Hybrid Evolutionary Algorithms based on PSO-GA for Training ANFIS Structure

Size: px
Start display at page:

Download "Hybrid Evolutionary Algorithms based on PSO-GA for Training ANFIS Structure"

Transcription

1 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): Hyrd Evolutonary Alorthms ased on PSO-GA for Trann ANFIS Structure S.Mlad.Nayyer Saet and MR.Deevand Department of Medcal Physcs and Bomedcal enneern, Shahd Behesht Unversty of Medcal Scences, Tehran, Iran Department of Medcal Physcs and Bomedcal enneern, Faculty of Medcne, Shahd Behesht Unversty of Medcal Scences, Tehran, Iran Astract Ths paper ntroduces a new approach for trann the adaptve network ased fuzzy nference system (ANFIS).In ths study we apply hyrd of Partcle swarm optmzaton wth Genetc Alorthm (PSOGA) n t to the trann the antecedent parameters and the concluson parameters ANFIS structure Fnally the method s appled to the dentfcaton of nonlnear dynamcal system and s compared wth asc BP, GA, PSO and showed qute satsfactory results. The proposed method s appled to dentfcaton of the nonlnear systems and predcton chaos systems Keywords: ANFIS, Genetc Alorthm, Swarm Intellent, Identfcaton, Trann.. Introducton TSK type [] [] s a fuzzy system wth crsp functons n consequent, whch perceved proper for complex applcatons [3].It has een proved that wth convenent numer of rules, a TSK system could approxmate every plant []. The TSK systems are wdely used n the form of a neural-fuzzy system called Adaptve Network-ased Fuzzy Inference System ANFIS [5]. ANFIS s a very effcent modeln method y comnn the attrutes of oth of fuzzy nference system and neural network. The comnaton of fuzzy loc wth archtectural desn of neural network led to creaton of neuro- fuzzy systems whch enft from feed forward calculaton of output and ack propaaton learnn capalty of neural networks, whle keepn nterpretalty of a fuzzy system [3].ANFIS has ood alty and performance n system dentfcaton, predcton and control and has een appled n many dfferent systems. The ANFIS has the advantae of ood applcalty as t can e nterpreted as local lnearzaton modeln and conventonal lnear technques for state estmaton and control are drectly applcale. The updatn and trann of ANFIS parameters that consst of the antecedent and concluson parameters s one of the man prolems. The trann ANFIS, n the antecedent parameters s more dffcult than the concluson parameters. Varous methods have een used to optmze the fuzzy memershp functons n recent years. These methods can e dvded nto two types ncludn dervatve ased and heurstc alorthms n eneral []. Shoorehdel et al [6-] proposed hyrd methods composed partcle swarm optmzaton (PSO) and et al [7] [8] used recursve least square (RLS) and extended Kalman flter (EKL) et al [9] for trann. In dfferent studes, they proposed forettn factor recursve least square for trann the concluson parameters and Lyapunov stalty theory to study the stalty of used alorthm [6]. In addton to these, they used NSGA-II the trann of all parameters of ANFIS structure []. Zaneneh et al [3] proposed a new type of trann ANFIS s applyn complex type (DE/current-toest/+/n & DE/rand//n) on predctn of Mackeylass tme seres. In ths paper, we propose trann ANFIS y usn hyrd evolutonary Alorthms ased on PSO-GA alorthm. The proposed method s appled to predcton of Mackey-lass tme seres and dentfcaton of a nonlnear dynamc system revealn the effcency of proposed structure. Ths paper s oranzed nto sx sectons. In the next secton, we revew ANFIS. In secton 3, we dscuss learnn alorthms for ANFIS structure. In secton, the trann ANFIS usn PSOGA s explaned. In secton 5, the smulaton results are ven.fnally, secton 6 presents our concludn.. The Concept of ANFIS ANFIS technque was ornally presented y Jan []. It comnes the advantaes of fuzzy loc and neural networks. The ANFIS s composed of two approaches neural network and fuzzy. If we compose these two ntellent approaches, t wll e acheve ood reasonn n qualty and quantty. In other words we have fuzzy reasonn and network calculaton. ANFIS s structure oranzes two parts. The frst part conssts of antecedent 05 Internatonal Journal of Computer Scence Issues

2 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): part and the second part conssts of concluson part. Antecedent part and concluson part are connected to each other y the fuzzy rules n network form. The updatn and trann of ANFIS depends on these parts. An adaptve network s a multlayer feed-forward network n whch each node performs a node functon on the ncomn snal as well as a set of parameters pertann to ths node of whch ts output depends. These parameters can e fxed or varale, and t s throuh the chane of the last ones that the network s tuned. ANFIS has nodes wth varale parameters, called square nodes whch wll represent the memershp functons of the antecedents, and the lnear functons for TSK-type consequent. The nodes n the ntermedate layers connect the antecedent wth the consequent. Ther parameters are fxed and they are called crcular nodes. Moreover, the network otaned ths way would not reman a lack ox, snce ths network would have fuzzy nference system capaltes to nterpret n terms of lnustc varales [5].The ANFIS structure s demonstrated n fve layers. It can e descred as a mult-layered neural network as shown n F.. Layer : The frst layer s called Fuzzfcaton layer. The parameters n ths layer are referred to as premse parameters. In fact, any dfferentale functon such as ell and tranular-shaped memershp functons (MFs) are vald for the nodes n ths layer. Every node n ths layer s a square node wth a node functon. Usually MFs are used as Gaussan wth maxmum equal to and mnmum equal to 0 such as: A (x) x c [( ) ] a B (x) x c [( ) ] a Where a,,c are the parameters of MFS whch are affected n shape of MFs. The parameters n ths layer are called the antecedent parameters. Layer : Ths layer s called rule layer. The rule layer represents the frn strenth for each rule en enerated n fuzzfcaton layer. They are crcular nodes wth a node functon: R O O, R O O, R O O, R O O (3) Layer 3: Ths layer s called normalzaton layer. rmalze layer calculates the normalzed frn strenth for each of the nputs. Ths normalzaton s the rato of the frn strenth of the th rule to the total of all frn strenths as ven (). R R ' R R R R 3 Layer : Ths layer s called defuzzfcaton layer. Every node n ths layer s a square node wth a node functon: F (x, x ) x x (5) 0 F (x, x ) R R R R R 3 (6) Layer 5: Ths layer s called sum layer. The snle node n ths layer computes the overall output as the sum of all ncomn snals. F (x, x ) R 3 O(x, x ) R R R R 3 F(x) (x ) A O (x, x ) (8) (x ) X X A A B B A O R R' Π N O Π Π Π R F.. The ANFIS network wth nputs and MF for an nput ANFIS approxmaton alty wll depend on the resoluton of the nput space parttonn whch s determned y the numer of MFs n ANFIS and the numer of layers. 3. Learnn Alorthms The susequent to the development of ANFIS approach, a numer of methods have een proposed for learnn the parameters and for otann an optmal numer of MFs. Jan [6] s ntroduced four methods to update the parameters of ANFIS structure, as lsted elow accordn to ther computaton complextes: N N N R' R'3 R' F F F3 F () (7) 05 Internatonal Journal of Computer Scence Issues

3 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): All parameters are updated y the radent descent. After the network parameters are set wth ther ntal values, the consequent part parameters s adjusted throuh the LSE whch t s appled only once at the very ennn. Then, the radent decent updates all parameters. The hyrd learnn comnn radent descent and LSE. Usn extended kalman flter to update all parameters. In ths paper we used a hyrd method whch has less complexty and fast converence 3. Genetc alorthms Genetc alorthms (GAS) are a famly of computatonal models developed y Holland who s nspred y evoluton. [7] [8]. These alorthms encode a potental soluton to a specfc prolem on a smple chromosome lke data structure and apply recomnaton operators to these structures so as to preserve crtcal nformaton. GAS are often vewed as functon optmzers, althouh the rane of prolems to whch GAS have een appled s qute road. Ths GA procedure s sketched n Alorthm. Alorthm (Pseudo code verson of the GA alorthm): {Intalzaton} 0 /* : eneraton counter */ for = to M do /* M: populaton sze*/ Intalze ndvduals x to random values F f (x ) /* f: ftness functon */ Pop {x, x... x M} F {F, F,..., F M} {Man Loop} whle (not termnaton condton) do {Genetc Operators} Pop Selecton (Pop, F) Pop Crossover (Pop, p c) /*pc: proalty of crossover*/ Pop Mutaton (Pop, p m)/* pm: proalty of mutaton */ {Evaluaton Loop} for = to M do F f (x) F {F, F,..., F M} + end whle In GA populatons are formulated as astract representatons (called chromosomes) of canddate solutons (called ndvduals or phenotypes) to an optmzaton prolem. Typcally, the alorthm mantans a populaton of M ndvduals Pop () ={x (),..., x M()} for each teraton (also called eneraton), where each ndvdual represents a potental soluton of the prolem. The alorthm s an teratve process n whch new populatons are otaned usn a selecton process ased on ndvdual adaptaton and some enetc operators (crossover and mutaton). In each eneraton, the ftness (a measure of the qualty of the represented soluton) of every ndvdual n the populaton s evaluated. The ndvduals wth the est adaptaton measure have more chance of reproducn and eneratn new ndvduals y crossn and mutn. The selecton process s repeated several tmes and the selected ndvduals form a tentatve new populaton for further enetc operator actons. After selecton some of the memers of the new tentatve populaton undero transformatons. A crossover operator creates two new ndvduals (off sprns) y comnn parts from two randomly selected ndvduals of the populaton. In GA the crossover operator s randomly appled wth a specfc proalty, p c. A ood GA performance requres the choce of a hh crossover proalty. Mutaton s a untary transformaton whch creates, wth certan proalty, pm, a new ndvdual y a small chane n a snle ndvdual. In ths case, a ood alorthm performance requres the choce of a low mutaton proalty (nversely proportonal to the populaton sze).the mutaton operator uarantees that all the search space has a nonzero proalty of en explored. Usn these enetc operators, the eneral structure of the alorthm s sketched n Alorthm. Ths procedure s repeated several tmes (thus yeldn successve eneratons) untl a termnaton condton has een reached. Common termnatn crtera are that a soluton s found that satsfes a lower threshold value, or that a fxed numer of eneratons has een reached, or that successve teratons no loner produce etter results. Ideally, the alorthm s expected to evolve over tme toward etter solutons, althouh converence to loal optma cannot e enerally assured. 3. Partcle swarm optmzaton Partcle swarm optmzaton (PSO) s also an evolutonary computatonal model whch s ased on swarm ntellence. PSO s developed y Kennedy and Eerhart who have een nspred y the research of the artfcal lfe [8]. Smlar to GA, PSO s also an optmzer ased on populaton. The PSO does not possess the crossover and mutaton processes adopted n GA. It fnds the optmum soluton y swarms follown the est partcle. Compared to GA, the PSO has much more profound ntellent ackround and could e performed more easly. In PSO alorthm, the soluton of each prolem s 05 Internatonal Journal of Computer Scence Issues

4 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): a rd n the search space called partcle. Every partcle has a ftness value whch s otann y ftness functon. The rd whch s closer to food has more ftness value. PSO starts wth a roup of random answers (partcles), and then t searches for fndn optmal answer n prolem space y updatn partcles poston. Every multdmensonal partcle (dependn on the prolem nature) s specfed y P d and V d whch denote the locaton poston and speed of dmenson d th of partcle. In every phase of swarm movement, poston of each partcle s updated y two est values. The frst value s the est answer accordn to ftness whch s otaned for each partcle separately untl now and s called P. Another value s the est value that s otaned y all of partcles throuh total swarms untl now and s called P. If a nehorhood s defned for every partcle, ths value s computed n that nehorhood. In ths case, ths value s called P. In every teratons of the alorthm, the new speed and poston of each partcle are updated y (9) and (0), after fndn and [9]: V ( s ) wv ( s) R [ P ( s) P( s)] R [ P ( s) P( s )] (9) d P ( s ) P ( s) V ( s ) (0) d d Where P d(s) and V d(s) are respectvely the poston and the velocty of partcle at tme s, w s called nerta weht and decde how much the old velocty wll affect the new one and coeffcents _and _ are constant values called learnn factors, whch decde the deree of affecton of P and P. In partcular, _ s a weht that accounts for the socal component, whle _ represents the contve component, accountn for the memory of an ndvdual partcle alon the tme. Two random numers, R and R, wth unform dstruton on [0, ], are ncluded to enrch the searchn space. Fnally, a ftness functon must e ven to evaluate the qualty of a poston. Ths PSO procedure s sketched n Alorthm. Alorthm (pseudo code verson of the PSO alorthm): {Intalzaton} s 0 /* s: tme varale*/ for = to N do /* N: sze of the swarm */ Intalze vectors V and P to random values P P P est { P ; =... N} /* ntal loal est */ {Man Loop} whle (not termnaton condton) do {Evaluaton Loop} for = to N do f f( P ) s etter than f ( P ) then P /* f: ftness functon*/ P /* partcle s est poston */ end f f f ( P ) s etter than f ( P ) then P P /* swarm s est poston */ end f {Update Loop} for = to N do V w.v +γ.r(0,).(p -P )+γ.r (0,).(P -P ) P P + V s s + end whle Ths procedure s repeated several tmes (thus yeldn successve eneratons) untl a termnaton condton s reached. Common termnatn crtera are that a soluton s found that satsfes a lower threshold value, or that a fxed numer of eneratons has een reached, or that successve teratons no loner produce etter results. 3.3 Hyrdzaton of GA and PSO The drawack of PSO s that stochastc approaches have prolem-dependent performance. Ths dependency usually results from the parameter settns n each alorthm. In eneral, no snle parameter settn can e appled to all prolems. Increasn the nerta weht (w) wll ncrease the speed of the partcles resultn n more exploraton (loal search) and less explotaton (local search). Thus fndn the est value for the parameter s not an easy task and t may dffer from one prolem to another. A further drawack s that the swarm may prematurely convere. Therefore, from the aove, t can e concluded that the PSO performance s prolem-dependent. The prolemdependent performance can e addressed throuh hyrd mechansm. It comnes dfferent approaches to e enefted from the advantaes of each approach. To overcome the lmtatons of PSO, hyrd alorthms wth GA are proposed. The ass ehnd ths s that such a hyrd approach s expected to have merts of PSO wth those of GA. One advantae of PSO over GA s ts alorthmc smplcty [0]. Ths smplcty wll result n the ncrease of speed calculatons and the reachn to the desred answer wth low volume of memory.another clear dfference etween PSO and GA s the alty to control converence. Crossover and mutaton rates can sutly affect the converence of GA, ut these cannot e analoous to the level of control acheved throuh 05 Internatonal Journal of Computer Scence Issues

5 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): manpulatn of the nerta weht. In fact, the decrease of nerta weht dramatcally ncreases the swarm's converence. The man prolem wth PSO s that t prematurely converes to stale pont, whch s not necessarly maxmum. To prevent the occurrence, poston update of the loal est partcles s chaned. The poston update s done throuh some hyrd mechansm of GA. By applyn crossover operaton, nformaton can e swapped etween two partcles to have the alty to fly to the new search area. The purpose of applyn mutaton to PSO s to ncrease the dversty of the populaton and the alty to have the PSO to avod the local maxma. There are three dfferent hyrd approaches are proposed n [0].n ths paper used PSO-GA-seres hyrd evolutonary alorthm (PSOGA). The PSO-GA performs N PSO s populatons smultaneously at frst. After M teratons the est partcles n each populaton are selected to encode nto chromosomes to consttute an N- ndvdual-populaton for GA operatons. Then the populaton should e run usn GA-operators. After M teratons the est soluton of GA should e transmtted ack to PSO populatons. We defne ap-pso, ap-ga and ap as the teraton numers of PSO susystem, GA su-system, and whole system, respectvely; M, M and GAP-MAX the correspondn permssle maxmum teraton numers, respectvely[].the flow chart of the PSO-GA s shown n F.. Learnn y PSOGA In ths secton, the trann and updatn of ANFIS parameters usn PSO-GA s explaned. The ANFIS has two types of parameters whch need trann, the antecedent part parameters and the concluson part parameters. The memershp functons are assumed Gaussan as n equaton (, ), and ther parameters are a,,c, where a s the varance of memershp functons and c s the center of MFs. Also s usually equal to. The parameters of concluson part are traned and here are represented wth p,q,r. There are 3 sets of tranale parameters n antecedent part, each of these parameters has N enes. Where, N represents the numer of MFs. The concluson parts parameters also are traned durn optmzaton alorthm. Each chromosome n concluson part has (I +) R enes that R s equal to Numer of rules and I denotes dmenson of data nputs. The ftness s defned as RMSE [9]. Parameters are ntalzed randomly n frst step and then are en updated usn PSO-GA alorthms. In each teraton, one of the parameters set are en updated..e. n frst teraton for example a s are updated then n second teraton are updated and then after updatn all parameters aan the frst parameter update s consdered and so on. Intalzaton of N PSO populaton Set ap=0 Set ap_pso=0 Do PSO operators for each populaton=0 Termnaton crteron met? If ap_pso=m? Gap_PSO++ Intalzaton of GA populaton Set ap_ga=0 DO GA operators Termnaton crteron met? If ap_ga=m? Gap_GA++ If ap>=gap_max? Best ndvdual=>pso populatons Gap++ Fnal soluton F.. Flow chart of PSO-GA 05 Internatonal Journal of Computer Scence Issues

6 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): Smulaton Results Ths secton presents the smulaton results of ANFIS trann usn ack propaaton (BP), enetc alorthm (GA), partcle swarm optmzaton (PSO) and PSOGA alorthm. In ths secton, we use two dfferent types of trann data set: a snle-nput and snle-output trann data set, four-nput and snle-output trann data set. A. Example (nlnear Functon Modeln) In ths applcaton, a snle-nput and snle-output trann data set s used defned y (). Defnn rane of x s [-, ], the system produces 5 roups of nput and output data. y 0.6sn( x) 0.3sn( x) 0.sn(5 x) () The parameters for trann the ANFIS y PSO-GA lsted n Tale. Tale: parameters PSOGA for trann ANFIS Swarm sze (populaton sze) 0 Num.of epochs 00 Crossover percentae(pc) 0.8 Mutaton percentae(pm) 0.3 Mutaton rate(µ) 0.0 Gamma(γ) 0. Selecton pressure(β) 8 Constrcton coeffcents(φ).05 The results are ven n Tale.Ths Tale shows tran, test RMSE and STD (standard devatons) for dfferent types of trann ANFIS structure. The RMSE of PSOGA s 0.0 for 0 rules, as ven n Tale. The comparsons wth ANFIS valdate the performance of the PSOGA. Num. rule Tale : Test, Tran RMSE and STD for Example Tran y Tran. Test. Tran. Test. RMSE RMSE STD STD BP PSO GA PSO- 8 GA B. Example (Predctn Chaos Dynamcs) The data s enerated from the Mackey-Glass tme-delay dfferental equaton whch s defned y: 0. x(t ) x 0 x t 0. x(t) () When x (0) =. and τ = 7, we have a non-perodc and non-converent tme seres that s very senstve to ntal condtons. (We assume x (t) = 0 when t < 0.) ANFIS structure has nputs and one output. We use 807/36 data as trann/test. F.3- s depcted predctn of Mackey-Glass tme seres usn PSOGA, GA, BP, and PSO to tran parameters n ANFIS structure. Ths raphc shows tran, test data for tarets and outputs and RMSE for errors usn y dfferent types of trann ANFIS structure. The numers of teratons of ths alorthm and populaton sze s 000 and 00 respectvely, we used 0 rules and other parameters s accordn Tale. 05 Internatonal Journal of Computer Scence Issues

7 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): F3.The ANFIS traned y GA for Tran data and RMSE F6. The ANFIS traned y PSO for Test data and RMSE F. The ANFIS traned y GA for Test data and RMSE F7.The ANFIS traned y BP for Tran Data and RMSE F5. The ANFIS traned y PSO for Tran data and RMSE F8.The ANFIS traned y BP for Test Data and RMSE 05 Internatonal Journal of Computer Scence Issues

8 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): Conclusons F9.The ANFIS traned y PSOGA for Tran Data and RMSE In ths paper, we proposed a novel method for trann the parameters of ANFIS structure. In our novel method we used PSOGA for updatn the parameters. The smulaton results show that PSOGA s successful for trann ANFIS for complex nonlnear systems and predcton chaos systems. Snce ths alorthm s free of dervaton whch s very dffcult to calculate for trann of antecedent and concluson part parameters complexty of ths new approach s less than other trann alorthms lke BP,GA and PSO. Also, the local mnmum prolem n BP alorthm for trann n ths novel approach s solved. The effectveness of the proposed PSOGA method was shown y applyn t to dentfcaton of nonlnear model. Acknowledments Ths work s part of my master s thess. Authors are ndeted to the Department of Medcal Physcs and Bomedcal enneern of Shahd Behesht Unversty for the support provded durn ths study. F0.The ANFIS traned y PSOGA for test Data and RMSE F.The RMSE-Iteraton ANFIS usn y PSOGA The smulatons results showed PSOGA optmzes ANFIS parameters etter than GA, BP and PSO for test and tran data. References [] M.Sueno and G. T. Kan, Structure dentfcaton of fuzzy model, Fuzzy sets and systems, pp.5-33, 998. [] T. Taka and M. Sueno, Fuzzy dentfcaton of systems and ts applcaton to modeln and control, IEEE Trans. Sys., Man& Cyernetcs, pp.6-3, 985. [3] R. Alcala, J. Casllas, O. Cordon and F. Herrera, Learnn TSK rule ased system from approxmate ones y mean of MAGUL methodoloy. Granada unversty of Span, Oct.000. [] M. Mannle, FSTM: Fast Taka-Sueno fuzzy modeln. Unversty of Karsruhe, 999. [5] Jyh-Shn Roer Jan, ANFIS: Adaptve Network Based Fuzzy Inference System, IEEE Trans. Sys., Man & Cyernetcs, vol.3, no.3, May-June 993. [6] M. A. Shoorehdel, M. Teshnehla and A. Sedh, "Trann ANFIS as an dentfer wth ntellent hyrd stale learnn alorthm ased on partcle swarm optmzaton and extended Kalman flter," Fuzzy Sets and Systems, vol. 60, pp. 9-98, 009. [7] M. A. Shoorehdel, M. Teshnehla and A. Sedh, "A novel trann alorthm n ANFIS structure," n Amercan Control Conference, IEEE, June 006. [8] M. A. Shoorehdel, M. Teshnehla and A. Sedh, Identfcaton usn ANFIS wth ntellent hyrd stale learnn alorthm approaches, Neural computn & applcatons, vol. 8(), pp. 57-7, 009. [9] M. A. Shoorehdel, M. Teshnehla and A. Sedh, vel hyrd learnn alorthms for tunn ANFIS parameters 05 Internatonal Journal of Computer Scence Issues

9 IJCSI Internatonal Journal of Computer Scence Issues, Volume, Issue 5, Septemer 05 ISSN (Prnt): ISSN (Onlne): usn adaptve wehted PSO, In Fuzzy Systems Conference, fuzzy-ieee 007, IEEE Internatonal, pp. -6, 007. [0] M. A. Shoorehdel, M. Teshnehla, A. K. Sedh and M. A. Khanesar, Identfcaton usn ANFIS wth ntellent hyrd stale learnn alorthm approaches and stalty analyss of trann methods, Appled Soft Computn, vol. 9(), pp , 009. [] V. S. Ghomsheh, M. A. Shoorehdel, M. Teshnehla, Trann ANFIS structure wth modfed PSO alorthm, In Control & Automaton, MED'07. Medterranean Conference on IEEE, pp. -6, June 007. [] A. Chatterjee and K. Watanae, An optmzed Taka- Sueno type neuro-fuzzy system for modeln root manpulators, Neural Computn & Applcatons, vol. 5(), pp. 55-6, 006. [3] A. Z. Zaneneh, M. Mansour, M. Teshnehla and A. K. Sedh, Trann ANFIS system wth DE alorthm, In Advanced Computatonal Intellence (IWACI), IEEE 0 Fourth Internatonal Workshop on, pp , Octoer 0. [] J. S. Jan, "ANFIS: Adaptve-network-ased fuzzy nference system," Systems, Man and Cyernetcs, IEEE Transactons on 3.3, pp , 993. [5] M. Kumar and P. G. Devendra, Intellent Learnn of Fuzzy Loc Controllers va Neural Network and Genetc Alorthm, Proceedns of 00 JUSFA 00 Japan-USA Symposum on Flexle Automaton Denver, Colorado, 00. [6] Jyh-Shn Roer Jan, ANFIS: Adaptve-Network-Based Fuzzy Inference System, IEEE Trans. Sys.Man and Cyernetcs. Vol. 3,.3, May/June 993. [7] J.H. Holland, Adaptaton n Natural and Artfcal System, the Unversty of Mchan Press, Ann Aror, 975. [8] D.E.Golder, Genetc Alorthms n Search, Optmzaton & Machne Learnn. Readn, MA: Addson Wesley, 989. [9] J. Kennedy, R.C. Eerhart, "Partcle swarm optmzaton", Proceedns of IEEE Internatonal Conference on Neural Networks, vol., 995. [0] K. Premalatha and A.M. Natarajan, "Hyrd PSO and GA for Gloal Maxmzaton", Int. J. Open Prolems Compt. Math., Vol.,., pp , Decemer 009. [] Sh, X.H.; Lu, Y.H.; Zhou, C.G.; Lee, H.P.; Ln, W.Z.; Lan, Y.C., "Hyrd evolutonary alorthms ased on PSO and GA," n Evolutonary Computaton, 003. CEC '03. The 003 Conress on, vol., no., pp Vol., 8- Dec MR.Deevand receved the B.Sc. deree from the shahd ahonar Kerman, Tehran, Iran, the M.Sc. deree from the Isfahan Unversty of Medcal Scence, Tehran, Iran, and the Ph.D. deree from Tarat Modaress Unversty, Tehran, Iran, n 998, all n medcal physc.from 0, he was a Faculty Memer at the Shahd Behesht Unversty of Medcal Scences and Health Servces, Tehran, Iran. He s currently an Assocate Professor wth the Faculty Medcal Physcs and Bomedcal enneern Department of Medcal Physcs and Bomedcal enneern. Hs nterest nterests nclude medcal mae processn, computatonal ntellence, ntellent optmzaton and radaton protecton. S. Mlad. Nayyer Saet was orn n Rasht, Iran, n 986. He receved the B.Sc. deree from the Unversty of Rasht, Iran, n electrcal enneern snce 008.and the M.Sc. deree from the Shahd Behesht Unversty of Medcal Scences and Health Servces, Tehran, Iran n omedcal enneern. He has een workn toward the M.Sc. deree n omedcal enneern at the Shahd Behesht Unversty. Hs man research nterests nclude o rootc, snal processn& medcal mae processn, computatonal ntellence, ntellent optmzaton, and mult ojectve optmzaton. He s currently don research n the feld of control o rootc systems, as M.Sc. thess. 05 Internatonal Journal of Computer Scence Issues

Training ANFIS Structure with Modified PSO Algorithm

Training ANFIS Structure with Modified PSO Algorithm Proceedngs of the 5th Medterranean Conference on Control & Automaton, July 7-9, 007, Athens - Greece T4-003 Tranng ANFIS Structure wth Modfed PSO Algorthm V.Seyd Ghomsheh *, M. Alyar Shoorehdel **, M.

More information

Affine Invariant Matching of Broken Boundaries Based on Differential Evolution

Affine Invariant Matching of Broken Boundaries Based on Differential Evolution Affne Invarant Matchn of Broken Boundares Based on Dfferental Evoluton Wuchao tu P.W.M. Tsan Department of Electronc Enneern Cty Unversty of Hon Kon Hon Kon Chna Abstract - Affne nvarant matchn of a par

More information

Decomposition of Grey-Scale Morphological Structuring Elements in Hardware

Decomposition of Grey-Scale Morphological Structuring Elements in Hardware Decomposton of Grey-Scale Morpholocal Structurn Elements n Hardware I Andreads, C Fyrndes, A Gasteratos and Y Boutals Secton of Electroncs & Informaton Systems Technoloy Department of Electrcal & Computer

More information

A Task Scheduling Algorithm Based on PSO for Grid Computing

A Task Scheduling Algorithm Based on PSO for Grid Computing Internatonal Journal of Computatonal Intellence esearch. ISSN 973-1873 Vol.4, No.1 (28, pp. 37 43 esearch Inda Publcatons http://www.jcr.nfo A Tas Scheduln Alorthm Based on PSO for Grd Computn Le Zhan

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Tuning of Fuzzy Inference Systems Through Unconstrained Optimization Techniques

Tuning of Fuzzy Inference Systems Through Unconstrained Optimization Techniques Tunng of Fuzzy Inference Systems Through Unconstraned Optmzaton Technques ROGERIO ANDRADE FLAUZINO, IVAN NUNES DA SILVA Department of Electrcal Engneerng State Unversty of São Paulo UNESP CP 473, CEP 733-36,

More information

CHAPTER 4 OPTIMIZATION TECHNIQUES

CHAPTER 4 OPTIMIZATION TECHNIQUES 48 CHAPTER 4 OPTIMIZATION TECHNIQUES 4.1 INTRODUCTION Unfortunately no sngle optmzaton algorthm exsts that can be appled effcently to all types of problems. The method chosen for any partcular case wll

More information

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS

EVALUATION OF THE PERFORMANCES OF ARTIFICIAL BEE COLONY AND INVASIVE WEED OPTIMIZATION ALGORITHMS ON THE MODIFIED BENCHMARK FUNCTIONS Academc Research Internatonal ISS-L: 3-9553, ISS: 3-9944 Vol., o. 3, May 0 EVALUATIO OF THE PERFORMACES OF ARTIFICIAL BEE COLOY AD IVASIVE WEED OPTIMIZATIO ALGORITHMS O THE MODIFIED BECHMARK FUCTIOS Dlay

More information

PC Cluster based Parallel PSO Algorithm for Optimal Power Flow

PC Cluster based Parallel PSO Algorithm for Optimal Power Flow The 4th Internatonal Conference on Intellent System Applcatons to Power Systems, ISAP 2007 ovember 4-8, 2007, Kaohsun, Tawan PC Cluster based Parallel PSO Alorthm for Optmal Power Flow Jon-Yul Km, Hee-Myun

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

A Notable Swarm Approach to Evolve Neural Network for Classification in Data Mining

A Notable Swarm Approach to Evolve Neural Network for Classification in Data Mining A Notable Swarm Approach to Evolve Neural Network for Classfcaton n Data Mnng Satchdananda Dehur 1, Bjan Bhar Mshra 2 and Sung-Bae Cho 1 1 Soft Computng Laboratory, Department of Computer Scence, Yonse

More information

K-means Optimization Clustering Algorithm Based on Hybrid PSO/GA Optimization and CS validity index

K-means Optimization Clustering Algorithm Based on Hybrid PSO/GA Optimization and CS validity index Orgnal Artcle Prnt ISSN: 3-6379 Onlne ISSN: 3-595X DOI: 0.7354/jss/07/33 K-means Optmzaton Clusterng Algorthm Based on Hybrd PSO/GA Optmzaton and CS valdty ndex K Jahanbn *, F Rahmanan, H Rezae 3, Y Farhang

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

An Influence of the Noise on the Imaging Algorithm in the Electrical Impedance Tomography *

An Influence of the Noise on the Imaging Algorithm in the Electrical Impedance Tomography * Open Journal of Bophyscs, 3, 3, 7- http://dx.do.org/.436/ojbphy.3.347 Publshed Onlne October 3 (http://www.scrp.org/journal/ojbphy) An Influence of the Nose on the Imagng Algorthm n the Electrcal Impedance

More information

Complexity Analysis of Problem-Dimension Using PSO

Complexity Analysis of Problem-Dimension Using PSO Proceedngs of the 7th WSEAS Internatonal Conference on Evolutonary Computng, Cavtat, Croata, June -4, 6 (pp45-5) Complexty Analyss of Problem-Dmenson Usng PSO BUTHAINAH S. AL-KAZEMI AND SAMI J. HABIB,

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

Classifier Swarms for Human Detection in Infrared Imagery

Classifier Swarms for Human Detection in Infrared Imagery Classfer Swarms for Human Detecton n Infrared Imagery Yur Owechko, Swarup Medasan, and Narayan Srnvasa HRL Laboratores, LLC 3011 Malbu Canyon Road, Malbu, CA 90265 {owechko, smedasan, nsrnvasa}@hrl.com

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation

Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation Internatonal Conference on Logstcs Engneerng, Management and Computer Scence (LEMCS 5) Maxmum Varance Combned wth Adaptve Genetc Algorthm for Infrared Image Segmentaton Huxuan Fu College of Automaton Harbn

More information

MHTalebpour@malcom [] W Ne j j L j W Nos a a a a T Nos W G Net f Net f G Ten T Ten T Ten T 4 Nec f Nec f G Com C Com C Com C 5 4 4 4 Ndof f Ndof f G 6 Ne L a Nos : G J NDN R Ne ID 7 ID J NDN R J Ne : G

More information

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

Estimation of Image Corruption Inverse Function and Image Restoration Using a PSObased

Estimation of Image Corruption Inverse Function and Image Restoration Using a PSObased Internatonal Journal of Vdeo& Image Processng and Network Securty IJVIPNS-IJENS Vol:10 No:06 1 Estmaton of Image Corrupton Inverse Functon and Image Restoraton Usng a PSObased Algorthm M. Pourmahmood,

More information

Analysis of Particle Swarm Optimization and Genetic Algorithm based on Task Scheduling in Cloud Computing Environment

Analysis of Particle Swarm Optimization and Genetic Algorithm based on Task Scheduling in Cloud Computing Environment Analyss of Partcle Swarm Optmzaton and Genetc Algorthm based on Tas Schedulng n Cloud Computng Envronment Frederc Nzanywayngoma School of Computer and Communcaton Engneerng Unversty of Scence and Technology

More information

An Efficient Genetic Algorithm with Fuzzy c-means Clustering for Traveling Salesman Problem

An Efficient Genetic Algorithm with Fuzzy c-means Clustering for Traveling Salesman Problem An Effcent Genetc Algorthm wth Fuzzy c-means Clusterng for Travelng Salesman Problem Jong-Won Yoon and Sung-Bae Cho Dept. of Computer Scence Yonse Unversty Seoul, Korea jwyoon@sclab.yonse.ac.r, sbcho@cs.yonse.ac.r

More information

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

Natural Computing. Lecture 13: Particle swarm optimisation INFR /11/2010

Natural Computing. Lecture 13: Particle swarm optimisation INFR /11/2010 Natural Computng Lecture 13: Partcle swarm optmsaton Mchael Herrmann mherrman@nf.ed.ac.uk phone: 0131 6 517177 Informatcs Forum 1.42 INFR09038 5/11/2010 Swarm ntellgence Collectve ntellgence: A super-organsm

More information

A Time-driven Data Placement Strategy for a Scientific Workflow Combining Edge Computing and Cloud Computing

A Time-driven Data Placement Strategy for a Scientific Workflow Combining Edge Computing and Cloud Computing > REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 1 A Tme-drven Data Placement Strategy for a Scentfc Workflow Combnng Edge Computng and Cloud Computng Bng Ln, Fangnng

More information

Application of Improved Fish Swarm Algorithm in Cloud Computing Resource Scheduling

Application of Improved Fish Swarm Algorithm in Cloud Computing Resource Scheduling , pp.40-45 http://dx.do.org/10.14257/astl.2017.143.08 Applcaton of Improved Fsh Swarm Algorthm n Cloud Computng Resource Schedulng Yu Lu, Fangtao Lu School of Informaton Engneerng, Chongqng Vocatonal Insttute

More information

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION 24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

Chinese Word Segmentation based on the Improved Particle Swarm Optimization Neural Networks

Chinese Word Segmentation based on the Improved Particle Swarm Optimization Neural Networks Chnese Word Segmentaton based on the Improved Partcle Swarm Optmzaton Neural Networks Ja He Computatonal Intellgence Laboratory School of Computer Scence and Engneerng, UESTC Chengdu, Chna Department of

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

Data Mining For Multi-Criteria Energy Predictions

Data Mining For Multi-Criteria Energy Predictions Data Mnng For Mult-Crtera Energy Predctons Kashf Gll and Denns Moon Abstract We present a data mnng technque for mult-crtera predctons of wnd energy. A mult-crtera (MC) evolutonary computng method has

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

Recommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm

Recommended Items Rating Prediction based on RBF Neural Network Optimized by PSO Algorithm Recommended Items Ratng Predcton based on RBF Neural Network Optmzed by PSO Algorthm Chengfang Tan, Cayn Wang, Yuln L and Xx Q Abstract In order to mtgate the data sparsty and cold-start problems of recommendaton

More information

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks

GA-Based Learning Algorithms to Identify Fuzzy Rules for Fuzzy Neural Networks Seventh Internatonal Conference on Intellgent Systems Desgn and Applcatons GA-Based Learnng Algorthms to Identfy Fuzzy Rules for Fuzzy Neural Networks K Almejall, K Dahal, Member IEEE, and A Hossan, Member

More information

Reliable and Efficient Routing Using Adaptive Genetic Algorithm in Packet Switched Networks

Reliable and Efficient Routing Using Adaptive Genetic Algorithm in Packet Switched Networks IJCSI Internatonal Journal of Computer Scence Issues, Vol. 9, Issue 1, No 3, January 2012 ISSN (Onlne): 1694-0814 www.ijcsi.org 168 Relable and Effcent Routng Usng Adaptve Genetc Algorthm n Packet Swtched

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

A Novel Deluge Swarm Algorithm for Optimization Problems

A Novel Deluge Swarm Algorithm for Optimization Problems A Novel eluge Swarm Algorthm for Optmzaton Problems Anahta Samad,* - Mohammad Reza Meybod Scence and Research Branch, Islamc Azad Unversty, Qazvn, Iran Soft Computng Laboratory, Computer Engneerng and

More information

An Evolvable Clustering Based Algorithm to Learn Distance Function for Supervised Environment

An Evolvable Clustering Based Algorithm to Learn Distance Function for Supervised Environment IJCSI Internatonal Journal of Computer Scence Issues, Vol. 7, Issue 5, September 2010 ISSN (Onlne): 1694-0814 www.ijcsi.org 374 An Evolvable Clusterng Based Algorthm to Learn Dstance Functon for Supervsed

More information

Hermite Splines in Lie Groups as Products of Geodesics

Hermite Splines in Lie Groups as Products of Geodesics Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the

More information

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

Using an Adaptive Neuro-Fuzzy Inference System (AnFis) Algorithm for Automatic Diagnosis of Skin Cancer

Using an Adaptive Neuro-Fuzzy Inference System (AnFis) Algorithm for Automatic Diagnosis of Skin Cancer Journal of Communcaton and Computer 8 (2011) 751-755 Usng an Adaptve Neuro-Fuzzy Inference System (AnFs) Algorthm for Automatc Dagnoss of Skn Cancer Suhal M. Odeh Department of Computer Informaton Systems,

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

Using Particle Swarm Optimization for Enhancing the Hierarchical Cell Relay Routing Protocol

Using Particle Swarm Optimization for Enhancing the Hierarchical Cell Relay Routing Protocol 2012 Thrd Internatonal Conference on Networkng and Computng Usng Partcle Swarm Optmzaton for Enhancng the Herarchcal Cell Relay Routng Protocol Hung-Y Ch Department of Electrcal Engneerng Natonal Sun Yat-Sen

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b

Clustering Algorithm Combining CPSO with K-Means Chunqin Gu 1, a, Qian Tao 2, b Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) Clusterng Algorthm Combnng CPSO wth K-Means Chunqn Gu, a, Qan Tao, b Department of Informaton Scence, Zhongka

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

A Two-Stage Algorithm for Data Clustering

A Two-Stage Algorithm for Data Clustering A Two-Stage Algorthm for Data Clusterng Abdolreza Hatamlou 1 and Salwan Abdullah 2 1 Islamc Azad Unversty, Khoy Branch, Iran 2 Data Mnng and Optmsaton Research Group, Center for Artfcal Intellgence Technology,

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Cracking of the Merkle Hellman Cryptosystem Using Genetic Algorithm

Cracking of the Merkle Hellman Cryptosystem Using Genetic Algorithm Crackng of the Merkle Hellman Cryptosystem Usng Genetc Algorthm Zurab Kochladze 1 * & Lal Besela 2 1 Ivane Javakhshvl Tbls State Unversty, 1, I.Chavchavadze av 1, 0128, Tbls, Georga 2 Sokhum State Unversty,

More information

AN IMPROVED GENETIC ALGORITHM FOR RECTANGLES CUTTING & PACKING PROBLEM. Wang Shoukun, Wang Jingchun, Jin Yihui

AN IMPROVED GENETIC ALGORITHM FOR RECTANGLES CUTTING & PACKING PROBLEM. Wang Shoukun, Wang Jingchun, Jin Yihui Copyrght 2002 IFAC 5th Trennal World Congress, Barcelona, Span A IPROVED GEETIC ALGORITH FOR RECTAGLES CUTTIG & PACKIG PROBLE Wang Shouun, Wang Jngchun, Jn Yhu Tsnghua Unversty, Beng 00084, P. R. Chna

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Identification of a Nonlinear System by Determining of Fuzzy Rules

Identification of a Nonlinear System by Determining of Fuzzy Rules Identfcaton of a Nonlnear System by Determnng of Fuzzy Rules Hodjatollah Hamd* Department of Industral Engneerng, K. N. Toos Unversty of Technology, Tehran, Iran h_hamd@kntu.ac.r Atefeh Darae Department

More information

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

More information

Hierarchical clustering for gene expression data analysis

Hierarchical clustering for gene expression data analysis Herarchcal clusterng for gene expresson data analyss Gorgo Valentn e-mal: valentn@ds.unm.t Clusterng of Mcroarray Data. Clusterng of gene expresson profles (rows) => dscovery of co-regulated and functonally

More information

Performance Evaluation of an ANFIS Based Power System Stabilizer Applied in Multi-Machine Power Systems

Performance Evaluation of an ANFIS Based Power System Stabilizer Applied in Multi-Machine Power Systems Performance Evaluaton of an ANFIS Based Power System Stablzer Appled n Mult-Machne Power Systems A. A GHARAVEISI 1,2 A.DARABI 3 M. MONADI 4 A. KHAJEH-ZADEH 5 M. RASHIDI-NEJAD 1,2,5 1. Shahd Bahonar Unversty

More information

(1) The control processes are too complex to analyze by conventional quantitative techniques.

(1) The control processes are too complex to analyze by conventional quantitative techniques. Chapter 0 Fuzzy Control and Fuzzy Expert Systems The fuzzy logc controller (FLC) s ntroduced n ths chapter. After ntroducng the archtecture of the FLC, we study ts components step by step and suggest a

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Structural optimization using artificial bee colony algorithm

Structural optimization using artificial bee colony algorithm 2 nd Internatonal Conference on Engneerng Optmzaton September 6-9, 2010, Lsbon, ortugal Structural optmzaton usng artfcal bee colony algorthm Al Hadd 1, Sna Kazemzadeh Azad 2, Saed Kazemzadeh Azad Department

More information

EECS 730 Introduction to Bioinformatics Sequence Alignment. Luke Huan Electrical Engineering and Computer Science

EECS 730 Introduction to Bioinformatics Sequence Alignment. Luke Huan Electrical Engineering and Computer Science EECS 730 Introducton to Bonformatcs Sequence Algnment Luke Huan Electrcal Engneerng and Computer Scence http://people.eecs.ku.edu/~huan/ HMM Π s a set of states Transton Probabltes a kl Pr( l 1 k Probablty

More information

Wavefront Reconstructor

Wavefront Reconstructor A Dstrbuted Smplex B-Splne Based Wavefront Reconstructor Coen de Vsser and Mchel Verhaegen 14-12-201212 2012 Delft Unversty of Technology Contents Introducton Wavefront reconstructon usng Smplex B-Splnes

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010 Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement

More information

A Novel Approach for an Early Test Case Generation using Genetic Algorithm and Dominance Concept based on Use cases

A Novel Approach for an Early Test Case Generation using Genetic Algorithm and Dominance Concept based on Use cases Alekhya Varkut et al, / (IJCSIT) Internatonal Journal of Computer Scence and Informaton Technologes, Vol. 3 (3), 2012,4218-4224 A Novel Approach for an Early Test Case Generaton usng Genetc Algorthm and

More information

Network Intrusion Detection Based on PSO-SVM

Network Intrusion Detection Based on PSO-SVM TELKOMNIKA Indonesan Journal of Electrcal Engneerng Vol.1, No., February 014, pp. 150 ~ 1508 DOI: http://dx.do.org/10.11591/telkomnka.v1.386 150 Network Intruson Detecton Based on PSO-SVM Changsheng Xang*

More information

USING MODIFIED FUZZY PARTICLE SWARM OPTIMIZATION ALGORITHM FOR PARAMETER ESTIMATION OF SURGE ARRESTERS MODELS

USING MODIFIED FUZZY PARTICLE SWARM OPTIMIZATION ALGORITHM FOR PARAMETER ESTIMATION OF SURGE ARRESTERS MODELS Internatonal Journal of Innovatve Computng, Informaton and Control ICIC Internatonal c 2012 ISSN 1349-4198 Volume 8, Number 1(B), January 2012 pp. 567 581 USING MODIFIED FUZZY PARTICLE SWARM OPTIMIZATION

More information

The Comparison of Calibration Method of Binocular Stereo Vision System Ke Zhang a *, Zhao Gao b

The Comparison of Calibration Method of Binocular Stereo Vision System Ke Zhang a *, Zhao Gao b 3rd Internatonal Conference on Materal, Mechancal and Manufacturng Engneerng (IC3ME 2015) The Comparson of Calbraton Method of Bnocular Stereo Vson System Ke Zhang a *, Zhao Gao b College of Engneerng,

More information

Biostatistics 615/815

Biostatistics 615/815 The E-M Algorthm Bostatstcs 615/815 Lecture 17 Last Lecture: The Smplex Method General method for optmzaton Makes few assumptons about functon Crawls towards mnmum Some recommendatons Multple startng ponts

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming CS 4/560 Desgn and Analyss of Algorthms Kent State Unversty Dept. of Math & Computer Scence LECT-6 Dynamc Programmng 2 Dynamc Programmng Dynamc Programmng, lke the dvde-and-conquer method, solves problems

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

More information

Lecture #15 Lecture Notes

Lecture #15 Lecture Notes Lecture #15 Lecture Notes The ocean water column s very much a 3-D spatal entt and we need to represent that structure n an economcal way to deal wth t n calculatons. We wll dscuss one way to do so, emprcal

More information

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented

More information

Genetic learning of accurate and compact fuzzy rule based systems based on the 2-tuples linguistic representation q

Genetic learning of accurate and compact fuzzy rule based systems based on the 2-tuples linguistic representation q Internatonal Journal of Approxmate Reasonng 44 (2007) 45 64 www.elsever.com/locate/jar Genetc learnng of accurate and compact fuzzy rule based systems based on the 2-tuples lngustc representaton q Rafael

More information

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

APPLICATION OF PREDICTION-BASED PARTICLE FILTERS FOR TELEOPERATIONS OVER THE INTERNET

APPLICATION OF PREDICTION-BASED PARTICLE FILTERS FOR TELEOPERATIONS OVER THE INTERNET APPLICATION OF PREDICTION-BASED PARTICLE FILTERS FOR TELEOPERATIONS OVER THE INTERNET Jae-young Lee, Shahram Payandeh, and Ljljana Trajovć School of Engneerng Scence Smon Fraser Unversty 8888 Unversty

More information

A soft computing approach for modeling of severity of faults in software systems

A soft computing approach for modeling of severity of faults in software systems Internatonal Journal of Physcal Scences Vol. 5 (), pp. 074-085, February, 010 Avalable onlne at http://www.academcjournals.org/ijps ISSN 199-1950 010 Academc Journals Full Length Research Paper A soft

More information

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT

THE PATH PLANNING ALGORITHM AND SIMULATION FOR MOBILE ROBOT Journal of Theoretcal and Appled Informaton Technology 30 th Aprl 013. Vol. 50 No.3 005-013 JATIT & LLS. All rghts reserved. ISSN: 199-8645 www.jatt.org E-ISSN: 1817-3195 THE PATH PLANNING ALGORITHM AND

More information

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like: Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

Optimizing SVR using Local Best PSO for Software Effort Estimation

Optimizing SVR using Local Best PSO for Software Effort Estimation Journal of Informaton Technology and Computer Scence Volume 1, Number 1, 2016, pp. 28 37 Journal Homepage: www.jtecs.ub.ac.d Optmzng SVR usng Local Best PSO for Software Effort Estmaton Dnda Novtasar 1,

More information

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples

More information

ARTICLE IN PRESS. Applied Soft Computing xxx (2012) xxx xxx. Contents lists available at SciVerse ScienceDirect. Applied Soft Computing

ARTICLE IN PRESS. Applied Soft Computing xxx (2012) xxx xxx. Contents lists available at SciVerse ScienceDirect. Applied Soft Computing ASOC-11; o. of Pages 1 Appled Soft Computng xxx (1) xxx xxx Contents lsts avalable at ScVerse ScenceDrect Appled Soft Computng j ourna l ho mepage: www.elsever.com/locate/asoc A herarchcal partcle swarm

More information

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton

More information

An Efficient Genetic Algorithm Based Approach for the Minimum Graph Bisection Problem

An Efficient Genetic Algorithm Based Approach for the Minimum Graph Bisection Problem 118 An Effcent Genetc Algorthm Based Approach for the Mnmum Graph Bsecton Problem Zh-Qang Chen, Rong-Long WAG and Kozo OKAZAKI Faculty of Engneerng, Unversty of Fuku, Bunkyo 3-9-1,Fuku-sh, Japan 910-8507

More information

An Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method

An Accurate Evaluation of Integrals in Convex and Non convex Polygonal Domain by Twelve Node Quadrilateral Finite Element Method Internatonal Journal of Computatonal and Appled Mathematcs. ISSN 89-4966 Volume, Number (07), pp. 33-4 Research Inda Publcatons http://www.rpublcaton.com An Accurate Evaluaton of Integrals n Convex and

More information

APPLICATION OF PREDICTION-BASED PARTICLE FILTERS FOR TELEOPERATIONS OVER THE INTERNET

APPLICATION OF PREDICTION-BASED PARTICLE FILTERS FOR TELEOPERATIONS OVER THE INTERNET APPLICATION OF PREDICTION-BASED PARTICLE FILTERS FOR TELEOPERATIONS OVER THE INTERNET Jae-young Lee, Shahram Payandeh, and Ljljana Trajovć School of Engneerng Scence Smon Fraser Unversty 8888 Unversty

More information

A reconstruction algorithm for electrical capacitance tomography via total variation and l 0 -norm regularizations using experimental data

A reconstruction algorithm for electrical capacitance tomography via total variation and l 0 -norm regularizations using experimental data A reconstructon alorthm for electrcal capactance tomoraphy va total varaton and l 0 -norm reularzatons usn expermental data Jaoxuan Chen 1,2, Maomao Zhan 1 and Y L 1,3 1 Graduate School at Shenzhen, Tsnhua

More information

NGPM -- A NSGA-II Program in Matlab

NGPM -- A NSGA-II Program in Matlab Verson 1.4 LIN Song Aerospace Structural Dynamcs Research Laboratory College of Astronautcs, Northwestern Polytechncal Unversty, Chna Emal: lsssswc@163.com 2011-07-26 Contents Contents... 1. Introducton...

More information

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

OPTIMIZATION OF FUZZY RULE BASES USING CONTINUOUS ANT COLONY SYSTEM

OPTIMIZATION OF FUZZY RULE BASES USING CONTINUOUS ANT COLONY SYSTEM Proceedng of the Frst Internatonal Conference on Modelng, Smulaton and Appled Optmzaton, Sharah, U.A.E. February -3, 005 OPTIMIZATION OF FUZZY RULE BASES USING CONTINUOUS ANT COLONY SYSTEM Had Nobahar

More information

Combining Cellular Automata and Particle Swarm Optimization for Edge Detection

Combining Cellular Automata and Particle Swarm Optimization for Edge Detection Combnng Cellular Automata and Partcle Swarm Optmzaton for Edge Detecton Safa Djemame Ferhat Abbes Unversty Sétf, Algera Mohamed Batouche Mentour Unversty Constantne, Algera ABSTRACT Cellular Automata can

More information

Target detection in SAR images via radiometric multi-resolution analysis

Target detection in SAR images via radiometric multi-resolution analysis Target detecton n SAR mages va radometrc mult-resoluton analyss JngwenHu a,, Gu-Song Xa, Hong Sun* a a School of Electronc Informaton, Wuhan Unversty, Wuhan, 43007, Chna State Key La. LIESMARS, Wuhan Unversty,

More information