Đang tải... (xem toàn văn)
KHOA HQC CbNO N 0 H | OVG DLJIMG PHU01VG P H A P QUY HOACH DdlXIG I X I G A U !\IHI£|\| T H | £ T L A P OUY TR iMH D l £ u T l ^ T d i U U HO CHUA T H U Y Dl£lV 50IV LA THEO T H d l G IAN THyC NguySnl[.]
KHOA HQC C b N O N H | OVG DLJIMG PHU01VG P H A P Q U Y HOACH DdlXIG I X I G A U !\IHI£|\| T H | £ T L A P OUY TRiMH D l £ u T l ^ Tdi U U HO C H U A T H U Y Dl£lV I V L A T H E O T H d l G I A N T H y C NguySnlTiiSng' T6MTAT Cdng trinh thiiy di^n Son La \i m()l du dn sa dd khai thdc nSng lugng h$ ih6ng sdng Dd, dui?c khft c6ng xay dyng vdo thdng 12/2006 v^ c6 t6 miy d k ti6n dua vAo v ^ hinh nftm 2010 T§p dodn Di^n luc Vi^t Nam lam chu dAu tu NOi dung bii bio gi6i thi$u k^t quA nghifen cuu ung dung quy ho^ch dOng ngau nhi^n v6i gia thi^t d6ng chay dffn tuSn theo Ij?tiiuy^txich Markov bjc d^ thi^t l^p quy Uinh v$n hanh hii t^i ini theo thdi gian thuc v6i him myc tifiu Ii eye d^ nSng lupng trung binh xic su^t nam K^t qua nghidn cuu dinTc trinh biy du6i d^ng d6 tlij tuong ung 12 thing nim, d6 ling v6i myc nii6c ho Aiu thing xem x^t vi gii tn luu luqng A6n dy bio s6 xic djnh dupe luu lugng di^u ti^t t6i uu cho v$n hinh Tir khoi; Quy ho&ch ddng ngdu nhidn, diSu Hdt tdi uu theo thdi gian thi/c, xich Markov b$c 1, du in thuy di$n Son La theo thdi gian Trong d^ tdi ndy dd tdp trung nghidn LBATVANB^ cuu ling dyng 1^ thuydt cua SDP vd thi^t lap thujt Theo k^t qua nghidn ciiu thuc hi^n boi C6ng ty todn d^ giai quy^t bdi todn vdn hdnh toi uu mpt ho Tu van Xay dung Di^n nim 2005, du ttn thuy di^n chiia thuy di^n v6i d6ng chay vdo hA dupe xem nhu Son La co cong su^t Idp may 2400 MW, difin lupng Id mpt bi^n s6 ngdu nhifin K^t qua tinh se cho ph6p trung binh nam 8963 tr.kWh Du ^ c6 nhi^m vu chu thi^t Idp cdc d6 thj bi^u thi su khai thdc toi mi yeu la phat di^n va tham gia m6t phin vao di^u ti^t IQ timg truong hpp cu th^ ciia muc nuoc ho dau didi liy cho h^ thong song Da Thi^t lap quy trinh van hanh vd gid trj luu lupng d^n du bdo Cdc d6 thj s6 tdi uu cho ho theo thoi gian thuc la mpt bai toin dupe dimg d^ xdc djnh vd dua cdc quy^t djnh v^ phuc tap bi^n dau vao ciia bai toan Id dong chay luu lupng liy khoi ho qud trinh vdn hinh hd d^n mang tinh chat ngSu nhi^n qud trinh khai toi uu theo thoi gian thuc K^t qud d^ng ndy cua bii thac Ly thuydt tat dinh khong th^ dp dung vdo bdi todn Id nOi dung khde co ban giua nghifin ciiu toan d^ thi^t lap quy trinh van hdnh toi uu ho chua cua d^ tdi ndy so v6i cdc Idi gidi bdi todn di^u ti^t toi theo thoi gian thuc v^n hdnh ho mpt thdi uukhdc dgan xem x6t phu thupc vao trang thdi myc nu6c h6 dau thoi ky vd gia tr; luu lirpng d^n d^ bdo ddy sir Cdc quy uoc vd djnh nghia sau se dupe su d^ng dung phuong phap quy hoach dpng ngau nhi^n md hinh hod bdi todn SDP: (Stochastic Dynamic Programming - SDP) d^ thi^t i, j : cdc chi s6 dimg cho ddng chay d^n, lan lupt Idp quy trinh vdn hdnh h6 toi im v6i hdm myc tieu Id chi ddng chdy d^n chu k;^ tvd (t+l) cue dai sdn lupng di^n trung binh xdc sudt ndm k , ! : cdc chi s6 diing cho dung tich ho, lin lupt Buoc ddu gidi ban nghifin cuu cho bdi todn don h6 Thudt todn vd loi giai bdi todn dupe xdc djnh v6i si; chi th^ tich nu6e h6 chu ky t vd (t+l) h6 trp cua chuong trinh tinh dupe l§p trinh bdng Qit & %*\- gi^ tri cigi bi^u cila t6ng d6ng chay ngon ngii Fortran 90 tdc gid thyc hi^n d^n chu k^ t vd (t+l) LLfTHUm Quy hoach dOng ngdu nhi^n Id mOt nhung ky thudt nhdm giai quy^t bdi todn toi uu c6 dang loi giai Id cdc quy^t dinh c6 dang tuin ti; Sfct & SH.I: gid trj d ^ bi^u v^ dung tich chu kytvd(t+l) Phuong trinh cdn b ^ th^ tich m6t chu k? di^u ti^t bit ky cd th^ vi^t nhu sau: ^kiii =Sk, +Qi, - E ^ , -S|,^., (1) Khoa Ky thu^t Xay dyng, Truong D^i hpc Bach khoa Dai hpc Qu6c gia Tp Ho Chi Minh 56 NONG NGHllP VA PHAT TRIEN N N G T H N - KY - T U A W ^ A/2ni2 KHOA HQC CbNG N G H | Vdi Rjiiit chi th^ tich lay khoi h6 tuong irng vdi Chu y gi^ tri Q,t v^ ph4i cua phuong trinh th^ tich ban dau cua ho Id S^,; th^ tich cuoi cua h6 Id (1) 1^ dai lugmg ngSu nhifin, d6 Rjjiit cung la mot S,t,i; th^ tich nude tdi Q,t; th^ tich m i t b6c hoi Id dai lugmg ng^u nhi^n Ky hi^u t Wy c4c gid trj l i i luqrt tir T d^n v4 ky hi$u n gia t5ng m^t cdch tudn tu theo timg bu6c ciia phuong phdp SDP (xem so d6 hinh 1) Gid tri EhitPhu tiiupc vdo th^ tich ban ddu vd cuoi chu ky cua ho chua (Si(, vd Sit,i) Chuk^ t=T- -H \ t= t=l \ n=t+ n=T+l t=2 t=T- ] n=T t=T h n=TBu-c^c truy h i i Hinh : So d6 bucic tray h6i v4 chu k^ tinh 1^ thuydt SDP Phuong trinh tniy h6i tong qudt cua ly thuydt SDP cho bat ky buoc tinh n va chu ky t ndo nhu sau : F'(k,i)=MaxB„„ + XP,;C:(l.j) Vk (2) {VI} P ' : he SO ma trdn xdc suat chuydn d bude tinh t dong chay dau chu ky d trang thai i vd cuoi chu k y d trang thdi j B^i,: bi^n nghien ciiu (chi di^n ndng nghien ciiu nay) d budc tinh t tuong ling dong chdy d^n i; mue nude ho d trang thai dau Id k vd cuoi Id f^*j: dien ndng tuong iing vdi xac suit ^ F^(k, i ) : ham muc tieu d budc tinh n tuong iing vdi luu luong d^n i vd miic nude h6 k Theo ly thuydt cua SDP Idi gidi bdi todn se dupe xdc dinh vdi thudt todn truy hdi ngupc cho phuong trinh (2) Tinh todn dupe gia thi^t bdt ddu tii cuoi chuoi ddng chdy di^u ti^t xem xet CD- T^^c hi^n tinh todn cho moi chu ky chuoi ddng chdy ndy dupe xem Id mot bude (n) tinh todn todn bd cua bdi todn quy hoach ddng tong th^ cua he thdng Di^u cd nghia ldn=l t=T, n=2 t=T-l, Vdi gia thi^t ma trdn chuyin xdc suit Pj) Id khong doi cho moi budc vd 6n djnh cho ca chuoi ddng chdy tinh todn, phuong trinh (2) se dupe giai bJing phuong phdp truy hdi ngupe Theo ly thuydt SDP, he sd cua ma tran chuydn P,J se dupe xdc djnh tren co sd gid thi^t dong chay d^n tudn theo xich Markov b?c dupe trinh bdy bdi phuong trinh nhu sau: P[X,,,/X„X,_p ,Xo]=P[X,,,/X,] (3) V^ trdi cua phuong trinh (3) chi xdc suit cd di^u ki^n cua bi^n du bdo X^^i phu thu^c vdo tat cd cdc gid tri hi^n tai Xj cQng nhu qud khii Xt.i, ,Xo V^ phai phuoTig trinh (3) chi xdc suit cd di^u ki^n cua bi^n du bdo, chi phu thuOc vdo gid tri hi^n tai Xt Vdi gia thi^t ndy chuoi ddng chdy d^n tinh todn dupe xem nhu mpt chuoi ngau nhien dn dinh Qud trinh tinh truy hoi ngupc tim loi giai bdi todn theo so dd hinh neu tren se dupe xem Id hdi tu di^u ki^n sau dupe thoa mdn: [CT(k,i)-fXk>i)] = hangso V(k,i,t) (4) Khi dd nhdn dupe Idi gidi dn djnh, cdc muc tru tdi uu vdo moi cuoi chu kj, 1', se dupe xdc dinh cho cdc gid trj xem x6t k vd i cho tit cd cdc chu k j t ndm Ket qua ndy cho phep xdc djnh chinh sdch v?u hdnh trang thdi dn djnh ky hi^u 1* (k, i, t ) Ddy Id ca sd d^ thuc hien quy trinh van hanh tdi uu hd theo ly thuydt SDP • jb> DVNG LY I H U Y ^ SDP T H i r U P QUY TRiTH V9N HMH Hd Tdi UU CUA SON U THEO THOI GIAN THVC Cdc thdng sd chinh cua dv dn thuy di$n Son La dupe tdm tdt bang Bdng Thdag so chinh di^ dnSonLa Donvi Gid tri Chi aeu m m 215 175 WH, tr.m' P Q» * mVs N,„ E„ tr.kWh 6504 13.6 1532 2400 8963 MNDBT MNC MW NguSn: Cdng fyCdphin Tu vin Xiy dung Biin nam 2004 N N G NGHllP VA PHAT TRIEN NONG T H N - KY - THANG 8/2012 57 KHOA HQC C N NOHi JjJO D^ thi/c hi$n m6 phdng SDP, muc nu6c ho cliiia duQc tir MNC-175,0 m din MNDBT.215,0 m dugc chia thdnh 51 thang (c6 t h i thay ddi) nhu sau (Bang 2): " 180 160 Chu6i d6ng chdy sir dung phdn tich di 108 nam c6 luu lu(mg trung binh thdng tai vj tri tuy& d$p chinh Id Q,b-1532 mVs (xem gid tri diin hinh bdng 3) , 100 1300 M •1 NguSn: Cdng tyTuvih — 7500 IOCOO 12S0O iv»«i Xiy dung Biin nim 2000 Hinh Du6ng cong ddc tinh h6 chira Son La Z(m) Nam 2005 2006 2006 2008 2009 VI 1587,6 1123,1 1511,6 2187,1 1292,0 175,0 Bdng M6 hinh hod cao trinh muc nude h6 chiia SDP 36 41 46 31 16 21 26 11 179,0 183,0 187,0 191,0 195,0 199,0 203,0 207,0 I 211,0 Bdng D6ng chdy trung binh X VII ItX K 11383,2 3884,5 1748,0 1080,9 1781,8 1528,5 2102,7 3952,0 4196,9 3977,3 2280,0 1064,0 3968,9 3622,7 2339,1 1258,2 3538,2 2879,6 1866,2 895,1 Ngudn: Cdng ty Cdphdn Tu van Xiy dung Didn Vdi l u u l u p n g d d n g c h d y t h e o d a n g t r u n g b i n h thdng tir 2005 d i n 2009, don vj mVs II III XII XI 844,4 539,6 377,5 345,4 377,5 783,6 402,0 401,1 331,0 296,4 628,3 400,3 347,1 263,5 167,2 1680,5 493,2 320,9 299,8 214,5 454,3 405,3 490,6 408,7 226,3 IV 378,3 184,1 203,5 243,2 294,7 V 318,4 332,7 556,5 525,2 798,8 3-8/2010 SDP t h d n g , d o d d q u y t r i n h v a n h d n h h d d u p e thi^t Idp B a n g t r i n h b a y k h o d n g gid trj n g a u nhien (xac t u o n g ling c h o 12 t h d n g T r e n c o s d sd lieu luu l u p n g djnh b d i g i d t q gidi hgji d u d i vd g i a trj gioi han d ^ n t i m g t h d n g q u a n sdt d u p e s e d u p e c h i a r a t h d n h tren) c u a l u u l u p n g d d n t u o n g i i n g vdi t h ^ dien 14 k h o a n g gid trj n g a u n h i e n t u o n g iing dd m o p h d n g h i n h vd t r o n g t d n g s d 12 t h d n g B d n g K h o d n g gid tri n g i u n h i e n c h o d d n g c h d y d d n cila t h d n g & t h d n g Thang 10 333,7 376,2 391,3 410,6 425,7 442,9 471,2 485,4 510,7 519,8 541,0 576,4 601,7 Gi(M h^n 333,7 trSn 376,2 391,3 410,6 425,7 442,9 471,2 485,4 510,7 519,8 541,0 576,4 601,7 782,8 10 11 12 13 14 Gioi ban 232,6 duoi i Thdng i GlM han 220,4 du6i Gi6ihan trSn 293,3 11 12 13 14 293,3 315,5 333,7 348,9 366,1 374,2 382,2 397,4 416,6 436,9 448,0 474,3 501,6 315,5 333,7 348,9 366,1 374,2 382,2 397,4 416,6 436,9 448,0 474,3 501,6 604,1 Tir g i d trj l u u lucjng q u a n txic 108 n d m vd c d c k h o a n g gid tri n g a u n h i e n ndy, c d c h i s o c u a m a t r § n p h u o n g t r i n h (2) B d n g t r i n h b d y m a t r d n c h u y i n d i i n h i n h tir tiiSflg chuyin xich Markov bdc se duqrc xdc djnh vd qua thdng tong s i 12 ma titln chuyin su dung md hinh SDP voi ky h i i u P,J chu k^ ndm 58 N N G NGHllP VA PHAT TRIEN N N G T H O N - K'? - THANG 8/2012 KHOA H C C H HGHI B d n g M a t r ^ c h u y i n P , tir t h d n g • » t h d n g 2 10 11 12 13 14 0,286 0,167 0,429" 0,143 0,167 0,143 0,167 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,167 0,000 0,000 0,000 0,000 0,167 0,167 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,000 0,167 0,000 0,000 0,1671 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,000 0,000 0,000 0,167 0,167 0,333 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,000 0,000 0,167 0,000 0,000 0,167 0,167 0,167 0,000 0,000 0,000 0,000 0,333 0,167 0,000 0,000 0,000 0,167 0,333 0,067 0,167 0,133 0,167 0,667 TiMJSn 10 11 12 13 14 0,333 0,167 0,167 0,333 0,000 0,167 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,000 0,000 0,167 0,167 0,333 0,167 0,000 0,333 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,000 0,167 0,000 0,000 0,167 0,167 0,000 0,167 0,000 0,000 0,000 0,000 0,000 0,333 0,000 0,000 0,000 0,167 0,000 0,167 0,167 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,167 0,000 0,000 0,000 0,000 0,000 0,000 (*) gid tri 0,429 cd nghia Id xdc suit dd luu lupng hien tai khoang gid trj ngau nhien vd luu lupng thdng kd tidp roi vdo khoang ngdu nhien thii Id Pi2=0,429 Vdi y nghia ndy cdc h^ sd P^ bit ky ma trdn chuydn phdi cd tinh chat Z p j = ^ ; ^ i (5) j=i Vdi i chi sd hang vd j chi sd cdt ma tran chuydn IV.KfTQUiiNGMByCUU 46 41 36 31 26 21 16 11 e k-61 —— |U^ 10 11 12 13 14 T-2 k=SB *-AA=- ^^^ ——Ic»t1^^- — k - i r '• S 0,000 0,000 0,000 0,333 0,067 0,167 0,000 0,167 0,000 0,167 0,167 0,000 0,000 0,167 0,000 0,333 0,167 0,167 0,167 0,000 0,167 0,333 0,000 0,000 0,067 0,000 0,000 0,000 0,000 0,000 0,167 (12 dd thj tuong iing quy trinh didu tidt 12 thdng ndm) trinh bdy kdt qud lien quan ddn cac bidn: luu lupng ddng chay ddn (Q), muc nude hd diu thdi doan didu tidt (kk=l*^51) vd muc nude hd tdi uu cho cudi thdi doan didu tidt (Z) nhu sau: Khai thdc kdt qua dd thj van hdnh tdi uu se dupe thyc hidn nhu sau (vi du xem dd thi 3, tuong ling chu ky didu tidt tir dau thdng ddn dau thdng 2; cdc thj khde tir ddn 14 cd ciing nguyen tic khai thdc van hdnh cho cdc thdng tuong iing): Du bdo luu lupng cho ddng chdy ddn (Q) Gid tri se xdc djnh tu mot mo hinh dir bdo thiiy vdn khu vuc diing du bdo luu lupng vd hd Tir dd thi T=l, vi dy luu lupng du bdo Id d vdo thang gid tri (dd thj 3) HinhS Jh«ng Z 0,000 0,167 0,167 ~~ B 10 11 12 13 14 Hinh4 Vdi su hd trp ciia mdy tinh, ket qua Idi giai tdi uu dupe thidt Idp sau ndm tinh l§p (n=96) Cdc thj Ghi nhdn gid tri niuc nude hd Z dau thdng Tir bdng xdc djnh gid tri k tuong iing Vi dy mue nude hd dau thdi kj^ Id d thang thii k=6 (dd thj 3) Vdi gid tri Q vd k neu tren, tra dd thi xdc dinh Z Diing bdng 2, xdc dinh dupe muc nude hd tdi tru cudi thdi dogn (ddu thdng 2) tuong iing Tir sd lieu ndy se xdc djnh dirpc luu lupng xa (di vdo nhd mdy phdt didn) c ^ thidt nhd vdo phuong trinh cdn bllng nude (1) Vdi luu Iirpng vdo nhd mdy vd muc nude hd dd xac dinh se xdc dinh dupe ndng lupng tuong iing Vdi sd lidu vj du neu tren, tra tir dd thj mtrc nude didu tidt tdi uu cuoi thdng se cd thang gid trj tuong ling Id k=17 Cdc dd thj sau tuong iing cho cdc chu kjf didu tidt ciia cdc thdng cdn lai mpt ndm nhu sau: N N G NGHIEP VA PHAT TRIEN NONG TH6N - KV • THANG 8/2012 59 KHOA HOC C N G N C H | T - _ _ _ h-fl Thano Z k-ia_iiôji ã lLl2ô k-lt _- - tmr , _- li"9 _^ Hinh5-> 14.S6tfatxdcdinhmvtcnufch6uynamic programming for optimal water resources systems analysis Prentice-Hall Englewood cliffs, New Jersey 07632 S Vedula & P P Mujumdar (2005) Water Resources Systems Modeling Techniques and Analysis Tata McGraw-Hill Publishing Company Limited NEW DELHI Loren P Meissner (1995) Fortran 90 PWS Publishing Company APPUCATION OF STOCHASTIC DYNAMIC PROGRAMMING FOR DETERMINING THE OPTIMAL OPERATIONAL REGULATION IN THE REALTIME CALCULATION WITH CASE STUDY OF SON LA HYDROPOWER PRJECT Nguyen Thong Summary Son La hydropower project is one of the most projects in the cascade scheme of Da river basin whose construction was started in 12/2006 The first unit of this project was put into service in 2010 This paper presents an application of Stochastic Dynamic Programming (SDP) with an assumption that the inflow was calculated based on a first order Markov chain The main goal of this study is to propose an optimal operational regulation in real-time calculation with the objective fimction for maximizing the annual mean value of probabilistic energy Research results are presented in graphs corresponding 12 months of the year in which the water level corresponding to the consideration month and value of expectedflowvrill be used to determine die optimalflowregulator to operate K^words: Stochastic dynamic programming, optimal regulation in real time, chain Markov of order 1, Son La hydropowerplant Ngudi phdn bi^n: TS L6 Hiiiig Nam Ngdy iih*a bdi: 15/5/2012 Ngdy thdng qua phan bi$n: 16/7/2012 Ngdy duy^t ddng: 23/7/2012 NdNG NGHIEP VA PHAT TRIEN N O N G THdN - KY - THANG 8/2012 61 ... quy trinh van hanh tdi uu hd theo ly thuydt SDP • jb> DVNG LY I H U Y ^ SDP T H i r U P QUY TRiTH V9N HMH Hd Tdi UU CUA SON U THEO THOI GIAN THVC Cdc thdng sd chinh cua dv dn thuy di$n Son La. .. FOR DETERMINING THE OPTIMAL OPERATIONAL REGULATION IN THE REALTIME CALCULATION WITH CASE STUDY OF SON LA HYDROPOWER PRJECT Nguyen Thong Summary Son La hydropower project is one of the most projects... determine die optimalflowregulator to operate K^words: Stochastic dynamic programming, optimal regulation in real time, chain Markov of order 1, Son La hydropowerplant Ngudi phdn bi^n: TS L6 Hiiiig