PSO based Congestion Management in Deregulated Power Systems using Optimal Allocation of TCSC

Transcription

1 Internatonal Journal of Innovaton and Scentfc Research ISSN ol. 25 No. 1 Jun. 2016, pp Innovatve Space of Scentfc Research Journals PSO based Congeston Management n Deregulated Power Systems usng Optmal Allocaton of K. etrvel 1, S. Ganapathy 2, and K. Uma Rao 3 1 Natonal Power Tranng Insttute /PSTI, Bangalore, Karnataka, Inda 2 Department Electrcal Engneerng, AnnamalaUnversty, AnnamalaNagar, Tamlnadu, Inda 3 Department Electrcal and Electroncs, R College of Engneerng, Bangalore, Karnataka, Inda Copyrght 2016 ISSR Journals. Ths s an open access artcle dstrbuted under the Creatve Commons Attrbuton Lcense, whch permts unrestrcted use, dstrbuton, and reproducton n any medum, provded the orgnal work s properly cted. ABSTRACT: Congeston n the transmsson lnes s one of the techncal problems that appear partcularly n the deregulated envronment. One of the congeston management methodologes s nstallng Thyrstor Controlled Seres Compensator () devces nto the system. The maor obectve n applyng s to ncrease power transfer capacty n crtcal te-lnes under contngency condtons. Under normal steady state condtons, t can be used to damp Sub-Synchronous Resonance by convertng a part of the fxed compensaton to controllable seres compensaton. In ths paper, a novel method usng Partcle Swarm Optmzaton (PSO) algorthm s proposed to determne the optmal allocaton of devces for maxmzng the Avalable Transfer Capablty (ATC) of power transactons between source and snk areas n the deregulated power system. The algorthm smultaneously searches the locaton, sze and cost of devces. ATC s calculated usng AC Power Transfer Dstrbuton Factors (ACPTDF). The effectveness of the proposed method s demonstrated usng a modfed IEEE 30 test system n normal and contngency condtons for the selected blateral, multlateral and area wse transactons. The smulaton results show that the ntroducton of devces n a rght locaton could enhance ATC, reduced total losses and mprove the lne congeston as compared to that of the system wthout devces. KEYWORDS: Avalable Transfer Capablty, Independent System Operator, Thyrstor Controlled Seres Compensator, Partcle Swarm Optmzaton, Power Transfer Dstrbuton Factors. 1 INTRODUCTION The power system operaton faces new challenges due to deregulaton and restructurng of the electrcty markets. The deregulated power system conssts of Generaton companes (Gencos), Dstrbuton companes (Dscos), Transmsson companes (Transcos) and Independent System Operator (ISO) wth an open assess polcy. The power flow pattern n deregulated envronment s dfferent from those n exstng regulated one. All partes wll try to get the benefts of cheaper source and greater proft margns leadng to overloadng and congeston of certan transmsson corrdors. The role of ISO s to releve that congeston so that the system s mantaned n a secure state. To releve the congeston, ISO can use ether cost-free or non-cost free means. In Cost-free means have advantages n that no extra cost s ncurred n relevng the congeston and Gencos and Dscos do not come n to pcture [1-3]. In order to facltate the electrcty market operaton and trade n the restructured envronment, ample transmsson capablty should be provded to satsfy the demand of ncreasng power transactons. The conflct of ths requrement and restrctons on the transmsson expanson n the restructured electrcty market has motvated the development of methodologes to estmate ATC of the exstng transmsson grds. For transmsson networks, one of the maor consequences of the non-dscrmnatory open-access requrement s a substantal ncrease of power transfers, whch demand adequate ATC to ensure all economc transactons. Suffcent ATC should be guaranteed to support free market tradng and mantan an economcal and secure operaton over a wde range of system condtons [4, 5]. However, tght restrctons on the constructon of new facltes due to the ncreasngly dffcult economc, Correspondng Author: K. etrvel 67

2 PSO based Congeston Management n Deregulated Power Systems usng Optmal Allocaton of envronmental, and socal problems, have led to a much more ntensve shared use of the exstng transmsson facltes by utltes and ndependent power producers (IPPs). These concerns have motvated the development of strateges and methodologes to boost the ATC of the exstng transmsson networks. arous mathematcal models have been developed by the researcher to determne the ATC of the transmsson system based on conventonal power system equatons. Recently the dstrbuton factors based on DC and AC power flow methods [6-9] have been proposed for calculatng ATC. In ths paper, the Power Transfer Dstrbuton Factor (PTDF) usng AC power flow are derved to calculate ATC usng senstvty propertes of Newton Raphson Load Flow (NRLF) Jacobean. Flexble Alternatng Current Transmsson Systems (FACTS) devces are proved to be very useful n achevng the control of power flows wthout dsturbng the generaton schedulng or topologcal changes and n addton these devces wll also enhance the secured operaton of power system [10, 11]. FACTS devces, despte mnmzng lne congeston and maxmzng the transfer capablty, assure that contractual constrants and targets are satsfed farly. FACTS devces not only provde solutons for effcently ncreasng transmsson system capacty but also ncrease ATC, releve congeston, mprove relablty and enhance operaton and control. Moreover, t s mportant to ascertan the locaton for placement of these devces because of ther consderable costs. However, t s hard to determne the optmal allocaton and parameters of FACTS devces due to the complcated combnatoral optmzaton. Thus, attenton s pad n ths current work to study a technque to optmally allocate the devces to enhance ATC. The task of calculatng ATC s one of man concerns n power system operaton and plannng. ATC s determned as a functon of ncrease n power transfers between dfferent systems through prescrbed nterfaces. The nserton of such devces n electrcal systems seems to be a promsng strategy to ncrease ATC [12, 13]. In ths paper, a s used for enhance the ATC and mprovng lne congeston. Some of the researchers have dscussed the heurstc optmzaton algorthms that are used to locate FACTS devces n power systems [14-16], such as the Smulated Annealng (SA, Tabu Search (TS), Evolutonary programmng (EP), Genetc Algorthm (GA) and recently Partcle Swarm optmzaton (PSO) to solve smple and complex problems effcently and effectvely. However, unlke SA, TS, EP and GA, each ndvdual n PSO fles n search space wth a velocty whch s dynamcally adusted accordng to ts own flyng experence and ts companons flyng experences. The man merts of PSO are ts fast convergence speed and t can be realzed smply as less parameters need adustng. The PSO algorthm was frst ntroduced by Eberhart and Kennedy [17, 18]. In ths paper, PSO technque s used to fnd optmal locaton and sze of to acheve maxmzaton of ATC, decrease the lne congeston and total power loss. Ths paper s dvded nto several sectons. Secton 2 elaborates the Avalable Transfer Capablty. Secton 3 descrbes computatonal procedure for ATC determnaton whle Secton 4 presents mathematcal model of the Devce. Secton 5 presents problem formulatons of optmum allocaton of devces. The procedure of PSO algorthm to allocate FACTS devces s dscussed n Secton 6. The smulaton results are presented and dscussed brefly n Secton 7. 2 AAILABLE TRANSFER CAPABILITY Avalable Transfer Capablty (ATC) s a measure of the transfer capablty remanng n the physcal transmsson network for further commercal actvty over and above the already commtted uses [7, 8]. ATC evaluaton s mportant because t s the pont where power system relablty meets electrcty market effcency. ATC can have a huge mpact on market outcomes and system relablty, so the results of ATC are of great nterest to all nvolved. ATC can be expressed as: ATC TTC TRM CBM ETC (1) Where Total Transfer Capablty (TTC) s the maxmum transfer power that does not reach the lmts, Exstng Transmsson Commtment (ETC) s the sum of the avalable transmsson commtment between 2 areas, Transmsson Relablty Margn (TRM) s the amount of transmsson capablty that s requred to ensure that the nterconnected system s secure under an acceptable range of uncertanty and Capacty Beneft Margn (CBM) s the amount of transmsson that s reserved by the load to ensure access to the generaton from nterconnected systems to meet the requrement of the generaton relablty. Utltes would have to determne adequately ther ATC s to guarantee that system relablty s mantaned whle servng a wde range of transmsson transactons. ATC between and wthn areas of the nterconnected power system and ATC for crtcal transmsson paths between these areas would be contnuously updated and posted about changes n scheduled power transfers between the areas. ATC at base case, between m and n usng lne flow lmt (thermal lmt) crteron s mathematcally formulated usng AC Power Transfer Dstrbuton Factors (ACPTDF). ISSN : ol. 25 No. 1, Jun

3 K. etrvel, S. Ganapathy, and K. Uma Rao 3 ATC DETERMINATION USING AC POWER TRANSFER DISTRIBUTION FACTORS AC Power Transfer Dstrbuton Factors (ACPTDF) determnes the lnear mpact of a transfer (or changes n power necton) on the elements of the power system. These values provde a lnearzed approxmaton of how the flow on the transmsson lnes and nterfaces change n response to transacton between the seller and buyer. For a sngle area ATC, the transacton wll be between seller and buyer present n the area and for mult-area ATC, the transacton wll be between two areas. The proposed AC power transfer dstrbuton factors method s used for calculaton of ATC for a change n MW transacton at dfferent operatng condtons. Consder a blateral transacton t k between a seller m and buyer n. Lne l carres the part of the transacted power and s connected between es and. For a change n real power, transacton among the above buyer and seller by t k MW, f the change n a transmsson lne quantty (real power flow) q 1 s q 1, power transfer dstrbuton factors can be defned as, PTDF q l, mn (2) t k The transmsson quantty q l can be ether real power flow from to (P ) (or) real power flow from to (P ). The above factors have been proposed to compute at a base case load flow wth results usng senstvty propertes of NRLF Jacoban. Consder full Jacoban n polar coordnates [J T ], defned to nclude all the es except slack (ncludng Q- equatons also for P es). We get the followng 1 P P P 1 P (3) J T Q Q Q Q In a base case load flow, f only one of the k th blateral transactons s changed by t k MW, only the followng two entres n the msmatch vector on RHS of (3) wll be non zero. P t k P t k (4) Wth the above msmatch vector elements, the change n voltage angle and magntude at all es can be computed from (3) and (4) and, hence, the new voltage profle can be calculated. These can be utlzed to compute all the transmsson quanttes q l and hence the correspondng changes n these quanttes, q l, from the base case. Once the q l for all the lnes correspondng to a change n transacton t k s known, PTDFs can be obtaned from (2). These ACPTDFs, whch are computed at a base load flow condton, have been utlzed for computng change n transmsson quanttes at other operatng condtons as well. ACPTDF s also calculated for multlateral transacton n whch group of sellers have a blateral contract wth group of buyers. The change n multlateral transacton can be assumed to be shared equally by each of the sellers and buyers. However, the transacton amount can be shared n any pre-decded rato n a deregulated envronment. The msmatch vector for the multlateral transactons wll have non zero entres correspondng to the buyer and seller es. The rest of the procedure for calculaton of ACPTDF wll be the same as outlned above the blateral transacton case. The ATC s then calculated as follows ATC mn T mn N L mn,, (5) Where T,mn denotes the transfer lmt values for each lne n the system. It s gven by T, mn max 0 P P ACPTDF, (nf nte ) max P P ACPTDF, mn 0 mn ; ; ; ACPTDF ACPTDF ACPTDF, mn, mn, mn Where, P max s the MW power lmt of a lne between and, P o s the base case power flow n the lne between and, ACPTDF,mn s the power transfer dstrbuton factor for the lne between and when a transacton s takng place between m and n. N L s the total number of lnes. ACPTDF as gven n equaton (6) s operatng pont dependent and computed usng Jacoban nverse. ACPTDF s reman farly constant for reasonable varatons n power nectons. (6) ISSN : ol. 25 No. 1, Jun

4 PSO based Congeston Management n Deregulated Power Systems usng Optmal Allocaton of 4 MODELING OF THYRISTOR CONTROLLED-SERIES CAPACITOR DEICES Fg 1. Basc structure of Thyrstor controlled-seres capactor Thyrstor controlled-seres capactor () s a seres connected FACTS devce n whch a capactor s connected n seres wth the transmsson lne and a parallel connecton of thyrstor-controlled nductor wth the capactor s shown n Fg 1. s connected n seres wth the lne conductors to compensate for the nductve reactance of the lne. It may have one of the two possble characterstcs namely capactve or nductve, respectvely to decrease or ncrease the reactance of the lne X lne respectvely. The ratng of s depends on transmsson lne where t s located. To prevent overcompensaton, reactance s chosen between-0.8 X lne to 0.2 X lne Moreover, n order not to overcompensate the lne, the maxmum value of the capactance s fxed at -0.8 X lne whle that for nductance, t s 0.2 X lne. The rated value of where t s located s gven by X X X (7) Lne X * X (8) Lne Where, X s the reactance of the transmsson lne and Lne s the compensaton factor of. The power flows n heavly loaded lne can be reduced by through power flow control n the network. The power flow control wth the s used to decrease or ncrease the overall lnes effectve seres transmsson mpedance, by addng a capactve or nductve reactve correspondngly. Fg.2 shows a model of transmsson lne wth one whch s connected between - and -. The power flow equatons of are gven by Fg.2 Sngle lne dagram of power flow control wth P Q 2 g ( g cos b sn ) (9) 2 b ( g sn b cos ) (10) Where, and are voltage of th and th respectvely 5 PROBLEM FORMULATION As stated n secton 2, ATC s defned as the addtonal power that can be transmtted through a specfed nterface over and above the already commtted transactons. The problem of ATC computaton n blateral and multlateral transacton can ISSN : ol. 25 No. 1, Jun

5 K. etrvel, S. Ganapathy, and K. Uma Rao be formulated as an optmzaton problem n whch the obectve s to maxmze the dfference between TTC and ETC wthout volatng the constrants. The am of the optmzaton s to perform the best utlzaton of the exstng transmsson lnes. The obectve s to maxmze the ATC.e., uncommtted actve transfer capacty of the prescrbed nterface, when a transacton s takng place between a seller (m) and buyer (n). The obectve functon to be maxmzed s expressed as J Maxmze ( ATCmn ) (11) Where ATC for each blateral transacton between a seller at (m) and power purchaser at (n) satsfes the followng power balance relatonshp: P D 0 D P 0, t k It s subected to Eqn (13) and (14) as equalty constrants such as the power flow equatons at, (12) P G P D nb 1 Y cos 0 (13) Q G Q D nb 1 Y sn 0 (14) P G, Q G s real and reactve power generaton at, P D, Q D s real and reactve loads at,, oltage magntude at and. Y, s magntude and phase parts of the th element of the admttance matrx., s voltage angle of and, respectvely. nb s the total number of es. In Eqn (15), (16) and (17) are nequalty constrants, whch ensure the system real, reactve power flow and voltage lmtatons, respectvely. mn G G max G P P P for 1,2... n (15) mn G G max G g Q Q Q for 1,2,... n (16) mn b b max b b g for b 1,2... n (17) The constrants on the devces used n ths work are gven below: 0.8 X X 0.2 X p. u. (18) Lne Lne Where, X s the reactance added to the lne by placng, X Lne s the Reactance of the lne where s located. To prevent overcompensaton, reactance s chosen between 0.8X Lne to 0.2X Lne. The constrants on the nstallaton cost of the correspondng devces are gven by, IC C * S *1000 (19) where IC denotes optmal nstallaton cost of devces n US$. C represents cost of nstallaton of devces n US $/Kar; S s the operatng range of devces n MAR and t s gven by S Q 1 Q 2 (20) Where Q 2 s the reactve power flow n the lne after nstallng devce n MAR and Q 1 represents reactve power flow n the lne before nstallng devce n MAR. The cost of nstallaton of are taken from Semens data base and reported n [21]. The cost of nstallaton of devces are gven by the followng equatons: 2 C S S (21) ISSN : ol. 25 No. 1, Jun

6 PSO based Congeston Management n Deregulated Power Systems usng Optmal Allocaton of 6 OPTIMAL PLACEMENT OF USING PARTICLE SWARM OPTIMIZATION ALGORITHM 6.1 OER IEW OF PSO ALGORITHM The Partcle Swarm Optmzaton algorthm s a smple, fast and effcent populaton based optmzaton method whch was proposed by by Kennedy and Eberhart [17]. The basc assumpton behnd the PSO algorthm s that brds fnd food by flockng and not ndvdually. Ths leads to the assumpton that nformaton s owned ontly n the flockng. The swarm ntally has a populaton of random soluton. Each potental soluton, called a partcle (agent) s gven a random velocty and s flown through the problem space. All the partcles have memory and each partcle keeps track of ts prevous best poston (P best ) and the correspondng ftness value. The swarm has another value called (G best ), whch s the best value of all P best. It has been found to be extremely effectve n solvng a wde range of engneerng problems and solves them very quckly. At each tme step, the partcle swarm optmzaton conssts of velocty changes of each partcle towards ts P best and G best [19, 20]. After fndng the best values, the partcle updates ts velocty and poston accordng to the followng equatons: k1 k k k w* C1 * rand1 *( Pbest S ) C2 * rand2 *( Gbest S ) (22) S S (23) k 1 k k 1 Where, k+1 s the velocty of th ndvdual at (k + 1) th teraton, k s the velocty of th ndvdual at k th teraton, W s the nerta weght, C 1 and C 2 are the postve constants havng values (0, 2.5), rand 1 and rand 2 are the random numbers selected between 0 and 1, P best s the best poston of the th ndvdual, G best s the best poston among the ndvduals (group best) and S k s the poston of th ndvdual at k th teraton. The acceleraton coeffcents C 1 and C 2 control how far a partcle wll move n a sngle teraton. Typcally, these are both set to a value of 2.5. The velocty of each partcle s modfed accordng to (22) and the mnmum and maxmum velocty of each varable n each partcle s set wthn the lmts of mn and max respectvely. The poston s modfed accordng to (23). The nerta weght factor w s modfed usng (24) to enable quck convergence. ( wmax wmn ) w wmax * ter ter (24) max Where w max s the ntal value of nerta weght equal to 0.9, w mn s the fnal value of nerta weght equal to 0.4, ter s the current teraton number and ter max s the maxmum teraton number. Small values of w result n more rapd convergence usually on a suboptmal poston, whle a too large value may prevent dvergence of soluton. The PSO system combnes two models; a socal-only model and a cognton-only model. These models are represented by the velocty update, shown n (22). 6.2 STEP BY STEP ALGORITHM TO OPTIMALLY LOCATE FOR MAXIMIZING ATC USING PSO Step 1: Step 2: Input the data of Transmsson lne, generators es and loads. Choose populaton sze of partcles, maxmum number of teratons and convergence crteron. Defne type of transactons. Select reactance settng and locaton (lne number) of as control varables. Step 3: Randomly generate populaton of partcles wth ther varables n normalzed form (.e., between 0 to 1) Step 4: Step 5: Randomly nstall one devces n the Transmsson lne and check that devce s not employed on the same lne more than once n each teraton. Fnd demoralzed value (actual value) of reactance and locaton of usng the followng Equaton. X X X X ) * X Denormalzed (25) mn ( max mn normalzed Where X mn, X max are mnmum and maxmum values of the varable X s reactance respectvely. Demoralzed value of locaton of s rounded to nearest nteger durng optmzaton. Modfy the admttance matrx. Run Newton-Raphson load flow to get lne flows, actve power generatons, reactve power generatons, lne losses and voltage magntude of all es. Step 6: Calculate the ATC of each partcle usng Eqn (5). Step 7: Calculate the obectve functon of each partcle subect to satsfy the constrants usng Eqn (11). ISSN : ol. 25 No. 1, Jun

7 K. etrvel, S. Ganapathy, and K. Uma Rao Step 8: Fnd out the global best (G best ) partcles havng maxmum value of obectve functon n the populaton and personal best (P best ) of all partcles. Step 9: Update the velocty and poston of each partcle usng Eqn (22) and (23). Step 10: Go to Step 4 untl maxmum number of teratons are completed. Step 11: The obectve functon of (G best ) partcle s the optmzed (maxmum) value of ATC. Coordnates of (G best ) partcle gve optmal settng and locaton of respectvely and also calculate cost of nstallaton of usng (21). 7 SIMULATION RESULTS AND OBSERATIONS Ths secton present the detals of the smulaton study carred out on a modfed IEEE 30- system for ATC computaton under normal operatng condton and lne outage condton used for proposed approach. The test system conssts of sx generators and forty one lnes as shown n Fg 3. The system data are n a per-unt system and taken from [21] and the base MA value s assumed to be 100 MA. Generators at es 8, 11 and 13 are consdered n area 1, whle the remanng generators at es 1, 2 and 5 are consdered n area 2. The te-lnes exstng between the two areas are shown n Fg 4. Transacton s carred out between Area 1 and Area 2. Three nequalty constrants such as voltage lmt, lne thermal lmt, and reactve power generaton lmt are consdered. The voltage magntude lmt of each s assumed to be wthn 0.95 pu and 1.05 pu. The optmal locaton and sze of devces are obtaned usng PSO Algorthm for maxmzng ATC based on the selected blateral, multlateral and area wse transactons. Installaton cost of devces has also been calculated for each transacton wth reference to ATC value and cost of nstallaton. The smulatons studes were carred out on Intel Pentum Dual Core, 2.40 GHz system n a MATLAB 2010a envronment. Case 1: Normal operatng condtons The test system results for dfferent blateral and multlateral transactons under normal operatng condtons usng proposed approach are gven n Table 1 and Table 2. In blateral transactons, fve transactons between a seller n source area and buyer n snk area such as (11-27, 2-10, 5-20, 2-23 and 8-30) wth the obectve functon (11) to maxmze the ATC wthout and wth devce usng PSO algorthm have been consdered. Table 1 shows the test results of blateral transactons for fve transactons. Consder a blateral transacton from 11 to 27, the ATC value s MW wthout nstallng, whereas after nstallng the ATC value s ncreased to MW wthout volatng system constrants. The actve power losses s 9.2 MW wthout placng, but t s reduced to 8.1 MW after placng and optmal locaton of s between 6 to 28. The optmal sze (reactance) of s p.u and negatve sgn ndcates that operates n capactve mode and the correspondng cost of nstallaton of devces s 6.33 x10 6 US $. From Table 1 and Fg 5, t can be clear that ATC values are ncreased for all possble blateral transacton and actve power losses are reduced after placng devces n rght locaton. In multlateral transactons between a seller es n source area and buyer es n snk area such as (8, 13-27, 20 and 2, 8, 13-23, 27) wth maxmze the ATC s consdered. A multlateral transacton from es 8, 13 to es 27, 20 consder. In ths case the ATC value s MW wthout nstallng, whereas after nstallng the ATC value s ncreased to MW wthout volatng system constrants and optmal locaton of s connected between 22 to 24 and sze (reactance) of s p.u and the correspondng cost of nstallaton of devces s 8.56 x10 6 US $ as shown n Table 2. The PSO convergence curve for ths transacton utlzng devces s shown n Fg 6. From Table 2, t s obvous that s maxmzng ATC and reducng power loss wth mnmum nstallaton cost there by mprovng lne condton for multlateral transactons also. Case 2: contngency operatng condtons For a contngency case, the branch lne outage between es 9 and 10 s consdered. In ths case, the lne connected between 9 and 10 s removed. The correspondng ATC values for the test system wthout and wth for blateral and multlateral transactons wth lne outage condtons are gven n Tables 3 and 4 respectvely. The actve power loss for dfferent blateral transactons wthout and wth devces s shown n Fg 7. From the Tables 3 and 4 and Fg 7 ndcate that optmally placed devces by PSO sgnfcantly ncrease ATC, reduced actve power losses under contngency condtons. ISSN : ol. 25 No. 1, Jun

8 PSO based Congeston Management n Deregulated Power Systems usng Optmal Allocaton of Fg 3. Sngle lne dagram of the modfed IEEE 30- system Fg 4. Te-lne between areas- IEEE 30 s Table 1. ATC enhancement results for blateral transactons under normal operatng condtons Transactons ATC n MW Placement of devces Settngs (X tcsc ) n p.u Installaton Cost ( x10 6 US $) No Source Snk wthout Wth From To T T T T T Table 2. ATC enhancement results for multlateral transactons under normal operatng condtons Transactons ATC n MW Placement of devces Settngs (X tcsc ) n p.u Installaton Cost ( x10 6 US $) No Source Snk wthout Wth From To T1 8, 13 27, T2 2, 8, 13 23, Table 3. ATC enhancement results for blateral transactons wth lne outage of 9-10 ISSN : ol. 25 No. 1, Jun

9 K. etrvel, S. Ganapathy, and K. Uma Rao Transactons ATC n MW Placement of devces Settngs (X tcsc ) n p.u Installaton Cost ( x10 6 US $) No Sourc e Snk wthout Wth From To T T T T T Table 4. ATC enhancement results for multlateral transactons wth lne outage of 9-10 Transactons ATC n MW Placement of devces Settngs (X tcsc ) n p.u Installaton Cost ( x10 6 US $) No Source Snk wthout Wth From To T1 8, 13 27, T2 2, 8, 13 23, Actve Power loss n MW T1 T2 T3 T4 T5 Transactons wthout wth Fg 5. Actve Power Loss (blateral Transacton) for IEEE 30 system Base case ATC n MW Number of teratons Fg 6. PSO convergence curve for multlateral transactons wth under normal operatng condton ISSN : ol. 25 No. 1, Jun

10 PSO based Congeston Management n Deregulated Power Systems usng Optmal Allocaton of Actve Power Loss n MW wthout wth 0 T1 T2 T3 T4 T5 Transactons Fg 7. Actve Power Loss (blateral Transacton) for IEEE 30 system wth lne outage of CONCLUSION To facltate the deregulated electrcty market operaton, control and tradng, suffcent transmsson lne capablty should be provded to satsfy ncreasng demand of power transactons relably. PSO based algorthm has been s used to fnd optmal placement and settng of devce for maxmzng ATC and mnmzng the actve power losses of the compettve electrcty market whch conssts of blateral and multlateral transactons. The smulaton results ndcate that optmally placed by PSO could sgnfcantly ncrease ATC and reduced power losses under normal and contngency condtons. Moreover, PSO exhbts rot convergence characterstc so t could be used effectvely to select optmal locaton of for enhancement of ATC, thereby mprovng transmsson servces of the compettve electrcty market. ACKNOWLEDGEMENT The authors wsh to thank the authortes of Annamala Unversty, Annamalanagar, Tamlnadu, Inda for the facltes provded to prepare ths paper. REFERENCES [1] R. D. Chrste, B. F. Wollenberg, and I. Wangensteen, Transmsson Management n the Deregulated Envronment, IEEE Proceedngs, ol. 88, No. 2, pp , [2] P. Kumar and S.C.Srvastava, Congeston Management n deregulated market- A case study on an Indan Power System, NPSC, Bangalore, Inda, pp , [3] R.S. Fang and A.K. Davd, Optmal Dspatch under transmsson contracts, IEEE Transactons on power system, ol.4, No.2, pp , [4] Y. Xao, Y.H. Song, C. Lu, and Y.Z. Sun, Avalable transfer capablty enhancement usng FACTS devces, IEEE Transactons on Power Systems, ol. 18, No.1, pp , [5] Prathba, B.Moses, D.Devara and. K. Pand, Evolutonary Algorthm for Estmaton of Avalable Transfer Capablty n Deregulated Envronment, Mddle-East Journal of Scentfc Research, ol.23, No.9, pp , 2015 [6] G.C.Eebe, J.G. Waght, J.G. Frame, Wang and W.F. Tnney, Avalable Transfer Capablty Calculatons, IEEE transactons on Power systems, ol.13, No. 4, pp , [7] G.Hamoud, Assessment of avalable transfer capablty of transmsson systems, IEEE Transactons on Power Systems, ol.15, No.1, pp , [8] B..Mankandan, S.Charles Raa, P.enkatesh and P.S.Kannan, Avalable Transfer Capablty Determnaton n the Restructured Electrcty Market, Electrc. Power Components and Systems, ol.36, No. 9, pp , [9] A.Kumar, S.C. Srvastava and S.N. Sngh, Avalable Transfer Capablty (ATC) Determnaton n a Compettve Electrcty Market Usng AC Dstrbuton Factors, Electrc Power Components and Systems, ol.32, No.9, pp , [10] P.R. Sharma, A. Kumar and N. Kumar, Optmal locaton for shunt connected FACTS devces n a seres compensated long transmsson lne, Turksh Journal of Electrcal Engneerng & Computer Scences, ol. 15, No. 3, pp , ISSN : ol. 25 No. 1, Jun

11 K. etrvel, S. Ganapathy, and K. Uma Rao [11] R.A. Hooshmand and M. Ezatabad, Correctve acton plannng consderng FACTS allocaton and optmal load sheddng usng bacteral foragng orented by partcle swarm optmzaton algorthm, Turksh Journal of Electrcal Engneerng & Computer Scences, ol. 18, No.4, pp , [12] N. Acharya and N. Mthulananthan, Locatng seres FACTS devces for congeston management n deregulated electrcty markets, Electrc Power Systems Research, ol. 77, No.3-4, pp , [13] K.S.erma, S.N.Sngh, H.O.Gupta, FACTS devce locaton for enhancement of total transfer capablty, IEEE power wnter meetng, Jan , Oho, USA. [14] S.Gerbex, R.Cherkaou and A.J. Germond, Optmal Locaton of FACTS Devces to Enhance Power System Securty, IEEE Bologna Power Tech Conference, June 23-26, 2003, Bologna, Italy. [15] P. enkatesh, R. Gnanadass and N. P. Padhy, Comparson and applcaton of Evolutonary programmng technques to combned economc emsson dspatch wth lne flow constrants, IEEE Transactons on Power systems, ol.18, No.2, pp , [16] Ippolto, Luco, L.Cortgla, A. Petrocell and Mchele, Optmal allocaton of FACTS devces by usng mult-obectve optmal power flow and genetc algorthms, Internatonal Journal of Emergng Electrc Power Systems, ol.7, No.2, pp. 1-19, [17] J.Kennedy and R.Eberhart, Partcle Swarm Optmzaton, Proceedng. IEEE nternatonal conference Neural Networks, ol.4, pp ,, Nov/Dec 1995, Perth, WA [18] Yuhu Sh and Russell C. Eberhart, Emprcal Study of Partcle Swarm Optmzaton, Proceedngs of the Internatonal Congress on Evolutonary Computaton, ol.3, pp , [19] H. Hashemzadeh and S. H. Hossen, "Locatng Seres FACTS Devces Usng Lne Outage Senstvty Factors and Partcle Swarm Optmzaton for Congeston Management", IEEE PES General Meetng, July, 2009, Calgary, Alberta, Canada. [20] A.A.John and T.Joseph, Optmal Allocaton of FACTS Devces usng Partcle Swarm Optmzaton for Congeston Relef, Proceedngs of Internatonal Conference on Materals for the Future - Innovatve Materals, Processes, Products and Applcatons, pp , 2013 kerala, Inda. [21] L.J. Ca, I. Erlch and G. Stamtss, Optmal choce and allocaton of FACTS devces n deregulated electrcty market usng genetc algorthms, IEEE PES Power system conference and Exposton, pp.10-13, 2004, USA. ISSN : ol. 25 No. 1, Jun