This is a mini project for the subject Artificial Intelligence whereby an intelligent microwave system is created by using fuzzy logic rules (Mamdani).
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0 ratings0% found this document useful (0 votes)
221 views23 pages
Intelligent Microwave System For Cooking
This is a mini project for the subject Artificial Intelligence whereby an intelligent microwave system is created by using fuzzy logic rules (Mamdani).
Prepared by: Name Matric No. Jolynn Samantha Martin 50213211509 Muhamad Sahih Bin Juso 50213211160 Ahmad Hazim Mohd Rusli 50213212559
Date: 14 May 2014 Prepared for: Mr. Zalhan Mohd Zin i
TABLE OF CONTENTS
Introduction to Fuzzy Logic....................................................................................................... 1 Overview of Mini Project .......................................................................................................... 3 Literature Review....................................................................................................................... 4 Problem Statement ..................................................................................................................... 6 Project Workflow ..................................................................................................................... 10 Development of Fuzzy Logic Applications ............................................................................. 12 Results and Discussion ............................................................................................................ 16 Conclusion ............................................................................................................................... 18 Recommendation ..................................................................................................................... 19 References ................................................................................................................................ 20 Appendix: Source Codes.......................................................................................................... 21 MATLAB Source Code for Fuzzy Inference System (FIS) ................................................ 21
1
INTRODUCTION TO FUZZY LOGIC
The concept of Fuzzy Logic was conceived by Lotfi Zadeh, a professor at the University of California at Berkley, and presented not as a control methodology, but as a way of processing data by allowing partial set membership rather than crisp set membership or non- membership. This approach to set theory was not applied to control systems until the 70's due to insufficient small-computer capability prior to that time. Professor Zadeh reasoned that people do not require precise, numerical information input, and yet they are capable of highly adaptive control. If feedback controllers could be programmed to accept noisy, imprecise input, they would be much more effective and perhaps easier to implement. Unfortunately, U.S. manufacturers have not been so quick to embrace this technology while the Europeans and Japanese have been aggressively building real products around it. In this context, fuzzy logic is a problem-solving control system methodology that lends itself to implementation in systems ranging from simple, small, embedded micro-controllers to large, networked, multi-channel PC or workstation-based data acquisition and control systems. It can be implemented in hardware, software, or a combination of both. Fuzzy logic provides a simple way to arrive at a definite conclusion based upon vague, ambiguous, imprecise, noisy, or missing input information. Fuzzy logic's approach to control problems mimics how a person would make decisions, only much faster. Fuzzy logic incorporates a simple, rule-based IF X AND Y THEN Z approach to a solving control problem rather than attempting to model a system mathematically. The fuzzy logic model is empirically-based, relying on an operator's experience rather than their technical understanding of the system. For example, rather than dealing with temperature control in terms such as "SP =500F", "T <1000F", or "210C <TEMP <220C", terms like "IF (process is 2
too cool) AND (process is getting colder) THEN (add heat to the process)" or "IF (process is too hot) AND (process is heating rapidly) THEN (cool the process quickly)" are used. These terms are imprecise and yet very descriptive of what must actually happen. Consider what you do in the shower if the temperature is too cold: you will make the water comfortable very quickly with little trouble. Fuzzy logic is capable of mimicking this type of behavior but at very high rate. Fuzzy logic requires some numerical parameters in order to operate such as what is considered significant error and significant rate-of-change-of-error, but exact values of these numbers are usually not critical unless very responsive performance is required in which case empirical tuning would determine them. For example, a simple temperature control system could use a single temperature feedback sensor whose data is subtracted from the command signal to compute "error" and then time-differentiated to yield the error slope or rate-of- change-of-error, hereafter called "error-dot". Error might have units of degs F and a small error considered to be 2F while a large error is 5F. The "error-dot" might then have units of degs/min with a small error-dot being 5F/min and a large one being 15F/min. These values don't have to be symmetrical and can be "tweaked" once the system is operating in order to optimize performance. Generally, fuzzy logic is so forgiving that the system will probably work the first time without any tweaking. Fuzzy logic was conceived as a better method for sorting and handling data but has proven to be an excellent choice for many control system applications since it mimics human control logic. It can be built into anything from small, hand-held products to large computerized process control systems. It uses an imprecise but very descriptive language to deal with input data more like a human operator. It is very robust and forgiving of operator and data input and often works when first implemented with little or no tuning. 3
OVERVIEW OF MINI PROJECT
Most household appliances use fuzzy logic for ease of use these days, especially for use in the kitchen. This allows consumers to simply heat/cook and go with their food and provides ease of use to those who are on the move, inept at cooking, or inept even at the heating of food. Microwave ovens usually operate on the basis of type of food and/or weight of food to heat and/or cook food as desired by the consumer. The use of microwave ovens have evolved extraordinarily due to the integration of fuzzy logic in its system, allowing ease of use and saves consumer time. The microwave oven system includes a power output controller that controls the applicator output power; weight sensor that automatically detects the weight of food on the microwave ovens turntable; sensor that automatically classifies the food type as soft, medium or hard; and fuzzy logic controller that receives a signal from the weight sensor indicating the current weight of food on the turntable as well as signal from the sensor indicating the current food type and adjusts the microwave output power based on the current weight of food and food type in accordance with fuzzy logic rules by sending a control signal to the power output controller. A membership function divides the expected range for the amount of weight and food type into multiple regions, each region having a pre-computed regional output settings. The regional output settings of the regions include the current weight of food and type of food are used to compute the control signal. The user controls the time that is needed to cook or heat the food. The microwave oven then adjusts its power output level accordingly to effectively cook or heat the food. Establishing the power output level in terms of percentage utilizes the microwaves magnetron well to evenly cook the food. The centroid method is used for defuzzification. 4
LITERATURE REVIEW
The microwave oven has added an innovative dimension to todays kitchen as a preparation and cooking appliance, differing from traditional cooking appliances. The first domestic microwave oven was introduced in the 1960s by Litton, followed by Amana which brought out the first countertop model in 1968 (Thompson & Sweaney, 1993). Today, the microwave oven is used in 90% of American households and 68% of workplaces (The microwave tidal wave, 1989). Microwave ovens operate on the basis whereby when the weight of the food and/or the type of food is detected by the controller, giving a fuzzy membership function. The fuzzy logic membership then establishes how much power is to be supplied to the magnetron in order to generate the microwave output power to cook and/or further heat the food. The microwave output power level is measured in terms of percentage because of the correlation between microwave oven output power percentages to conventional oven temperatures in order to cook food evenly. It must also be taken into consideration that different microwave ovens have different wattages. Therefore, a 1000 W microwave oven has a different operating power when compared to a 750 W microwave oven. Unlike the conventional oven that utilizes dry heat to cook food, the microwave oven cooks food by exciting the water molecules in the food by using microwaves. Microwaves are a type of wave that are sandwiched between radio waves and infrared radiation on the electromagnetic spectrum. In the case of microwave ovens, the commonly used wave frequency is roughly 2,450 megahertz (2.45 gigahertz). Waves in this frequency range have an interesting property whereby they are absorbed by water, fats and sugars. Once absorbed, they are converted directly into atomic motion which is heat. These waves boast another 5
interesting, related property too as they are not absorbed by most plastics, glass or ceramics. Metal reflects microwaves, however, which is why metal pans do not work well in a microwave oven. It is also why the devices have metal walls for reflection of the microwaves In microwave cooking, the radio waves penetrate the food and excite water and fat molecules almost evenly throughout the food. No heat has to migrate toward the interior by conduction. There is heat everywhere all at once because the molecules are all excited together. There are limits, of course. Microwaves penetrate unevenly in thick pieces of food and there are also hot spots caused by wave interference. The entire heating process is different because of the process of exciting atoms rather than the conventional process of conducting heat. Using a microwave oven with fuzzy logic principles aids in cooking time and promotes ease of use because of the superiority of the fuzzy logic controller especially when combined with one or more sensors. For example, if one were to bake a whole potato in a conventional oven with a conventional controller, it would take about an hour to fully cook the potato. However, if one were to cook a whole potato in a microwave oven with a fuzzy logic controller, it would take about 10 minutes to fully cook the potato. This clearly illustrates the superiority of the fuzzy logic controller. It not only cooked the potato much faster, it also saved energy. Fuzzy logic principles are applied to microwave ovens whereby by combining the input values of the weight and type of food, the fuzzy logic controller then adjusts the output power level accordingly to cook the food thoroughly. Solid foods require a longer cooking time and thus, the fuzzy logic controller will signal to the microwaves magnetron to cook the solid food at a lower power to prevent the food from becoming too hard on the outside yet uncooked in the inside. Similarly, for foods in a soft or liquid state will require a shorter cooking time. Therefore, a short burst of higher power is suitable to quickly boil the liquid which is decided accordingly by the fuzzy logic controller. 6
PROBLEM STATEMENT
From an engineering standpoint, both microwave heating and cooking of foods are complex. Thus, microwave cooking is certainly complex and a complete engineering description of this process that includes physical, chemical, biological, and sensory aspects is simply not available. Cooking depends on such a myriad of oven and food factors. Thus, there are situations where cooking is more of an art than science. Patterns begin to emerge from persistent testing of actual food products and processes. These can lead to the development of simple, basic microwave techniques, developed from the other end of the spectrum, such as the cooked final product. Microwave ovens have various cooking patterns, and it is important to determine if there are hot or cold areas which could cause uneven cooking of foods. To help ascertain where these areas are, dampen a paper towel, place it on your turntable or the floor of your oven, and heat it for a minute or two. Hot or cold spots will show up as damp and dry patterns on the towel. Sometimes these areas can actually be used to an advantage by placing certain types of foods within the areas to help cook them faster or slower. For example, a leg of lamb roast will cook more evenly when the small end of the leg is placed in the cooler area, and the larger end of the roast in the warmer area in the oven. Although newer microwaves have fewer hot and cold spots, this can be a problem in many of the older models. Electrical power in some areas may vary throughout the day. This can explain why food cooked exactly the same way will turn out differently at times, which can be very frustrating to microwave cooks. To help determine the peak time of power usage in a certain area, it is helpful to boil a cup of room-temperature water (25C) at three separate times throughout the day. It is best to do this test on a weekday when schedules are more similar than on weekends 7
or holidays. At certain times throughout the day, time for the water to reach boiling point may vary as much as 1 or 2 min. Therefore, cooking times may need to be adjusted upward or downward at certain times in order to achieve the same result in the food cooked. Microwave oven power is measured in watts, which indicates the cooking power of each new oven. Although the wattage information can usually be found in the owners manual, actual wattage varies depending on the age of the oven and a number of other oven food. Experience has shown that foods cook more consistently by correlating microwave percentages of power to conventional oven temperatures. Table 1 shows this rule of thumb for microwave power levels as they relate to conventional temperatures, for a full-size microwave oven (600750 W). Higher wattage ovens will need to lower the power or lessen cooking time. For example, 70% power on a microwave oven having 1000 W or more of power would be equivalent to 100% power on a 600750 W oven. Use similar techniques as in conventional heating, such as to cover or not to cover, and then correlate microwave power levels to the proper conventional temperatures at which a particular food would cook conventionally. Refer to power levels as 100% or 70% rather than medium, low, simmer, or bake, which may not be consistent from manufacturer to manufacturer. For example, one microwave manufacturer may have a medium of 50% power while another may have a medium of 70% power. By using percentages of power rather than inconsistent terminology, accuracy will be greater when cooking foods in a microwave.
8
Table 1: Approximate Adaptations for Conventional Cooking Temperatures to Microwave Power Levels for 750 W Oven Microwave Power Level, % Equivalent Setting on Conventional Oven, C 10 65 90 (low stovetop) 20 90 110 30 40 110 150 50 60 150 180 (medium stovetop) 70 180 195 (medium-high stovetop) 80 195 220 90 100 220 260 (deep-fat-fry, broil, high stovetop)
Below is a list of factors affecting cooking times in a microwave oven: 1. Starting temperature of the food (cold vs. room temperature or warm). 2. Quantity of food (large or small amount). Since small pieces of food cook quicker than larger pieces, items of similar size cook more uniformly. Large foods such as turkeys or roasts, which cannot be stirred, should be turned over rotisserie style while cooking. This promotes uniform, even heating in the microwave just as a rotisserie cooks large pieces of meat more evenly on a charcoal grill. 3. Shape of food (thin vs. thick; round vs. irregular). 4. Shape of container in which food is cooked (square vs. round vs. ring). 5. Composition of food (high sugar-fat content heats faster than lower sugar-fat content; higher sugar-fat content heats faster than water). Fresh vegetables contain more water and cook faster than those that are less fresh. 6. Density of food. Two foods weighing the same with different densities cook differently; a dense or compact roast takes longer to cook than a porous loaf of bread of the same weight. 7. Microwave power being usedpower level (e.g., 100% vs. 30%) and oven wattage. 9
8. Electrical power fluctuation (power supply in your home may vary throughout the day and may affect the microwave cooking times). 9. Preference as to doneness (crisp vegetables vs. soft, tender vegetables). 10. Liquid being added to food (vegetables in sauce vs. vegetables cooked alone). Foods, such as soups, with a high liquid content will take longer to cook than those that cook from their own moisture content. Liquid ingredients slow down cooking. In other words, 500 grams of food with 50% moisture added will take less time to heat to a certain temperature than 500 grams of food with 75% moisture added. If meat is cooked in liquid, cooking time will be longer than meat not cooked in liquid; a potato cooked in water will take longer to cook than one cooked by itself. Also, if meat or vegetables are partially cooked prior to adding the liquid, cooking time will be shorter. Combine partially cooked food with hot liquid, and the cooking time will be shorter still.
Due to these factors, the weight sensor and material sensor must accurately extract the weight of food and type of food respectively so that the fuzzy logic controller will adjust the output power accordingly to prevent the food from overcooking. However, there is no sensor to determine that the food has been thoroughly cooked. Thus, the user must select an appropriate cooking time to effectively cook the food in a microwave oven. There is also a tendency for liquids to superheat when cooking in a microwave oven whereby it exceeds its boiling point. Because of that, the fuzzy logic controller is adjusted accordingly so that it signals the microwave ovens magnetron to supply a lower power when cooking liquids (or soft food type).
10
PROJECT WORKFLOW
The basic structure of this mini project consists of a microwave oven with a fuzzy logic controller. Two sensors are used. One is used to measure the weight of food and the second is used to detect the type of food. The sensors with amplification are connected with the two fuzzifiers of the fuzzy logic control system. Outputs of defuzzifiers are to control the range.
Figure 1: Basic model of microwave oven with fuzzy logic controller
This simplified design algorithm is used to design the fuzzifier, rules and defuzzifier for the microwave oven to achieve an effective range (power) of cooking food in a microwave oven. The design uses three membership functions: 1. Input: Weight of food with a range of 0 to 100% of the microwave ovens turntables maximum acceptable weight. 2. Input: Type of food with a range of 0 to 100 % of solid food state. 3. Output: Output power with a range of 0 to 100 % of the microwave ovens maximum operating power. 11
The fuzzifier uses the data of two input variables, weight of food and type of food to determine the data of the output variable, output power. The membership functions and range are given in the tables below. Table 2: Membership function for weight of food Membership Function for Weight of Food Range of Food Weight (%) Light 0 40 Medium 30 75 Heavy 65 100
Table 3: Membership function for type of food Membership Function for Type of Food Range of Food Type (%) Soft 0 45 Medium 40 75 Hard 60 100
Table 4: Membership function for output power Membership Function for Output Power Range of Output Power (%) Very Low Power 0 25 Low Power 20 45 Medium Power 35 65 High Power 55 85 Very High Power 75 100
From the membership functions and range above, it is now possible to develop the rules for the microwave ovens fuzzy logic controller. The centroid method is used for defuzzification.
12
DEVELOPMENT OF FUZZY LOGIC APPLICATIONS
Before developing the rules, the inputs range are determined. X1 = Weight of Food = {Light, Medium, Heavy}, ranging from 0 to 100 percent. X2 = Type of Food = {Soft, Medium, Hard}, ranging from 0 to 100 percent. The outputs range is determined as well. Y = Output Power = {Very Low, Low, Medium, High, Very High}, ranging from 0 to 100 percent output power. The rules are then developed based on the inputs. Table 5: Table of rules developed by two inputs WEIGHT TYPE LIGHT MEDIUM HEAVY SOFT Medium High Very High MEDIUM Low Medium High HARD Very Low Low Medium
The rules developed are as follows: 1. If Weight is LIGHT and Type is SOFT, then Power is MEDIUM 2. If Weight is LIGHT and Type is MEDIUM, then Power is LOW 3. If Weight is LIGHT and Type is HARD, then Power is VERY LOW 4. If Weight is MEDIUM and Type is SOFT, then Power is HIGH 5. If Weight is MEDIUM and Type is MEDIUM, then Power is MEDIUM 6. If Weight is MEDIUM and Type is HARD, then Power is LOW 7. If Weight is HEAVY and Type is SOFT, then Power is VERY HIGH 8. If Weight is HEAVY and Type is MEDIUM, then Power is HIGH 9. If Weight is HEAVY and Type is HARD, then Power is MEDIUM
13
The rule selector receives three crisp values of weight of food and type of food, distributes the universe of discourse into regions, fires the rules, and then gives the output values corresponding to each variable. The plots of membership functions generated in MATLAB are provided in the figures below.
Figure 2: Plot of membership function for input variable Weight
14
Figure 3: Plot of membership function for input variable Type
Figure 4: Plot of membership function for output variable Power
15
Figure 5: Establishing the rules in MATLAB's Rules Editor
16
RESULTS AND DISCUSSION
The value for range of the microwave ovens output power is calculated using the Centroid of Gravity method. If the weight of food is 65% of the microwave ovens maximum allowable weight and the type of food is 23% of being solid, then the microwave ovens output power is around 70% of its maximum operating power. The figures below display the result by MATLABs Rule Viewer as well as the surface generated.
Figure 6: Rule viewer of MATLAB for Power Output
17
Figure 7: Surface generated by MATLAB's Rule Viewer
Figure 8: Output generated by MATLAB Simulink's scope
18
CONCLUSION
The convenience of the mini projects microwave oven model is to allow users to easily operate a microwave oven without having to worry about having to set the food weight and/or set the foods type/state. This is safer and may even prolong the life of the microwave oven as well as ensure user safety by preventing the superheating of foods (especially in soft or liquid state). Due to this, one can safely use the microwave oven without having the knowledge of internal construction and technical knowledge of microwave ovens. In conclusion, this microwave oven will help to save electricity in the long term due to operating at its optimal output power level in correlation with the weight of food and type of food initially detected by the microwave ovens sensors. The user only needs to set the microwave cooking/heating time and can go on about other chores while waiting for the food to be cooked or heated by the microwave oven. Therefore, the integration of fuzzy logic control systems in everyday applications improve the quality of life and promote ease of use.
19
RECOMMENDATION
Despite the microwave ovens ability to detect the weight of food and type of food, it must be taken into consideration that the microwave oven does not brown food like its conventional oven counterpart. The use of an inverter and current microwave oven technologies when combined with this microwave oven may achieve the desired effect of browned foods. In addition to that, microwave ovens cook food from the inside as opposed to external heat by its conventional oven counterpart. This involves a very deep understanding about the molecular structure of various types of foods as well as the moisture content of different types of foods since microwave ovens cook food by exciting the water molecule within the food. Therefore, the use of a moisture sensor may seem applicable in this context. Converting this microwave oven to a combinational microwave oven (which is the current trend in microwave oven technology) is also ideal because it allows users to cook food much like a conventional oven.
20
REFERENCES
[1] Abramovitch, D. (1994). Some Crisp Thoughts on Fuzzy Logic. American Controls Conference (p. 5). Baltimore: Hewlett-Packard Laboratories. [2] Altrock, C. v. (2002). Fuzzy Logic and NeuroFuzzy Applications Explained. [3] Datta, A. K., & Anantheswaran, R. C. (2001). Handbook of Microwave Technology for Food Applications. New York: Marcel Dekker, Inc. [4] Karambe, R. K., & Gahane, D. H. (2014). Automation of Grinder - An Introduction of Fuzzy Logic. IOSR Journal of Electrical and Electronics Engineering (IOSR-JEEE), 5. [5] Kim, J. W. (1994). Seoul, Republic of Korea Patent No. 5,283,410. [6] Kipphan, H. (2001). Handbook of print media: technologies and production methods (Illustrated ed.). In H. Kipphan, Handbook of print media: technologies and production methods (Illustrated ed.) (pp. 130 - 144). Springer. [7] Lee, I. K. (1995). Kyungsangnam-Do, Republic of Korea Patent No. 5,382,775. [8] Perkowitz, M., & Etzioni, O. (2000). Adaptive Websites. In M. Perkowitz, & O. Etzioni, Adaptive Websites (pp. 152 - 158). Commun ACM. [9] Spott, M., & Nauck, D. (2006). Towards the Automation of Intelligent Data Analysis. Appl. Soft Comput.