Kamis, 25 Januari 2007
Energy Analysis of An Air Conditioning System Using PID And Fuzzy Logic Controllers
ABSTRACTReducing energy consumption and ensuring thermal comfort are two important considerations in designing an air conditioning system. Alternative approach to reduce energy consumption proposed in this study is to use a variable speed compressor. Two control strategies were proposed, which are proportional plus integral plus derivative (PID) and fuzzy logic controllers. An air conditioning system, originally operates on an On/Off control mechanism, was retrofitted to enable the implementation of the controllers. Measurements and computer interface systems were designed and software to implement the controller algorithms was developed using Visual Basic. The system was installed to a thermal environmental room together with a data acquisition system to monitor the temperature of the room, coefficient of performance, energy consumption and energy saving. Measurements were taken during the two hours experimental period at a time interval of five minutes for temperature setpoints of 20, 22 and 24 C with internal heat loads of 0, 500, 700 and 1000W. Each controller was tuned for the best performance. The results indicate that thermal comfort of the room together with significant energy saving can be obtained through a proper selection of controller parameters. Energy analysis shows that PID and fuzzy logic controllers are better than On/Off control mechanism. Generally, fuzzy logic controller is better than PID controllers. However, conventional controllers such as PID or its combinations are still capable of controlling the space temperature with some amount of energy saving but at the expense of the time to tune the controller parameters. A new PID tuning method based on trial and error was therefore proposed. This study shows that using variable speed compressor and choosing suitable control strategy, the space temperature is able to be controlled with significant energy saving.
CONCLUSION
The basic operating principle of air conditioning remains relatively unchanged over the years, that is to regulate indoor temperature with a cyclic On/Off operation. The air conditioner will automatically turns itself Off once the room temperature drops below the set temperature, and will then turns back On when the room temperature rises above the set temperature. This makes it necessary for the air conditioner to repeatedly turn itself On and Off, and a great deal of electricity is wasted due to fluctuations in the room temperature above and below the temperature setting. In the case of a variable speed control, however, it is possible to maintain a fixed temperature without having to repeatedly turn the motor of the compressor On and Off. The compressor changes the compression rate of the refrigerant, thereby making it possible to adjust the room temperature. In actuality, this adjustment is carried out by changing the rotational speed of the motor of the compressor. Since the rotational speed can be smoothly controlled over wide range by varying the frequency of the motor, the controller makes it possible for air conditioners not only energy efficient, but also capable of a fine-tuned temperature control.
The inverter allows having more than one temperature setting. Using this system, the inverter can be set for more conditions such as raining season, summer, night and morning. For example, during raining season the ambient temperature is lower and the room is unexposed to solar radiation. This results with a lower room temperature. Although air conditioner system remains to be operated then the inverter can be selected to low or medium frequency to maintain a comfortable room condition. The operating frequency of the motor increases with the increment of motor power, and consequently, the energy consumption of the motor increases. For every 5 Hz decrease in the motor frequency it was found that the increase in energy save is about 10.5%.
The change in motor frequency affects the room temperature, compressor performance and energy consumption in an air conditioning. The effects of the increase in motor frequency are :
- Decrease in room temperature.
- Decrease in pressure at discharge in compressor, suction in condenser and suction in expansion valve.
- Increase in pressure at suction in compressor.
- Increase in cooling capacity.
- Increase the compression ratio.
- Decrease actual and Carnot COP.
- Increase motor performance such as speed, voltage, current and power.
- Increase the energy consumption.
Compared with a conventional air conditioning (On/Off control), when the motor compressor is turn Off: the room temperature slowly increases, the power requirements by the compressor and the motor are zero. Significant energy saving is obtained when the motor is not working. The longer the motor compressor is Off more energy saving is obtained.
The problem with designing a PID controller, is selecting the values for the PID parameters i.e. Kp, Ki and Kd. When the plant is unknown, the PID design problem is essentially one of trial and error, and it becomes increasingly more difficult in the presence of large dynamic interactions, nonlinearities and parameter variation in the plant. In particular, the PID controller parameters must be continuously adjusted (or tuned) in order to meet the performance criteria.
The tuning of PID controller parameters such as the proportional, integral and derivative gains affects the room temperature and the energy consumption in air conditioning applications. High proportional gains will decrease the room temperature oscillation at steady state and energy consumption. An effect is observed with derivative gains. The room temperature oscillation at steady state and energy consumption vary in a quadratic manner when the integral term is increased. The controller gains to obtain a good response of the room temperature and optimum energy consumption for the system studied were found to be 3.3, 0.180 and 0.040 for the proportional, integral and derivative terms, respectively. PID controller ensures a lower energy consumption and gives the highest saving followed by PD, P and PI in comparison with the conventional On/Off control.
Fuzzy logic control is one of the many types of controller that have been investigated. The implementations of variable speed controller for energy saving and thermal comfort using fuzzy logic control have been conducted. The optimum performance of fuzzy logic control can be obtained by modifying the initial set of rule parameters. The rule set can be modified by evaluating the dynamic process curve of the controlled parameters. The modification of the rules affects the thermal comfort and energy consumption. In comparison with PID and On/Off control (conventional controllers), the following can be deduced :
- The design of the fuzzy logic control does not require complicated mathematical equations. It requires only a set of basic rules to form the decision table. Basically all of the rules in the decision table are based on human experience (not available compared with On/Off control).
- A faster respond in reaching the setpoint temperature is obtained using fuzzy logic control (was found to be five minute for all temperature setting).
- Tuning fuzzy logic controller is faster or easier than tuning PID controller.
- Fuzzy logic control gives the highest saving, while PID controller is better at temperature setting of 24 C (i.e. at internal heat loads of 0 and 500 W).
Fuzzy logic control systems have been successfully applied to a wide variety of applications. These controllers perform at least as well, and in most cases better, than conventional controllers, especially when the plant is difficult to model, and when there is significant knowledge from human engineers is available. The study has shown that the performance of fuzzy logic control is as good as or better than the conventional control in term of thermal comfort and energy saving for an air conditioning application. Fuzzy logic control is easy to understand and it is simple to implement. However, conventional controllers such as P, PI, PD and PID is still capable to control the space temperature with some amount of saving but at the expense of the time required to tune the controller parameters.
Dr. Ir. Henry Nasution, M. T
Staff Jurusan Teknik Mesin
Fakultas Teknologi Industri
Universitas Bung Hatta