TE LKOM NI KA  Te le c om munica tion,  C omp u tin g,   El e ctroni cs and  Contr ol   Vo l.   19 ,  No.   3 June   2021 ,  pp.  1001 ~ 1009   IS S N: 16 93 - 6930, acc red it ed   First G ra de by  Keme nr ist ek di kti, D ec ree  N o: 21/E/ KP T/ 2018   DOI: 10. 12 928/ TELK OMN I KA.v1 9i3 . 18777     1001       Journ al h om e page http: // jo ur nal. uad.ac .id /i nd ex. php/TE LKOMNIKA   Extend ed stat e obs erver  based lo ad freq uenc y contr oller fo three ar ea inte rconnecte d pow er syst em       Van  Van H uynh 1 P hong  Th an h  Tran 2 , Tu an   Anh Tr an 3 ,  Dao Huy Tu an 4 V an - Du c P ha n 5   1 ,2, 3 ,4 Fa cul ty   of Electrical  and   E l ec tron ic s E ng ineeri ng,   Ton   Duc Thang  Univ ersity Ho Chi   Minh  Cit y,   Viet n am   5 Facul ty   of  Auto mobi le Technol o gy,   Van   La ng   U nive rsity ,   Ho Ch Minh C it y,   Vie tna m       Art ic le  In f o     ABSTR A CT   Art ic le  history:   Re cei ved   J un  15, 2 020   Re vised  N ov 23, 2 020   Accepte d   Dec  5,   2020       In  thi p ape r ,   w dev el op   a   new   extende d   state  var ia b le  observe base d   loa d   fre quenc y   cont r oll er   ( LFC )   sc hem e   for   thr ee - are a   int er connect ed   power   sys te ms.  T he   e xte nded   sta te  o bserve rba sed   lo ad  fr eque n cy   co ntrol lers   a r e   deve lop ed   whic ut il i ze  distur banc e   esti m at io t ec hni ques.   T he  propos e   cont rol   appr o ach  assures  t hat  th flu ct u at ing   th i ngs  of  th e   loa d   fre quencie s   rea ch es  to   saf e ran ge   and   the  l oad  fre qu encie ca al so  be   m ad a a   ver y   mi nimal   not   to   h ave  an   eff ect  on   power  qua li ty   an power   flow   in   mul ti - area   int er conne c te d   power  sys te m .   Th r esult s   of  th si mul a t ions  using  MA TL AB/S IM ULINK   done  did  not  only  add r ess  tha th pro posed  newly   cont rol   method  works   eff ectivel but  al so  cha n g powerful ly  th par amete r   var iations  of  th interc onne cted   are as  of  th p ower  sys te m.   E spec ially,   it  works   ver y   well  to   l im i t   disturb anc es   im p act   on   in te rco nn ecte d   are as   in   th e   sys te m.  The r efo re,  the  p erf ormance   of  int er con nec t ed  power   s ystem   und er   diffe r en mu lt i - condi ti ons  is  simul ated  with   t he  new  con trol  me thod   t o   dem onstra te   the  fea sibi li ty   of the s ystem .   Ke yw or d s :   Ar ea  contr ol erro r   Fu ll - order st at e obser ve r desig n   Loa f re qu e nc c on t ro   M ulti - a rea  power syste m   Tie - li ne  area  in  pow e s ys te m   This   is an  open   acc ess arti cl e   un der  the  CC  BY - SA   l ic ense .     Corres pond in Aut h or :   Dao H uy T ua n   Faculty  of Elec tric al   and   Ele ct ronics E nginee rin g   To n Du c  Th a ng  Un i ver sit y,   19 Ng uy e n H uu T ho Street,  T an  P hong  Wa r d,   Distric t 7 , H C hi M i nh Ci ty, Vie tnam   Emai l:   daoh uy t uan@tdt u. e du. vn       1.   INTROD U CTION     In  m ulti - area   interc onnected   powe s ys te m ,   sta bili ty  of  frequ e nc is   very  sig nif i nd ic at or   of   powe qual it as  the  load s   of   the  ti e - li ne   mu lt iple  are as  kee incre asi ng   a nd  cha ng i ng   pro gr e s sively.   M ea nwhile , loa distu r ban ce s can  occ ur  s udde nly   w hich  c an  cause  de viati on s in  ti e - li ne powe area e xc hange   and  nomi nal  f r equ e ncies  i ns ta bili ty  [ 1 2].  Loa f re qu e nc c on tr ol  for  ye ars  now  has  been  one  of  th basic   rob us co ntr ol  mecha nisms   i la r ger   sc al e   powe el ect ric  sy ste ms   with   interco nnect e area   an t he   mai n   crit eria  of  t he   load   f reque ncy  co ntr ol   is  t keep  a nd  mai nt ai the   syst em   f re qu e nc un if or m   at   it nomina l   value   duri ng  a nd  w he the re  is   load   cha nge.   la rg e   po wer  el ect ric  sy ste can   be   di vide a nd  se par at ed  int sever al   loa f r equ e nc a rea   c on t ro l   that  is   i nterc onnected   by  ti e - li nes   a nd  it   relat es   to   operati onal   proc edures   to   be  f ollow e d   in   the  e ve nt  of   majo r   fau lt or   of   tie - li ne  powe r.   Ge ner al l y,   t he  mai go al   and  duty  of   load   fr e qu e nc c on t ro is  sim ply  to  a dju st  the   f r equ e ncies  of  the  se par at e a reas  a nd   t si mu lt ane ously  modu la te   powe fl ow i ng  across  the   ti e - li nes  acco rd i ng  with  the  ag r eement  of   an  i nter  a rea  powe s ys te m Mor eov e r,  load  fr e quenc co ntr ol  nor ma li zes  fr e qu e nc y,  mainta in dynamics,   an m akes  qual it as su ra nce  of  the  powe su ppl y,   require s the  us of  loa d fr e qu e nc c ontr oller ( L FC )   scheme  in   [3 - 6].    Evaluation Warning : The document was created with Spire.PDF for Python.
                          IS S N :   1693 - 6930   TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l V ol.  19 ,  No.  3 June  20 21 :    10 01   -   100 9   1002   Loa fr e quenc c ontrol   is  al s gr eat l si gn if ic ant  in   po wer  el ect ric  sy ste m   operati on al   f or  deliveri ng   reli able  a nd  e ffi ci ent  po wer  w it ho ut   poor  qu al it y.   T her e f ore,  a   novel   co nt ro l   te ch nique   ne eds   to   be  de ve lop e in  ot her   t achi eve  LFC  ai m s,  there by,  maint ai nin an s us t ai nin reli abili ty  of  the  el ect ri cal   power  sy st em  in   mu lt i - areas I order  to  s olv e  ab ove pro blem s,  resea rc her s   a nd  co ntr ol  e ng i neer s   ha ve  pro po s ed  wide   ra nge o f   methods   of  st at e - of - the - art   l oad  fr e quenc c on tr ollers   to   b a ppli ed   in   m ulti - area   int ercon nected   powe sy ste [7 - 15 ].   That  is,  the  me thod  of  a dap ti ve   con t ro propose to  ta ke  ca re  pa rameter  va riat ion   in  [7 - 9].  Bu t   it   cannot   le ad   strongl t a   ge ner al   so l utio to   the   pr ob le faced   by  L F in   po wer  s yst em.  M a ny  kin d s   of   con t ro te ch ni qu s uc as  pr oport ion al   inte gr al   ( PI )   or  pr oport ion al   inte gr al   de rivati ve   ( PID )   c ontr oller  f or   LFC  we re  us e d.   I the   presc ribe e nv i ron ment,   some   fa ct or li ke  un ce rtai nties  ma ke  it   diff ic ult  to  a pp l th e   above - me ntioned LFC  tech niq ue s i n p racti ce, which  impli es [10 - 12 to i nt rodu ce  some i nter nal m od el   con t ro l   to  desig PID   ty pe  L FC  c ontr ollers.  I [10]  pr opos e f uzzy  P rop or ti onal   I nteg ral  c ontr ollers  for  L FC  of  powe el ect ric  sy ste ms W hile  the   di ff e ren t   methods   s how   that  it   is   po s sible  to   e nh a nce   the   pe rfo rma nc of   LFC  in  s pecifi ed  en vir onme nt s.  The  P I co ntr oller  pro pos ed  in  [11]   co m bin e with  ne str uctu re  show to   be  rob us tl a nd  to   en ha nce  t he  dam ping  of  the  powe el e ct ric  sy ste m   tr aci ng  with   s mall   sign i fican ste changes  i loa d.   T he  st rateg of  tu ni ng   is   base s olely  on   t he  ma xim um   pea res onance   of  the  s ys te m   sp eci ficat io ns   in  [ 12].  Howe ver,  m os c on ven ti onal   PID   co ntr ollers  with  gai ns   fixe was   desi gned   unde r   nominal  sy ste op e rati ng  c onditi ons,   the   se le ct ed  m ode  of  it gains   i usual ly  al ways  on  t rial   an er ror  wit no  anal ytica ll meth ods  of  de te rmin in it par a mete rs  opti mall w hich   s om et imes  fail to  giv e   out  th be s t   and   acc ur at con t ro sc hem es  and   perf ormance  ov e a   wider   range  of   s ys te op e rati ng   c onditi ons  an exh i bits,   t her e by,  a   poor  dyna mic  res ponse   and  performa nc du rin oper at ion s.   S om e   of  th c ontrol   m et hods  need  to   a pp l the   f ull - sta te   of  t he  c ontro area  a feedback   in put  whil so me   will   le ad  to   higher - orde r   con t ro ll ers th ese  facto rs  we re  to co mp le to  be  under s tood  an c ompre he nd e by   con t ro a nd   el ect rical   eng i neer s  in [ 13 14]   Among  va rio us  co ntr ol  te ch ni qu es   me ntioned,  t he  opti mal  co ntr ol  with   s ta te   feedback   t echn i qu e   is  on of   t he  best   op ti ons.  Als o,  robu st  dece nt rali zed  li near   con t ro ll er  was  app li ed   in   [ 15 - 19].   Anothe c on t ro te chn iq ue   to   lo ok  at   is   the   sl iding  m ode  c ontr ol  ( SM C )”.   Sli din m ode  con t ro l   te ch niques  is   a no t her  bette way  a nd  ap pro ach  t s olv e   L FC  pro blem S M has  ofco ursed  bee a ppli ed  for  LFC  in   powe el ect ric  sy ste m   in [19 - 25]  to  a chie ve fast  re s pons e a nd  rob ust ly p e rforman ce in the  po wer netw ork, t his  method is a  no nlinear   con t ro l   strat eg with   fam ous  r ule.  It  is   al so   i ns e ns it ive  perha ps   t c ha ng e of  the   plant  par a mete rs   and  as   well   impro ves  sy ste tra ns ie nt  co ntr ol  pe rformance T he   above  a ppr oac hes  a re  achie ve under  assu mp ti on   that  al s ys te m   sta te   va riables   are  t be   me as ur a ble   and  re a di l avail able   for  fee db ac k.  I fact,   no t   al s ys te sta te   var ia bles   are  meas ur a bl for  fee dback s,  an the we   need   t est im at the  sta te   va riables  that  ar no t   unmeas ur a ble.   Esti mati ng   un measu rab le   sta te   var ia bles  is  of te cal le obser vatio in   [ 21 - 25].   T his  s cheme   can  a dap t he  unknow uppe bo unds   of  m at ched   nonline arit an distu rb a nce.  It  gets  not  only  t he  s ys te m   sta te   traj ect or ie accom plish ment   but  al so  sat isfie in  pa rameters  of   t he  syst em  sta te   error s The  work   il lustrate a bove,   achie ve  a   sign ific a nt  re s ult  relat ed   to  LFC’s  of  inte r connecte power  syst ems  a pp l ying   var i ou s  contr ol  techn i qu e s.    Howe ver,  the r are   some   li mit at ion of  the  a bove   a pp r oach e s.  Firstl y,  the   dist urbance are   not   tru ncated  from   the  outp ut  poi nts  in  ste ad s ta te Seco nd l y,  the  co ntr oller   gains  a re  not  set   to  be  e xtre mely  high  to  at te nu at disturba nce of   unkn own  bounda ries.  T hi rd ly the  c on t r oller  is  desi gned  in  acc orda nc to  the  nomin al   transf e f unct io of  the  plant  wh e no - loa dist urban ce  is  con si der e d.   In   orde to  s olv t he   above  li mit at ion s,  in  t he  pa pe we  dev el op  ne wly  e xten ded  sta te   ob se rv e r - base L FC  scheme  for  th ree - are a   interco nnect ed powe el ect ric  sy ste m.  T he  m ai co ntri bu ti ons  of the  pa per are as  foll owing:     Exten de sta te   obser ver  is  fir st  desig t es ti mate   the  un measu rab le   syst em  sta te   va riables  a nd   al s th e   load u ncer ta int ie s.     The  e xten de sta te   ob s er ver - base loa fr e qu e nc c ontrol le rs  are   de vel op e w hi ch   ut il iz disturban ce   est imat ion   te ch niques.  T hus,  t he  co ntr oller ga ins  are not  set   to  be  e xtre mely  hi gh   t at te nuat distu r ban c es   of un known  bo unda ries,  wh ic is  very  us e fu l  in  loa d fr e que ncy co ntr oller  desig n.     The   simulat io res ults  i nd ic at that  t he  pr op os e ne wly  m et hod  im pro ve the   s ys te m   dyna mic  respo nse   wh ic h pro vid es  a s ys te m c on t r ol to  a da pt a nd meet u t he  L FC re qu i reme nt   The  oth er  pa rts  of   t he  pro po sed  pa pe are  structu re in  t he  f ollo wing.  mathe mati cal   dynamics   model  of   t hr e e - area  power   el e ct ric  sy ste is  pr ese nte in  sect io 2.  The  f ollo wi ng   pro posed  e xten ded   ob s er ver  c on tr oller  is   sho wed  in   sect io 3.  T he  res ults  of   th va rio us  simu la ti on of  th ree - area   po wer  s yst em   app l ying   the   pro posed   new l ba sed   c on t ro l   ap proac hes   ar de scri bed  in   sect io 4.  La stl y,   c oncl us i ons   ar e   discusse d i se ct ion   5.           Evaluation Warning : The document was created with Spire.PDF for Python.
TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l         Ext end e st ate  ob s erver  based  load  fre quenc y co ntro ll er f or   ( V an Van  H uynh )   1003   2.   THE  MATHE MA TI C AL MODEL  O F   T HREE - AR E A   INTER CON NECTED  PO WER S YS TE MO DEL   At  first,  we  analyze  L FC  of   powe s ys te of   m ul ti - areas.  Fi gur il lustrate three - area   interco nnect ed   sy ste ms  [16]   a desm os trat e d.  The  s ole  miss ion   in  t he  rese arch   wor is  simply  t exa mine  th e   var i ou s   ti e - li ne   area i n   pow e s ys te ms   i ot her  to   c ontrol   t he   f reque ncies  of   the   f ollo wi ng  mu lt i - area   and  t regulat sim ultaneo us l powe fl ow i ng  th rought  the   ti e - li ne acco rd i ng  t a inte r - a rea   op e rati ng   a gr e ement.   In  th ree - a rea  netw orks,  eac c ontrol   area   is  in dicat ed  by  a   t urbi ne,   ge ner at or ,   a nd  gove rnor  s ys te m.  T he     tie - li ne  po wer  in  the   inte rcon nected   th ree - a r ea  m us be  c onside red  to   ge ner at e   the   incr ement  powe s ta bili ty   equ at io n o eac h powe s ys te m ar ea , s inc e t her e  is po wer f low  i eac a re a thro ugh  t he  ti e li ne.     Fr om   [3 4], th e ti e li ne  po we inc reme nt tha t i s out of a rea  is:      ( . ) = 2  (   )           ( 1 )     To  a chie ve  ba la nce  betwee i nterc onnect ed  c ontr ol  are as,  the   fr e qu e ncy  dev ia ti on   an ti e - li ne  powe r   fluctuati on  are  detect ed  in  othe to  dete rmin e   the  area  co ntr ol  error   (A CE )   of   eac co ntr ol   area.  The  AC can   be  e xpres sed   f or  eac c ontr ol   area   as   li ne ar it c ombina ti on   of  ti e - li ne   po wer  flu nctu at ion  an f requen c dev ia ti on.      =  +                 ( 2 )           Figure  1.  A  simpli fied  sket ch of  3 - area  inte rcon nected  po wer syste m       po wer   netw ork  is  sta ble  if   and   only  if  A CE 0.   I ot he w ords:   and  te nd   t ow a r ds   z ero   ov e r   s hort   ti me   fr ame I pr act ic e,   t ypic al   interco nne ct ed  gen e rati on   s ys te m   is   nonlinea r ity   an d   al s o   dynamic s a pp li cat ion   of  the   li near it mod el   is   al lowa ble   in  the   LFC   powe r   pro blem s   that  is,  i m oder n’s   powe s ys te m,   small  ch a ng e in loa is al wa ys  a ntici pated unde r normal  operati on s i n [ 18,  26 - 28 ].      ̇ 1 ( ) = 1 1 1 ( ) + 1 1 1 ( )  1  1  1 ( )  1  1  3 ( )  1  1 1   ( 3 )     ̇ 1 ( ) = 1 1 1 ( ) + 1 1 1 ( )             ( 4 )     ̇ 1 ( ) = 1 1 1 1 ( ) 1 1 1 ( ) + 1 1 1           ( 5 )     ̇ 1 ( ) = 1 1 ( ) +  1 ( ) +  3 ( )             ( 6 )     ̇ 1 ( ) = 2 12 1 ( ) 2 12 2 ( ) + 2 13 1 ( ) 2 13 3 ( )     ( 7 )     ̇ 2 ( ) = 1 2 2 ( ) + 2 2 2 ( )  2  2  1 ( )  1  1 2 ( )  1  1 2   ( 8 )     ̇ 2 ( ) = 1 2 2 ( ) + 1 2 2 ( )             ( 9 )     ̇ 2 ( ) = 1 2 2 2 ( ) 1 2 2 ( ) + 1 2 2           ( 10 )   Evaluation Warning : The document was created with Spire.PDF for Python.
                          IS S N :   1693 - 6930   TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l V ol.  19 ,  No.  3 June  20 21 :    10 01   -   100 9   1004   ̇ 2 ( ) = 2 2 ( ) +  1 ( ) +  2 ( )             ( 11 )     ̇ 2 ( ) = 2 12 2 ( ) 2 12 1 ( ) + 2 23 2 ( ) 2 23 3 ( )     ( 12 )     ̇ 3 ( ) = 1 3 3 ( ) + 3 3 3 ( )  3  3  2 ( )  3  3  3 ( )  3  3 3   ( 13 )     ̇ 3 ( ) = 1 3 3 ( ) + 1 3 3 ( )             ( 14 )     ̇ 3 ( ) = 1 3 3 3 ( ) 1 3 3 ( ) + 1 3 3           ( 15 )     ̇ 3 ( ) = 3 3 ( ) +  2 ( ) +  3 ( )             ( 16 )     ̇ 3 ( ) = 2 23 3 ( ) 2 23 2 ( ) + 2 13 3 ( ) 2 13 1 ( )     ( 17 )     The  mat rix  form  s hows  i t he  dy namic  e qu at io ns   from   (3)  to  ( 17),   t he  the re - a rea  i nterc onnected   powe sy ste desc rib ed by  Fig ure  2 wh ic ca n   be writt en  a nd e xpress ed  in  sta te - sp ace  r e pr e se ntati on   bel ow :     ̇ ( ) = ̃ ( ) + ̃ ( ) + ̃  ( )               ( 18 )     wh e re  ( )   is  the  sta te   vector,  ( )   is  t he  co ntr ol  vect or,  an d   ̃ , ̃ , ̃ is  con st ant  matri e quivale nt   i th   of  eac a rea.   ( ̃ × ̃ × × ).     ̃ = [         0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 3 0 0 ]             ̃ = [         1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 3 0 0 0 0 ]             ̃ = [                                                       1  1  1  1 0 0  1  1 0 0 0 0 0 0 0 0 0  1  1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 1 2 ( 12 + 13 ) 0 0 0 0 2 12 0 0 0 0 2 13 0 0 0 0 0 0 0 0  2  2 1  2  2  2 0 0  2  2 0 0 0 0 0 0 0 0 0 0 0 1 2 1 2 0 0 0 0 0 0 0 0 0 0 0 0 1 2 2 0 1 2 1 2 0 0 0 0 0 0 0 0 0 0 1 2 0 0 0 1 0 0 0 0 0 2 21 0 0 0 0 2 ( 21 + 23 ) 0 0 0 0 2 23 0 0 0 0 0 0 0 0 0 0 0 0 0  3  3 1  3  3  3 0 0  3  3 0 0 0 0 0 0 0 0 0 0 0 1 3 1 3 0 0 0 0 0 0 0 0 0 0 0 0 1 3 3 0 1 3 1 3 0 0 0 0 0 0 0 0 0 0 1 3 0 0 0 1 2 13 0 0 0 0 2 23 0 0 0 0 2 ( 31 + 32 ) 0 0 0 0 ]                                                         ̇ ( ) = [ 1 ( ) 1 ( ) 1 ( ) 1 ( )  1 ( ) 2 ( ) 2 ( ) 2 ( ) 2 ( )  2 ( ) 3 ( ) 3 ( ) 3 ( ) 3 ( )  3 ( ) ]     Be cause   it   is   te dio us  to   determine   the   va lues  of  the   s yst em  exa ct   pa rameters   of   ̃ , ̃ , ̃ du t nonlinea rity  a nd   dyna mics  of  po wer   el ec tric   sy ste m,  t he   dyna mic  model  ( 18)  is  re vi sed  to  the   no minal   par a mete rs  a nd  p a rameter  v a riat ion s se pa rati on s  in  the  fo ll owin g:     Evaluation Warning : The document was created with Spire.PDF for Python.
TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l         Ext end e st ate  ob s erver  based  load  fre quenc y co ntro ll er f or   ( V an Van  H uynh )   1005   ̇ ( ) = [ +  ( , ) ] ( ) + [ +  ( , ) ] ( ) + ̃ ( )   =  ( ) +  ( ) + ( , )   ( ) =  ( )                     ( 19 )     wh e re ,   is  the   e xact  values   of  ̃ , ̃ the  unknow matri ces  ( , )   an d  ( , )   de no te s   by   their  ti me - var ia nt  syst em  of   pa rametric  va riat ion s;  a nd  ( , )   is  cal le the  l umpe un ce rtai nties  an we  c an  al s denote   by   ( 20).     ( , ) =  ( , ) ( ) +  ( , ) ( ) + ̃ ( )           ( 20 )           Figure  2. Bl oc k diag ram of  3 - area  LFC s ys t em       3.   PROP OSE EXTE ND E D OB SER VER - BASED  LOA F REQ UEN CY CONT ROL LE R   The   pro pose new l c ontr ol  te chn iq ue   that   is,  sta te obser ver  performs   t he  f un ct io ns  by  e sti mati ng  the  sta te   va riables  of  th s yst ems  ty pical ly  the  ou t pu t   a nd  co ntr ol  va riab le s.  Stat obse rv e rs   can   be  de sig ne and  a pp li e on ly  wh e the   ob serv a bili ty  require c onditi on  is   sat isfie d.   First,  we   exte nd  l umped   unce rtai nty   as  an  ad diti on a sta te   var ia ble  to  desig a nd   init ia te   an  extend e ob se r ver - base load  fr e qu e nc co ntr ol le in   the s ys te m as   ( 21).     + 1 ( ) = ( , )                 ( 21 )     The n,  t he  th ree - area  powe s yst em in (1 9) ca n be  wr it te as:     Evaluation Warning : The document was created with Spire.PDF for Python.
                          IS S N :   1693 - 6930   TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l V ol.  19 ,  No.  3 June  20 21 :    10 01   -   100 9   1006   ̄ ̇ ( ) = ̄ ̄ ( ) + ̄ ( ) + ( ) ̄ ( ) = ̄ ̄ ( )           ( 22 )     ( ) =  ( , )                    ( 23 )     wh e re ̄ ( ) = [ ( ) + 1 ( ) ]   ̄ = [ ×     × 0 ×      0 × ] ̄ = [ × 0 × ] = [ 0 × × ] ̄ = [ × ,      0 × ]   With  re gards   to  t he  sta te   obs erv e rs   disc us s ed,  we   will   ap ply  the   nota ti on  ̄ ̂   to   in dicat the  vecto r   ob s er ved   sta te .   The  vect or   ̄ ̂   o t h e   obs e rv ed  sta t e   is  us ed   and   a ppli ed  in  the  sta te   fee db ac to  i niti at the  desire a nd r e quire c on tr ol   ve ct or .  If  we  cal l t he  sta te ̄ is ap pro ximate t o st at e,   ̄   ̂ the dy na mica l mo del:     ̄ ̂ ̇ ( ) = ̄ ̄ ̂ ( ) + ̄ ( ) + ( ̄ ( ) ̄ ̂ ( ) ) ̄ ̂ ̇ ( ) = ̄ ̄ ̂ ( )         ( 24 )     The   sta te s obse rv e ha ve  u   a nd  ̄   as   t he  i nput  an d o utput  si gnal .   The  g ai L   of   sta te  observ er  is   c hosen   s o t hat   the eige nval ue of   ̄ ̄ li e in the  de sired  l ocati ons  in the le ft - half s - pla ne.   The  contr ol in pu t i c ho s en  as ;     ( ) = ̄ ̂ = [ ̃     ̂ ] ̄ ̂                 ( 25 )     wh e re ,   ̃ is  the  f eedb ac c ontro gain  to  be  c hose so   t hat  th ei genvalues   of   ̃ li in  sp eci f ic   locat ion s   in the le ft - half s - pla ne  a nd the  lum ped unce rtai nty  c ompe nsa ti on   gain  ̃ is de sign e d:     ̂ = [ ( ̃ ) 1 ] 1 ( ̃ ) 1             ( 26 )     Com bin (22 ) a nd (2 4) ,  the e sti mati on  e rror o sta te  obse rvers  ( ) = ̄ ( ) ̄ ̂ ( )   can  be re vised b y:     ̇ ( ) = ̄ ( ) ( ̄ ( ) ̄ ̂ ( ) ) + ( ) = ( ̄ ̄ ) ( ) + ( )       ( 27 )     Denote   ( )   by   ( )   and u si ng f inal - val ue  the orem,  w e  h a ve:      ( ) =  ( (  ( ̄ ̄ ) ) 1 ( ) =  (  ( ̄ ̄ ) ) 1 ×   ( )   = (  ( ̄ ̄ ) ) 1 ×  ( )               ( 28 )     Sine  (  ( ̄ ̄ ) ) 1 is  bounde a nd    ( ) = 0 .   The refor e est imat ion  error  of  sta te   obser vers  is:   ( ) = ̄ ( ) ̄ ̂ ( )   is asy mp t otica ll sta ble.   Re mark  1:  I t he  s ys te m   sta te are   not  mea su ra ble,   the t he  est imat io of  the   lum pe uncertai nt a nd  the   par a mete rs   of  sy ste m   sta te c an  be   ap ply i ng  in  t he  de si gn  con t ro l.   T her e f or e the   c omposi te   con t ro l   la will   be desig ne as  in (2 1) .   Re mark  2:   It  is   note t hat  t he   lu mp e unce rtai nty  can not  be   at te nuat ed   c omplet el an t otall from   th e   sta te  equ at io no  matt er  wh at   c on t ro ll er   was   desig ne d.  I this  a ppr oac h,  on e   of  th m os rece nt  ac hi evab le   ob je ct ives  is  si mp ly  t tr un ca te   the  disturba nces  at   the  out pu po i nt  in  ste ady   sta te   by   t he  ap plica ti on   of   th e   com posit co nt ro la w T her e fore,  t he  li mit at ion recor ded  by  oth e c on t ro st rategies  i this   pa pe [ 21 - 25]  h as  bee s olv e d.       4.   SIMULATI O N RESULTS   In   t he  case  to  evaluate  the  ne wly  e xten ded  sta te   ob se rv e r   appr oach,  tw sim ulati on by   us i ng   t he   M A TLAB/S I M U LI NK  s of t war e  are give n as f ollow i ng :   Simulat io n 1:  The para mete r s of the  th ree - a rea inter co nnec te po wer sy ste m w e re s ee a s g i ven in  [16 17].    Ca se  1.   F or  th sim ulati on  at   this   insta nce   in  c ase  1,  t he  pa rameters   with   the re  nomi nal  val ues   of   the   3 - a rea   powe netw ork   are  a ppli ed.   At  this  point,  we  ass um e   zer disturba nces   occ ur i ng  on  t he  giv e s ys te m,  i.e. ,   ( , ) = 0 The refor e th fr e qu e nc fl un ct uation  or  dev ia ti ons  of   t he  3 - area  i nter connecte pow er  net work   accor ding  t t he  insta nce  of  case  w he re by,  a pp l ying  t he  new l e xtend e sta te   obs erv e c on t ro ll e are  disp la yed  in   th res ults  of  sim ulati on  in   Fi gure  3   to   Fig ure  4.  I Fig ur e   3,  the  fr e quenc dev ia ti on   ap proach e s   to  zer mar at   exactl 1.5 s.  Con se quently Figure  sho ws   ti li ne  power  dev ia ti on  getti ng  to   zer mar wit the  desig ne exten ded  sta te   obse r ver  c ontrolle r.  By   c ompa rin t he   s imulat ion  resul ts   from   t he  new l exten ded  obse rv e c on t ro ll er   with  res ults  giv e s how in  [ 16 17],  the  new l e xtend e sta te   ob serv e r   con t ro ll er  was   able to a ssure  f ast  r es pons e t o t he  s ys te m a nd also ca pa ble of tr un cat smal le over sho ots.   Evaluation Warning : The document was created with Spire.PDF for Python.
TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l         Ext end e st ate  ob s erver  based  load  fre quenc y co ntro ll er f or   ( V an Van  H uynh )   1007         Figure  3.   Fr e quenc y de viati ons  (H z ) of t he  t hr ee - area   without  disturb ances     Figure  4.   Tie  li ne powe r devia ti on   for powe sy ste with ou t  d ist urba nces       Ca se  2:   Desig ning  c ontrol le with  t he  a im  an ca pacit to  perf or m   excell ent  within  a uncert ai nt env i ronme nt  o po wer   netw orks  a re  al wa ys  the  main  goal   of   s eve ral  el ect rical   and   c on trol  en gin ee rs.  In   t his   case,  the  pro po s ed  e xten de sta te   obse r ver   c ontr oller   was  a ppli ed   unde un ce rtai nties  with  matc hed  par a mete rs   an l oad  disturb ance  i oth e to  e xamine   th net wor performa nce  unde m at che pa ramet e r   un ce rtai nties  a nd  loa distu r ban ce s.  The   load   dist urba nc e;   1 ( ) = 0 . 02   pu 2 ( ) = 0 . 015   pu an 3 ( ) = 0 . 01   pu   wer e   pres um to  ta ke  place  at   area  1,   a rea  2,  an area   acc ordin gly.  T he  r esp on ses   in  t he   c losed - lo op  f or  every one  of  t he   co ntr ol  area ap pl ying  the   e xten ded  sta te   obser ve c on tr ol le an th co nt ro ll er   giv e in  [1 6 17]   are  sho wn in F i gure  a nd  Figure  6.   Figure  5,  Fi gure  a nd  Ta ble   s how  cl ea rly  t hat  the   res ponse of  the   s ys te a re   not  on l gr eat   t deal  with  ove rsho ot - prob le ms,  but  al s ens ur es   quic an fast  set tl ing  pe rio a matc hed - to   th rece nt   appr oach  in   [ 16 17] In  the   same  c onditi on,   it   is  see t ha fr e quenc de viati on  co nve rg e t ze r in   ab out  1.6s wit t he n ewly p rop os ed   ex te nded  stat ob s er ver that  is , s at isfie d re qu i ment  of LFC  pro blems.                 Figure  5.   Fr e quenc y de viati ons  (H z ) of t he    three - a rea  unde matc hed unc ertai nties an l oad  disturba nces     Figure  6.   Tie  li ne powe r devia ti on   unde mat ched  un ce rtai nties a nd loa d dist urb ances       Table  1.   Sett ing  ti me  -     and Ma x.O.S ( maxim un -   over - sho ot - cal culat ion)  of ELFC a nd  DL FC   Kin d s o f   co n troller   Exten d ed  state ob serv er - b ased  load   fr eq u en cy  con troller   ( EL FC )   Decentraliz ed  load  f requ en cy   co n troller ( D LFC)   [16 ]   Para m eters     (s)   Ma x.O.  S  (pu )     (s)   Ma x.O.  S  (pu )     1   1 .5   2 .1 × 1 0 3   7 .5   3 .7 × 1 0 3     2   2 .0   1 .45 × 1 0 3   7 .5   3 .8 × 1 0 3     3   1 .8   2 .0 × 1 0 3   7 .5   4 .0 × 1 0 3       Simulat io 2:   The   pract ic al   powe sy ste m wi th  loa distu rbance  is   c on si de red  in   this   exa mp le .   T he  c on diti on s   and  par am et ers   us i ng  in   this  s imulat ion  are   the  sa me  with   the  recent   resea rch  in   [ 18].  Fi gure  an Fig ur e   8   sh ow   that  t he   pro pose e xtend e sta te   ob serv e c ont r ol  sche me  has   f ast er  re spo ns e   an le sse sign ic a nt   ov e rs hoot in  c ompari ng the  previ ou s  contr ol  in [1 8].   Evaluation Warning : The document was created with Spire.PDF for Python.
                          IS S N :   1693 - 6930   TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l V ol.  19 ,  No.  3 June  20 21 :    10 01   -   100 9   1008   Re mark  3:   B matc hing - up  t he  re su lt of  the  sim ulati on  for  the   tw si mu la ti ons  a bove,   t he  new l exten ded  sta te   ob se rv e r - base loa f re qu e nc c on tr ol le r disp la yed  rob us tness   an fast  res ponse   to  dist or ti on an disturba nce  oc currin on  t he  s ys te m   co rr el at ed   with   var ia nt  of  t he   matc he unc ertai nties  an loa disturba nces  use d for sim ulati on s .             Figure  7.   Fr e quenc y de viati ons  of the  t hr ee - area  unde matc hed uncertai nties a nd loa d dist urb ances     Figure  8.   Tie  li ne powe r devia ti on   unde mat ched  un ce rtai nties a nd loa d dist urb ances       5.   CONCL US I O N   In  this  pa per,  t he  ne wly  e xten ded  sta te   ob se r ver  f or  LFC   f or  a i ntercon ne ct ed  s ys te is   performe d.  In   real  en vir on ment,  s om va rio us   sta te   va riables  are  not  measu rab le   i load  fr e quenc con t ro s ys te f or  instance  a rea  c on t ro e rror   or  com bin at io of  area  co ntr ol  er rors.   T res olve   this  unmeas urable  sta te   va ri ables  pro blem,  the  exten ded   sta te   ob se r ver   is  pro posed  for  e sti mati ng   the  unmeas ur a ble  sta te   var ia bles The   exten ded  sta te   ob s er ver - ba sed   load   f reque nc c on tr ollers  ut il iz distur ba nc est imat ion   t echn i qu e s;  th us,  the  con t ro ll er   gai ns  are  not  set   t be  extre mely   high  to   at te nua te   disturba nces   of  unknow boun dar ie s,   w hi ch  is   very   use f ul  i load   f re qu e nc co ntr oller   de s ign .   T he refor e ,   it   ca be   c oncl ud e that  t he   a pp li cat ion  of  the   pro po se exte nd e sta te   obs erv e for  l oad  fr e qu e nc c ontrols  of  inte rconn ect e powe r   sy ste ca operate   eff ect i vel i pr act ic al   si gh t.   By   us in M A TLAB/SI MUL INK,   the   sim ul at ion  res ults  a bove   pr ese nt  t hat  t he   new l meth od   imp r ov es   t he  dynamics   r esp on s es  of  t he  s ys te m   an pr ovide   de sig ns  f or  new  LFC ’s   sy ste that  sat isfie s   th LFC   require ments.   In  the   f uture  w ork,  we   te nd   to   desig e xten ded  sta te   obse r ver  f or  r obus LFC’s i m ulti - area  powe s yst ems combi ne d wit h rene wa ble en e r gy syst ems.       ACKN OWLE DGME NT   This  re searc i fun ded  by  F oundat io f or  S ci ence  an Tec hnolog D evel opment  of  To Du c   Th an Un i ver sit ( FOSTEC T) , webs it e:  h tt p:// fo ste ct .tdtu. e du. vn, unde r Gra nt F OS TECT .20 17.BR .05       REFERE NCE S   [1]   Fu  C. ,   &   Ta n   W. ,   " De ce ntr ali sed   Lo ad   Frequ enc y   Control  fo Pow er   Sys te m with   Comm un ic a ti on   Del ays  via  Acti ve   Disturba nce   R ej e ction, "   I ET  Gene ration ,   Tr ansm i ss ion  & D istribut ion,  vol .   12 no .   6 ,   pp .   1 751 - 8687,   2018 .   [2]   Zha ng  Y. ,   Ya ng  T.,  " De ce ntr a li z ed  Sw it ch ing  Control   Stra te gy   for  Lo ad   Frequ enc Con trol   in  Multi - Area   Pow er   Sys te ms wi th Ti me   De la y   and   Pa cke t   Losse , IEEE  A ccess ,   vol .   8 ,   pp.   15838 - 1585 0,   Jan   2020 .     [3]   Deve ndra   K.  Ch at urve d i " Techn ique s a nd   it s   Applicati ons  in El e ct ri ca l   Eng ine er i ng "   Springer ,   20 08.   [4]   Vija Vi tt a l ,   Ja me D.   McCalle y ,   Paul   M.   Ande rson ,   A.  A .   Foua d   P. ,   " Pow er  Sys te Con trol  an Stabilit y , "   Wi l ey   3rd E dit io n. Oct ober   2019 .   [5]   Guh D. ,   Roy   P.  K . ,   &   B ane r jee  S. ,   "Lo ad   Fre quenc y   Contro of  Int erc onn ecte Pow er   Sys te m   Us ing   Grey   W olf   Optim izati o n, S warm   and  Ev olu ti onary  Computa ti on,   vol. 27 ,   pp .   97 - 115,   Apr   20 16 .   [6]   Yous ef  H.  A. ,   AL - Kharusi  K. ,   Albadi  M.   H.,  &   Hos seinz ade h   N. ,   "Lo ad  Frequ en cy  Contro of   Multi - Area   Pow er   Sys te m:  An  Ada pti ve   Fuz zy   Log ic   Approac h , IE EE   Tr ansacti ons  on  Powe System s, vol.   29,   no.   4,   pp. 1822 - 183 0 ,   Jan  2014 .   [7]   Ze ng   G.   Q.,  Xi X .   Q . ,   &   Ch en  M.   R . ,   " An   Adapti ve   Model   Predi ct iv e   Lo a Freque n cy   Co ntrol   Method   fo   Multi - Area   In te r conne c te Pow e Sys te ms  with  Photovolt aic  Ge ner ations, "   Elec t rical   Pow er  and  Ene rgy  S yste m ,   vol.   10 ,   no .   11 ,   p p.   1 - 23 ,   Nov 201 7.   [8]   Beni   Reh ia r a   A. ,   Y orino  N . ,   Sas aki   Y. ,   Zok Y . ,   "A Adapt iv Lo ad  Frequ en cy  Control  Base on  L ea st  Squa re   Method , "   Adv an ce s in   Mode ll ing   and  Control   of   Wind  and  H ydroge nerators ,   vol .   49,   pp .   220 ,   202 0.   Evaluation Warning : The document was created with Spire.PDF for Python.
TELK O M N IKA   Tel ec om m un   C ompu t El  C on t ro l         Ext end e st ate  ob s erver  based  load  fre quenc y co ntro ll er f or   ( V an Van  H uynh )   1009   [9]   Dong  L .   L ,   Zhang  Y . ,   Gao   Z .   Q. ,   " A   Robust  Dec ent r al i ze d   L oad  Frequ enc y   Control le r   for   I nte rco nn ec t ed   Pow er   Sys te ms, "   ISA   Tr ansacti ons ,   vol .   51,   no.   3,   pp.   41 0 - 419,   May   201 2.   [10 ]   Gheisa rne j ad  M . ,   Khooban ,   M.  H.,  " Design  an  Opt im a F uzz Fra ct ion al  Proportiona l   I nte gra Deri v ati ve  Control le r   with   Deri vative   Filt er   for  L oad   Frequ enc Con trol  in  Pow er  Sys te ms, Tr ansacti ons  of  th Insti tute  of  Me asur eme nt  an Control ,   vol .   4 1,   no .   9 ,   pp .   1 - 1 9,   Jan   2019.   [11 ]   Anw ar   M.   N. ,   a nd  Pan   S.,  " N ew  PID   Loa d   Freque ncy   Contro l le r   Design   Meth od  in   Frequ enc y   Doma in  Throug Dire ct Synthe sis   Approac h , "   E lectric P ower   and  Ene rgy  S yste ms ,   vol.   67 ,   pp .   560 - 569,   May   2015.     [12 ]   Sonkar  P.,  &   R a hi  O.   P. ,   " Tun in of  Modi fie d   PI Loa d   Freque n cy  Contro ller  for   Int erc onne cted  Sys te with   Wind  Pow er  Plant  vi IMC   Tuni ng   Method , "   201 4th   IE EE  Ut tar  Pradesh   Se ct ion   Int ernati o nal  Conf ere nc e   o n   El e ct rica l Computer  and  Elec tr onic s ,   Jan   2018 .   [13 ]   K.  L ia o   and   Y.   Xu,  " Robust   Loa d   Freque n cy   Control  Sch em e   for   Pow er  Sys tems  Based   on   Se cond - Order  Sl iding  Mode  and  Extended  Disturba nce   Obs erv er , "   IEE Tr ansac ti on on  Industrial  Informatic s ,   vol.   14 ,   no .   7,   pp.   3076 - 3086 ,   J uly  2018.   [14 ]   Zhe ng   Y . ,   Li u   J. ,   L iu  X. ,   Fang   D . ,   Wu  L. " Ad apt iv Second - Order  Slid ing  Mo de  Contro Desi gn  for  Cl ass  o f   Nonline ar   Sys tems wit Unkno wn Input , "   Ma th emati ca Probl e ms   in  Engi n ee ri ng vol .   2015 ,   no .   1 ,   pp . 1 - 7,   2015 .   [15 ]   Y.  Sun,   Y.   Wa n g,   Z.  Wei  and   X.   Wu ,   " Robust   H∞  Lo ad   Freque n cy  Con trol  of   M ult i - Are a   Pow er  Sys te wi th   Tim e   Dela y:  Slidi n Mode  Control  Approac h, "   IE EE /CAA  Journa of  Aut omat ic Sini ca ,   vol.   5 ,   n o.   2,   pp .   610 - 61 7,   2018.   [16 ]   Ya ng  M . ,   Yang   F . ,   Chengshan   W .,   and  Peng   W. ,   " Dec en tra l iz ed  Sl idi ng  M ode  Loa Fr eq uenc Contro f or    Multi - Area   Pow er  Sys te ms, "   IE EE   Tr ansacti ons   on  Pow er  Syst e m ,   vol .   28 ,   no .   4 ,   pp.   4301 - 4309,   Aug 2013.     [17 ]   Muthana   T.  Alr ifa i ,   Moha me d   F.  Hass an,  Mohame d   Zr ibi . " Dec ent r al i ze d   L oad  Frequ enc y   Control le r   for     Multi - Area   Int er conne c te d   Pow er  Sys te m, "   Elec t rical   Powe r and   Ene rgy  S yste ms ,   vol.   33 ,   no .   2 ,   pp . 198 - 209,   2011.   [18 ]   Yang  Mi  et   a l. ,   " The  Slidi ng   Mode  Loa d   Fre quenc Con trol  for  Hybrid   Pow er  Sys te B a sed  on  Disturb a nce   Obs erv er, "   E lect rical   Powe r and   Ene rgy  S yste ms ,   vol.   74 ,   pp .   446 - 452,   Jan   2016.     [19 ]   D.  Qian ,   and  G.  Fan,   "N eur a l - Network - Base Te r mi n al   Sl id ing  Mode   Cont rol  for   Freque n cy  Stab il i zation   of   Rene wabl Pow er  Sys te ms, "   IE EE /CAA  Journal   of Aut omati ca  S i nic a ,   vol .   5 ,   no .   3,   pp .   706 - 717 ,   Apr 2018.   [20 ]   S.  Tri e al. ,   "P assivit y - Based  Design  of  Slidi n Modes  for  Optim al   Loa Frequ enc Contro l, "   I EE Tr ansacti o ns  on  Control  S ystem s Tec hnology ,   vol.   27 ,   no .   5 ,   pp . 1893 - 1906,   201 9.   [21 ]   Li   H .   Y . ,   Shi  P . ,   Yao  D .   Y . ,   Wu   L .   G . " Obs erv e r - Based  Adapt iv Slidi ng   Mode  Control   of   Nonl ine ar   Markovi an  Jump System s, "   Aut omatic a ,   vo l. 64, pp. 133 - 142 ,   Mar  2016 .   [22 ]   Khaya ti   K. ,   " Multi var ia b le   Ada pti ve  Slid ing - Mode  Obs erv er - B ase Control   for   Mec hanica l   Sys te ms,”   Canad ia n   Journal  of   Elec t rical   and   Computer  Eng ine ering ,   vol.   38 ,   no .   3 ,   pp . 253 - 265,   Nov 2 015.   [23 ]   Wa ng  B . ,   Shi   P.,  Kari m H .   R . ,   L im  C.   C,   " Obs erv er - Based   Slidi ng   Mode  Co ntr ol  for   Stab il i z at ion   of  Dyn a mi c   Sys te with   Del aye Outpu Fe e dbac k, " Ma the ma ti cal P robl ems i Engi n ee ring ,   v ol.   3 ,   pp .   1 - 6,   Se 2013.   [24 ]   Ouass ai M.,  Maa roufi   M.,  &   C her kaoui  M,   " O bserve r - Based   Nonlinear  Contro l   of  Pow er  Sys tem  Us ing  Slid ing   Mode  Control St rat egy , "   E lectric  Powe r Sy st ems  Re search ,   vol .   8 4,   no .   1 ,   pp .   135 - 143,   Jan   2012.   [25 ]   Yang  B. ,   Yu   T.,   Shu  H . ,   Yao   W . ,   &   Ji ang  L. ,   " Slidi ng - Mode   Perturbation   Obs e rve r - Based   Slid i ng - Mode  Contro Design  for   Stabilit y   Enh anc e me n of   Multi - Mac hi ne  Pow er   Sys tems, Tr ansacti o ns  of  the  Insti tute  of   M easurement   and  Control ,   vol .   41 ,   no .   15 ,   pp .   1418 - 1434,   Jul   2 018.   [26 ]   Prasad  S.,   Purw ar  S.,  Kishor   N.,   " Non - Li n e ar  Slidi ng   Mode  Loa d   Freque n cy  Control  in  Multi - Area   Pow er  Sys te m, Contro Eng ine ering   Pr act i ce ,   vol .   61 ,   p p.   81 - 92 ,   De 20 17 .   [27 ]   Dianwe Q. ,   Shi wen  T. ,   Xiangjie   L. ,   " Loa Frequ enc Contro for   Micro  Hydro  Pow er  Plant by  Slidi ng  Mode  an Model  Order   Re duct ion , "   Au tomati ka ,   vol .   56 ,   no .   3 ,   pp .   318 - 330 ,   Jan  2017 .   [28 ]   Dianwe Q. ,   Shi wen  T. ,   Hong  L. ,   Xiang ji e   L . ,   " L oad  Freque n cy  Control   by  N eur al - Network - Bas ed  Int egr al  Slidi ng   Mode  for  Nonl in ea Pow er   Sys tems wit Wi nd   T urbine s, "   N euroc omputing ,   vo l.   1 73,   pp .   875 - 885 ,   Jan  2016 .     Evaluation Warning : The document was created with Spire.PDF for Python.