The research paper published by IJSER journal is about Intelligent Air Conditioning System using Fuzzy Logic 1

ISSN 2229-5518

Intelligent Air Conditioning System using Fuzzy Logic

Sanjit Kumar Dash, Gouravmoy Mohanty, Abhishek Mohanty

Depaprtment of Information Technology College of Engineering and Technology Biju Patanaik University of Technology Bhubaneswar, Odisha, India

Abstract— W ith the exponential increase in the use of cooling device, the air conditioning systems are becoming an essential part of ou r day to day life. Data suggest an exponential rise in the use of air conditioners in urban as well as rural India. W ith the increase in the usage of air conditioners, there is a simultaneous increase in the electrical power consumption. In this paper a design has been pr oposed considering various input parameters and applying Fuzzy Logic System to the Air Conditioner. By consid ering the input parameters we can greatly modify the functioning of the AC and reduce the electrical energy intake of the AC compressor/Fan while utilizing all available resources in the efficient manner. For better implementation, we have considered the c limatic condition of coastal region namely Bhubaneswar area in the state of Odisha, India. Bhubaneswar being in the coastal area, the values of temperature and humidity are higher in comparison to non-coastal areas of India.

Index Terms— Fuzzy, Air Conditioning System, Defuzzification, Dew Point, Fuzzy base class, Fuzzy rule base

——————————  ——————————

1 INTRODUCTION

uzzy logic is a form of many-valued logic or probabilistic logic; it deals with reasoning that is approximate rather than fixed and exact [2]. Fuzzy logic was formulated by Lofti Zadeh of the University of California at Berkeley in the mid-1960s. Zadeh also formulated the notion of fuzzy control that allows a small set of 'intuitive rules' to be used in order to control the operation of electronic devices. Countries like Ja- pan have home appliances those have integrated fuzzy logic such as vacuum cleaners, microwave ovens and video camer- as. Such appliances can adapt automatically to different condi- tions; for instance, a vacuum cleaner would apply more suc- tion to an especially dirty area. One of the benefits of fuzzy control is that it can be easily implemented on a standard computer. In contrast with traditional logic theory, where bi- nary sets have two-valued logic: true or false, fuzzy logic vari- ables may have a truth value that ranges in degree between 0 and 1. Fuzzy logic has been extended to handle the concept of partial truth, where the truth value may range between com- pletely true and completely false. Fuzzy logic imitates the log- ic of human thought, which is much less rigid than the calcu- lations computer generally perform. Intelligent control strate- gies mostly involve a large number of inputs. The objective of using fuzzy logic has been to make the computer think like people. Fuzzy logic can deal with the vagueness intrinsic to human thinking and natural language and recognize its nature is different from randomness. Using fuzzy logic algorithm, we could enable machines to understand and respond to vague
human concept such as hot, cold, large, small, etc
Air conditioners and air conditioning systems are integral
part of almost every institution. They contribute significant
part of total energy consumption. Studies suggest that in loca- tions like auditoriums, indoor stadiums and conference halls, air conditioning can contribute as much as 75% of total energy intake. For example, a Survey in the Pantaloons mall, Bhuba- neswar, Odisha in summer season shows a daily average con- sumption of 2300 units of electricity by their 320 ton central air
conditioning system. Even in homes and offices, power con- sumed by air conditioners is significant.
With the present rate of electrification of urban and rural In- dia, it is not long before we have to think of efficient ways to reduce electrical power consumption, especially in devices those are specifically designed for heating or cooling purposes as these are the devices those have the maximum power inake. There will be an exponential growth in the number of such devices both in urban and rural areas as per the data available in official portal of Ministry of Environment & Forestry, Govt. of India [1] given in Table 1.

TABLE1

POPULATION OF HEATING / COOLING APPLIANCES (MILLIONS)

In this paper we have devised a scheme implementing Fuzzy Logic in the Air- Conditioning system. With the help of this logic the Air-Conditioner would assess the environmental fac- tors like temperature, humidity, etc and thereby provides com- fortable levels of cooling and optimized electricity consump- tion

http://www.ijser.org

The research paper published by IJSER journal is about Intelligent Air Conditioning System using Fuzzy Logic 2

ISSN 2229-5518

2 BASIC DESIGN OF AN AIR-CONDITIONER

An air conditioner (often referred to as AC) is a home appli- ance, system, or mechanism designed to dehumidify and ex- tract heat from an area [5]. The cooling is done using a simple refrigeration cycle which consists basically of the following steps [6]:
1. The compressor compresses cool Freon gas, causing it to
become hot, high-pressure Freon gas.
2. This hot gas runs through a set of coils so it can dissipate
its heat, and it condenses into a liquid.
3. The Freon liquid runs through an expansion valve, and in
the process it evaporates to become cold, low-pressure
Freon gas.
4. This cold gas runs through a set of coils that allow the gas
to absorb heat and cool down the air inside the building.

3 FUZZY LOGIC CONTROLLER

Fuzzy Logic controller forms the base of the Fuzzy Control System. It basically consists of the heuristics rules those define the parameters of the problem [4]. It consists of:
 Data Base: It normalizes the input crisp values and con- tains the fuzzy partitions of the input and output space.
 Fuzzy Rule Base: It contains the type of fuzzy rules and the source and derivation of the fuzzy control rules

 Fuzzy Inference Machine: The basic function is to compute the overall output of the control output variable based on the individual contribution of each rule in the Fuzzy Rule Base.
 Defuzzification: It converts the set of modified control out- put values into single point-wise (crisp) values and denormalizes the output onto its physical domain.

4 FUZZY VARIABLES

The various variables for the Fuzzy Controller are:

4.1 Fuzzy Input Variables

4.1.1 User Temperature (Ut)

User Temperature (Ut) is the temperature provided by the user through remote controller or thermostat. The range of this thermostat should vary between 18°C and 30°C. So the user set the temperature accordingly.
the values approximated from the data provided by IMD, Bhubaneswar [2].

Fig 4: User Temperature Preference (Ut) Membership Functions

4.1.2 Temperature Difference (Td)

Temperature Difference (Td) is measure of the difference in the actual room temperature and the temperature which is pro- vided by the user .The difference range is between -6C to +6C. Also AC cannot work as a heat pump and reverse its opera- tion, so it is switched of once the difference go out of range.

Fig 5: Temperature Difference (Td) Membership Functions

4.1.3 Dew Point (TDew)

Dew point temperature is the temperature at which water va- por in the air will condense into dew, frost, or water droplets given a constant air pressure. It can be defined alternately as the temperature at which the saturation vapor pressure and actual vapor pressure are equal [5]. Human reaction towards change in dew point temperature can be generally established. Based on the data provided by Indian Meteorological Depart- ment, stationed at Bhubaneswar, a standard Dew Point - Hu- man Reaction table is generated. Based on this table, the mem- bership function for Dew Point is determined and given in Fig
6.

Fig 3: Arrangements to Accept User Temperature Preference (Ut)

Membership functions for Ut are shown in Fig. 4. The values taken into consideration during the membership Function are the values approximated from the data provided by IMD, Bhubaneswar [2].
Membership functions for Ut are shown in Fig. 4. The values taken into consideration during the membership Function are

TABLE 2

STANDARD DEW POINT TEMPERATURES

http://www.ijser.org

The research paper published by IJSER journal is about Intelligent Air Conditioning System using Fuzzy Logic 3

ISSN 2229-5518

of User Temperature and Dew Point Temperature are ranged keeping in mind the data provided by Indian Meteorological Department, Bhubaneswar [2]. The ranges of these values can be adjusted according to the specification of the area of opera- tion of the AC.

Fig 6: Dew Point Temperature (TDew) Membership Functions

4.1.4 Occupancy (Occ)

Occupancy is number of people exposed to air conditioner. The range of people will decide the level of occupancy as low, medium or high. In the absence of people the compressor as well as the fan remains off. We have taken into account the condition in a medium sized room. Level of Occupancy can also be applied to shopping malls where if it lies between 1-

100 then it’s considered as low else between 101-300 as medi- um or else above 300 as high. The ranges can be varied accord- ing to various scenarios like indoor stadiums, auditoriums, etc.

Fig 7: Occupancy (Occ) Membership Functions

4.1.5 Time of Day (TDay)

4.2 Fuzzy Output Variables

The various outputs of the Fuzzy Controller are:

4.2.1 Compressor Speed (Sc)

The speed of compressor is varied between 30 to 100%. Ac- cordingly it will affect the room temperature as per to the giv- en input.

Fig 9: Compressor Speed (Sc) Member Function

4.2.2 Fan Speed (Sf)

The fan speed gives the information about the fan running inside the air conditioner .The speed of fan is accordingly var- ied between 30 to 100%.

Fig 10: Fan Speed (Sf) Membership Functions

Fig 8: Time of Day (TDay) Membership Functions

Time of Day is the period during which the AC would be working. The temperature and dew point values vary signifi- cantly during morning/night time with that of afternoon time as per the data provided by IMD [2]. Also the value of Rela- tive Humidity changes nearly between 15% to 20% at
0830hours and 1730 hours. Accordingly the range of require- ment can be decided for an optimum cooling and power con- sumption. The range would be varied as 00:00 – 13:00 as morn- ing, 09:00- 18:00 as afternoon, and 16:00-24:00 as night. Values

http://www.ijser.org

Fig 11: Operation mode (Mo) Membership Functions

The research paper published by IJSER journal is about Intelligent Air Conditioning System using Fuzzy Logic 4

ISSN 2229-5518

4.2.3 Mode of Operation (Mo)

Air conditioning system can act as a cooler as well as dehu- midifier. In the cooling state it will regulate the air to release cool air. But as dehumidifier it can absorb the humid content of the air by passing dry air into the room. This process does not increase the temperature of the room. This setting prefer- ence is usually not given to the user and is performed implicit- ly by the AC. Considering this parameter leads to greater effi- ciency and comfort levels.

4.2.4 Fin Direction (Fn)

The fins are the set of blades attached to the air conditioner to ensure a swift flow of air in a particular direction. The direc- tion of these fins will define the flow of air either towards or away from the user. The angle of propagation of blades is set accordingly considering 0° as ”towards” and 90° as “away”.

Fig 12: Fin direction (Fn) Membership Functions

5 FUZZY RULE BASE

Rules are formed keeping in mind heuristic relationship be- tween input and output parameters. The inputs give rule base matrixes with size 2X3X3=18 matrices. Every cell has four out- puts, each for compressor speed, fan speed, mode of operation and fin direction. This equates to total sets of 216 IF-THEN. For simplicity of understanding 4 of the 18 rule base output matrix is show below instead of the If-then statements.

TABLE 3

FUZZY BASE RULES FOR DEW POINT TEMPERATURE AT

OPTIMAL VALUE AND OCCUPANCY AT LOW AND TIME OF DAY IS AFTERNOON

FUZZY BASE RULES FOR DEW POINT TEMPERATURE AT OPTIMAL POINT VALUE, OCCUPANCY AT HIGH AND TIME OF DAY IS NIGHT

TABLE 5

FUZZY BASE RULES FOR DEW POINT TEMPERATURE AT

HUMID VALUE, OCCUPANCY AT HIGH AND TIME OF DAY AT AFTERNOON

TABLE 6

FUZZY BASE RULES FOR DEW POINT TEMPERATURE AT HUMID VALUE, OCCUPANCY AT MEDIUM AND TIME OF DAY AT MORNING

TABLE 4

http://www.ijser.org

The research paper published by IJSER journal is about Intelligent Air Conditioning System using Fuzzy Logic 5

ISSN 2229-5518

6 IMPLEMENTATION IN MATLAB

6.1 Fuzzy Base Class

We have used Mamdani fuzzy system for the illustration which uses max aggregation and centroid method for defuzzi- fication. The FIS Editor defines the Fuzzy Base Class, the vari- ous inputs, i.e. User Choice (ut), Temperature Difference (Td), and Occupancy (Occ), Time of Day (TDay) and the various output variables like Compressor Speed (Sc), Fan Speed (Sf ), Mode of Operation (Mo) and Fin Direction (Fn) as shown in Fig. 13.

Fig 13: Fuzzy Base Class

6.2 Fuzzy Rule Base

Fuzzy rules can be designed manually by a user, or automati- cally, i.e. the Rule Editor generates rules for all combinations of selected input variable and a user fills consequent fuzzy terms. Every rule can be deactivated and independent rule weight can be defined for each consequent variable. The rules are defined by selecting the right sequence in the If-then se- quence. These rule sets form the basis upon which the fuzzy rules are implemented upon the provided inputs and sub- sequentially provide the fuzzy outputs.

Fig 14: Fuzzy Base Rules

6.3 Simulated Graphs

With the help of the Surface Viewer of the Fuzzy Logic Toolbox, the following Graphs are generated:

Fig 15: Gradient Graph of No. of Occupant vs Time of Day vs. Fan Speed

Fig 16: Gradient Graph of No. of Occupant vs. User Choice vs.

Fin Direction

Fig 17: Gradient Graph of Temp Diff vs No. of Occupants vs.

Compressor Speed

http://www.ijser.org

The research paper published by IJSER journal is about Intelligent Air Conditioning System using Fuzzy Logic 6

ISSN 2229-5518

Fig 18: Gradient Graph of Time of Day vs. Number of Occupants vs.

Mode of Operation

Fig 19: Fuzzy Rule Viewer

The Rule Viewer shown in Fig 19 gives us the insight about the application of the fuzzy member function to input values of the simulation. A set of inputs are taken, i.e. Ut=24, Td=1, Occ=5, TDay= 1200 hours, TDew=25 and the outputs are Sc=66.4, Fn=29, Mo=0.614, Sf=65.3.

7 Conclusions and Future Work

Previously the Air-Conditioning systems which were used to simply cool the rooms now can perform a variety of functions. By adding intelligence to the Air-Conditioning system we do not have to worry about the cooling process. The analysis clearly maps out advantage of fuzzy logic in dealing with problems that are difficult to study analytically yet are easy to solve intuitively in terms of linguistic variables. In case of the Air-Conditioning system, fuzzy logic helped solve a complex problem without getting involved in intricate relationships between physical variables. Intuitive knowledge about input and output parameters was enough to design an optimally performing system. With most of the problems encountered in day to day life falling in this category, like washing machines, vacuum cleaners, etc, fuzzy logic is sure to make a great impact in human life. In future we will come up with a device that implements the Fuzzy Logic controller in an embedded system which can be used for increasing the efficiency of Air Condi- tioners.

REFERENCES

[1] http://moef.nic.in/downloads/public-information/ Residen- tialpowerconsumption.pdf

[2] Indian Meterological Department, Bhubhaneswar, Odisha, Govern- ment of India.

[3] http://en.wikipedia.org/wiki/Fuzzy_logic US Patent – 5,921,099; Air conditioner temperature control apparatus; Inventor: Seon Woo Lee; Assignee: Samsung Electronics Co., Ltd. Issue date: Jul 13, 1999

[4] I J Nagrath and M Gopal, Control system engineering; New Age I n- ternational Publisher 2010.

[5] http://en.wikipedia.org/wiki/Air_conditioner. [6] http://www.alcornaire.com/howdo.shtml

http://www.ijser.org