A Novel Gain Tuning of Anti-Windup PID Controller using Ant Lion Optimization

Similar documents
Experimental Verification of Conduction Noise of Three-level V-connection Rectifier-Inverter System

Novel Analytic Technique for PID and PIDA Controller Design. Seul Jung and Richard C. Dorf. Department of Electrical and Computer Engineering

Design of composite digital filter with least square method parameter identification

Computer Graphic with Matrices Math 308A Project

Graphs fundamental to many problems. Web graphs. Biology. Other.

Bending Geometry Factor For Profile Corrected Involute Gear Tooth With Trochoidal Fillet

About this lecture. Representation of Graph Adjacency List, Adjacency Matrix. Breadth First Search

CSC 263 Lecture 3. October 4, 2006

LECTURE 24 INDUCTION MACHINES (1)

A Cooperative MIMO Mobile Multihop Relay for Cellular Networks

Phase Locked Loop based Pulse Density Modulation Scheme for the Power Control of Induction Heating Applications

7. Inverter operated induction machines

Fuzzy control of a doubly fed induction generator based wind power system

ELG3175 Introduction to Communication Systems. VSB and Introduction to Angle Modulation

ISSN: [Reddy & Rao* et al., 5(12): December, 2016] Impact Factor: 4.116

Rotating Field Voltage Analysis on the Stator and Rotor of the Inverted Rotor Induction Motor

12th WSEAS International Conference on SYSTEMS, Heraklion, Greece, July 22-24, Feedback Signals Estimation of an Induction Machine Drive

A New Buck-Boost DC/DC Converter of High Efficiency by Soft Switching Technique

EEO 401 Digital Signal Processing Prof. Mark Fowler

FROM ANALOG TO DIGITAL

4. Space-Time Block Coding

Analysis of Occurrence of Digit 0 in Natural Numbers Less Than 10 n

Lecture 5: DC-DC Conversion

Flow Meter / Monitor FLEX-HD2K

1 Performance and Cost

Driver Assistance Technical Report D Analytical Performance Considerations of p. Date:

A New Method for Anti-Noise FM Interference

Experimental Analysis of Parameter Limitations in High-Frequency Resonant Gate Driver

where and are polynomials with real coefficients and of degrees m and n, respectively. Assume that and have no zero on axis.

A Complexity Cost Function for the Signal Processing in a WCDMA. Basestation for dimensioning of a Software Defined Radio.

TOTAL LUMINOUS FLUX AND CHROMACITY MEASUREMENT FOR LED LUMINAIRES USING ABSOLUTE INTERATING SPHERE METHOD

LECTURE 1 CMOS PHASE LOCKED LOOPS

DC electrical circuits

Simulation and Analysis of Indirect Field Oriented Control (IFOC) of Three Phase Induction Motor with Various PWM Techniques

Near-field Computation and. Uncertainty Estimation using Basic. Cylindrical-Spherical Formulae

Pointwise Image Operations

Performance Evaluation of an Enhanced Distance-based Registration Scheme Using the Normal Distribution Approximation

Supply Chain Coordination with Buy-back and Wholesale-price Contracts under Random Demand

A New Hierarchical Approach for Modeling Protection Systems in EMT-type Software

DESIGN OF A ROBUST LINEAR CONTROLLER FOR PROPER TRACKING IN FREQUENCY DOMAIN

Universal microprocessor-based ON/OFF and P programmable controller MS8122A MS8122B

A Modified Bow Tie Antenna for RFID Application

Wireless Channels Path Loss and Shadowing

APL: Autonomous Passive Localization for Wireless Sensors Deployed in Road Networks. Technical Report

N2-1. The Voltage Source. V = ε ri. The Current Source

ECMA st Edition / June Near Field Communication Wired Interface (NFC-WI)

Image Enhancement. To process an image so that output is visually better than the input, for a specific application.

P. Bruschi: Project guidelines PSM Project guidelines.

Frequency Separation Actuation Resonance Cancellation for Vibration Suppression Control of Two-Inertia System Using Double Motors

An Efficient Control Approach for DC-DC Buck-Boost Converter

10! !. 3. Find the probability that a five-card poker hand (i.e. 5 cards out of a 52-card deck) will be:

Communication Systems. Department of Electronics and Electrical Engineering

Cross-Layer Design for Combining Cooperative Diversity with Truncated ARQ in Ad-hoc Wireless Networks

Development of On-Board Orbit Determination System for Low Earth Orbit (LEO) Satellite Using Global Navigation Satellite System (GNSS) Receiver

HIGH PERFORMANCE CONTROLLERS FOR SPEED AND POSITION INDUCTION MOTOR DRIVE USING NEW REACHING LAW

Software solutions to mitigate the EM emission of power modules

Journal of Applied Science and Agriculture

Antenna fundamentals: With answers to questions and problems (See also Chapter 9 in the textbook.)

Journal of Kerbala University, Vol. 7 No.2 Scientific. 2009

Switch Basics and Applications. Selection Service Speed

ECE3204 Microelectronics II Bitar / McNeill. ECE 3204 / Term D-2017 Problem Set 7

EE201 Circuit Theory I Fall

Performance optimization of a MAC protocol with multiple. contention slots in MIMO ad hoc networks

Efficient Power Control for Broadcast in Wireless Communication Systems

Variable Speed Drive Volumetric Tracking (VSDVT) for Airflow Control in Variable Air Volume (VAV) Systems

Post-Fault Speed Sensor-Less Operation Based on Kalman Filter for a Six-Phase Induction Motor

Experimental Prototype of Open-End Stator Winding Induction Machine Fed by Three Phase Inverters

Phase-Shifting Control of Double Pulse in Harmonic Elimination Wei Peng1, a*, Junhong Zhang1, Jianxin gao1, b, Guangyi Li1, c

x O O 3 O 05. Questions on Conditional Probability Q1. The probability that it will rain on a day in June is 0.

Study and Analysis of Various Tuning Methods of PID Controller for AVR System

100G and 200G single carrier transmission over 2880 and 320 km using an InP IQ modulator and Stokes vector receiver

Control Limits of Three-Phase AC Voltage Controller under Induction Motor Load A. I. Alolah Ali M. Eltamaly R. M. Hamouda

Variation Aware Cross-Talk Aggressor Alignment by Mixed Integer Linear Programming

Maximum Likelihood Detection for Detect-and-Forward Relay Channels

M2 3 Introduction to Switching Regulators. 1. What is a switching power supply? 2. What types of switchers are available?

Question 1 TELE4353. Average Delay Spread. RMS Delay Spread = = Channel response (2) Channel response (1)

I. SUMMARY II. NETWORK LEVEL ANALYSIS

Design of FIR Filter using Filter Response Masking Technique

2) Calibration Method and Example

Fuzzy-based Direct Power Control of Doubly Fed Induction Generator-based Wind Energy Conversion Systems

MATLAB/SIMULINK TECHNOLOGY OF THE SYGNAL MODULATION

International Journal of Advance Engineering and Research Development

EXPERIMENT #4 AM MODULATOR AND POWER AMPLIFIER

Low-Complexity Time-Domain SNR Estimation for OFDM Systems

Information Hiding Method Based on Block DWT Sub-Band Feature Encoding

Segmentation of Fluorescence Microscopy Cell Images Using Unsupervised Mining

Lab 7 Rev. 2 Open Lab Due COB Friday April 27, 2018

Sliding Mode Control for Half-Wave Zero Current Switching Quasi-Resonant Buck Converter

A Comparison of Modulation Schemes in Bandlimited AWGN Channels

A 10 MHz GaNFET Based Isolated High Step-Down DC-DC Converter: Design and Magnetics Investigation

7 th International Conference on DEVELOPMENT AND APPLICATION SYSTEMS S u c e a v a, R o m a n i a, M a y 27 29,

49.8. Control Relays and Timers. Contents. Product Selection Guide. Timing Relays

Memorandum on Impulse Winding Tester

Interconnect Planning, Synthesis, and Layout for Performance, Signal Reliability and Cost Optimization. Project Overview

THE UNIVERSITY OF NEW SOUTH WALES. School of Electrical Engineering & Telecommunications

Investigation and Simulation Model Results of High Density Wireless Power Harvesting and Transfer Method

EXPERIMENT #9 FIBER OPTIC COMMUNICATIONS LINK

Analysis of the optimized low-nonlinearity lateral effect sensing detector

IEEE Broadband Wireless Access Working Group < Modifications to the Feedback Methodologies in UL Sounding

Chapter 2 Summary: Continuous-Wave Modulation. Belkacem Derras

Transcription:

Inenaional Jounal of Conol Theoy and Applicaion ISSN : 974-557 Inenaional Science Pe Volume Numbe 6 7 A Novel Gain Tuning of Ani-Windup PID Conolle uing An Lion Opimizaion Ayyaao SLV Tummala a, Ravi Kian Inapakuhi b and P.V. Ramanaao c a,b Depamen of Elecical and Eleconic Engineeing, GMR Iniue of Technology Rajam, Sikakulam, AP, India. Email: ayyaao. lv@gmi.og, avikian.i@gmi.og c Depamen of Elecical and Eleconic Engineeing, Nagajuna Univeiy, Gunu, AP, India. Abac: PID conolle i upeio in em of fa epone, immune o paamee vaiaion, diubance ejecion ec. Bu when conol inpu o he plan auae, he epone of he yem i ignificanly affeced and in ome cae hi may lead o inabiliy. PID conolle wih ani-windup i popoed in he lieaue ha can handle auaion in conol inpu. The challenging ak in implemenaion i he uning he paamee of ani-windup PID. To adde hi iue, a implified mehod o deign he paamee of PID wih ani-windup i popoed in hi pape. The paamee of ani-windup ae uned by An Lion opimizaion echnique (ALO). The popoed mehod i validaed fo diffeen cae udie including baic econd ode yem, peed conol of DC moo and inducion moo wih diec oque conol. The popoed deign i found o be upeio in dynamic epone in he cae udied. A Novel Gain Tuning of Ani-Windup PID Conolle i popoed uing An Lion Opimizaion. Keywod: Ani-windup PID, PI conol, An Lion opimizaion echnique, Gain uning. Inoducion PI and PID conolle ae he mo popula linea conolle due o he alien feaue like zeo eady ae eo, good dynamic epone, high diubance ejecion. Becaue of hee feaue, he applicaion of PI conolle ae va and ae no only limied o vaiable peed dive, poce conol, wind enegy conveion yem, ola powe plan, convee. PID conolle i widely ued in eveal induial applicaion. Mo of he pacical yem poe non-linea chaaceiic and hee yem ae lineaized aound an opeaing poin. Hence he conol inpu o he plan i limied o ceain ange. In ome cae he conol inpu geneaed may no be ealized in pacical cae. Conideing he cae of peed conol of choppe fed DC moo, he conol volage canno exceed inpu baey volage. Fo mo of he cae, he conol inpu i limied by plan funcioning equipmen. Conide anohe example of invee fed inducion moo, cuen inpu o he moo i eiced by device aing. Theefoe a auaion block i conneced in imulaion udie o limi he conol inpu o he plan. When auaion i e in, non-lineaiy i enfoced in plan dynamic. Thu he funcionaliy of PID 85 Inenaional Jounal of Conol Theoy and Applicaion

Ayyaao SLV Tummala, Ravi Kian Inapakuhi and P.V. Ramanaao conolle i deviaed fom he deied. To adde hi iue, PID conolle wih ani-wind up i popoed in he lieaue. The paamee of he PID conolle mu be deigned o impove he abiliy, anien epone of he yem and o educe eady ae eo. The uned yem mu alo be immune o lage diubance, modeling eo and paamee vaiaion[] [4]. Thee ae many echnique popoed in he lieaue o une he paamee. The unable paamee fo PID conolle ae K P, K I and K D. Ohe han convenional mehod like Ziegle Nichol mehod, Pole placemen echnique, Nyqui mehod, conolle can alo be uned by naue inpied opimizaion algoihm. The mo ecen, upeio and eaie one i uning he paamee uing An Lion Opimizaion. The anfe funcion of PI conolle i given a T PI () = K The anfe funcion of PID conolle i given a T PID () = K P P K i KI K The conol oupu of PID conolle wih ani-windup i D (.) (.) whee, e = v - u v = K Ê ˆ e K Ú e e d K Ë Á T P I D de d (.3) KD y - e K P v Acuao u Plan G() y K I - e T Figue : Ani-windup PID conolle If conolle oupu v i equal o plan inpu u i.e. unde non-auaion condiion, hi conolle wok like an odinay PID conolle and T i called a coecion faco[5] [9]. Inenaional Jounal of Conol Theoy and Applicaion 86

A Novel Gain Tuning of Ani-Windup PID Conolle uing An Lion Opimizaion Thi pape deail he deign of ani-windup PID conolle by uning he paamee uing ALO algoihm[]. The ALO uned ani-windup PID conolle i compaed wih PID conolle fo econd ode yem. Lae ani-windup PI i uned fo peed conol of DC moo. The pefomance of ani-windup PI conolle i compaed wih uned PID conolle. In he lae cae, ALO uned ani-windup PI conolle i compaed wih PID conolle fo peed conol of inducion moo wih diec oque conol.. An Lion Opimizaion Algoihm ALO opimizaion i a bio-inpied heuiic algoihm inpied fom he anlion apping he an in he poce of huning hem. Anlion make inveed cone haped caviie on and wih hei jaw. Thee caviie ae ued fo apping he an. Anlion will hide ino ha ip of he cone and wai fo he an o fall in ha caviy. If he an ap in hoe caviie hen he anlion y o cach hem. Afe huning, anlion will make he channel clean fo anohe hun. The ALO algoihm imiae he behavio of an lion and an in a ap. In uch modeling an ae made o cawl ove he each pace and an lion ae made o cach hem uing ap. A andom walk i iniiaed fo hi model a an wande aound fo food andomly. Apa fom an, an lion ae alo conideed o occupy ome pace and hiding. Fo he applicaion of ALO, eveal conideaion like diffeen andom walk choen by an in he each pace, ap of an lion and ize of he caviy ae made. The ap buil by he an lion ae baed on hei fine and ae alo affeced by hei abiliy o cach. In evey ieaion each anlion hun an an and he ange of andom walk will be diminihed o a o model he liding of he an. The elocaion of he anlion depend on he fine of he an hun by i. The andom walk of an i uch ha hey updae hei poiion in each ep of opimizaion maching wih he poiion of an lion. Fo building a ap, ALO algoihm ue a oulee wheel fo elecion of fie an lion duing opimizaion which enue a highe pobabiliy fo an lion in caching he pey. Once he an fall in he ap, an lion dag he an o he epicene. Once he an eache o boom of he ap, he an lion conume i body and hen i move o a new locaion o enhance i ap o cach he ohe pey. In hi poce he be of all an lion fom each ieaion i aken a elie. In ALO algoihm, diffeen oluion (an and an lion) ae andomly geneaed. Fo evey ieaion, he poiion of he an change accoding o he an lion eleced by oulee wheel opeao. Iniially he bounday of poiion i accoding o he cuen ieaion and hen i will depend on he eleced an lion and elie. Fine of all he an will be evaluaed and he new locaion of he an lion i mached o he highe fi an lion in he peviou ieaion. The be an lion obained in cuen ieaion i compaed wih he elie and he fie of he wo i conideed a he new elie. The Anlion algoihm i given below [] I. Iniialize he maximum ieaion, populaion of anlion and an epaaely fo a given dimenion d. II. Geneae he fine value of an and anlion fom he objecive funcion. ( d ) ( d ) È YL,, L,,..., L, YL,, L,,..., L, Q OA = : : YLn,, Ln,,...., Î, ( L n d ) (.) 87 Inenaional Jounal of Conol Theoy and Applicaion

Ayyaao SLV Tummala, Ravi Kian Inapakuhi and P.V. Ramanaao (,,, d ) (,,, d ) È YLL, LL,..., LL YLL, LL,..., LL Q OAL = : : YLLn,, LLn,..., LL Î LL n, d epeen d h vaiable of n h anlion. L n, d epeen d h vaiable of n h an. (, n, d ) Y(L n,, L n,,..., L n, d ) epeen fine value of n h an. Y(LL n,, LL n,,..., LL n, d ) epeen fine value of n h anlion. III. The anlion wih opimum fine value i eaed a elie. while i < Max ie fo j < n (maximum # an) IV. Chooe anlion uing Roulee wheel V. Updae min-max of vaiable a given in (.3) and (.4) (.) q = q I (.3) VI. Geneae a andom walk of an uing (.5) whee, i a andom value. VII. Updae anlion poiion uing (.6) p = p I ( ) ( - ) ( - ) (.4) X( ) = È Î, S ( ) -, S ( ),..., S ( n ) (.5) end fo loop VIII. Geneae he fine value of an An i R = IX. Replace an wih anlion if f An f Anlion X. Modify elie if fine value i opimum. end while loop Diplay elie A R E ( i )> ( j ) (.6) 3. Cae Sudie and Reul Analyi Cae udy : A Second ode yem: Le u conide econd ode linea yem Inenaional Jounal of Conol Theoy and Applicaion 88

A Novel Gain Tuning of Ani-Windup PID Conolle uing An Lion Opimizaion G () = w xw w Hee he objecive i o une he paamee K P, K I and K D fo PID conolle. Conain: K P K I K D x =.8 and w = (3.) To une he paamee of PID conolle, an objecive funcion i fomulaed in uch a way ha eady ae eo, eling ime, ie ime and peak ovehoo i educed. whee, y i a efeence oupu, y i he meaued oupu, y p i he peak value of he oupu. ( ) Ú F = min y - y d yp - y (3.) PI and PID conol paamee ae uned uing ALO algoihm. The paamee obained fo PID conolle afe uning ae given below K P = 9.9, K I =.49 and K D = 3.66 The anien epone of he plan i hown in Figue ().When compaed o PI conolle, PID conolle i upeio in em of peak ovehoo, eling ime, Figue (3) how he conol inpu o he plan wihou auaion (acuao). The conol effo i vey high which may no be poible in eal-ime condiion. Hence hee i a need o limi he conol inpu o he plan. To adde hi iue a auaion block i added o imulaed yem o limi he conol inpu o ±5. oupu y.5.5 PID PI conol inpu 4 x 4 3 4 6 8 ime in Figue : Oupu of econd ode yem.5.5 ime in Figue 3: Conol inpu The epone how ha peak ovehoo, ie ime and eling ime ae inceaed unde auaed condiion. Thi i becaue when he conol inpu o he plan canno exceed auaion limi and hi inoduce nonlineaiy in o he yem. In ome cae hi may lead o inabiliy. To adde hi iue, PID conolle i eplaced wih ani-windup PID conolle. Thee ae many deign of PID conolle wih ani-windup. Ou of hee he convenional mehod, Tacking Back calculaion cheme i adoped hee. The anfe funcion of ani-windup PID conolle i given in ecion-i. The deign of ani-windup PID conolle include elecion of K P, K I, K D and T. 89 Inenaional Jounal of Conol Theoy and Applicaion

Ayyaao SLV Tummala, Ravi Kian Inapakuhi and P.V. Ramanaao The limi of PID conolle ae eleced ame a befoe. The ange of coecion faco i eleced a [. ]. Thu he numbe of paamee o be uned i fou. Following he AnLion opimizaion algoihm given in ecion, he objecive funcion (3.) i minimized and he coeponding uned paamee ae given below. K P = 9.9, K I =.49, K D = 3.66 and T =.43.4. wihou auaion wih auaion oupu y.8.6.4. 4 6 8 ime in Figue 4: Oupu of he yem Figue (5) how he pefomance of he yem wih ani-windup conolle. The oupu ele fae wih le peak ovehoo fo PID wih ani-windup..4. PID wih Ani-windup PID oupu y.8.6.4. Cae Sudy-: Speed Conol of DC moo: 4 6 8 ime in Figue 5: Oupu wih ani-windup conolle Hee he objecive i o une he paamee K P, K I and K D fo PID conolle. Conain: K P K I K D To une he paamee of PID conolle, an objecive funcion i fomulaed in Eq.(3.3) o miigae eady ae eo, eling ime, ie ime and peak ovehoo. Ê * - ˆ * wmax w F = F = min ÁÚ w - w d * (3.3) Ë w Inenaional Jounal of Conol Theoy and Applicaion 9

A Novel Gain Tuning of Ani-Windup PID Conolle uing An Lion Opimizaion V S ω - PID wih Ani-windup i a Hyeii conolle DC Choppe DC Moo ω ω The open loop anfe funcion of DC moo i: Figue 6: Speed conol of DC moo T() = Ka J B ( L R ) K (3.4) ( ) a a a In hi ecion, we wan o analyze he pefomance of DC moo wih ani-windup PID conolle. The objecive of hi ecion i o deign a peed conolle fo DC moo. The DC moo opeae unde cloed loop condiion opeaing wih vaiable load and i peed i egulaed uing duy aio of ep-down choppe which i deived fom he conolle. Acual peed w i ened and i compaed wih i efeence value w *. Thi peed eo i given o a peed egulao which geneae efeence amaue cuen i a * and hi cuen efeence i given o hyeei cuen conolle wih hyeei band of ± A. Iniially PID conolle i eleced a peed conolle. Wih PID conolle, peak ovehoo i.83%. Thi i becaue a he ime of aing he aing cuen i limied o 3A. To enhance he pefomance of peed conolle, PID conolle i eplaced wih PI conolle wih ani-wind up. The majo challenge in he deign of ani-windup i elecion of T. Figue (7) how he epone of he yem wih diffeen value of T. 5 peed in ad/ 5...3.4 9.5.5.5 ime in Figue 7: Vaiaion of peed wih coecion faco Thu hee i a need o elec a pope value of T. The value of coecion faco i uned uing ALO algoihm. Figue (8) how he pefomance of he conolle wih ani-windup. The peed ovehoo i daically educed a he ime of aing. Unde unning condiion ince he conol cuen i le han 3A, he conolle i imila o PI conolle. A load oque of 3Nm i added a. and becaue of hi peed i educed and again Inenaional Jounal of Conol Theoy and Applicaion

Ayyaao SLV Tummala, Ravi Kian Inapakuhi and P.V. Ramanaao 5 PI PI wih ani-windup peed in ad/ 5.5.5.5 ime in Figue 8: Speed oupu wih ani-windup conolle egained o efeence peed wihin.5. Bu hee i no effec of coecion faco on hi becaue he conol cuen equied unde load change condiion i le han 3A. 35 3 acual cuen efeence cuen 5 cuen in A 5 5.5.5.5 ime in Figue 9: Amaue cuen Cae udy-3: Speed Regulaion of Inducion moo: The dynamic of inducion moo ae given by [].The noaion ae ame a given in lieaue. V S Swiching Table Volage Souce Invee Inducion Moo ω Volage ω - ω PID wih Ani-windup τ e Toque and Flux Hyeei Compaao Flux & Toque Eimao Cuen * ψ Figue : Speed conol of inducion moo Inenaional Jounal of Conol Theoy and Applicaion 9

A Novel Gain Tuning of Ani-Windup PID Conolle uing An Lion Opimizaion dw npm = d JL TL ( yaib - yaia )- (3.5) J dy a R = - R ya - d L L M ia npwy b (3.6) dy b R = - R yb d L L M ib npwy a (3.7) di a = d di b = d MR LL MR LL y y a b npm Ê MR LR ˆ wy b - ia u LL Á a Ë LL L (3.8) npm Ê MR LR ˆ - wy a - ib u LL Á b Ë LL L (3.9) The meaued peed w i compaed wih he efeence value w * and peed eo i fed o peed egulao. The oupu of he peed egulao i a efeence oque e *. Flux and Toque eimao eimae flux and oque fom ao volage and cuen. Thee efeence oque and efeence fluxe ae given o Toque and hyeei compaao. Figue () how he peed epone wih PID conolle and PI conolle wih ani-windup. Reul how ha he peed change unde udden change in load oque condiion i low wih ani-windup conolle. 4. Concluion Figue : Speed epone of Inducion moo The funcionaliy of PID conolle i deviaed fom he deied when conol inpu o he plan auae. Thi lead o inoducion of conolle wih ani-windup. In adiional mehod, coecion faco i eleced a a andom value which effec he pefomance of he conolle. A i i difficul o une he ani-windup conolle paamee, we popoed he deign of ani-windup PID conolle paamee uing An Lion opimizaion echnique. The pefomance of ALO uned ani-windup conolle i evaluaed on diffeen yem. In cae of econd ode yem, ALO uned ani-windup PID conolle oupefom ani-windup PID conolle. In he econd cae ALO uned ani-windup PI conolle oupefom PID conolle. In he hid cae, ALO uned ani-windup PID conolle howed upeio pefomance compaed o PID conolle. 93 Inenaional Jounal of Conol Theoy and Applicaion

Ayyaao SLV Tummala, Ravi Kian Inapakuhi and P.V. Ramanaao Refeence [] K. H. Ang, G. Chong, and Y. Li, PID conol yem analyi, deign, and echnology, IEEE Tan. Conol Sy. Technol., Vol. 3, No. 4, pp. 559 576, 5. [] [3] [4] [5] [6] [7] [8] [9] [] [] Y. Li, K. H. Ang, and G. C. Y. Chong, PID conol yem analyi and deign, IEEE Conol Sy., Vol. 6, No., pp. 3 4, 6. K. J. Åöm and T. Hägglund, The fuue of PID conol, Conol Eng. Pac., Vol. 9, No., pp. 63 75, Nov.. G. P. Liu and S. Daley, Opimal-uning PID conol fo induial yem, Conol Eng. Pac., Vol. 9, No., pp. 85 94,. J. W. Choi and S. C. Lee, Aniwindup aegy fo PI-ype peed conolle, IEEE Tan. Ind. Elecon., Vol. 56, No. 6, pp. 39 46, 9. H.-B. Shin and J.-G. Pak, Ani-Windup PID Conolle Wih Inegal Sae Pedico fo Vaiable-Speed Moo Dive, IEEE Tan. Ind. Elecon., Vol. 59, No. 3, pp. 59 56,. F. Cupeino, E. Mininno, D. Nao, B. Tuchiano, and L. Salvaoe, On-line geneic deign of ani-windup unucued conolle fo elecic dive wih vaiable load, IEEE Tan. Evol. Compu., Vol. 8, No. 4, pp. 347 364, 4. J. M. Gome da Silva J. and S. Tabouiech, Ani-windup deign wih guaaneed egion of abiliy: an {LMI}-baed appoach, Vol. 5, No., pp. 445 4456, 3. L. Zaccaian and A. R. Teel, A common famewok fo ani-windup, bumple anfe and eliable deign, Auomaica, Vol. 38, No., pp. 735 744,. S. Mijalili, The An Lion Opimize, Adv. Eng. Sofw., Vol. 83, pp. 8 98, 5. T. S. L. V Ayyaao, G. I. Kihoe, and M. Rambabu, Adapive Conol of Sauaed Inducion Moo wih Unceain Load Toque, In. J. Recen Tend Eng. Technol., Vol. 3, No. 4, pp. 6,. Inenaional Jounal of Conol Theoy and Applicaion 94