ILLUMINATION CONTROL USING FUZZY LOGIC PRESENTED BY: VIVEK RAUNAK reg: 13090260 CONTENTS INTRODUCTION OF FUZZY LOGIC HISTORIC BACKGROUND ILLUMINATION CONTROL SYSTEM ARCHITECTURE OF FLC DESIGN STEPS OF FLC HARDWARE DESCRIPTION ADVANTAGE OF FLC DISADVANTAGE OF FLC APPLICATION OF ILLUMINATION CONTROL SYSTEM • CONCLUSION • • • • • • • • • HUMAN LIKE THINKING “THINKING”……………… * DIGITAL LOGIC * FUZZY LOGIC 0 OR 1 (Y OR N) FUZZY LOGIC: [0,1] DIGITAL LOGIC: HISTORIC BACKGROUND LOTFI ZADEH • Fuzzy logic was born in 1965 father of fuzzy logic – LOTFI ZADEH Fristly used in control system in 1974 by - EBRAHAM MAMDANI • The international fuzzy system association (IFSA) was established in 1984 • It is too much famous in japan. laboratory of international fuzzy engineering (LIFE) was inaugurated in 1989. ARCHITECTURE OF FLC DESIGN OF FLC CLASSIFICATION AND SCALING OF INPUT(FUZZY PLANE) FUZZIFICATION RULE FORMATION RULE FIRING DEFUZZIFICZTION CLASSIFICATION AND SCALING OF INPUT input error = set point – actual Change in error = pre error - current error Ep=(error / setpoint)100 ∆Ep =(change in error / pre. error ) 100 DYNAMIC RANGE Ep [-100,100] ; ∆Ep Z [0,100]; [-100,100] LINGUAL VARIABLE Fuzzy variable are called lingual variable. It may have infinite no. of values, each value is associated with distinct membership value. LINGUAL VARIABLES Input NB NM NS ZE PS PM -Negative Big -Negative Medium -Negative Small -Zero -Positive Small -Positive Medium PB -Positive Big Output DK -Dark ST -Streak SP -Spark M -Minimum MD -medium H -High Brightness VH -Very High Brightness RANGES OF LINGUAL VARIABLE Input lingual NB NS ZE PS PB output lingual VH HI MD M DK range -100 - -45 -90 - 0 -45 - 45 0 - 90 45 - 100 range 0 - 35 20 - 50 35 - 65 50 - 80 65 - 100 Membership function • It is function through which we get membership value of the element of lingual variable. Ranges from 0 to 1. types… Triangular Gaussion function ϒ function S function Generally trianguler membership function is used. FUZZY PLANE FUZZIFICATION It is process to change crisp input into fuzzy input. Rule formation “if(A=x) then (z=y)” antecedent conclusion Rule formation needs knowledge and experiment. 4 rules in single iteration If If If If (l1 = x1 AND l3 = y1) then U = Z1 (l1 = x1 AND l4 = y2) then U = Z2 (l2 = x2 AND l3 = y1) then U = Z3 (l2 = x2 AND l4 = y2) then U = Z4 •For the given input the lingual variable in which output will lie is determined by knowledge and experience. •Total 49 possible rule •Rule firing mean…to apply the pre-determined rule to get the output. There are many methods for rule firing Minimum composition Product of maximum composition Maximum of minimum composition Minimum of minimum composition Maximum of maximum composition •We use max-min composition for inferring output. Max-min composition Defuzzification It is process to convert fuzzy output into crisp output. Various method: Centre of gravity defuzzification Centre of sums defuzzification Centre of largest area defuzzification First of maxima defuzzification Middle of maxima defuzzification Height defuzzification most commonly used COG = ∫zµdz ∫µdz defuzzification method. ADVANTAGES OF FLC Humen like thinking Efficient design for non-linear control system Cheaper Reduces tedious mathematical calculation Reliable DISADVANTAGES FORMATION OF RULE IS VERY TEDIOUS OBEYS NEW LOGIC APPLICATION OF ILLUMINATION CONTROLLER sensitive photosynthesis LCD brightness control Street light Automatic room light control CONCLUSION The Presentation aimed towards fuzzy logic control system. we saw all aspects of FLC by taking a control system used for illumination control. Illumination control system controls the environment wherevere unpredictable change in illumination is expected.