constant temperature is supplied to the indoor unit when the indoor cooling load is small (2010, Daikin Industrial Co., Ltd., Osaka, Japan). As shown in Figure 1a, the fixed-pressure control turns off the operation of compressor when the set temperature is reached, and when the room temperature is 3 to 5 °C away from the set temperature, the compressor turns on again and repeats the operation with fixed-pressure control. Thus, as the system repeats on-off, the power consumption of the system increases. As shown in Figure 1b, the cooling-load-estimation control adjusts the operating frequency of the compressor by calculating the cooling load after the room temperature reaches the initial set temperature. Because the compressor operates at low frequencies, the temperature change in the room is minimized, and because the system operates continuously without the compressor on-off at low cooling loads, the power consumption of the system is minimized.

Various studies examined control techniques for maintaining constant indoor temperatures and saving energy under changes in a wide range of external environments. Mohammed et al. [9,10] suggested an optimal on–off control strategy for improving the AC utilization process in deserts with extremely hot weather conditions. The proposed control solution stated that Elman neural network-based estimators should be proposed to estimate the real value of outdoor temperatures, which allow the adaptive adjustment of indoor temperatures regardless of outdoor temperature changes. Conti et al. [11] presented the relevance of the dynamic analysis of a building HVAC system and potential of the hardware-in-the-loop simulation approach in assessing actual part-load heat pump performance. The authors conducted simulations by collecting continuous real-world climate data from common dwellings in Pisa, Italy, and described methods for optimizing the performance of heat pump systems in buildings under intermittent conditions. Yang et al. [12] presented a novel model predictive control (MPC) method developed for a separate sensible and latent cooling (SSLC) system with a dedicated outdoor air-system (DOAS) support, showing results compared with those of a conventional feedback-control-based building management system (BMS). MPC systems can achieve 18% and 20% energy efficiencies for single-coil AHUs and DOAS-enabled SSLCs, respectively, compared with single-coil air handling units (AHUs) controlled using a BMS. Moreover, MPC systems can improve indoor thermal comfort better than a BMS. Gao et al. [13] conducted an experimental study on control methods by combining setpoint reset techniques with bilinear control techniques. The authors presented a method to efficiently reduce the on/off frequency of a cooler and simultaneously guarantee the robustness of the spatial temperature control in the control role. Moreover, Lin et al. [14] presented HVAC system optimization control strategies related to fan coil unit (FCU) temperature control. The strategies included a dynamic FCU temperature setting, refrigerated tone conversion based on FCU temperature, and pump operating frequency regulation algorithms. The results of the study showed that the proposed strategy can achieve significant energy efficiency compared with HVAC systems operating at full load. Belic et al. [15] reviewed methods investigated in scientific papers in recent years. Most studies focused on one of the following trends: improving “classical” control methods, the use of predictable controls based on models, and the use of intelligent control methods. Krinidis et al. [16] proposed an alternative to the optimization of HVAC control systems. The authors implemented a multicriteria algorithm to integrate occupants’ thermal comfort and HVAC systems’ electricity consumption into the same function and optimized the operational control of HVAC systems. Mossolly et al. [17] employed optimal control strategies using a genetic algorithm to maintain thermal comfort and indoor air quality (IAQ). Compared with the base control strategy of the fixed temperature setpoint control method, the method of adjusting the fresh air supply rate and air temperature in each zone demonstrated up to 30.4% energy efficiency during the summer period. Song et al. [18] presented a real-time optimal control strategy for variable air volume (VAV) air conditioning in HVAC systems using genetic algorithms and simulated large-scale office buildings. The authors set supplied air temperature and duct static pressure as control variables to provide optimal control for the VAV air conditioning system.

In the previous study, the system control method was applied by controlling compressor revolutions per minute (RPM) so that evaporative and condensation pressures followed certain targets and controlled the flow rate of refrigerant. In order to cope with changes caused by heat load or ambient air environment, the compressor was operated while controlling the evaporative pressure so that the evaporative pressure was constant. Therefore, this method of controlling the compressor has a problem of high energy consumption due to the lack of operation control of the system due to changes in the cooling load in the room due to fixed target pressure.

In this work, we apply differentiated control methods in two respects compared to conventional control methods that do not take into account the variation of cooling loads in the room. First, we examined how to control the energy consumption by varying the pressure of the system according to the cooling load in the room, and secondly, we conducted a study to control the number of turns of the FAN in the room while simultaneously varying the pressure according to the cooling load in the room. It also wanted to control the indoor fan and evaporative pressure at the same time to optimize the indoor evaporative temperature, and to control the indoor space to reach the set temperature smoothly to further improve the energy consumption efficiency. Therefore, it is a study to increase the efficiency of the air conditioning system under the conditions used by actual users. We compared the accuracy and energy consumption of spatial temperature control according to compressor control methods, air volume control of indoor machines, and target pressure control of the system. The results of this study confirm that the method of changing the operating pressure according to the difference between the target temperature and ambient temperature can control the temperature more efficiently than using fixed operating pressure and save energy consumption. In addition, we find that energy savings are further increased when air volume control of indoor machines is applied simultaneously.

2. Experimental Methods

2.1. Experimental Conditions

As shown in

Table 1. Test conditions.

Ambient Condition			Indoor Fan	System Control
Target Temp.	Outdoor Temp.	Beginning Indoor Temp.
24.0	27.0	27.0	Fixed	Fixed pressure
			Fixed	Cooling load estimation
			Variable fan control	Cooling load estimation fan control
24.0	30.0	30.0	Fixed	Fixed pressure
			Fixed	Cooling load estimation
			Variable fan control	Cooling load estimation fan control
24.0	35.0	35.0	Fixed	Fixed pressure
			Fixed	Cooling load estimation
			Variable fan control	Cooling load estimation fan control

, the experiment began with the outdoor temperatures maintained at 27 °C, 30 °C, and 35 °C and the room temperature set to 24 °C. The initial temperature in the room was the same as the temperature outside. For the cycle control method, fixed-temperature, load-estimation, and fan-interlocking controls with load estimation were employed to evaluate temperature stability speed and energy efficiency according to the control method.

2.2. Experiment Equipment

The outer wall of the environmental chamber was made of expanded polystyrene, with a thickness of 70 mm. The inner/outer sides of the steel plate had a thickness of 1.0 mm, and the inner wall of the PVC with a thickness of 6.0 mm was made of aluminum.

Furthermore, the temperature control system of the environmental chamber comprised an AHU and air sampler controlling the temperature and humidity to set the environmental temperature and temperature in the space where the outdoor unit was installed. Normally, the outdoor constant temperature and humidity requirements can be met by a dry bulb temperature of −20–60 °C and a relative humidity of 30–95%.

An air sampler is a device that measures the dry and wet bulb temperatures in the outdoor unit of an air conditioning system. The AHU is a device that supplies or removes heat and moisture to set the temperature and humidity. The analog signals transmitted through the outdoor measuring instrument and sensor are collected and sent to and processed using a computer as discrete data.

3. Control Methods

The initial temperature of the chamber is Table 1′s beginning indoor temperature in both the yellow and white sections of Figure 2. When the chamber temperature in all areas stabilizes to the Beginning Door temperature, close the door of the test room and operate the test unit. During the experiment, the yellow area of the chamber is maintained at the outdoor temperature of Table 1, and the temperature of the test room (white area) is controlled by the test unit. It measures the power consumption and the temperature change of the test room while the test room reaches the target temperature and maintains that temperature.

3.1. Fixed-Pressure Control

Among the components of an air conditioner, the compressor consumes the most power. Thus, in the case of constant-speed air conditioners, the efficiency of the compressor is an important factor determining the efficiency of the product. However, in recent years, numerous inverter controls were applied to vary the operating frequency of compressors. However, if the operating frequency of a compressor is controlled by comparing the evaporator temperature with the target value without considering the cooling load, then the operation temperature of the compressor may not be controlled precisely, thereby resulting in the compressor operating frequency being turned off repeatedly. In this way, the controller degrades the energy efficiency of the system. Figure 3 shows an example of the fixed-pressure control method.

$$\Delta P=P_{Target}-P_{Eva.}$$

(1)

ΔHz = KHzp × ΔP + KHzi × dΔP/dt

(2)

The fixed-pressure control method sets the target pressure (p Target) as one value and uses the current pressure (p Eva) and error of the target pressure as the pressure error (ΔP). The pressure error (ΔP) and pressure error conversion rate (dΔP/dt) at each control cycle are calculated and multiplied using the p gain (KHzp) and I gain (KHzi), respectively, to obtain the frequency increment (ΔHz). If the error is large, then a large frequency increment (ΔHz) is obtained, and the rate at which the pressure error rate (dΔP/dt) approaches the target pressure is checked. A controller is designed to obtain a small frequency increment (ΔHz) as its speed approaches quickly. When the fixed-pressure control method is used, the operating frequency of the compressor is determined by the difference between the indoor and evaporation temperatures at a fixed target pressure independent of the set temperature. In other words, the target pressure is set to a fixed value; thus, the temperature of the space determines the operating frequency.

3.2. Load-Estimation Control

The heat-load estimation control method comprises a logic that estimates the cooling load using the integrated value of the error between the current temperature and target value in the indoor room as the control value. The steady state error is reduced using the accumulated error value, and the operating frequency of the compressor is varied while comparing the temperature of the evaporator with the target value to stably and quickly control system response. Through the use of this control logic, the result of further improving system energy efficiency is obtained. Figure 4 shows the logic of the load-estimation control method.

$$\Delta T=T_{Target}-T_{Room.}$$

(3)

ΔP = KPp × ΔT + KPi × dΔT/dt.

(4)

The cooling-load-estimation controller obtains the target pressure from the difference between the set and room temperatures. The temperature error ($\Delta T$) and temperature error rate (dΔT/dt) at each control cycle are calculated and multiplied using the p gain (KPp) and I gain (KPi), respectively, to obtain the target pressure error (ΔP). The pressure error (ΔP) and pressure error rate (dΔP/dt) from the target pressure varying with the cooling load are calculated and multiplied using the p gain (KHzp) and I gain (KHzi), respectively, to obtain the frequency increment (ΔHz). The frequency control p gain (KHzp) and I gain (KHzp) are applied similar to the fixed-pressure controllers. When the target pressure is determined from the set temperature and room temperature error, the operating frequency is determined from the set temperature and speed closest to the set temperature.

3.3. Load-Estimation-with-Fan Control

Compared with the fixed-pressure control method, which can control simple compressor operating frequency, the cooling-load-estimation control method can better reduce power consumption. However, if the cooling-load-estimation control method can precisely control the amount of heat required for cooling, then the changes in the compressor operating frequency can be reduced, thereby further improving energy efficiency. When the rotation speed of the indoor unit fan changes according to the change in the compressor operating frequency, changes in the room temperature can be reduced, thereby improving room space comfort and system energy efficiency. The following flowchart shows the control logic corresponding to Figure 5.

ΔRPM = KRPMp × ΔT + KRPMi × dΔT/dt.

(5)

The cooling-load-estimation control method is applied to the compressor, and the additional indoor air fan is reflected for control in response to the load. Fan control in the indoor air unit is achieved through the difference between the set and room temperatures rather than through the fixed RPM control method according to the set wind volume. The RPM increment (ΔRPM) is obtained by multiplying the p gain (KRPMp) and I gain (KRPMi) with the temperature error (ΔT) and temperature error rate (dΔT/dt) at each control cycle, respectively. The cooling load can be controlled effectively according to the RPM controller design (control cycle, gain size, and so on) if the load-response speed is slow or exceedingly fast owing to the compressor-control cycle and 1-Hz unit control. However, if the RPM controller reacts quickly or changes to an exceedingly large value, then the air conditioner will not cool as much as the cooling load but will demonstrate a large cooling capacity or be controlled by insufficient cooling capacity.

4. Results and Discussion

Figure 6 shows the experimental results for the fixed-pressure control, cooling-load-estimation control, and cooling-load-estimation-with-interlocking-fans control methods. With regard to compressor operating frequency fluctuations resulting from an outdoor temperature of 30 °C, the operating frequency fluctuation range decreased in the fixed-pressure control, cooling-load-estimation control, and cooling-load-estimation-with-interlocking-fans control methods. Particularly, in the fixed-pressure control method, the on/off operation of the compressor was repeated seven times. This repetitive on/off operation in the fixed-pressure control method resulted in an excessive cooling capacity. Figure 7 shows that the temperature in the room was 1.1 °C lower than the set temperature. In the cooling-load-estimation control method, the error between the indoor room temperature and target value was 0.1 °C, and the fan temperature control in the cooling-load-estimation-with-fan control method could maintain the room temperature at the level of the set temperature.

In conclusion, obtaining a result in which the temperature in the indoor room can be stably maintained at the same level of the set temperature by determining the cooling load and receiving the feedback of the difference between the set and evaporation temperatures to fluctuate the indoor fan in consideration of the change in the compressor operating frequency is possible.

Figure 8 displays the energy consumption according to the control method. Specifically, 882.0 W was used in the fixed-pressure control method; 567.7 W was consumed in the cooling-load-estimation control method; and 527.7 W was consumed in the fan-load-control condition of the cooling-load-estimation control method.

Figure 9 shows the total power consumption for 2 h after the start of the operation according to the control method for environmental temperatures of 27 °C, 30 °C, and 35 °C. Power consumption was reduced in the fixed-pressure control, cooling-load-estimation control, and cooling-load-estimation-with-fan control methods, sequentially. The greater the cooling load, such as when the environment temperature was 35 °C, the greater the benefit of energy efficiency through control.

As shown in Figure 10, the fan-load-control condition of the cooling-load-estimation control method had an energy savings effect of 16.1% to 48.7% compared with the fixed-pressure control method and 1.2% to 37.7% compared with the cooling-load-estimation control method. At the environmental temperature of 27 °C, the fan-load-control condition of the cooling-load-estimation control method had an energy savings effect of 48.7% compared with the fixed-pressure control method and 37.7% compared with the cooling-load-estimation control method. As the environmental temperature increased, the energy conservation effect decreased rapidly. When the environmental temperature was 35 °C, the fan-load-control condition of the cooling-load-estimation control method had an energy savings effect of 16.1% compared with the fixed-pressure control method and 1.2% compared with the cooling-load-estimation control method.

5. Conclusions

Fixed-pressure control methods use compressor frequency and flow control of expansion valves to fix evaporative pressure, which limits the response to cooling load fluctuations such as rising and decreasing room temperature. In this study, the cooling-load-estimation control method and the fan-load-control condition of the cooling-load-estimation control method were reviewed to overcome the limitations of cooling load fluctuation control. In the case of the fan-load-control condition of the cooling-load-estimation control method, the heat transfer rate of the heat exchanger can be varied by varying the air volume of the fan, while controlling the temperature difference between the room temperature and the evaporative temperature. These two controls enable sophisticated regulation of heat transfer rates in response to fluctuations in cooling loads in the room. The cooling-load-estimation control method can adjust the heat transfer rate only by the temperature difference between the room temperature and the evaporative temperature. As a result, cooling-load-estimation control has a lower ability to adjust the heat transfer rate due to cooling load fluctuations compared to fan-load control of the cooling-load control method. Finally, fixed-pressure control methods perform the worst in terms of energy efficiency, as they have little control over cooling load fluctuations compared to the two previously mentioned control methods.

For outdoor temperatures of 27 °C, 30 °C, and 35 °C, the room temperature fluctuations and power consumption according to the control method were confirmed through the experiments. Sequentially, temperature fluctuations were stable in the fixed-pressure control, cooling-load-estimation control, and cooling-load-estimation-with-fan control methods. When the fan air volume and number of compressor rotations in conjunction with changes in the load were controlled, comfort could be improved. Moreover, sequentially, the power consumption was high in the fixed-pressure control, cooling-load-estimation control, and cooling-load-estimation-with-fan control methods. The fan-control technique of the cooling-load-estimation control method is linked to the difference between the indoor room and evaporation temperatures to control the fan air volume to reduce indoor temperature fluctuations to improve indoor-space comfort and system energy efficiency.

The fan-load-control condition of the cooling-load-estimation control method showed an energy savings effect of 16.1–48.7% compared with the fixed-pressure control method and 1.2–37.7% compared with the cooling-load-estimation control method. Increase in environmental temperatures and the effects of energy savings tend to be inversely proportional. When the environmental temperature was 27 °C, the fan-load-control condition of the cooling-load-estimation control method had an energy savings effect of 48.7% compared with the fixed-pressure control method and 37.7% compared with the cooling-load-estimation control method. When the environmental temperature was 35 °C, the fan-load-control condition of the cooling-load-estimation control method had an energy savings effect of 16.1% compared with the fixed-pressure control method and 1.2% compared with the cooling-load-estimation control method.