# Introduction Ow a days, the brushless dc (BLDC) motors are very widely used in many applications because of high efficiency, high power density, silent operation, long operating life, low maintenance and good dynamic performance [1][2][3][4][5]. Lot of research work is progressing towards the BLDC motor drive due to the increased demand of such systems in the market [6][7][8][9]. However, the use of BLDC motors is restricted due to the complexity involvement in the speed control for applications with the variable speed and varying loads [10][11][12][13]. The classical controllers with fixed gains may not work properly due to the changes in plant dynamics with ageing process, parameter variations and external disturbances [14][15][16]. Fuzzy control is one of the most interesting techniques that can be well applied for the control of imprecise, nonlinear and ill-defined systems [17][18][19][20][21]. The selection of membership functions and the development of fuzzy control rules is a difficult task without the information from the skilled operator. More over the fuzzy control rules could not be defined properly if the skilled operator is unable to give the exact instructions about the process operation. The artificial neural networks have the learning capability which gains the knowledge of a process automatically from the sample input and output relationship of the process [22][23]. This paper introduces the implementation fuzzy logic controller (FLC) and adaptive neuro fuzzy inference system (ANFIS) controller for the speed control of BLDC motor drive system to improve the performance at different operating conditions. This paper is organized in the following manner. Section II illustrates the description of BLDC motor drive that consists of speed controller, current controller, BLDC motor and inverter. Section III presents the simulation results of the drive with FLC as speed controller. Section IV explains the application of ANFIS speed controller to the BLDC motor drive system along with the simulation results. Section V gives the comparative performance of BLDC motor drive for different operating conditions between FLC and ANFIS controller. Finally the main concluding remarks are given in section VI. # II. # Description of B Motor Drive System The block diagram of BLDC motor drive system with the combination of FLC and the neural network as speed controller is shown in figure 1. The drive system consists of BLDC motor, rotor position sensor, combined FLC and neural network known as ANFIS controller and three phase voltage source inverter. There are two basic control loops in the description of the drive system known as speed control loop and current control loop [24][25]. The speed control loop is the outer control loop which consists of BLDC motor, rotor position sensor, speed controller, reference current generator, hysteresis controller and three phase voltage source inverter. The current control loop is the inner control loop which consists of reference current generator, hysteresis current controller and three phase voltage source inverter. The back electro motive force (EMF) wave forms of BLDC motor are in trapezoidal shape. These trapezoidal waveforms for all the three phases are built based on the rotor position (?). sudden load of 0.662 N-m is applied at 0.05 second and withdrawn at 0.12 second. The motor speed is changed at 0.15 second to a value equal to the set speed but in the opposite direction. # LDC # Implementation of Motor Drive with A Controller The properties of both fuzzy control and artificial neural networks are combined to produce the ANFIS controller. The FLC is provided with the learning capability by the structure of neural network that helps the FLC to select the best possible ranges for the input and output membership functions that match the sample input and output data. The response of BLDC motor drive with the proportional-integral (PI) controller is used as sample input and output data to train FLC. The basic structure of ANFIS network obtained through the MATLAB is shown in figure 11. The trained membership functions with the best possible ranges for error (e) and change in error (Î?"e) are shown in figure 12. The chosen fuzzy inference system for FLC is Takagi-Sugeno-Kang type of fuzzy inference system. The selected membership functions are the generalized bell shaped one. The parameters of membership functions are tuned by both the back propagation and least squares method through the neural networks structure. The structure of neural network is helping in tuning and selection process of membership functions and fuzzy control rules respectively thereby eliminating the use of human expert. The updating of parameters through the learning process is facilitated by the gradient vector. # Global Journal of Researches in Engineering ( ) F Volume XIV Issue # Comparison of Performance Indices between and Controller A comparison of FLC and ANFIS controller is carried out in terms of different performance indices at various set speeds for different operating conditions. The table 2 shows the comparison for the performance specifications of settling time, steady state error, peak overshoot and time taken to reach the set speed in the reverse direction between FLC and ANFIS controller. The table 2 shows the comparison for the error constants of integral of absolute error (IAE), integral of squared error (ISE), integral of time weighted absolute error (ITAE) and integral of time weighted squared error (ITSE) between FLC and ANFIS controller. The time taken by the BLDC motor to settle at the set speed from standstill is observed for FLC and ANFIS controller at different set speeds of 3500, 3100, 2600, 2100, 1600 and 1100 rpm. It is found that the drive is taking less time with ANFIS controller for all the speeds. ANFIS controller is giving less overshoot at all speeds except at 3100 rpm as compared to FLC. The steady state error is almost equal in both the cases. # b) Speed Reversal and Load Perturbation The drive rotating at set speed is suddenly reversed to rotate in the opposite direction. The ANFIS controller is attaining the set speed in the reverse direction quickly at most of the speeds. When the drive is running under steady state condition, a rated load of 0.662 N-m is applied at time of 0.05 second and withdrawn at 0.12 second. The dip in the speed is slightly high with ANFIS controller as compared to FLC. # VI. # Conclusion The simulation model of BLDC motor drive is developed to study the transient and steady state performance with FLC and ANFIS controller. The use of expert in selection of fuzzy control rules and tuning the membership functions is eliminated. The ANFIS controller is giving better performance during the different operating conditions at all speeds as compared to FLC except during the load perturbation where the performance is slightly decreased. 1![Fig.1 : Block diagram of BLDC motor drive with the combined FLC and neural network based speed controller](image-2.png "Fig. 1 :") 2![Figure 2 : Simulation model of drive system using MATLAB and Simulink](image-3.png "Figure 2 :") 3![Figure 3 : Membership functions for the inputs of error and change in error](image-4.png "Figure 3 :") 4![Figure 4 : Membership functions for the output of FLC](image-5.png "Figure 4 :") ![a) Simulation Results of Drive with FLC The simulation results of BLDC drive with FLC are shown in figures 5 to 10 for various operating conditions at a set speed of 3500 rpm. The parameters of the BLDC motor considered are 4 poles, 1 HP, star connected, 3500 rpm, 160 V DC supply, 5 A, resistance of 0.75 ? per phase, self inductance minus mutual inductance of 0.00305 H per phase, torque constant of 0.21476 N-m/A, moment of inertia of 0.82614 e-4 Kg-m2 and back EMF constant of 0.10743 V-s per radian. The](image-6.png "") 567![Figure 5 : Speed response of drive with FLC](image-7.png "Figure 5 :Figure 6 :Figure 7 :") 89![Figure 8 : Torque response of drive with FLC](image-8.png "Figure 8 :Figure 9 :") 11![Figure 11 : The network of ANFIS controller](image-9.png "Figure 11 :") 12![Figure 12 : Membership functions for error and change in error a) Simulation Results of Drive with ANFIS Controller The simulation results of BLDC drive with ANFIS controller are shown in figures 13 to 18 for different operating conditions as done with FLC at a set speed of 3500 rpm.](image-10.png "Figure 12 :") 101315![Figure 10 : Magnified view of back EMFs with FLC](image-11.png "Figure 10 :Fig. 13 :Fig. 15 :") 1618![Fig. 16 : Torque response of drive with ANFIS controller](image-12.png "Fig. 16 :Fig. 18 :") 1LNLNLNLNMN SNSN ZEMNLNMNMN MN SNZESPSNLNMNSNSNZESP MPZELNMNSNZESPMP LPSPMNSNZESPSPMP LPMPSNZESPMP MP MP LPLPZESPSPMPLPLPLP 2 3 Year 2014 © 2014 Global Journals Inc. (US) © 2014 Global Journals Inc. (US) * Electric motor drives: Modelling, Analysis, and control RKrishnan 2001 Prentice-Hall Upper Saddle River, NJ * Analysis of Electric Machinery and Drive Systems PCKrause OWasynczuk SDSudhoff 2002 IEEE Press Piscataway, NJ * A new approach for minimum-torque-ripple maximum-efficiency control of BLDC motor SJPark HWPark MHLee FHarashima IEEE Trans. Ind. Electron 47 1 Feb. 2000 * A new torque control method for torque ripple minimization of BLDC motors with un-ideal back EMF HLu LZhang WQu IEEE Trans. Power Electron 23 2 Mar. 2008 * A control strategy for fourswitch three phase brushless dc motor using single current sensor CXia ZLi TShi IEEE Trans. Ind. Electron 56 6 Jun. 2009 * A new approach to sensorless control method for brushless DC motors T.-SKim B.-GPark D.-MLee J.-SRyu D.-SHyun Int. J. Control, Autom. Syst 6 4 Aug. 2008 * Sensorless brushless DC motor drive based on the zerocrossing detection of back electromotive force (EMF) from the line voltage difference PDamodharan KVasudevan IEEE Trans * DC motor drives without current and hall sensors IEEE Trans. Power Electron 26 6 Jun. 2011 * Direct torque and indirect flux control of brushless DC motor SBOzturk HAToliyat IEEE/ASME Trans. Mechatronics 16 2 Apr. 2011 * On the feasibility of four-switch three-phase BLDC motor drives for low cost commercial applications topology and control Byoung-KukLee Tae-HyungKim MEhsani IEEE Trans. Power Electron 18 1 2003 * Integrated design of speed-sensorless and adaptive speed controller for a brushless DC motor Hsiu-PingWang Yen-TsanLiu IEEE Trans. Power Electron 21 2 2006 * Digital control strategy for four quadrant operation of three phase BLDC motor with load variations CSJoice SRParanjothi VJ SKumar IEEE Trans. Ind. Informatics 9 2 2013 * Implementation and performance analysis of Digital controllers for brushless DC motor drives RShanmugasundaram KMZakariah NYadaiah IEEE/ASME Trans. Mechatronics 19 1 2014 * Robust and optimal tuning of PI and PID controllers BKristiansson BLennartson IEE proceedings on control theory and applications 149 1 2002 * An educational tool for fuzzy logic controlled BDCM MAliAkcayol ACetin CElmas IEEE Trans. Education 45 1 Feb. 2002 * Fuzzy PI control design for an industrial weigh belt feeder YZhao EGCollins IEEE Trans. Fuzzy Systems 11 3 Jun. 2003 * Performances of fuzzy-logic-based indirect vector control for induction motor drive MNUddin TSRadwan M AzizurRahman IEEE Trans. Industry Applications 38 5 2002 * A comparative analysis of fuzzy logic and PI speed control in highperformance AC drives using experimental approach ZIbrahim ELevi IEEE Trans. Industry Applications 38 5 2002 * New selftuning fuzzy PI control of a novel doubly salient permanent-magnet motor drive MingCheng QiangSun EZhou IEEE Trans. Ind. Electron 53 3 2006 * Performance analysis of an FLC-based online adaptation of both hysteresis and PI controllers for IPMSM drive RSRebeiro MNUddin IEEE Trans. Industry Applications 48 1 2012 * Development and implementation of a new adaptive intelligent speed controller for IPMSM drive MM IChy MNUddin IEEE Trans. Industry Applications 45 3 2009 * Development and implementation of a simplified self-tuned neuro-fuzzy-based IM drive MNUddin ZhiRuiHuang AB M SHossain IEEE Trans. Industry Applications 50 1 2014 * Advanced simulation model for brushless DC motor drives BKLee MEhsani Electric power components and systems 2003 31 * Performance evaluation of classical and fuzzy logic control techniques for brushless DC motor drive MSurya Kalavathi CSubba Rami Reddy IEEE international power modulator and high voltage conference June 2012 * Energy conv 25 3 Sep. 2010 * Fuzzy-logic-controller-based cost-effective fourswitch three-phase inverter-fed IPM synchronous motor drive system YSLai YK NLin ; M TSUddin MRadwan AzizurRahman IEEE Trans. Industry Applications 42 1 2006 A unified approach to zerocrossing point detection of back EMF for brushless 20