VIEWS: 1 PAGES: 16 CATEGORY: Business POSTED ON: 4/18/2013
INTERNATIONAL JOURNAL OF ELECTRICAL ENGINEERING International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME & TECHNOLOGY (IJEET) ISSN 0976 – 6545(Print) ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), pp. 229-244 IJEET © IAEME: www.iaeme.com/ijeet.asp Journal Impact Factor (2013): 5.5028 (Calculated by GISI) ©IAEME www.jifactor.com A COMPARATIVE ANALYSIS OF SPEED CONTROL OF SEPARATELY EXCITED DC MOTORS BY CONVENTIONAL AND VARIOUS AI TECHNIQUE BASED CONTROLLERS Debirupa Hore* (M.Tech Power and Energy systems, B.E Electrical Engineering) *Assistant Professor Electrical Engineering Department KJ Educational Institutes” KJCOEMR Pune Maharashtra India ABSTRACT This paper presents a comparison of speed control of a separately excited DC motor using different types of controllers. Conventional controllers are generally used to control the speed of the separately excited DC motors in various industrial applications. It is found to be simple and high effective if the load disturbances are small. But during high load or large variation of load the AI technique based controllers such as proves to be fast and reliable. To control the speed of the motor a step down chopper is also used. The control scheme of the motor was tested with convention PI controller following the Fuzzy controller and then ANFIS Based speed controller. All the responses were analyzed in MATLAB/SIMULINK environment. The simulation results show that the Artificial Intelligence based speed controllers gives good performance and high robustness in large load disturbances. I. INTRODUCTION Development of high performance motor drives is very essential for industrial applications. A high performance motor drive system must have good dynamic speed command tracking and load regulating response. Depending on the application, some of them have fixed speed and some have variable speed. The variable speed drives, have various limitations such as poor efficiencies, lower speeds etc. With the advent of power electronics today we have variable drive systems which are not only smaller in size but also very efficient, highly reliable and meeting all the stringent demands of the various industries of modern era. DC motors provide excellent control of speed for acceleration and deceleration. The power supply of a DC motor connects directly to the field of the motor which allows for 229 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME precise voltage control, and is necessary for speed and torque control applications. DC drives, because of their simplicity, ease of application, reliability and favorable cost have long been a backbone of industrial applications. DC drives are less complex as compared to AC drives system. DC drives are normally less expensive for low horsepower ratings. DC motors have a long tradition of being used as adjustable speed machines and a wide range of options have evolved for this purpose. Cooling blowers and inlet air flanges provide cooling air for a wide speed range at constant torque. DC regenerative drives are available for applications requiring continuous regeneration for overhauling loads. AC drives with this capability would be more complex and expensive. Properly applied brush and maintenance of commutator is minimal. DC motors can provide high starting torques which is required for traction drives. They are also used for mobile equipment such as golf carts, quarry and mining applications. DC motors are conveniently portable and well fit to special applications, like industrial equipments and machineries that are not easily run from remote power sources. With the advent of thyristors and thyristor power converters the variable voltage to the dc motor is obtained from static power converters. Phase controlled rectifiers provide variable dc voltage from constant voltage, constant frequency mains. The static apparatus is very efficient, compact and has a very good dynamic behavior. It is very easy to provide a four quadrant drive with slight modifications in the converter. A dc chopper can be used to obtain a variable voltage from a constant dc voltage. The average value of the output voltage can be varied by varying the time ratio of the chopper. II. CHOPPERS A DC chopper is a static power electronic device that converts fixed dc input voltage to a variable dc output voltage. A Chopper may be considered as dc equivalent of an AC transformer since they behave in an identical manner. As chopper involves one stage conversion, these are more efficient. Choppers are now being used all over the world for rapid transit systems. These are also used in trolley cars, battery-operated vehicles, traction- motor control, and control of induction motors, marine hoists, forklift trucks and mine haulers. The future electric automobiles are likely to use choppers for their speed control and braking. Besides, the saving in power, the DC chopper offers greater efficiency, faster response, lower maintenance, small size, smooth control, regeneration facility and for many applications, lower cost, than motor-generator sets or gas tubes approaches. A. PRINCIPLE OF STEP-DOWN CHOPPER (BUCK-CONVERTER) OR CLASS A CHOPPER A chopper is a high speed ON or OFF semiconductor switch which is consists of power semiconductor devices, input dc power supply, elements (R, L, C, etc.) and output load. The average output voltage across the load is controlled by varying on-period and off- period (or duty cycle) of the switch In fig.1 when chopper CH1 is ON, V0 = and current i0 flows in the arrow direction shown. When CH1 IS OFF, V0 = 0 but i0 in the load continues flowing in the same direction through freewheeling diode FD. Hence average values of both load voltage and current, i.e. and I0 are always positive as shown by the hatched area in the first quadrant of - plane in fig.1 (b). The power flow in type-A chopper is always from source to load. This chopper is also called step-down chopper as average output voltage V0 is always less than the input dc voltage. 230 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME Fig. 1(a) Class A Chopper Circuit Fig. 1(b): Voltage and Current Directions The variations in on and off periods of the switch provides an output voltage with an adjustable average value. Fig 1(c): Voltage Waveforms Average voltage, =( /( + ))* =( /T)* =α (1) = on-time = off-time T= + = chopping period Thus the voltage can be controlled by varying duty cycle. = f* * (2) f = 1/T = chopping frequency III. MODELLING OF SEPARATELY EXCITED DC MOTOR A separately excited dc motor is very versatile as a variable speed motor. Its speed can be varied by varying the applied voltage to the armature or field current. The speed control using the variation of armature voltage can be used for constant torque application in the speed range from zero to rated speed (base speed). Speeds above base speed are obtained by means of field weakening, the armature voltage being kept at the rated value. The speed control in this case is at constant power. In both cases the speed control is smooth and step-less 231 lectrical International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME 2 Fig 2(a): Separately Excited DC motor Fig 2(b): Complete layout for DC motor speed control A. OPERATION OF SEPARATELY EXCITED DC MOTOR When a separately excited dc motor is excited by a field current of if and an armature balance the load current of ia flows in the circuit, the motor develops a back EMF and a torque to ba torque at a particular speed. The field current if is independent of the armature current ia.. Each winding is supplied separately. Any change in the armature current has no effect on the field current. In general ia is much less than if. B. FIELD AND ARMATURE EQUATIONS Instantaneous field current: dif f = Rf if + Lf (3) dt , resis Where, Rf and Lf are the field resistor and inductor, respectively. Instantaneous armature current: dia a = Raia+La + Eg (4) dt armature Where, Ra and La are the armatu resistor and inductance respectively. The motor back emf, which is also known as speed voltage, is expressed as: E g = Kv ω i f (5) Kv is the motor voltage constant (in V/A-rad/sec and ω is the motor speed (in rad/sec). ( in 232 lectrical International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME C. BASIC TORQUE EQUATION The torque developed by the motor is: d = Kt if ia (6) where, (Kt=Kv) is the torque constant.(in V/A-rad/s) V/A equal For normal operation, the developed torque must be equal to the load torque plus the friction and inertia, i.e. dω d =J + Bω + L (7) dt , (N.m/rad/s) Where, B: viscous friction constant, ( : load torque (N.m) (kg.m J: inertia of the motor (kg.m2) STATE D. STEADY-STATE TORQUE AND SPEED The motor speed can be easily derived: ω = ( a - aRa)/Kv f (8) motor If Ra is a small value (which is usual), or when the motor is lightly loaded, i.e. Ia is small, ω= a /Kv f (9) That is if the field current is kept constant, the motor speed depends only on the supply voltage. The developed torque is: d = Kt f a = Bω + L (10) The required power is: Pd = d ω (11) E. TORQUE AND SPEED CONTROL steady An important fact can be deduced for steady-state operation of DC motor. For a fixed field current, or flux ( f) the torque demand can be satisfied by varying the armature current a. The motor speed can be controlled by controlling a (voltage control) or controlling f (field control). F. VARIABLE SPEED OPERATION Fig. 3(a): Torque Vs Speed Characteristic for Different Armature Voltages 233 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME Family of steady state torque speed curves for a range of armature voltage can be drawn as above. The speed of DC motor can simply be set by applying the correct voltage. The speed variation from no load to full load (rated) can be quite small. It depends on the armature resistance. G. BASE SPEED AND FIELD-WEAKENING Fig. 3(b): Torque Vs Speed and Power Vs Speed Characteristic of Separately Excited DC Motor The motor speed can be varied by- (a) Controlling the armature voltage , known as voltage control; (b) Controlling the field current , known as field control; and (c) Torque demand, which corresponds to an armature current , for a fixed field current . The speed corresponds to the rated armature voltage, rated field current and rated armature current which is known as the rated (or base) speed. In practice, for a speed less than the base speed, the armature current and field currents are maintained constant to meet the torque demand, and the armature voltage is varied to control the speed. For speed higher than the base speed, the armature voltage is maintained at the rated value and the field current is varied to control the speed. However, the power developed by the motor (power = torque * speed) remains constant. Base speed ( ) -The speed which correspond to the rated , rated and rated . Constant torque region (W< ) -Ia and If are maintained constant to met torque demand. is varied to control the speed as power increases with speed. Constant power region (W> ) - is maintained at the rated value and If is reduced to increase speed. However the power developed by the motor (= torque x speed) remains constant. This phenomenon is known as Field weakening. Fig 3(c): Typical Operating Regions of Separately Excited DC Machines 234 lectrical International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME IV. DESIGN OF CONTROLLER Three types of controllers were used and the result was analyzed. Design of Controller involves the design of three types of controller 1. Conventional PI Controller 2. Fuzzy Controller 3. ANFIS Controller A. DESIGN OF PI CONTROLLER 1. DESIGN CURRENT CONTROLLER Fig4: Block diagram for design of Current Controller d KtKp(1+ is)*( Ra ) a is(1+s a) = (12) aref KtKp(1+ is)*(Ra) K1 is(1+s a) 1+ 1s Para Here, (Current Controller Parameter) can be varied as when required. should be chosen the such that it cancels the largest time constant in the transfer function in order to reduce order of the system in 7.2.and, the response will be much faster. Assuming Ti=Ta Now, putting the value in equation (12) we have a KtKp/Ras a = (13) aref K1KtKp/Ra a s(1+ 1s) KtKp Let Ko = RaTa 235 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME So, / = = ( ) ( ) = (14) ² Therefore, the characteristic equation of the above equation is given as s² + s + K K =0 s² + + =0 s² + 2€ ω + ω ²=0; ω= €=1/(2 )= For a second order system value of €=0.707 to have a proper response. 0.707= K K =1/2 = (15) K = (16) From Eq.14 we have ( ) = ² ) ( )/ = (17) ² ² The zero in the above equation may result in an overshoot. Therefore, we will use a time lag filter to cancel its effect. The current loop time constant is much higher than filter time constant. For a small delay we can write (1 + s) = (18) ² = (19) ² Neglecting s² term we have = (20) This equation is used as a response for the speed controller. 236 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME 2. DESIGN OF SPEED CONTROLLER Fig5: Block diagram for design of Speed Controller ω(s)/ ω(s)(ref) = (K /K )(Ra/K )(1 + /(1 + ) ²)/ 1 + (K Ra/K K )(1 + /(1 + ) ²)(K1/(1 + 1 )) (21) Here, we have the option to Tn such that it cancels the largest time constant of the transfer function So, Tn=2T2 Hence, equation 21 will be written as: ω(s)/ω(s)(ref. ) = (K Ra/K K )(1 + 1 )/ K K (1 + 1 ) + K RaK1 (22) Ideally, ω(s) =1/S (S²+αs+β) The damping constant is zero in above transfer function because of absence of S term, which results in oscillatory and unstable system. To optimize this we must get transfer function whose gain is close to unity Then using Modulus og Hugging method and deducing the above equation we finally get ( )( )/ ( )( ) = 1/( 1 + 1+ 1+ 1) (23) 3. DESIGN OF FUZZY CONTROLLER The fuzzy controller used in this scheme is a Speed Controller. The Conventional Speed PI Controller is replaced by a Fuzzy Logic Based speed Controller for providing more reliable controller outputs for the Speed control of The Induction Motor. The main objective of the fuzzy controller is that the actual speed response of the induction motor must track the reference speed response The design of a Fuzzy logic system includes the design of a rule base, the design of the member-ship functions determination of the Linguistic values. Here, the inputs of fuzzy controllers are the error in speed and the rate of change of this error at any time interval. The output of the fuzzy controller is the Active Power. Here, five fuzzy sets (NB, N, Z, P, and PB) are adopted for each input and output variables. The Character NB, N Z, P, PB represents Negative big, Negative, Zero, Positive, Positive Big. The membership functions of the two input variables and the one output variable has normalized universe of discourse over the interval [- 1, 1].For the implementation of FLC, firstly, the universe of discourse of input and output variables of FLC are determined. In practice, each universe is restricted to an interval that is related to the maximal and minimal possible values of the respective variables. That is to the operating range of the variable. The universe of discourse of the input and output variables of the Mamdani type (PI like) FLC can be determined as- The universe of the error is defined by the maximal and minimal values of the variables. It {emin, emax] is the interval where: = = 237 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME Analogously the change in error and the change of the output have operating ranges between [ ] and [ ]. Where, ` = = = = The operating ranges are defined with respect to the external values of the relevant variables. They can be further adjusted by taking into account the dynamics of the controlled systems and the sampling intervals. For simplification and unification of the design of the FLC and its computer implementation, however it is more convenient to operate with normalizes universe of discourse of the input and output variables of the FLC. The normalized universes are well defined domains; the fuzzy values of input and output variables are fuzzy subsets of these domains. In general, the normalized universes can be identical to the real operating ranges of the variables, but in most applications they coincide with the closed interval [-1 1].Otherwise ,scaling of both input and output variables are done in order to bring the values within prescribed limit. The Rule Table is formulated based on which the Fuzzy Controller is designed. Table:1 Rule base for speed controller 4. DESIGN OF ANFIS CONTROLLER Steps Involved in the making of the ANFIS Controller Step1: Get the Error and Change in Error values from the previous model and save it in workspace and finally define these values in a matrix in the M-file of Matlab. Step2: Run the matrix file. Step3: Open the ANFIS editor. Step4: Load data from workspace in the ANFIS editor. Step5:Generate FIS using Grid Partition. Here, the number of Member Functions and type of Member Functions are taken. Step6: Train the data using training FIS. Here the Error tolerance and epochs are also given. Step7: The data is trained using Train FIS. Step8: Step2 to 7 are performed using Checking Data. Step9: The file is saved in both workspace and matlab-file by exporting it. Step10: The Simulink model for ANFIS is opened and the controller is named after the name of the file saved in Step9. The ANFIS Speed Controller is obtained using the above steps. 238 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME V. SIMULATION RESULTS AND DISCUSSIONS The graph responses of the Fuzzy and ANFIS controllers has been compared with the. conventional PI controller. With the implementation of AI techniques (Fuzzy & ANFIS), the response shows that there is much flexibly with the variation of load found as compared to conventional model. In all the responses it has been found that the actual speed of the DC motor is as par with the reference speed of the motor. With Conventional PI speed controller the actual speed of the motor is tracking the reference speed of the motor up to the rated speed and even beyond it. But with Fuzzy & ANFIS Controller the actual speed of the motor is tracking the reference speed of the motor upto the rated speed and when the motor is overloaded the responses falls and error goes on increasing concluding that the motors fails to start or the motor stops. Thus the AI technique based controllers provides much better and precise control of the DC motor. FIG 6:-Block Representation of Motor Model with Fuzzy & PI Controller Fig 7:-Block Representation of Motor Model with ANFIS & PI Controller 239 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME Appendix Table:2 I. Performance of DC motor using PI speed Controller It is observed from the responses that the actual speed of the DC motor tracks the reference speed of the motor up to the rated speed of the motor (55rpm). Beyond the rated speed the PI controller has limitations in controlling the speed of the dc motor. Reference Speed 60 40 S p e e d in rp m 20 0 -20 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 60 S p e e d in rp m 40 20 0 -20 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 60 S p e e d in rp m 40 20 0 -20 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 8: Speed Responses of a Separately Excited dc Motor with Step Source as Input Using PI Controller Reference Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 60 S p e e d i n rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 1 S p e e d i n rp m 0.5 0 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 9: Speed Responses of a Separately Excited dc Motor with Ramp Source as Input Using PI Controller 240 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME II. Performance of DC motor using Fuzzy speed Controller A. With motor running up to rated speed Reference Speed 60 S p e e d in r p m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 9: Speed Responses of a Separately Excited dc Motor with Step Source as Input with Rated Load Using Fuzzy Controller Reference Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 60 S p e e d in r p m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 1 S p e e d in r p m 0.5 0 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig10: Speed Responses of a Separately Excited dc Motor with Ramp Source as Input with Rated Load Using Fuzzy Controller 241 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME B. With motor running at overloaded condition Reference Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 0 S p e e d i n rp m -200 -400 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 500 400 S p e e d in rp m 300 200 100 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 11: Speed Responses of a Separately Excited dc Motor with Step Source as Input with Over Loading Condition Using Fuzzy Controller Reference Speed 100 S p e e d in rp m 50 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 0 S p e e d in rp m -200 -400 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed S p e e d in rp m 300 200 100 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 12: Speed Responses of a Separately Excited dc Motor with Ramp Source as Input with Over Loading Condition III. Performance of DC motor using ANFIS based speed Controller A. With motor running up to rated speed Reference Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 13: Speed Responses of a Separately Excited dc Motor with Step Source as Input with Rated Load Using ANFIS Controller 242 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME Reference Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 60 40 S p e e d in rp m 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 60 40 S p e e d in r p m 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 14: Speed Responses of a Separately Excited dc Motor with Ramp Source as Input with Rated Load Using ANFIS Controller B. With motor running at overloaded condition Reference Speed 60 S p e e d in r p m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 0 S p e e d in rp m -20 -40 -60 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed S p e e d in rp m 100 50 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Fig 15: Speed Responses of a Separately Excited dc Motor with Step Source as Input with Over Loading Condition Using ANFIS Controller Reference Speed 60 S p e e d in rp m 40 20 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Actual Speed 0 S p e e d in rp m -100 -200 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time in Sec Error Speed 200 S p e e d in rp m 100 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 Time in Sec Fig 16: Speed Responses of a Separately Excited dc Motor with Ramp Source as Input with Over Loading Condition Using ANFIS Controller . 243 International Journal of Electrical Engineering and Technology (IJEET), ISSN 0976 – 6545(Print), ISSN 0976 – 6553(Online) Volume 4, Issue 2, March – April (2013), © IAEME VI. CONCLUSION In this paper the speed of a dc motor has been done using three different types of Controllers viz Speed PI controller, Fuzzy logic based controller and ANFIS based controller. The responses of the motor with three different types of controllers are simulated and studied and compared with step and ramp responses. With varying load responses it is observed that the AI technique based controllers’ gives a flexible and precise control of speed both on normal loaded and overloaded condition as compared to the conventional PI controller. REFERENCES [1].Waleed I. Hameed1 and Khearia A. Mohamad2” Speed control of separately excited dc motor using fuzzy neural model reference controller” International Journal of Instrumentation and Control Systems (IJICS) Vol.2, No.4, October 2012 PP 27-39. [2] Dr.P.S Bimbhra “Power Electronics”,Khanna Publishers. [3]Ronald R Yager, Dimitar P Filev, “Essentials of Fuzzy Modeling and Control “Wiley Interscience Publications. [4]Jun Yan, Michael Ryan, James Power “Using Fuzzy Logic towards Intelligent systems” [5]Singari.v.s.r Pavankumar, Sande.krishnaveni, Y.B.Venugopal, Y.S.Kishore Babu, “A Neuro- Fuzzy Based Speed Control of Separately Excited DC Motor”, IEEE Transactions on Computational Intelligence and Communication Networks, pp. 93-98, 2010. [6] Basma A. Omar, Amira Y. Haikal, Fayz F. Areed, “An Adaptive Neuro-Fuzzy Speed Controller for a Separately excited DC Motor”, International Journal of Computer Applications, pp. 29-37, Vol. 39 No.9, February 2012. [7] Vandana Jha, Dr. Pankaj Rai and Dibya Bharti, “Modelling and Analysis of Dc-Dc Converter Using Simulink for Dc Motor Control”, International Journal of Electrical Engineering & Technology (IJEET), Volume 3, Issue 2, 2012, pp. 58 - 68, ISSN Print : 0976- 6545, ISSN Online: 0976-6553. [8] M.Gowrisankar and Dr. A. Nirmalkumar, “Implementation & Simulation of Fuzzy Logic Controllers for the Speed Control of Induction Motor and Performance Evaluation of Certain Membership Functions”, International Journal of Electrical Engineering & Technology (IJEET), Volume 2, Issue 1, 2011, pp. 25 - 35, ISSN Print : 0976-6545, ISSN Online: 0976- 6553. BIBLIOGRAPHY Debirupa Hore was born in Guwahati Assam India on April 19th 1983.She received her B.E Degree in Electrical Engineering from Assam Engineering College Guwahati in 2006 and M.Tech in Energy and Power Systems in 2010 from NIT Silchar. She worked in GIMT Guwahati for 5 years as Assistant Professor. Currently she is working as an Assistant Professor in Electrical Engineering Department in KJ Educational Institutes, KJCOEMR, Pune (Maharashtra).Her research areas of interest includes Power Systems, AI Techniques, Power Electronics and Drives etc. 244