Qi Kang*, Lei Wang and Qidi Wu

Size: px
Start display at page:

Download "Qi Kang*, Lei Wang and Qidi Wu"

Transcription

1 In. J. Bo-Inspred Compaon, Vol., Nos. /, Swarm-based approxmae dynamc opmzaon process for dscree parcle swarm opmzaon sysem Q Kang*, Le Wang and Qd W Deparmen of Conrol Scence and Engneerng, ongj Unversy, 4800 Caoan Gongl, Shangha 0804, Chna E-mal: q.kang07@gmal.com E-mal: wangle_j@6.com E-mal: wqd@moe.ed.cn *Correspondng ahor Absrac: hs paper presens a convergence analyss of parcle swarm opmzaon sysem by reang as a dscree-me lnear me-varan sysem frsly. And hen, based on he resls of sysem convergence condons, dynamc opmal conrol of a deermnsc PSO sysem for parameers opmzaon s sded by sng dynamc programmng; and an approxmae dynamc programmng algorhm swarm-based approxmae dynamc programmng (swarm-adp) s proposed n hs paper. Fnally, nmercal smlaons proved he valdaed of hs presened dynamc opmzaon mehod. Keywords: parcle swarm opmzaon; PSO; approxmae dynamc programmng; dynamc opmzaon. Reference o hs paper shold be made as follows: Kang, Q., Wang, L. and W, Q.D. (009) Swarm-based approxmae dynamc opmzaon process for dscree parcle swarm opmzaon sysem, In. J. Bo-Inspred Compaon, Vol., Nos. /, pp Bographcal noes: Q Kang s crrenly a PhD sden of he Conrol Deparmen n ongj Unversy, Shangha, Chna. Hs crren research neress nclde nellgen compaon and approxmae dynamc programmng. Le Wang receved hs PhD n Conrol Deparmen from ongj Unversy n Shangha n 998. He s crrenly a Professor of he Conrol Deparmen n ongj Unversy, Shangha, Chna. Hs crren research neress nclde nellgen conrol, nellgen compaon, CIMS and sysem engneerng. Qd W s a Professor of Conrol Engneerng and Managemen Scence a ongj Unversy, Shangha, Chna. She receved her PhD n Conrol Scence and Engneerng n 986 from EH, Zrch and Swzerland. Her crren research focses on nellgen conrol, nellgen compaon, CIMS and managemen scence. Inrodcon he parcle swarm opmzaon (PSO) sysem s a poplaon-based hersc global opmzaon echnology frs nrodced by Kennedy and Eberhar (995). I belongs o he caegory of swarm nellgence mehods, and s basc dea s based on he smlaon of smplfed anmal socal behavors sch as fsh schoolng, brd flockng, ec. Drng he las decade, PSO ganed ncreasng poplary de o s effecveness n performng dffcl opmzaon asks (Eberhar and Sh, 00), as well as smplcy of mplemenaon and he ably o qckly converge o a reasonably good solon. here has been a consderable amon of work done n developng PSO, hrogh emprcal smlaons (Wang e al., 004; Yang e al., 007; Asanga e al., 004), n he negraon of s self-adapaon, parameer selecon, swarm opology and negrang wh oher nellgen mehods. In addon, has been effecvely appled o power sysem opmzaon (Esmn e al., 005; Abdo, 00), prodc schedlng (Xa e al., 004), raffc plannng (Zhang and L, 007), comper nework opmzaon (Yan e al., 004), neral nework ranng (Cha-feng, 004), and mlple objecves opmzaon (Carlos e al., 004), ec. Recenly, he analyss of he sably and convergence of he parcle swarm sysem s geng more aenon by researchers n compaonal nellgence. Kennedy carred o he frs analyss of he smplfed parcles behavor (Kennedy, 998), who showed he dfferen parcle rajecores for a range of desgn choces for he gan Copyrgh 009 Inderscence Enerprses Ld.

2 6 Q. Kang e al. hrogh smlaons. Ozcan and Mohan (999) showed ha a parcle n a smple one-dmensonal PSO sysem follows a rajecory defned by a snsodal wave, randomly decdng on boh s amplde and freqency, nder he premse of me-nvaran. Under he same premse, Clerc and Kennedy (00) formally sded he analyss of he sably properes of he algorhm, whch presened hree models of consrcon facor mehod and analysed a parcle s rajecory n dscree me, hen progressed o he vew of n connos me. Emara and Faah (004) presened a knd of connos PSO sysem model, and proved s sably sng Lyapnov mehod. Vsakan e al. (006) provded a sably analyss of he sochasc parcle dynamcs, by represenng he parcle dynamcs as a non-lnear feedback conrol sysem, who he assmpon ha all parameers are non-random, as formlaed by Lre. Jn e al. (007) presens a sffcen condon for he PSO sysem mean-sqare o be sable, based on he assmpon of he prevos arcles and he heory of sochasc processes. In hs paper, we wll provde convergence analyss of PSO by reang as a dscree-me lnear me-varan sysem grond on above resls frsly; and hen, based on he resls of sysem convergence condon, dynamc opmal conrol for PSO sysem s sded; and an approxmae dynamc opmzaon mehod swarm-based approxmae dynamc programmng (swarm-adp) s proposed n hs paper. Fnally, nmercal smlaons proved he valdaed of hs presened dynamc opmzaon mehod. he paper s organsed as follows: n Secon, he sandard PSO algorhm s gven. In Secon 3, he convergence analyss of he PSO sysem s dscssed o oban he admssble conrol space. In Secon 4, he swarm-based approxmae dynamc opmzaon sed o parameer selecon for a dscree PSO sysem s sded. In Secon 5, llsrave nmercal smlaons and resls are gven, followed by he conclson of he paper. he PSO sysem he PSO sysem s a poplaon-based hersc global opmzaon echnology frs nrodced by Kennedy and Eberhar n 995. he algorhm manans a poplaon of parcles, where each parcle represens a poenal solon o an opmzaon problem. In he parcle swarm algorhm, he rajecory of each ndvdal n he search space s adjsed by dynamcally alerng he velocy of each parcle, accordng o s own flyng experence and he flyng experence of he oher parcles n he search space. Assme an m-dmensonal ( m N ) search space m S R, and le n denoe he swarm sze ( n N ). Each parcle ( = n) can be represened as an objec wh several characerscs. hen, he flgh velocy of parcle ( = n) for he nex fness evalaon n dd ( = m) dmensonal sbspace s calclaed sng eqaon (): v ( + ) = ω v ( ) + c r ( ) p ( ) x ( ) + d, d,, d d, d, ^ cr, d() pgd, () x, d() Where, v () s he parcle velocy a he h eraon, x () s he parcle poson a he h eraon. p s he personal bes poson of parcle acheved p o he crren eraon. p g s he global bes poson obaned so far by all ndvdals, represens he collaborave effec of he ndvdals. ω > 0 s called nera wegh. he acceleraon coeffcens c and c deermne he relave nflence of he socal and cognon componens, and are ofen boh se o he same vale o gve each componen (he cognon and socal learnng raes) eqal wegh. r U (0,) and r U (0,) are elemens from nform random seqences n he range (0,). he new poson for ndvdals s he addon of he poson a me and he dsance ha ndvdals wll fly wh he new velocy. he synchronos pdae of poson s hs: () x ( + ) = x () + v ( + ) () Each parcle wll compe her fness vale ( ()) f x, and hen, he personal bes poson of each ndvdal s pdaed sng he followng eqaon: p(), f f( x( + )) f( p()) p ( + ) = (3) x( + ), f f( x( + )) < f( p( )) he global bes poson fond by any parcle drng all prevos seps, p s defned as: g p ( + ) = arg mn f ( p ( + )), n (4) g p he psedo code for PSO algorhm s descrbed n Fgre. Fgre he psedo code of PSO algorhm Begn Inalse poplaon: boh poson x and velocy v; Se parameers, ω, c and, c ec; Ieraon=; Whle (ermnaon creron s no me) Do Evalae swarm sng correspondng fness fncon; Updae he personal bes poson p (personal hnkng) and he global bes poson p (swarm collaboraon); g Updae he velocy v sng eqaon (); Updae he poson x sng eqaon (); Ieraon = Ieraon + ; End Do (nl ermnaon creron s me) End

3 Swarm based approxmae dynamc opmzaon process 63 3 he convergence analyss of PSO sysem In hs secon, we wll presen convergence analyss of PSO by reang as a dscree-me lnear me-varan sysem. o smplfy he noaon, he dervaon wll be performed n only one dmenson n he whole paper, sng a sngle parcle. hs allows s o drop boh and j sbscrps, frher mplyng ha x denoes he vale of x a me, raher han he vale x assocaed wh parcle nmber. he sochasc componen wll also be removed emporarly wh he sbsons φ = cr (), φ = cr () and φ = φ+ φ. Frher, he poson of parcle wll be consdered n dscree me only, reslng n he followng eqaons: φ p + φ pg x+ = x + ω v + φ ( x ) φ φ p + φ pg v+ = ω v + ( φ )( x ) φ Consderng he varey of parameer ω, φ, a dscree-me lnear me-varan PSO sysem s derved. Here, we rea he PSO dynamc sysem as a feedback conrol sysem, n whch s ω sll consan. he eqaons governng he dynamcs can be expressed as: x ω x = v+ ω v y x = ( 0) v φ = φ ( y p), where, p = p + φ pg φ + φ heorem : (Vsakan e al., 006): Le he parcle dynamcs be represened by eqaons (6) (8) and sasfy wh an eqlbrm pon a he orgn. hen, he orgn s asympocally sable f ( ω + ω ) ω <, ω 0, Φ : = sp( φ ) < + ω Proof: Consder he Lyapnov fncon V( X) = X QX (0) Where, Q s a symmerc posve defne marx. he decrease n he sysem energy as represened by he Lyapnov fncon beween wo dscree-me nsans s gven by ( ) ( ) ( ) ( ) φ φ Δ V = V X V X = X QX X QX = X A QA Q X y B QAX + y B QB (5) (6) (7) (8) (9) () Snce φy( φy Φy) 0, f s added o he rgh-hand sde of he eqaon, we ge ( ) φ + φ y BB φ y ( φ y Φy ) ( ) ( φ ) ( φ y ) ( B QB) ΔV X A QA Q X y B QAX = X A QA Q X y B QA ΦC X () he rgh-hand sde s negave by compleng a sqare erm f he followng marx eqaons are sasfed: A QA Q = U U B QA =ΦC W U B QB = W W (3) Comparng hese wh he relaonshp esablshed n he Dscree-me Posve Real Lemma ndcaes ha f and only f he lnear sysem wh he ransfer fncon H ( z ) sasfes all he condons saed n he Posve Real Lemma. H( z) = ΦC( zi A) B+ D, D = (4) I s sraghforward hen o show ha H ( z) sasfes he condons n he Posve Real Lemma f j { Ge θ } ω <, ω 0 and +ΦR ( ) > 0 (5) { } j Ge ( θ + ω R ) > ( ω + ω ) Here, for each θ [0, π ). R{} ndcaes he real par of s argmen. I hen leads o ( ω + ω ) K < + ω Conseqenly, ς ς ( ) φ ς φ ( Uς φ y W) ( Uς φ y W) Uς φ y W 0 ΔV U U yw U y W W = = (6) (7) Snce he dfference Δ V n he Lyapnov fncon s nonncreasng, he parcle dynamc are garaneed o be sable, accordng o he Lyapnov sably heorem. 4 Dynamc opmzaon of PSO sysem In Secon 3, he convergence of me-varan PSO sysem s dscssed; and correspondng parameer admssble regons ensre sysem sable s worked o. In hs secon, nder he correspondng parameer resrcon, we wll perform he dynamc parameer opmzaon based on he feedback PSO sysem sng dynamc programmng heory, on prpose o fnd he opmal rajecory of PSO parameers ncldng feedback conrol for parameer seng. Dynamc programmng was developed by R.E. Bellman n he laer 950s (Bellman, 957). I can be sed o solve conrol problems for non-lnear, me-varyng sysems and

4 64 Q. Kang e al. s sraghforward o program, whch s based on Bellman s prncple of opmaly. 4. Fne horzon dynamc opmzaon for deermnsc nbonded PSO sysem Consder he PSO dynamc sysem descrbed by eqaons (6) (8), whch has he assocaed performance ndex: N J; N( X) = mn Obj( X N) + Obj( X) + (8) = Where, Obj() s defned as he objecve fncon, N denoes he maxmm erave mes. he framework of dynamc opmser for PSO sysem dynamc opmzaon s descrbed as Fgre. Fgre o begn, le PSO sysem opmzaon sng dynamc programmng (see onlne verson for colors) N N = N and wre J ( X ) = Obj( X ) (9) I s he evalaon of objecve fncon o be opmsed a me N. Decremen o and N wre J Obj X Obj X N = ( N) + ( N ) + N N (0) Fnd N () by mnmsng (0). o do hs, se he sae eqaon o wre JN = mn { Obj( XN ) + N N } + Obj( XN) N = mn { Obj( X N ) + N N } N + Obj( AX + B ) N N } () If here are no consrans, he mnmm of J N s fond by J Obj( AX + B ) 0 = = + N N N N N N () Obj( AX N + BN ) N = arg N + = 0 (3) N N If we conned o decremen and apply he opmaly prncple, he resl for each = N,,, 0 s Obj( AX B) arg + = + = 0 4. Swarm-ADP for deermnsc PSO sysem (4) Dynamc programmng s a very sefl ool n solvng opmzaon and opmal conrol problems. However, s ofen compaonally nenable o rn re dynamc programmng de o he backward nmercal process reqred for s solon,.e., as a resl of he well-known crse of dmensonaly (Powell, 007). One has o fnd a seres of conrol acons ha ms be aken n seqence. hs seqence wll gve he opmal performance cos, b he oal cos of hose acons s nknown nl he end of ha seqence. Consderng he crse of dmensonaly of dynamc programmng, many approxmae mehods are proposed. A very general algorhm s based on he approxmaon of he cos-o-go fncon by means of some fxed-srcre paramerc archecre, whch s raned on he bass of sample pons comng from he dscresaon of he sae space. here are many examples of sch approach, where dfferen srcres are employed, polynomal approxmae, splnes, mlvarae adapve regresson splnes and neral neworks. here are several synonyms sed ncldng adapve crc desgns (ACD), approxmae dynamc programmng, neral dynamc programmng, renforcemen learnng and so on. For he comper mplemen of boh dscree-me PSO sysem and connos-me PSO sysem, we canno ge he opmal general solon lke eqaon (4), b a seral * dscree daa of φ k, whch s sed o f o he dynamc of * φ k. In addon, s hard o oban he resls by solvng he HJB eqaons drecly, becase dfferen problems have dsnc objecve fncons, and s sally non-lnear; frhermore, he conrol resrcon redces he compaon complexy, b canno avod crse of dmensonaly radcally. herefore, we propose a swarm-adp for opmzaon of PSO sysem. In he swarm-adp, he hersc sochasc poplaon searchng mehod s adoped o ry conrol laws n s admssble errory o fnd a near opmal conrol for PSO dynamc sysem. If conrol varable s defned n a ml-dmensonal space, swarm-adp wll p p more advanages becase of s characerscs of sochasc search and swarm nellgence. Insead of solvng HJB dfferenal eqaon, swarm-adp fnds he mnmal by drec calclaons sng some learnng cells, sch as neral neworks, ec. We wll sdy he sysem sep by sep o fnd he relaonshp of opmal conrols and opmal

5 Swarm based approxmae dynamc opmzaon process 65 performance coss sng parcle swarm o deermne he opmal conrol seqence. Defnon : Admssble sae space Ω AS : he admssble rajecores ms belong smlaneosly o hese reachable n n domans R and conrollable domans R n ΩR ΩC n-dmensonal space. ha s, a any gven me, he orgn ms be reachable from he presen sae, and he presen sae ms be reachable from he se of gven nal saes. hs, he admssble sae se s defned as n Ω : =Ω Ω R. AS R C Defnon : Admssble polcy space Ω : correspondng o he admssble sae space Ω AS, he polcy appled o he sysem make p of a l-dmensonal admssble polcy space l Ω R. Inally, defne parcle s codng forma and nalse he l poplaon n admssble polcy space Ω R. A nal sage 0, nalse V ( x ) and parcle s fness vale s calclaed by cos fncon Vx ( 0) = mn { rx ( 0, 0) + Vx ( ( 0, 0)) }. 0 Ω Once he opmal polcy * s fond, s saved by he 0 learnng cell 0 ( x 0 ),one b of he opmal polcy nework OPN 0 ( x ), composng opmal polcy seqence nework OPSN ( x. ) Meanwhle, V( x 0) s sored o he opmal cos nework OCN J ( x ), a hs sage, s recorded as J 0 ( x ). herefore, when solvng he Bellman eqaon V( x ) = mn r( x, ) + V( ( x, )), he vale of { } Ω Vx ( (, )) for correspondng sae x = ( x, ) can be + derved from J ( x ). However, only fne nmercal vales for dscree saes are recorded by neworks x ( ) and J ( x ), n each sage of solvng he Bellman eqaon. Usally, he ransfer saes () canno mach wh he saes n J ( x ) accraely. herefore, a knd of lnear nerpolaon approxmae mehod s adoped n ranng neworks. Usng he same mehod, we compe V( x ) by parcle swarm o fnd he near-opmal conrol law n accordngly, we can ge * and V( x ) Ω,, recorded by cell ( x ) and OCN J ( x ) sng lnear nerpolaon, respecvely. he res may be dedced by analogy, we ge J ( x), J ( x), J ( x),, nl he seqence converge. Meanwhle, he opmal polcy seqence x ( ) = { ( x), ( x), ( x), } 0 0 mappng { J ( x), J ( x), J ( x), } J( x) s obaned. Afer geng he opmal polcy seqence ( x ), we shold carry o he daa fng accordng o feedback conrol mode of he PSO dynamc sysem, o approxmae opmal parameers accordngly. he resls wll help o nsrc he parameer seng for achevng beer algorhm performance. Algorhm : Swarm-ADP Sep Inalse he sae vecor X 0 R n one-dmensonal admssble sae space AS randomly, ( ) X : = X (), X (),, X ( n) ncldes n sample saes, and se he nal cos fncon V ( X 0 ) = 0 n ; accordngly, he OCN J : J 0 ( X ) = 0 (nework nalsaon ) for connos sae X Ω AS. here are wo nalsaon sraeges: V( X 0 ) = 0 n ; esmae V( X 0) by applyng fne sable conrols. Sep Perform sae ranson X X + for each sample sae X () s, s =,,, n, n he general sage accordng o he followng PSO dynamc fncon (sae fncon): ω X+ () s = X() s + () s 0 ω [ φ ] (5) () s = φ y () s = F X () s = 0 X () s (6) Here, conrol polcy () s Ω brngs a sae ranson. Sep 3 Hersc sochasc opmzaon of conrol polcy () s Ω for he PSO dynamc feedback sysem descrbed by eqaons (5) (6) sng PSO: 3. Parcle codng and nalsaon. For sample sae X () s, defne parcle swarm PS() s, whch ncldes N parcles. he poson x () s of parcle ( =,,, N ) represens he polcy appled o sae () s a hs sage, whch s nalsed randomly n X Ω. For a l-dmensonal conrol sysem, parcle shold be represened as x(, s j): =, (, s j), h j =,,, l denoes he j sb-conrol n admssble space Ω. he velocy of parcle v () s s defned as

6 66 Q. Kang e al. he reglaon of conrol v(): s =Δ, (), s hrogh whch o carry o polcy search n space l. 3. A he general sage, mplemen swarm opmzaon of bonded conrol () s Ω for each sample sae X (), s s =,, n. he flgh velocy of parcle and s synchronos pdae of poson are calclaed as: Δ, (, sk+ ) = ω Δ, (, sk) + cr (, sk)[ p, (, sk), (, sk)] ^ + cr (, sk) pg, (, sk),(, sk) (7), (,k s + ) =, (,k) s +Δ, (,k s + ), Where, a maxmm velocy Δ max () s for each modls of he velocy vecor s defned n order o conrol excessve roamng of parcles osde he ser defned space. Whenever Δ, () s exceeds he defned lm, s se as Δ max () s. For parcle, s fness fncon s defned as follows: (, (, sk) ) = mn { rx ( ( s),, ( sk, )) + VX ( ( ( s),, ( sk, ))) } F (,) s k Ω (8) Here, rx ( ( s), ( sk, )) : = ObjX ( ( s)) + ( sk, ) ( sk, );,,, n whch, Obj() represens he objecve fncon; and ( ( ( ),, (, ))) ( ( ( ),, (, ))) V X s s k = J X s s k, whch s obaned from he OCN J ( X ) descrbed laer. Compe each parcle s fness vale (, (, )) F s k, and hen, he personal bes poson of each ndvdal s pdaed sng he followng eqaon: p, (, s k), f F(, (, s k+ )) F( p, (, s k)) p, (, s k+ ) = (9), (, s k+ ), f F(, (, s k+ )) < F( p, (, s k)) he global bes poson fond by any ndvdal drng all prevos seps, p, g( s, k ) s defned as: g,, p, ( ) p (, s k) = argmn F p (, s k), n (30) Perform swarm opmzaon eraons, le k = k+. Afer schedled eraons k = C, he global opmal max * conrol s fond, ( s) = p (, max ) g s C for all sample saes X (), s s =,, n. I s sored by he opmal polcy cell () s, whch s nclded n he OPN : =. () () ( n) For = 0,,,, he opmal cos V+ ( X ( s )) s pdaed accordng o he Bellman dynamc programmng fncon: { ( ) ( )} V ( X ( s)) = mn r X ( s), ( s) + V ( X ( s), ( s)) +,,, Ω ( ) * * * = Obj( X()) s + () s () s + V ( (), ()) X s s Here, V ( ( ( ), * ( ))) ( ( ( ), * X s s J X ( ))) s s =, whch s obaned from he opmal cos nework J ( X ). A he general sage, he opmal cos V( X( s )) s obaned and sored as J( X( s )), by applyng he opmal polcy seqence * * * * (): s = 0(), s (), s, () s n rn, whch s { } sored as he OPSN. Accordngly, * * * * : = () () ( n) * * * 0() () () * * * 0() () () = * * * 0( n) ( n) ( n) Sep 4 Approxmae he cos for all sae X, denoed by he OCN J ( X ) (nework learnng), hrogh lnear nerpolaon of V ( X ( s)), s =,, n. (3) (3) Le = +, repea Sep 3, nl V+ ( X ) V ( X ) < ε, ε s an arbrary posve real nmber; accordngly, J ( X ) J( X ), or he maxmm sage nmber s me = max. Op he opmal polcy seqence * * * * : () () ( ) = n. For nfne horzon opmal conrol, s possble o approxmae he J ( X ) hrogh few saes afer adeqae eraons, becase V( X ) can raverse almos all saes when. Sep 5 Daa fng and analyss of he polcy seqence * * * * : = () () ( n), and he resls and crves wll edce he correspondng opmal parameers, whch wll be help o nsrc parameer seng of PSO algorhm. he PSO sysem adopng he opmsed parameers wll be opmsed crclarly n hs same way.

7 Swarm based approxmae dynamc opmzaon process 67 Fgre 3 he framework of swarm-adp for -D PSO sysem (see onlne verson for colors) 5 Nmercal smlaons y x Le p = 0, y = x p = x, X =, v = v deermnsc PSO sysem can be represened: ω = 0.7 X+ = X + = ( X, ) 0 ω = 0.7 he (33) Accordng o he resls of relaonshp, assre of sysem convergence, ( ω + ω ) ω <, ω 0, Φ : = sp( φ ) <, we can + ω ge Φ ( ω = 0.7) < Accordngly, we can oban he admssble bond by = φx : Ω : = sp( ) = x (34)

8 68 Q. Kang e al. Fgre 4 he admssble bond Ω (see onlne verson for colors) * ( f) ( 0.x x x 3.54x 0.x + 6.4x 0.06) = * ( f) 0 = (0.4 x + 0.6x x 3.34x 0.7x x+ 0.0) (36) (37) Fgre 5 Smlaon resls for Sphere fncon (a) cos vale crve (b) opmal polcy rajecory (c) polcy fng (see onlne verson for colors) For he objecve fncon, defned n he assocaed performance ndex, wo benchmarks are adoped n he nmercal smlaon, whch s presened n able. able Benchmarks for smlaons Fncon Mahemacal represenaon Range of search (a) Sphere n f( x) = x = [,] Generalsed Grewank n n x f() x = cos( ) 4000 x x + = = [,] Bellman s dynamc programmng eqaon: J+ ( X( s)) = (35) mn Obj( X( s)) + ( s) + J ( ( ( ), ),, ( )), X s s Ω he parameers of or swarm dynamc programmng algorhm s se as follows: he sze of he parcle swarm s N = 50; he range of nera wegh facor ω s se as [ ωmn, ω max ] = [0.3,.0], and s lnear dynamc redced drng he whole opmzaon process; acceleraon consans c = c =.0; Vmax = Xmax ;.he maxmm eraon s C max = 00. he algorhm ermnae condon s sage = max, and he maxmm sage nmber for dynamc programmng s max = 00. Fgres 5 6 presens he smlaon resls for Sphere and Generalsed Grewank fncon, n whch, he resls nclde he cos vale, opmal polcy rajecory and polcy fng resls, ec. he fng fncons for Sphere ( f ) and Grewank ( f ) are presened respecvely as follows: (b) (c)

9 Swarm based approxmae dynamc opmzaon process 69 Fgre 6 Smlaon resls for Generalsed Grewank (a) cos vale crve (b) opmal polcy rajecory (c) polcy fng (see onlne verson for colors) he PSO opmal conrol problem dscssed n hs paper s parameer opmzaon, n whch he reqremen of resl precson s no so src; herefore, hs knd of precson sacrfce s deserved, whch redce he complexy grealy. In he work of hs paper, random varables n PSO sysem are gnored; n addon, s grond on he premse of sngle ndvdal flyng n one-dmensonal space. herefore, s or fre work and promsng projec o sdy approxmae dynamc opmzaon for a sochasc PSO sysem, as well as ml-dmensonal PSO sysem. References 6 Conclsons (a) (b) (c) hs paper provdes he convergence analyss of PSO by reang as a dscree me-varan feedback sysem. Based on he resls of sysem convergence condons, dynamc opmal conrol for a deermnsc PSO sysem s sded for parameers opmzaon; and a swarm-adp s proposed. Nmercal smlaons proved he valdaed of hs presened mehod. In swarm-adp, approxmae s sed o redce he compaon complexy; he fnal resls may have devaon from he precse vales. However, he objecve of Abdo, M.A. (00) Opmal power flow sng parcle swarm opmzaon, Elecrcal Power and Energy Sysem, Vol. 4, pp Asanga, R., Saman, K.H. and Harry, C.W. (004) Self-organzng herarchcal parcle swarm opmzer wh me-varyng acceleraon coeffcens, IEEE ransacon on Evolonary Compaon, Vol. 8, No. 3, pp Bellman, R.E. (957) Dynamc Programmng, Prnceon Unversy Press, Prnceon, 957. Carlos, A.C., Gregoro,.P. and Maxmno, S.L. (004) Handlng mlple objecves wh parcle swarm opmzaon, IEEE ransacons on Evolonary Compaon, Vol. 8, No. 3, pp Cha-feng, J. (004) A hybrd of genec algorhm and parcle swarm opmzaon for recrren nework desgn, IEEE ransacon on Sysems, Man and Cybernecs, Par B: Cybernecs, Vol. 34, No., pp Clerc, M. and Kennedy, J. (00) he parcle swarm-exploson, sably and convergence n a mldmensonal complex space, IEEE ransacons on Evolonary Compaon, Vol. 6, No., pp Eberhar, R.C. and Sh, Y. (00) Parcle swarm opmzaon: developmens, applcaons and resorces, Proc. Congress on Evolonary Compaon, IEEE, Pscaaway, pp Emara, H.M. and Faah, A.H.A. (004) Connos swarm opmzaon echnqe wh sably analyss, Proceedngs of Amercan Conrol Conference, Caro, Egyp, pp Esmn, A., Lamber-orres, G. and Zambron de Soza, A.C. (005) A hybrd parcle swarm opmzaon appled o loss power mnmzaon, IEEE ransacons on Power Sysems, Vol. 0, No., pp Jn, X.L., Ma, L.H. and W,.J. (007) Convergence analyss of he parcle swarm opmzaon based on sochasc processes, ACA Aomaca Snca, Vol. 33, No., pp Kennedy, J. (998) he behavor of parcle, Proc. 7h Ann. Conf. Evolon Programmng, March, San Dego, CA, pp Kennedy, J. and Eberhar, R. (995) Parcle swarm opmzaon, Proc. Inernaonal Conference on Neral Ne-works, IEEE, Perh, Asrala, pp Ozcan, E. and Mohan, C.K. (999) Parcle swarm opmzaon: srfng he waves, Proceedngs of Congress on Evolonary Compa, IEEE, Washngon D.C., USA, pp Powell, W.B. (007) Approxmae dynamc programmng: solvng he crses of dmensonaly, Wley-Inerscence, 007.

10 70 Q. Kang e al. Vsakan, K., Krsnaplla, S. and Flemng, P.J. (006) Sably analyss of he parcle dynamcs n parcle swarm opmzer, IEEE ransacons on Evolonary Compaon, Vol. 0, No. 3, pp Wang, L., Kang, Q. and W, Q.D. (004) Groped-and-delayed broadcasng mechansm for opmm nformaon n parcle swarm opmzaon, Proceedngs of he IEEE Inernaonal Conference on Cybernecs and Inellgen Sysems, Dec. 3, Sngapore, pp Xa, W.J., W, Z.M., Zhang, W., e al. (004) Applyng parcle swarm opmzaon o job-shop schedlng problem, Chnese Jornal of Mechancal Engneerng (Englsh Edon), Vol. 7, No. 3, pp Yang, X., Yan, J., Yan, J. and Mao, H. (007) A modfed parcle swarm opmzer wh dynamc adapaon, Appled Mahemacs and Compaon, Vol. 89, No., pp Yan, P., J, C.L. and Zhang, Y.Y. (004) Opmal mlcas rong n wreless ad hoc sensor neworks, IEEE Inernaonal Conference on Neworkng, Sensng and Conrol, pp Zhang, Z.X. and L, C.W. (007) Applcaon of he mproved parcle swarm opmzer o vehcle rong and schedlng problems, IEEE In. Conf. on Grey Sysems and Inellgen Servces, pp.50 5.

CONSISTENT EARTHQUAKE ACCELERATION AND DISPLACEMENT RECORDS

CONSISTENT EARTHQUAKE ACCELERATION AND DISPLACEMENT RECORDS APPENDX J CONSSTENT EARTHQUAKE ACCEERATON AND DSPACEMENT RECORDS Earhqake Acceleraons can be Measred. However, Srcres are Sbjeced o Earhqake Dsplacemens J. NTRODUCTON { XE "Acceleraon Records" }A he presen

More information

Stochastic Programming handling CVAR in objective and constraint

Stochastic Programming handling CVAR in objective and constraint Sochasc Programmng handlng CVAR n obecve and consran Leondas Sakalaskas VU Inse of Mahemacs and Informacs Lhana ICSP XIII Jly 8-2 23 Bergamo Ialy Olne Inrodcon Lagrangan & KKT condons Mone-Carlo samplng

More information

, t 1. Transitions - this one was easy, but in general the hardest part is choosing the which variables are state and control variables

, t 1. Transitions - this one was easy, but in general the hardest part is choosing the which variables are state and control variables Opmal Conrol Why Use I - verss calcls of varaons, opmal conrol More generaly More convenen wh consrans (e.g., can p consrans on he dervaves More nsghs no problem (a leas more apparen han hrogh calcls of

More information

Observer Design for Nonlinear Systems using Linear Approximations

Observer Design for Nonlinear Systems using Linear Approximations Observer Desgn for Nonlnear Ssems sng Lnear Appromaons C. Navarro Hernandez, S.P. Banks and M. Aldeen Deparmen of Aomac Conrol and Ssems Engneerng, Unvers of Sheffeld, Mappn Sree, Sheffeld S 3JD. e-mal:

More information

Hierarchical Sliding Mode Control for Series Double Inverted Pendulums System

Hierarchical Sliding Mode Control for Series Double Inverted Pendulums System Herarchcal Sldng Mode Conrol for Seres Doble Invered Pendlms Sysem Danwe Qan, Janqang Y, Dongbn Zhao, and Ynxng Hao Laboraory of Complex Sysems and Inellgence Scence Inse of Aomaon, Chnese Academy of Scences

More information

Dynamic Model of the Axially Moving Viscoelastic Belt System with Tensioner Pulley Yanqi Liu1, a, Hongyu Wang2, b, Dongxing Cao3, c, Xiaoling Gai1, d

Dynamic Model of the Axially Moving Viscoelastic Belt System with Tensioner Pulley Yanqi Liu1, a, Hongyu Wang2, b, Dongxing Cao3, c, Xiaoling Gai1, d Inernaonal Indsral Informacs and Comper Engneerng Conference (IIICEC 5) Dynamc Model of he Aally Movng Vscoelasc Bel Sysem wh Tensoner Plley Yanq L, a, Hongy Wang, b, Dongng Cao, c, Xaolng Ga, d Bejng

More information

by Lauren DeDieu Advisor: George Chen

by Lauren DeDieu Advisor: George Chen b Laren DeDe Advsor: George Chen Are one of he mos powerfl mehods o nmercall solve me dependen paral dfferenal eqaons PDE wh some knd of snglar shock waves & blow-p problems. Fed nmber of mesh pons Moves

More information

Real-Time Trajectory Generation and Tracking for Cooperative Control Systems

Real-Time Trajectory Generation and Tracking for Cooperative Control Systems Real-Tme Trajecor Generaon and Trackng for Cooperave Conrol Ssems Rchard Mrra Jason Hcke Calforna Inse of Technolog MURI Kckoff Meeng 14 Ma 2001 Olne I. Revew of prevos work n rajecor generaon and rackng

More information

Variants of Pegasos. December 11, 2009

Variants of Pegasos. December 11, 2009 Inroducon Varans of Pegasos SooWoong Ryu bshboy@sanford.edu December, 009 Youngsoo Cho yc344@sanford.edu Developng a new SVM algorhm s ongong research opc. Among many exng SVM algorhms, we wll focus on

More information

Lagrangian Relaxation-Based Unit Commitment Considering Fast Response Reserve Constraints *

Lagrangian Relaxation-Based Unit Commitment Considering Fast Response Reserve Constraints * Energy and Power Engneerng 2013 5 970-974 do:10.4236/epe.2013.54b186 Pblshed Onlne Jly 2013 (hp://www.scrp.org/jornal/epe) Lagrangan Relaxaon-Based Un Commmen Consderng Fas Response Reserve Consrans *

More information

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS

V.Abramov - FURTHER ANALYSIS OF CONFIDENCE INTERVALS FOR LARGE CLIENT/SERVER COMPUTER NETWORKS R&RATA # Vol.) 8, March FURTHER AALYSIS OF COFIDECE ITERVALS FOR LARGE CLIET/SERVER COMPUTER ETWORKS Vyacheslav Abramov School of Mahemacal Scences, Monash Unversy, Buldng 8, Level 4, Clayon Campus, Wellngon

More information

Solution in semi infinite diffusion couples (error function analysis)

Solution in semi infinite diffusion couples (error function analysis) Soluon n sem nfne dffuson couples (error funcon analyss) Le us consder now he sem nfne dffuson couple of wo blocks wh concenraon of and I means ha, n a A- bnary sysem, s bondng beween wo blocks made of

More information

Different kind of oscillation

Different kind of oscillation PhO 98 Theorecal Qeson.Elecrcy Problem (8 pons) Deren knd o oscllaon e s consder he elecrc crc n he gre, or whch mh, mh, nf, nf and kω. The swch K beng closed he crc s copled wh a sorce o alernang crren.

More information

John Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany

John Geweke a and Gianni Amisano b a Departments of Economics and Statistics, University of Iowa, USA b European Central Bank, Frankfurt, Germany Herarchcal Markov Normal Mxure models wh Applcaons o Fnancal Asse Reurns Appendx: Proofs of Theorems and Condonal Poseror Dsrbuons John Geweke a and Gann Amsano b a Deparmens of Economcs and Sascs, Unversy

More information

J i-1 i. J i i+1. Numerical integration of the diffusion equation (I) Finite difference method. Spatial Discretization. Internal nodes.

J i-1 i. J i i+1. Numerical integration of the diffusion equation (I) Finite difference method. Spatial Discretization. Internal nodes. umercal negraon of he dffuson equaon (I) Fne dfference mehod. Spaal screaon. Inernal nodes. R L V For hermal conducon le s dscree he spaal doman no small fne spans, =,,: Balance of parcles for an nernal

More information

HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD

HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Journal of Appled Mahemacs and Compuaonal Mechancs 3, (), 45-5 HEAT CONDUCTION PROBLEM IN A TWO-LAYERED HOLLOW CYLINDER BY USING THE GREEN S FUNCTION METHOD Sansław Kukla, Urszula Sedlecka Insue of Mahemacs,

More information

Solution of a diffusion problem in a non-homogeneous flow and diffusion field by the integral representation method (IRM)

Solution of a diffusion problem in a non-homogeneous flow and diffusion field by the integral representation method (IRM) Appled and ompaonal Mahemacs 4; 3: 5-6 Pblshed onlne Febrary 4 hp://www.scencepblshnggrop.com//acm do:.648/.acm.43.3 olon of a dffson problem n a non-homogeneos flow and dffson feld by he negral represenaon

More information

On One Analytic Method of. Constructing Program Controls

On One Analytic Method of. Constructing Program Controls Appled Mahemacal Scences, Vol. 9, 05, no. 8, 409-407 HIKARI Ld, www.m-hkar.com hp://dx.do.org/0.988/ams.05.54349 On One Analyc Mehod of Consrucng Program Conrols A. N. Kvko, S. V. Chsyakov and Yu. E. Balyna

More information

MANY real-world applications (e.g. production

MANY real-world applications (e.g. production Barebones Parcle Swarm for Ineger Programmng Problems Mahamed G. H. Omran, Andres Engelbrech and Ayed Salman Absrac The performance of wo recen varans of Parcle Swarm Opmzaon (PSO) when appled o Ineger

More information

Lecture 6: Learning for Control (Generalised Linear Regression)

Lecture 6: Learning for Control (Generalised Linear Regression) Lecure 6: Learnng for Conrol (Generalsed Lnear Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure 6: RLSC - Prof. Sehu Vjayakumar Lnear Regresson

More information

Approximate Analytic Solution of (2+1) - Dimensional Zakharov-Kuznetsov(Zk) Equations Using Homotopy

Approximate Analytic Solution of (2+1) - Dimensional Zakharov-Kuznetsov(Zk) Equations Using Homotopy Arcle Inernaonal Journal of Modern Mahemacal Scences, 4, (): - Inernaonal Journal of Modern Mahemacal Scences Journal homepage: www.modernscenfcpress.com/journals/jmms.aspx ISSN: 66-86X Florda, USA Approxmae

More information

Volatility Interpolation

Volatility Interpolation Volaly Inerpolaon Prelmnary Verson March 00 Jesper Andreasen and Bran Huge Danse Mares, Copenhagen wan.daddy@danseban.com brno@danseban.com Elecronc copy avalable a: hp://ssrn.com/absrac=69497 Inro Local

More information

Variational method to the second-order impulsive partial differential equations with inconstant coefficients (I)

Variational method to the second-order impulsive partial differential equations with inconstant coefficients (I) Avalable onlne a www.scencedrec.com Proceda Engneerng 6 ( 5 4 Inernaonal Worksho on Aomoble, Power and Energy Engneerng Varaonal mehod o he second-order mlsve aral dfferenal eqaons wh nconsan coeffcens

More information

An introduction to Support Vector Machine

An introduction to Support Vector Machine An nroducon o Suppor Vecor Machne 報告者 : 黃立德 References: Smon Haykn, "Neural Neworks: a comprehensve foundaon, second edon, 999, Chaper 2,6 Nello Chrsann, John Shawe-Tayer, An Inroducon o Suppor Vecor Machnes,

More information

Existence and Uniqueness Results for Random Impulsive Integro-Differential Equation

Existence and Uniqueness Results for Random Impulsive Integro-Differential Equation Global Journal of Pure and Appled Mahemacs. ISSN 973-768 Volume 4, Number 6 (8), pp. 89-87 Research Inda Publcaons hp://www.rpublcaon.com Exsence and Unqueness Resuls for Random Impulsve Inegro-Dfferenal

More information

Robustness Experiments with Two Variance Components

Robustness Experiments with Two Variance Components Naonal Insue of Sandards and Technology (NIST) Informaon Technology Laboraory (ITL) Sascal Engneerng Dvson (SED) Robusness Expermens wh Two Varance Componens by Ana Ivelsse Avlés avles@ns.gov Conference

More information

Cubic Bezier Homotopy Function for Solving Exponential Equations

Cubic Bezier Homotopy Function for Solving Exponential Equations Penerb Journal of Advanced Research n Compung and Applcaons ISSN (onlne: 46-97 Vol. 4, No.. Pages -8, 6 omoopy Funcon for Solvng Eponenal Equaons S. S. Raml *,,. Mohamad Nor,a, N. S. Saharzan,b and M.

More information

GENERATING CERTAIN QUINTIC IRREDUCIBLE POLYNOMIALS OVER FINITE FIELDS. Youngwoo Ahn and Kitae Kim

GENERATING CERTAIN QUINTIC IRREDUCIBLE POLYNOMIALS OVER FINITE FIELDS. Youngwoo Ahn and Kitae Kim Korean J. Mah. 19 (2011), No. 3, pp. 263 272 GENERATING CERTAIN QUINTIC IRREDUCIBLE POLYNOMIALS OVER FINITE FIELDS Youngwoo Ahn and Kae Km Absrac. In he paper [1], an explc correspondence beween ceran

More information

Robust and Accurate Cancer Classification with Gene Expression Profiling

Robust and Accurate Cancer Classification with Gene Expression Profiling Robus and Accurae Cancer Classfcaon wh Gene Expresson Proflng (Compuaonal ysems Bology, 2005) Auhor: Hafeng L, Keshu Zhang, ao Jang Oulne Background LDA (lnear dscrmnan analyss) and small sample sze problem

More information

( ) () we define the interaction representation by the unitary transformation () = ()

( ) () we define the interaction representation by the unitary transformation () = () Hgher Order Perurbaon Theory Mchael Fowler 3/7/6 The neracon Represenaon Recall ha n he frs par of hs course sequence, we dscussed he chrödnger and Hesenberg represenaons of quanum mechancs here n he chrödnger

More information

Online Supplement for Dynamic Multi-Technology. Production-Inventory Problem with Emissions Trading

Online Supplement for Dynamic Multi-Technology. Production-Inventory Problem with Emissions Trading Onlne Supplemen for Dynamc Mul-Technology Producon-Invenory Problem wh Emssons Tradng by We Zhang Zhongsheng Hua Yu Xa and Baofeng Huo Proof of Lemma For any ( qr ) Θ s easy o verfy ha he lnear programmng

More information

XIII International PhD Workshop OWD 2011, October Three Phase DC/DC Boost Converter With High Energy Efficiency

XIII International PhD Workshop OWD 2011, October Three Phase DC/DC Boost Converter With High Energy Efficiency X nernaonal Ph Workshop OW, Ocober Three Phase C/C Boos Converer Wh Hgh Energy Effcency Ján Perdľak, Techncal nversy of Košce Absrac Ths paper presens a novel opology of mlphase boos converer wh hgh energy

More information

from normal distribution table It is interesting to notice in the above computation that the starting stock level each

from normal distribution table It is interesting to notice in the above computation that the starting stock level each Homeork Solon Par A. Ch a b 65 4 5 from normal dsrbon able Ths, order qany s 39-7 b o b5 from normal dsrbon able Ths, order qany s 9-7 I s neresng o noce n he above compaon ha he sarng sock level each

More information

FTCS Solution to the Heat Equation

FTCS Solution to the Heat Equation FTCS Soluon o he Hea Equaon ME 448/548 Noes Gerald Reckenwald Porland Sae Unversy Deparmen of Mechancal Engneerng gerry@pdxedu ME 448/548: FTCS Soluon o he Hea Equaon Overvew Use he forward fne d erence

More information

Comparison of Differences between Power Means 1

Comparison of Differences between Power Means 1 In. Journal of Mah. Analyss, Vol. 7, 203, no., 5-55 Comparson of Dfferences beween Power Means Chang-An Tan, Guanghua Sh and Fe Zuo College of Mahemacs and Informaon Scence Henan Normal Unversy, 453007,

More information

OPTIMIZATION OF A NONCONVENTIONAL ENGINE EVAPORATOR

OPTIMIZATION OF A NONCONVENTIONAL ENGINE EVAPORATOR Jornal of KONES Powerran and Transpor, Vol. 17, No. 010 OPTIMIZATION OF A NONCONVENTIONAL ENGINE EVAPORATOR Andre Kovalí, Eml Toporcer Unversy of Žlna, Facly of Mechancal Engneerng Deparmen of Aomove Technology

More information

( t) Outline of program: BGC1: Survival and event history analysis Oslo, March-May Recapitulation. The additive regression model

( t) Outline of program: BGC1: Survival and event history analysis Oslo, March-May Recapitulation. The additive regression model BGC1: Survval and even hsory analyss Oslo, March-May 212 Monday May 7h and Tuesday May 8h The addve regresson model Ørnulf Borgan Deparmen of Mahemacs Unversy of Oslo Oulne of program: Recapulaon Counng

More information

ISSN MIT Publications

ISSN MIT Publications MIT Inernaonal Journal of Elecrcal and Insrumenaon Engneerng Vol. 1, No. 2, Aug 2011, pp 93-98 93 ISSN 2230-7656 MIT Publcaons A New Approach for Solvng Economc Load Dspach Problem Ansh Ahmad Dep. of Elecrcal

More information

Research Article Cubic B-spline for the Numerical Solution of Parabolic Integro-differential Equation with a Weakly Singular Kernel

Research Article Cubic B-spline for the Numerical Solution of Parabolic Integro-differential Equation with a Weakly Singular Kernel Researc Jornal of Appled Scences, Engneerng and Tecnology 7(): 65-7, 4 DOI:.96/afs.7.5 ISS: 4-7459; e-iss: 4-7467 4 Mawell Scenfc Pblcaon Corp. Sbmed: Jne 8, Acceped: Jly 9, Pblsed: Marc 5, 4 Researc Arcle

More information

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005

Dynamic Team Decision Theory. EECS 558 Project Shrutivandana Sharma and David Shuman December 10, 2005 Dynamc Team Decson Theory EECS 558 Proec Shruvandana Sharma and Davd Shuman December 0, 005 Oulne Inroducon o Team Decson Theory Decomposon of he Dynamc Team Decson Problem Equvalence of Sac and Dynamc

More information

PARTICLE SWARM OPTIMIZATION BASED ON BOTTLENECK MACHINE FOR JOBSHOP SCHEDULING

PARTICLE SWARM OPTIMIZATION BASED ON BOTTLENECK MACHINE FOR JOBSHOP SCHEDULING Proceedng 7 h Inernaonal Semnar on Indusral Engneerng and Managemen PARTICLE SWARM OPTIMIZATION BASED ON BOTTLENECK MACHINE FOR JOBSHOP SCHEDULING Rahm Mauldya Indusral Engneerng Deparmen, Indusral Engneerng

More information

Chapter Lagrangian Interpolation

Chapter Lagrangian Interpolation Chaper 5.4 agrangan Inerpolaon Afer readng hs chaper you should be able o:. dere agrangan mehod of nerpolaon. sole problems usng agrangan mehod of nerpolaon and. use agrangan nerpolans o fnd deraes and

More information

Particle Swarm Optimization Algorithm with Reverse-Learning and Local-Learning Behavior

Particle Swarm Optimization Algorithm with Reverse-Learning and Local-Learning Behavior 35 JOURNAL OF SOFTWARE, VOL. 9, NO. 2, FEBRUARY 214 Parcle Swarm Opmzaon Algorhm wh Reverse-Learnng and Local-Learnng Behavor Xuewen Xa Naonal Engneerng Research Cener for Saelle Posonng Sysem, Wuhan Unversy,

More information

A Functional-Link-Based Fuzzy Neural Network for Temperature Control

A Functional-Link-Based Fuzzy Neural Network for Temperature Control Proceedngs of he 7 IEEE Symposm on Fondaons of Compaonal Inellgence (FOCI 7) A Fnconal-Ln-Based Fzzy Neral Neor for emperare Conrol Cheng-Hng Chen *, Chn-eng Ln, Fello, IEEE, and Cheng-Jan Ln, ember, IEEE

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 9 Hamlonan Equaons of Moon (Chaper 8) Wha We Dd Las Tme Consruced Hamlonan formalsm H ( q, p, ) = q p L( q, q, ) H p = q H q = p H = L Equvalen o Lagrangan formalsm Smpler, bu

More information

Lecture VI Regression

Lecture VI Regression Lecure VI Regresson (Lnear Mehods for Regresson) Conens: Lnear Mehods for Regresson Leas Squares, Gauss Markov heorem Recursve Leas Squares Lecure VI: MLSC - Dr. Sehu Vjayakumar Lnear Regresson Model M

More information

New M-Estimator Objective Function. in Simultaneous Equations Model. (A Comparative Study)

New M-Estimator Objective Function. in Simultaneous Equations Model. (A Comparative Study) Inernaonal Mahemacal Forum, Vol. 8, 3, no., 7 - HIKARI Ld, www.m-hkar.com hp://dx.do.org/.988/mf.3.3488 New M-Esmaor Objecve Funcon n Smulaneous Equaons Model (A Comparave Sudy) Ahmed H. Youssef Professor

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 9 Hamlonan Equaons of Moon (Chaper 8) Wha We Dd Las Tme Consruced Hamlonan formalsm Hqp (,,) = qp Lqq (,,) H p = q H q = p H L = Equvalen o Lagrangan formalsm Smpler, bu wce as

More information

Is it necessary to seasonally adjust business and consumer surveys. Emmanuelle Guidetti

Is it necessary to seasonally adjust business and consumer surveys. Emmanuelle Guidetti Is necessar o seasonall adjs bsness and consmer srves Emmanelle Gde Olne 1 BTS feares 2 Smlaon eercse 3 Seasonal ARIMA modellng 4 Conclsons Jan-85 Jan-87 Jan-89 Jan-91 Jan-93 Jan-95 Jan-97 Jan-99 Jan-01

More information

Short-Term Load Forecasting Using PSO-Based Phase Space Neural Networks

Short-Term Load Forecasting Using PSO-Based Phase Space Neural Networks Proceedngs of he 5h WSEAS In. Conf. on SIMULATION, MODELING AND OPTIMIZATION, Corfu, Greece, Augus 7-9, 005 (pp78-83) Shor-Term Load Forecasng Usng PSO-Based Phase Space Neural Neworks Jang Chuanwen, Fang

More information

Mechanics Physics 151

Mechanics Physics 151 Mechancs Physcs 5 Lecure 0 Canoncal Transformaons (Chaper 9) Wha We Dd Las Tme Hamlon s Prncple n he Hamlonan formalsm Dervaon was smple δi δ Addonal end-pon consrans pq H( q, p, ) d 0 δ q ( ) δq ( ) δ

More information

Wronskian Determinant Solutions for the (3 + 1)-Dimensional Boiti-Leon-Manna-Pempinelli Equation

Wronskian Determinant Solutions for the (3 + 1)-Dimensional Boiti-Leon-Manna-Pempinelli Equation Jornal of Appled Mahemacs and Physcs 0 8-4 Pblshed Onlne ovember 0 (hp://www.scrp.org/jornal/jamp) hp://d.do.org/0.46/jamp.0.5004 Wronskan Deermnan Solons for he ( + )-Dmensonal Bo-Leon-Manna-Pempnell

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 4 CS434a/54a: Paern Recognon Prof. Olga Veksler Lecure 4 Oulne Normal Random Varable Properes Dscrmnan funcons Why Normal Random Varables? Analycally racable Works well when observaon comes form a corruped

More information

Fall 2010 Graduate Course on Dynamic Learning

Fall 2010 Graduate Course on Dynamic Learning Fall 200 Graduae Course on Dynamc Learnng Chaper 4: Parcle Flers Sepember 27, 200 Byoung-Tak Zhang School of Compuer Scence and Engneerng & Cognve Scence and Bran Scence Programs Seoul aonal Unversy hp://b.snu.ac.kr/~bzhang/

More information

Notes on the stability of dynamic systems and the use of Eigen Values.

Notes on the stability of dynamic systems and the use of Eigen Values. Noes on he sabl of dnamc ssems and he use of Egen Values. Source: Macro II course noes, Dr. Davd Bessler s Tme Seres course noes, zarads (999) Ineremporal Macroeconomcs chaper 4 & Techncal ppend, and Hamlon

More information

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and Ths arcle appeared n a jornal pblshed by Elsever. The aached copy s frnshed o he ahor for nernal non-commercal research and edcaon se, ncldng for nsrcon a he ahors nson and sharng h colleages. Oher ses,

More information

Single-loop System Reliability-Based Design & Topology Optimization (SRBDO/SRBTO): A Matrix-based System Reliability (MSR) Method

Single-loop System Reliability-Based Design & Topology Optimization (SRBDO/SRBTO): A Matrix-based System Reliability (MSR) Method 10 h US Naonal Congress on Compuaonal Mechancs Columbus, Oho 16-19, 2009 Sngle-loop Sysem Relably-Based Desgn & Topology Opmzaon (SRBDO/SRBTO): A Marx-based Sysem Relably (MSR) Mehod Tam Nguyen, Junho

More information

Comb Filters. Comb Filters

Comb Filters. Comb Filters The smple flers dscussed so far are characered eher by a sngle passband and/or a sngle sopband There are applcaons where flers wh mulple passbands and sopbands are requred Thecomb fler s an example of

More information

Existence of Periodic Solution for a Non-Autonomous Stage-Structured Predator-Prey System with Impulsive Effects

Existence of Periodic Solution for a Non-Autonomous Stage-Structured Predator-Prey System with Impulsive Effects Appled ahemacs 55-6 do:.6/am.. Pblshed Onlne arch (hp://www.scrp.org/jornal/am) Exsence o Perodc Solon or a Non-Aonomos Sage-Srcred Predaor-Prey Sysem wh Implsve Eecs Absrac eng W Zolang Xong Ypng Deng

More information

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!") i+1,q - [(!

In the complete model, these slopes are ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL. (! i+1 -! i ) + [(!) i+1,q - [(! ANALYSIS OF VARIANCE FOR THE COMPLETE TWO-WAY MODEL The frs hng o es n wo-way ANOVA: Is here neracon? "No neracon" means: The man effecs model would f. Ths n urn means: In he neracon plo (wh A on he horzonal

More information

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC

CH.3. COMPATIBILITY EQUATIONS. Continuum Mechanics Course (MMC) - ETSECCPB - UPC CH.3. COMPATIBILITY EQUATIONS Connuum Mechancs Course (MMC) - ETSECCPB - UPC Overvew Compably Condons Compably Equaons of a Poenal Vecor Feld Compably Condons for Infnesmal Srans Inegraon of he Infnesmal

More information

3. OVERVIEW OF NUMERICAL METHODS

3. OVERVIEW OF NUMERICAL METHODS 3 OVERVIEW OF NUMERICAL METHODS 3 Inroducory remarks Ths chaper summarzes hose numercal echnques whose knowledge s ndspensable for he undersandng of he dfferen dscree elemen mehods: he Newon-Raphson-mehod,

More information

Relative controllability of nonlinear systems with delays in control

Relative controllability of nonlinear systems with delays in control Relave conrollably o nonlnear sysems wh delays n conrol Jerzy Klamka Insue o Conrol Engneerng, Slesan Techncal Unversy, 44- Glwce, Poland. phone/ax : 48 32 37227, {jklamka}@a.polsl.glwce.pl Keywor: Conrollably.

More information

Chapter 6 DETECTION AND ESTIMATION: Model of digital communication system. Fundamental issues in digital communications are

Chapter 6 DETECTION AND ESTIMATION: Model of digital communication system. Fundamental issues in digital communications are Chaper 6 DCIO AD IMAIO: Fndaenal sses n dgal concaons are. Deecon and. saon Deecon heory: I deals wh he desgn and evalaon of decson ang processor ha observes he receved sgnal and gesses whch parclar sybol

More information

Ordinary Differential Equations in Neuroscience with Matlab examples. Aim 1- Gain understanding of how to set up and solve ODE s

Ordinary Differential Equations in Neuroscience with Matlab examples. Aim 1- Gain understanding of how to set up and solve ODE s Ordnary Dfferenal Equaons n Neuroscence wh Malab eamples. Am - Gan undersandng of how o se up and solve ODE s Am Undersand how o se up an solve a smple eample of he Hebb rule n D Our goal a end of class

More information

Outline. Probabilistic Model Learning. Probabilistic Model Learning. Probabilistic Model for Time-series Data: Hidden Markov Model

Outline. Probabilistic Model Learning. Probabilistic Model Learning. Probabilistic Model for Time-series Data: Hidden Markov Model Probablsc Model for Tme-seres Daa: Hdden Markov Model Hrosh Mamsuka Bonformacs Cener Kyoo Unversy Oulne Three Problems for probablsc models n machne learnng. Compung lkelhood 2. Learnng 3. Parsng (predcon

More information

Optimal Production Control of Hybrid Manufacturing/Remanufacturing Failure-Prone Systems under Diffusion-Type Demand

Optimal Production Control of Hybrid Manufacturing/Remanufacturing Failure-Prone Systems under Diffusion-Type Demand Appled Mahemacs, 3, 4, 55-559 hp://dx.do.org/.436/am.3.4379 Pblshed Onlne March 3 (hp://www.scrp.org/jornal/am) Opmal Prodcon Conrol of Hybrd Manfacrng/Remanfacrng Falre-Prone Sysems nder ffson-type emand

More information

Graduate Macroeconomics 2 Problem set 5. - Solutions

Graduate Macroeconomics 2 Problem set 5. - Solutions Graduae Macroeconomcs 2 Problem se. - Soluons Queson 1 To answer hs queson we need he frms frs order condons and he equaon ha deermnes he number of frms n equlbrum. The frms frs order condons are: F K

More information

CS286.2 Lecture 14: Quantum de Finetti Theorems II

CS286.2 Lecture 14: Quantum de Finetti Theorems II CS286.2 Lecure 14: Quanum de Fne Theorems II Scrbe: Mara Okounkova 1 Saemen of he heorem Recall he las saemen of he quanum de Fne heorem from he prevous lecure. Theorem 1 Quanum de Fne). Le ρ Dens C 2

More information

EEL 6266 Power System Operation and Control. Chapter 5 Unit Commitment

EEL 6266 Power System Operation and Control. Chapter 5 Unit Commitment EEL 6266 Power Sysem Operaon and Conrol Chaper 5 Un Commmen Dynamc programmng chef advanage over enumeraon schemes s he reducon n he dmensonaly of he problem n a src prory order scheme, here are only N

More information

Dual Approximate Dynamic Programming for Large Scale Hydro Valleys

Dual Approximate Dynamic Programming for Large Scale Hydro Valleys Dual Approxmae Dynamc Programmng for Large Scale Hydro Valleys Perre Carpener and Jean-Phlppe Chanceler 1 ENSTA ParsTech and ENPC ParsTech CMM Workshop, January 2016 1 Jon work wh J.-C. Alas, suppored

More information

M. Y. Adamu Mathematical Sciences Programme, AbubakarTafawaBalewa University, Bauchi, Nigeria

M. Y. Adamu Mathematical Sciences Programme, AbubakarTafawaBalewa University, Bauchi, Nigeria IOSR Journal of Mahemacs (IOSR-JM e-issn: 78-578, p-issn: 9-765X. Volume 0, Issue 4 Ver. IV (Jul-Aug. 04, PP 40-44 Mulple SolonSoluons for a (+-dmensonalhroa-sasuma shallow waer wave equaon UsngPanlevé-Bӓclund

More information

Advanced Macroeconomics II: Exchange economy

Advanced Macroeconomics II: Exchange economy Advanced Macroeconomcs II: Exchange economy Krzyszof Makarsk 1 Smple deermnsc dynamc model. 1.1 Inroducon Inroducon Smple deermnsc dynamc model. Defnons of equlbrum: Arrow-Debreu Sequenal Recursve Equvalence

More information

DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL

DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL DEEP UNFOLDING FOR MULTICHANNEL SOURCE SEPARATION SUPPLEMENTARY MATERIAL Sco Wsdom, John Hershey 2, Jonahan Le Roux 2, and Shnj Waanabe 2 Deparmen o Elecrcal Engneerng, Unversy o Washngon, Seale, WA, USA

More information

P R = P 0. The system is shown on the next figure:

P R = P 0. The system is shown on the next figure: TPG460 Reservor Smulaon 08 page of INTRODUCTION TO RESERVOIR SIMULATION Analycal and numercal soluons of smple one-dmensonal, one-phase flow equaons As an nroducon o reservor smulaon, we wll revew he smples

More information

FAIPA_SAND: An Interior Point Algorithm for Simultaneous ANalysis and Design Optimization

FAIPA_SAND: An Interior Point Algorithm for Simultaneous ANalysis and Design Optimization FAIPA_AN: An Ineror Pon Algorhm for mlaneos ANalyss an esgn Opmzaon osé Hersos*, Palo Mappa* an onel llen** *COPPE / Feeral Unersy of Ro e anero, Mechancal Engneerng Program, Caa Posal 6853, 945 97 Ro

More information

Let s treat the problem of the response of a system to an applied external force. Again,

Let s treat the problem of the response of a system to an applied external force. Again, Page 33 QUANTUM LNEAR RESPONSE FUNCTON Le s rea he problem of he response of a sysem o an appled exernal force. Agan, H() H f () A H + V () Exernal agen acng on nernal varable Hamlonan for equlbrum sysem

More information

Method of Characteristics for Pure Advection By Gilberto E. Urroz, September 2004

Method of Characteristics for Pure Advection By Gilberto E. Urroz, September 2004 Mehod of Charaerss for Pre Adveon By Glbero E Urroz Sepember 004 Noe: The followng noes are based on lass noes for he lass COMPUTATIONAL HYDAULICS as agh by Dr Forres Holly n he Sprng Semeser 985 a he

More information

[ ] 2. [ ]3 + (Δx i + Δx i 1 ) / 2. Δx i-1 Δx i Δx i+1. TPG4160 Reservoir Simulation 2018 Lecture note 3. page 1 of 5

[ ] 2. [ ]3 + (Δx i + Δx i 1 ) / 2. Δx i-1 Δx i Δx i+1. TPG4160 Reservoir Simulation 2018 Lecture note 3. page 1 of 5 TPG460 Reservor Smulaon 08 page of 5 DISCRETIZATIO OF THE FOW EQUATIOS As we already have seen, fne dfference appromaons of he paral dervaves appearng n he flow equaons may be obaned from Taylor seres

More information

Linear Response Theory: The connection between QFT and experiments

Linear Response Theory: The connection between QFT and experiments Phys540.nb 39 3 Lnear Response Theory: The connecon beween QFT and expermens 3.1. Basc conceps and deas Q: ow do we measure he conducvy of a meal? A: we frs nroduce a weak elecrc feld E, and hen measure

More information

2/20/2013. EE 101 Midterm 2 Review

2/20/2013. EE 101 Midterm 2 Review //3 EE Mderm eew //3 Volage-mplfer Model The npu ressance s he equalen ressance see when lookng no he npu ermnals of he amplfer. o s he oupu ressance. I causes he oupu olage o decrease as he load ressance

More information

NATIONAL UNIVERSITY OF SINGAPORE PC5202 ADVANCED STATISTICAL MECHANICS. (Semester II: AY ) Time Allowed: 2 Hours

NATIONAL UNIVERSITY OF SINGAPORE PC5202 ADVANCED STATISTICAL MECHANICS. (Semester II: AY ) Time Allowed: 2 Hours NATONAL UNVERSTY OF SNGAPORE PC5 ADVANCED STATSTCAL MECHANCS (Semeser : AY 1-13) Tme Allowed: Hours NSTRUCTONS TO CANDDATES 1. Ths examnaon paper conans 5 quesons and comprses 4 prned pages.. Answer all

More information

SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β

SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β SARAJEVO JOURNAL OF MATHEMATICS Vol.3 (15) (2007), 137 143 SOME NOISELESS CODING THEOREMS OF INACCURACY MEASURE OF ORDER α AND TYPE β M. A. K. BAIG AND RAYEES AHMAD DAR Absrac. In hs paper, we propose

More information

Appendix to Online Clustering with Experts

Appendix to Online Clustering with Experts A Appendx o Onlne Cluserng wh Expers Furher dscusson of expermens. Here we furher dscuss expermenal resuls repored n he paper. Ineresngly, we observe ha OCE (and n parcular Learn- ) racks he bes exper

More information

Chapter 6: AC Circuits

Chapter 6: AC Circuits Chaper 6: AC Crcus Chaper 6: Oulne Phasors and he AC Seady Sae AC Crcus A sable, lnear crcu operang n he seady sae wh snusodal excaon (.e., snusodal seady sae. Complee response forced response naural response.

More information

TSS = SST + SSE An orthogonal partition of the total SS

TSS = SST + SSE An orthogonal partition of the total SS ANOVA: Topc 4. Orhogonal conrass [ST&D p. 183] H 0 : µ 1 = µ =... = µ H 1 : The mean of a leas one reamen group s dfferen To es hs hypohess, a basc ANOVA allocaes he varaon among reamen means (SST) equally

More information

Real Time Hybrid Simulation using Shaking Tabels

Real Time Hybrid Simulation using Shaking Tabels Real Tme Hybrd Smlaon sng Shakng Tabels Deparmen o Cvl and Envronmenal Engneerng Unversy o Kassel, Germany Olne Inrodcon A ndamenal sbsrcre algorhm wh sb seppng Applcaons o he algorhm Conclsons Inrodcon

More information

Solving Parabolic Partial Delay Differential. Equations Using The Explicit Method And Higher. Order Differences

Solving Parabolic Partial Delay Differential. Equations Using The Explicit Method And Higher. Order Differences Jornal of Kfa for Maemacs and Compe Vol. No.7 Dec pp 77-5 Solvng Parabolc Paral Delay Dfferenal Eqaons Usng e Eplc Meod And Hger Order Dfferences Asss. Prof. Amal Kalaf Haydar Kfa Unversy College of Edcaon

More information

Li An-Ping. Beijing , P.R.China

Li An-Ping. Beijing , P.R.China A New Type of Cpher: DICING_csb L An-Png Bejng 100085, P.R.Chna apl0001@sna.com Absrac: In hs paper, we wll propose a new ype of cpher named DICING_csb, whch s derved from our prevous sream cpher DICING.

More information

. The geometric multiplicity is dim[ker( λi. number of linearly independent eigenvectors associated with this eigenvalue.

. The geometric multiplicity is dim[ker( λi. number of linearly independent eigenvectors associated with this eigenvalue. Lnear Algebra Lecure # Noes We connue wh he dscusson of egenvalues, egenvecors, and dagonalzably of marces We wan o know, n parcular wha condons wll assure ha a marx can be dagonalzed and wha he obsrucons

More information

Appendix H: Rarefaction and extrapolation of Hill numbers for incidence data

Appendix H: Rarefaction and extrapolation of Hill numbers for incidence data Anne Chao Ncholas J Goell C seh lzabeh L ander K Ma Rober K Colwell and Aaron M llson 03 Rarefacon and erapolaon wh ll numbers: a framewor for samplng and esmaon n speces dversy sudes cology Monographs

More information

Attribute Reduction Algorithm Based on Discernibility Matrix with Algebraic Method GAO Jing1,a, Ma Hui1, Han Zhidong2,b

Attribute Reduction Algorithm Based on Discernibility Matrix with Algebraic Method GAO Jing1,a, Ma Hui1, Han Zhidong2,b Inernaonal Indusral Informacs and Compuer Engneerng Conference (IIICEC 05) Arbue educon Algorhm Based on Dscernbly Marx wh Algebrac Mehod GAO Jng,a, Ma Hu, Han Zhdong,b Informaon School, Capal Unversy

More information

. The geometric multiplicity is dim[ker( λi. A )], i.e. the number of linearly independent eigenvectors associated with this eigenvalue.

. The geometric multiplicity is dim[ker( λi. A )], i.e. the number of linearly independent eigenvectors associated with this eigenvalue. Mah E-b Lecure #0 Noes We connue wh he dscusson of egenvalues, egenvecors, and dagonalzably of marces We wan o know, n parcular wha condons wll assure ha a marx can be dagonalzed and wha he obsrucons are

More information

10. A.C CIRCUITS. Theoretically current grows to maximum value after infinite time. But practically it grows to maximum after 5τ. Decay of current :

10. A.C CIRCUITS. Theoretically current grows to maximum value after infinite time. But practically it grows to maximum after 5τ. Decay of current : . A. IUITS Synopss : GOWTH OF UNT IN IUIT : d. When swch S s closed a =; = d. A me, curren = e 3. The consan / has dmensons of me and s called he nducve me consan ( τ ) of he crcu. 4. = τ; =.63, n one

More information

A New Generalized Gronwall-Bellman Type Inequality

A New Generalized Gronwall-Bellman Type Inequality 22 Inernaonal Conference on Image, Vson and Comung (ICIVC 22) IPCSIT vol. 5 (22) (22) IACSIT Press, Sngaore DOI:.7763/IPCSIT.22.V5.46 A New Generalzed Gronwall-Bellman Tye Ineualy Qnghua Feng School of

More information

Lecture 11 SVM cont

Lecture 11 SVM cont Lecure SVM con. 0 008 Wha we have done so far We have esalshed ha we wan o fnd a lnear decson oundary whose margn s he larges We know how o measure he margn of a lnear decson oundary Tha s: he mnmum geomerc

More information

3.2 Models for technical systems

3.2 Models for technical systems onrol Laboraory 3. Mahemacal Moelng 3. Moels for echncal sysems 3.. Elecrcal sysems Fg. 3. shows hree basc componens of elecrcal crcs. Varables = me, = volage [V], = crren [A] omponen parameers R = ressance

More information

Lecture 18: The Laplace Transform (See Sections and 14.7 in Boas)

Lecture 18: The Laplace Transform (See Sections and 14.7 in Boas) Lecure 8: The Lalace Transform (See Secons 88- and 47 n Boas) Recall ha our bg-cure goal s he analyss of he dfferenal equaon, ax bx cx F, where we emloy varous exansons for he drvng funcon F deendng on

More information

This document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore.

This document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore. Ths documen s downloaded from DR-NTU, Nanyang Technologcal Unversy Lbrary, Sngapore. Tle A smplfed verb machng algorhm for word paron n vsual speech processng( Acceped verson ) Auhor(s) Foo, Say We; Yong,

More information

A NOVEL NETWORK METHOD DESIGNING MULTIRATE FILTER BANKS AND WAVELETS

A NOVEL NETWORK METHOD DESIGNING MULTIRATE FILTER BANKS AND WAVELETS A NOVEL NEWORK MEHOD DESIGNING MULIRAE FILER BANKS AND WAVELES Yng an Deparmen of Elecronc Engneerng and Informaon Scence Unversy of Scence and echnology of Chna Hefe 37, P. R. Chna E-mal: yan@usc.edu.cn

More information