Thermal comfort and energy conservation based automatic air conditioner control system

In this thesis, a smart climate control system that operates like a user who controls an air conditioner (AC) via its remote control, is introduced. The smart climate control system sets AC to users thermal comfort temperature automatically, also increase in sustainability is achieved by the design. Today, people are using the air conditioning systems more intensely than ever to provide a comfortable, modern and quality living environment. While use of air conditioners is increasing with increase in income, electricity spending is also increasing in parallel. In addition, according to the simulation results, it is predicted that global warming will reduce the need for heating by 34% in the year 2100, but the cooling requirement is expected to increase by 72%. It is therefore worthwhile to investigate how to increase thermal comfort of air conditioning usage and reducing the energy requirement from intensive use. The smart climate control system is composed of software and hardware that contains accessibility over internet, communication with sensors, nonlinear equation solving, optimisation algorithms, AC control algorithm and hardware which is responsible for running these software factors at the same time without having performance issues. As hardware, a popular and powerful development board Raspberry Pi is used as controller to have fast development schedule by means of its open software and community support. In addition, three temperature and humidity sensor and 1 infrared transmitter are used as nodes of the controller. As software platform, Linux is used as operating system and mentioned software parts are implemented separately as programs. Programs like optimisation algorithm and nonlinear equation solving are written in C language to reduce calculation time. On the other hand, web server and PI controller algorithm are written in Python language for fast development because calculation time is not the main priority in that case. The smart controller is designed to be able to work with any kind of AC, which has an infrared remote control. Such compatibility is achieved by recording infrared signals for every button that is sent to AC when pressed. Therefore, smart controller acts as a real person who has the remote control and sending temperature commands. Decision making algorithm for determining which temperature command is going to be sent to AC makes the controller smart. The algorithm, determines thermal comfort temperature with information that comes from temperature and humidity sensors under constant test conditions. Povl Ole Fanger's Predicted Mean Vote (PMV) function, which is derived from the results obtained from the thermal comfort experiments on humans and the mathematical model of the thermal relationship of the human body with its environment is the core component of the thermal comfort temperature algorithm. The PMV value obtained as result of the PMV function, represents a point in thermal comfort scale which is defined in ISO 7730 and ASHRAE Standard 55 and it predicts the thermal comfort of a large number of people. Thermal comfort scale varies between +3 and -3 which means very hot conditions and very cold conditions, respectively. A value of 0 according to the PMV formula indicates the most favourable ambient conditions and even at this value on the scale, statistically 5% of people do not feel comfortable in a thermal sense. PMV formula is a very popular method in thermodynamic science and thermal comfort measurement. In ISO, it is specified that PMV value generated by the formula should be between +0.5 and -0.5 in order to have thermal comfort for human beings. When designing the control algorithm to determine thermal comfort temperature, staying inside of the specified ISO range and energy efficiency are ensured by selecting target PMV as +0.25 for summer conditions and -0.25 for winter conditions. Because air temperature is an input argument to find PMV value, the PMV formula is needed to be reversed. Taking reverse of PMV formula is a nonlinear task as the function itself. Therefore, Particle Swarm Optimisation (PSO) is used for finding PMV value and while taking reverse of the function to find thermal comfort temperature. PSO is a population based stochastic optimization technique inspired by social behaviour of bird flocking or fish schooling. The smart climate control system manages the air temperature of the test environment to stay in thermal comfort temperature by PI controller that is designed to have zero overshoot while being as fast as possible. Firstly, thermal transfer function of the test environment is obtained by applying the stepped temperature inputs to the air conditioner. After the thermal transfer function of the test environment is achieved, a PI controller was designed to eliminate the steady state errors due to the shape of the room and the position of the air conditioner. In addition, temperature oscillations made by the air conditioner which are related to the position of the room and the air conditioner are reduced. Thanks to designed thermal comfort and energy conservation based automatic air conditioner control system, AC temperature can be set with the resolution of 0.1 ⁰C. In the experiment with the designed control system, when the PMV value is selected as -0.25 and 0, the comfort value is found as 21.7 ⁰C and 22.8 ⁰C respectively. Since the air conditioner affects the temperature and humidity of the environment at the same time, the comfort temperatures oscillated by ± 0.1 ° C. In the experiment, for PMV -0.25 value, air temperature of the test environment is managed to be hold on at 21.7 ⁰C with error ±0.4 ⁰C. For PMV 0 value, air temperature of the test environment is managed to be hold at 22.8 ⁰C with error ±0.5 ⁰C. The reason behind the decrease of performance for PMV 0 value is that heating of the room is slow but cooling down is faster. In addition, when the outside air temperature is 20 ⁰C, the resistance against the error input of the system with the window opening scenario is measured and it is seen that the control system successfully suppresses the error signal in the test results. According to the data, when people are sitting in winter clothes and working with their computer and when the comfort temperature is determined according to PMV -0.25 value, the users are disturbed thermally from the environment between 6% and 8%. When the comfort temperature was determined according to the PMV 0 value in the same conditions, the users were disturbed thermally from the ambient by an average of 5.1%. The thermal comfort must be between +0.5 and -0.5 as it is specified in the ISO standards. Therefore, PMV-0.25 is preferred for winter months in terms of both energy efficiency and thermal comfort.