- About this Journal
- Abstracting and Indexing
- Aims and Scope
- Annual Issues
- Article Processing Charges
- Articles in Press
- Author Guidelines
- Bibliographic Information
- Citations to this Journal
- Contact Information
- Editorial Board
- Editorial Workflow
- Free eTOC Alerts
- Publication Ethics
- Reviewers Acknowledgment
- Submit a Manuscript
- Subscription Information
- Table of Contents
Advances in Mechanical Engineering
Volume 2013 (2013), Article ID 481213, 7 pages
Experimental Study on Intelligent Control Scheme for Fan Coil Air-Conditioning System
The College of Architecture and Civil Engineering, Beijing University of Technology, Beijing 100124, China
Received 3 July 2013; Accepted 5 October 2013
Academic Editor: Song Pan
Copyright © 2013 Yanfeng Li et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
An intelligent control scheme for fan coil air-conditioning systems has been put forward in order to overcome the shortcomings of the traditional proportion-integral-derivative (PID) control scheme. These shortcomings include the inability of anti-interference and large inertia. An intelligent control test rig of fan coil air-conditioning system has been built, and MATLAB/Simulink dynamics simulation software has been adopted to implement the intelligent control scheme. A software for data exchange has been developed to combine the intelligence control system and the building automation (BA) system. Experimental tests have been conducted to investigate the effectiveness of different control schemes including the traditional PID control, fuzzy control, and fuzzy-PID control for fan coil air-conditioning system. The effects of control schemes have been compared and analyzed in robustness, static and dynamic character, and economy. The results have shown that the developed data exchange interface software can induce the intelligent control scheme of the BA system more effectively. Among the proposed control strategies, fuzzy-PID control scheme which has the advantages of both traditional PID and fuzzy schemes is the optimal control scheme for the fan coil air-conditioning system.
The flexibility of fan coil unit (FCU) air-conditioning system makes it able to use the building space more efficiently and reduce fan noise. It is suitable for personalized rooms, facilitating flexible and easy adjustment according to different requirements for energy saving. In China, it has been commonly used in office buildings; hotels; hospitals; commercial, residential, and research establishments; and recreation and other public buildings . The most widely used form of fan coil units is an electric water valve of twin tube, three gears, and on-off control. The traditional control method of FCU can meet energy-saving requirement to a certain degree. Optimal control method can not only improve the control precision of FCU air-conditioning system but also increase energy efficiency of centralized air-conditioning systems.
Researches on optimal control of FCU air-conditioning system have mainly focused on the control strategy and the development of thermostat for FCU. Liao , Mou , and Tian and Liu  proposed the improvement of fuzzy control by introducing the optimization and feedback into the process of logical reasoning. Through combination of fuzzy control with traditional PID control form, a new type of fuzzy inference method was developed, which can automatically reason using different parameters. This method has been used to control the temperature of the air-conditioned rooms. The results of simulations showed that this new type of fuzzy inference method in the temperature control system has good control quality and robustness. Zhao and Wang  proposed a rule of adaptive fuzzy controller which consisted of alpha factor and ku-factor adjustment. The simulation results showed that the adaptive fuzzy controller is more effective than traditional PID control system in steady-state accuracy, dynamic characteristics, and robustness of the control system. Guo et al.  introduced air-conditioning genetic algorithm into the fuzzy control system, which can find the optimal value automatically by collaborating with the input variable membership functions and control rules. The test results showed that this type of control system had succeeded in achieving a stable, agile, energy saving based multiobjective optimization. Chi et al.  and Zhao et al.  used the duty cycle fuzzy control method in the electric valve control of the fan coil. The results showed that this strategy could achieve a good control effect and is more energy efficient than the traditional control method.
Most previous studies were based on the simulation and experimental results in laboratory. The proposed methods have not been applied in practical projects. At present, building automation (BA) system is widely used in fan coil air-conditioning systems. In order to better implement the intelligent control programs in practical engineering, an experiment rig was built by combining the intelligent control with the traditional control. Based on the measured data of BA control system and Simulink simulation software which has the advantages of the flexibility in realizing various control strategies, a data exchange interface was designed. It could serve as a bridge between the build control system and simulation software. By using the data exchange interface, the fuzzy controller could obtain real-time data in the MATLAB simulation. If we can read or write the AV/BV points which are supported by building automation control network (BACnet) protocol in the building control system, traditional control method can be combined with the intelligent control strategy. Thus, without additional hardware device, a control method based on the BA system can be realized.
In this paper, experiments on fan coil air-conditioning systems characteristics with traditional PID control method, fuzzy control method, and fuzzy-PID control method would be carried out in the experiment rig.
2. Intelligent Control Experiment Rig of Fan Coil Air-Conditioning System
2.1. Experimental Subjects
Fan coil air-conditioning system intelligent control experiment rig is based on the actual air-conditioning system in the Building Environment and Equipment Engineering (BF&EP) Laboratory at Beijing University of Technology. Eleven air-conditioned rooms on the second floor, as shown in Figure 1, were chosen as experimental subjects. The total construction area of eleven rooms is 206.3 m2 and the height of the room is 3 m. The air-conditioning system is implemented by fan coil units with twin tube, three gears, and electric two-way water valve. On the test day, the total cooling load of the office was 22970 W and the maximum air flow was 6266 m3/h. Room number 204 which is northward was selected as a sample. The gross floor area of room number 204 was 10.23 m2 (3.3 m × 3.1 m). The fan had three different air volumes corresponding to three fan gears, that is, 340 m3/h for high volume, 260 m3/h for middle volume, and 170 m3/h for low volume. The corresponding input fan powers were 31 W, 26 W, and 21 W, respectively. The largest cooling load of room number 204 was 1718.05 W, which was recorded at 2:00 pm on the test day.
2.2. Mechanisms of Experiment Rig
Figure 2 presents the schematic diagram of the fan coil air-conditioning intelligent control experiment rig. The basic structure of the experiment rig can be divided into three parts. The first part is the control system that consisted of Alerton building automation control system and Simulink simulation software. The second part is the data exchange system constituted by the building automation control Windows software. The third part is the equipment system that is composed of the air-conditioned rooms (including fan coil units, indoor cooling equipment, and envelope structure), heating and cooling systems (the air source heat pump, circulating water pump), temperature sensor, valve and valve actuator.
In the Alerton building automation control system, the indoor air temperature can be measured by the indoor temperature sensor Microset II. Data can be exchanged between Alerton and Simulink through the data exchange interface, that is, BAC Window. The deviation of the indoor temperature and change rate of temperature deviation are sent to fuzzy controller as input signals. These input signals go through the process of fuzzification, fuzzy logic decision making, and defuzzification. The results of the process are sent back to the Alerton system. Then, the Alerton system can control the indoor temperature of the air-conditioned room by adjusting fan gears including high, middle, low, and stop.
In the rig, the data exchange system serves as a bridge and can introduce the scheme of intelligent control into BA system. Its main part is BAC Window data exchange interface software. The interface of BAC Window is shown in Figure 3.
3. Traditional PID Control Experiment
3.1. Traditional PID Control Programming
The algorithm of traditional PID control used in FCU air-conditioning system intelligent control experiment rig is given as follows : where is the controller output, is the scale factor, is the integral coefficient, is the differential coefficient, and is the difference between the feedback signal and the set point.
The traditional PID control strategy built by the Alerton system is shown in Figure 4. PI module means PI controllers; SP means the set room temperature; FB means the measured temperature. AV-26 is the output value of PID controller; its value is the actual air volume according to the difference between the actual temperature and set temperature value. The diagram of traditional PID control is shown in Figure 5.
3.2. Analysis of Experimental Results by the Traditional PID Control
In the experiment of traditional PID control method, the following control parameters were selected by optimized adjustment. was 15 and was 9. The calculated results were given as AV-26, as shown in Figure 4. For the VA-26, four regions are divided and each interval corresponds to a fan gear. For example, the region from 0 to 33 corresponds to off, the region from 33 to 50 corresponds to low fan gear, the region from 50 to 75 corresponds to the middle fan gear, and the region from 75 to 100 corresponds to high fan gear. For different temperatures environments, the indoor air temperature of the conditioned room could be controlled by adjusting fan gear automatically according to the partition of region.
A day of typical ambient temperature variation was selected and the experimental results are shown in Figure 6. The initial room temperature was 29.9°C and the outdoor air temperature was 30.9°C. The target indoor air temperature was set at 24.0°C and the temperature fluctuation range of the room was 24 ± 1°C which meets the requirements of air-conditioned comfort in an office room. At night, the indoor temperature ranged from 23.5°C to 24.5°C, with the steady-state deviation of ±0.5°C. In the daytime the temperature fluctuation increased during office hours (from 9:00 am to 9:00 pm). It can be seen from Figure 6 that the indoor temperature and the outdoor ambient temperature have similar curve trends. It can also be seen in this figure, from 13:02:00 to 15:18:00, when the outdoor temperature dropped by 2.4°C, the indoor temperature decreased correspondingly.
It means that the traditional PID control system does not resist outside interference effectively, especially interference caused by outdoor ambient temperature fluctuation.
Traditional PID control is based on the accurate mathematical model of control system. However, the fan coil air-conditioning system has the characteristics of time delay, being nonlinear, and a variety of interference factors. Therefore, the traditional PID control system cannot resist outside interference effectively especially for the system with high parameter degeneration [10, 11].
4. Intelligent Control Experiment
The fuzzy PID control system combines the fuzzy controller with the traditional PID system. The task of fuzzy PID control system is to find the relationship between three parameters of PID control system and error and error change rate. The fuzzy controller keeps adjusting the three parameters online to meet the requirement of error and change rate of error according to the fuzzy control rules.
4.1. Design of Fuzzy Controller
The fuzzy controller is established by fuzzy logical toolbox in MATLAB. The air-conditioned room temperature deviation () and error change rate () are set as the fuzzy controller input variables, and the fan flow () is set as the fuzzy controller output variable . It is a two-dimensional controller with dual input and single output. The triangle membership function was selected. Nearest rounding method is used for fuzzy exact amount, and MIN-MAX-gravity method is used for defuzzification of blur amount. The domain of temperature deviation of input variable was 6. The input variable has a discourse domain and a set of three linguistic variables (PB, PM, PS, O, NS, NM, and NB) containing the information about the degree of the accelerating or decelerating in the linguistic terms of the change in error. Seven fuzzy grades are divided as NB (negative big), NM (negative medium), NS (negative small), O (zero), PS (positive small), PM (positive medium), and PB (positive big); refers to the fuzzy inference value of the fan speed. Defuzzification of fan speed is carried out according to the specified range of calculated which is adopted as . NB, NM, and NS corresponded to −2, −1.5, and 0.5, respectively. PS, PM, and PB corresponded to 0.5, 1.5, 2, respectively. When was lower than 0.5, fan speed gear was low; when was between 0.5 and 1.5, fan speed gear was medium; when was larger than 1.5, fan speed gear was high. The rules design of the fuzzy controller is listed in Table 1.
The schematic diagram of fuzzy control system is shown in Figure 7. In this system, the traditional PID control controller (Figure 5) is replaced by the fuzzy controller. The change rate of the temperature deviation is set as input signals to the fuzzy controller, and then the flow volume of fan can be adjusted according to the fuzzy control rules stated in Table 1.
The fuzzy PID control system shown in Figure 8 is a combination of fuzzy control and traditional PID control mode. The threshold interval is used to switch between these two modes of control; therefore, the advantages of two control modes can be integrated.
4.2. Analysis for Fan Coil Air-Conditioning System with Different Control Schemes
For room number 204, target indoor temperature was set at 24.5°C. The actual indoor temperatures were shown in Figure 9 when different control schemes were applied. The processed results of data are listed in Table 2.
The different features of three control system can be analyzed from four aspects.(i)Stability. Trends of indoor and outdoor air temperature show that, when an outside disturbance occurs, fuzzy PID control system and fuzzy control system can maintain the relative stability and invariance of indoor air temperature better than traditional PID control system. This means that fuzzy PID control system and fuzzy control system have a stronger anti-interference ability.(ii)Static Characteristics. As shown in Table 2, certain temperature deviation exists in fuzzy control system. Its higher deviation is 0.3°C and lower one is zero. For fuzzy PID control system, both the higher and lower deviation are 0.3°C. However, the range is smaller than that of traditional PID control system. It can be concluded that the fuzzy PID control system can reduce the temperature deviation compared with the traditional PID control system. Compared with the fuzzy control system, it can eliminate the static deviation and achieve high precision.(iii)Dynamic Characteristics. The range of overshoot of traditional PID control is 3.06%. For fuzzy PID control, overshoot is 1.02%. It was reduced by 66.7% compared with the traditional PID control. Likewise, the peak time of fuzzy PID control is 42.5% shorter than that of traditional PID control, the transition time is 44.0% shorter, and the delay time is 45.0% shorter. The results indicate that fuzzy PID control has the characteristics of short transition time, small overshoot, and fast response.(iv)Economy. The power consumption of fan coil system with fuzzy PID control is much lower than that of traditional PID control. For the traditional PID control system, the power consumption is about 17 W. In fuzzy PID control system, the value is 10.29 W. The power consumption was reduced by 39.5%–43.3%. It means that, under the premise of the suitable fan gears distribution, the fuzzy PID control system is more energy efficient and more economical compared with traditional PID control and fuzzy control system.
A data exchange system that combines the traditional control and intelligent control schemes has been developed for the FCU system in an office building. It introduced the combined schemes into BA system, and a test rig for controlling the FCU system was built accordingly. The performance of the traditional PID control, the fuzzy control, and the fuzzy PID control was tested. Some conclusions have been drawn from the test.
Compared with the traditional PID control, the fuzzy PID control is characterized by small overshoot, fast responses, high precision, and a strong anti-interference performance. Compared with fuzzy control, the fuzzy PID control is characterized by elimination of the steady-state error and high precision. For a reasonable distribution of fan gears, compared to traditional PID control and fuzzy control, the fuzzy PID control system is more energy efficient and more economical in terms of fan power consumption. Combining the advantages of fuzzy control and traditional control, the fuzzy PID control is suited to the FCU system which is a strong time variant, nonlinear system with multi-interference. It will make the FCU system operate more efficiently at a lower energy cost. The energy saving effect will be enhanced significantly if on-off controller in FCU is replaced by the fuzzy PID controller.
With the data exchange system composed by the BA control Windows interface software, the FCU system intelligent control rig can achieve seamless integration between the intelligent control and the BA system without additional hardware.
This work is supported by the open project of Key Laboratory of Urban Security and Disaster Engineering, Beijing University of Technology (Grant no. 0040005466120160), and the Innovation Team Plan of Beijing Institute of Science and Technology (Grant no. IG201206N).
- L. Han, The study on the fan-coil unit with fresh air system on dry condition [M.S. thesis], Harbin Institute of Technology, 2011 (Chinese).
- Z. Liao, “The Fuzzy-PID control technique of parameter auto-tuning based on self-learning and its application in central air conditioner system,” Programmable Controller & Factory Automation, vol. 10, pp. 131–138, 2006 (Chinese).
- Y. Mou, “Temperature control based on optimal Fuzzy-PID,” Computer Simulation, vol. 26, no. 5, pp. 195–197, 2009 (Chinese).
- L. Tian and X. Liu, “The design of temperature fuzzy controller for central air-conditioning room,” Application of Electronic Technique, vol. 4, pp. 40–41, 2002 (Chinese).
- R. Zhao and X. Wang, “Application of fuzzy-PID control in central air-conditioning system,” Computer Simulation, vol. 23, no. 11, pp. 311–313, 2006 (Chinese).
- X. Guo, Y. Rui, J. Li, and S. Zhu, “Research on the control system of fuzzy intelligence based on genetic algorithm,” Journal of Soochow University Engineering Science Edition, vol. 24, no. 5, pp. 38–41, 2004 (Chinese).
- G. Chi, J. Zhang, and T. Zhao, “Simulation of duty ratio fuzzy control of fan coil units,” Heating Ventilation & Air Conditoning, vol. 40, no. 12, pp. 80–84, 2010 (Chinese).
- T. Zhao, L. Ma, and J. Zhang, “Experimental analysis on duty ratio fuzzy control for fan-coil unit,” Journal of Civil, Architectural and Environmental Engineering, vol. 12, pp. 92–99, 2010 (Chinese).
- Z. Huang, Principle of Building Automation System, China Architecture & Building Press, Beijing, China, 2003 (Chinese).
- D. L. Loveday, G. S. Virk, J. Y. M. Cheung, and D. Azzi, “Intelligence in buildings: the potential of advanced modeling,” Automation in Construction, vol. 6, pp. 447–461, 1997.
- P. Riederer, D. Marchio, J. C. Visier, and R. Lahrech, “Room thermal modeling adapted to the test of HVAC control systems,” Building and Environment, vol. 37, pp. 777–790, 2002.
- S. Li, Fuzzy Control, Harbin Institute of Technology Press, Harbin, China, 2011 (Chinese).