Transmission Congestion Management in Electricity Market Restructured and Increases the Social Welfare on the System IEEE 14-Bus

Transcription

1 Australan Journal of Basc and Appled Scences, 5(): , ISSN Transmsson Congeston Management n Electrcty Market Restructured and Increas the Socal Welfare on the System IEEE 4-Bus Masoud Mohammad Rahm Fard and Aref Jall Iran Department of Electrcal Engneerng, Ardabl Branch, Islamc Azad Unversty, Ardabl, Iran. Abstract: Restructurng of power systems, to create competton and better u of resources and reduce costs, ncrea profts, reduce loss and the methods of modern ndustry was establed. Several studes n order to create a healthy compettve market, to earn the hghest profts and lowest prces offered. FACTS devces usng one of the most commonly ud equpment to reduce producton costs and reduce the power densty lne systems s consdered. In ths artcle locaton optmal UPFC on the system bad on IEEE 4 bus PSO algorthm for congeston management market power transmsson lnes s ud. Key words: locaton optmzaton; FACTS devces; congeston management; UPFC; PSO. INTRODUCTION Prvatzaton n the electrcty market, leadng to fundamental changes n performance and power systems are organzed. Electrcty markets and the types of transactons s one of them. The markets are large and complex systems ntegraton engneerng that requres a certan level of performance up to that focus. One of the most mportant ssues restructured systems, network densty s transferred. In dealng wth ths problem varous methods from basc rearchers and tho nvolved gve power ndustry. And nclude that the contract path method and crtcal factors, and cost transfer prcng method, method of usng FACTS devces, etc. could be mentoned. The purpo of the procedures under the propod congeston management are the frst nformaton to create mechansms for attractng partcpaton subscrbers and prevent the occurrence of network congeston or densty, and then resolve potental network denstes usng the methods are far and transparent. Restructurng of power systems n many countres followng ncome had an mportant and has led to the u of common equpment to be more FACTS day. Snce the power transmsson network lmtatons esntally controlled flow removable or network s reduced, so u of the elements densty for the management ems very uful. The applcaton ssues FACTS, a place for the true and accurate to mze profts FACTS expensve equpment and management s optmum densty lnes. The most mportant studes done n the feld of FACTS devces for the management, densty, UPFC can be ud n dfferent ways to manage the densty. The methods bad on market and generally are ud n Europe. Bad method, dvded nto geographc market and the u of FACTS elements between areas dvded accordng to specfc methods of densty management. The methods nclude usng mutual exchange, are market and aucton market n the border area between the two s den. But the most mportant ssues n the feld placement FACTS devces are optmzed. Placement for problem solvng many optmzaton methods FACTS devces such as lnear and nonlnear programmng, neural networks, fuzzy logc, steel platng and ntellgent algorthms are ud. Intellgent algorthms such as genetc algorthms (Saguan, M. et al., 4; Reddy, K. et al., 6; Nabav, S.M.H. et al., 6) and PSO algorthm due to the nherent characterstcs of natural and smple to understand, be understood, usng knowledge of human and lvng, etc. them to make approprate problem solvng dscusson. Algorthm PSO, for the frst tme n 995 by Eberhart and Kennedy as an optmzaton method was ntroduced. Ths algorthm s a socal arch algorthm that models the socal behavor of brds has clubs and governng n order to dscover patterns and sudden change n flght path of brds at the same tme they change and ape optmzaton s appled to categores. The work of PSO s bad on the prncple that every moment of each partcle n your locaton arch space the best place so far has been the best of hs neghborhood where there s total, s t. In reference (Chen, Z.. et al., 5) of the PSO method for solvng OPF market for ared congeston management s ud. The man purpo of ths reference for managng congeston at least cost and demand on actve buyng power wth system relablty s prerved. In reference (Meena, T., K. Selv, 5) wth the prence of TCSC, PSO algorthm for optmzed playback on wth the am to mnmze the congeston charge has been ud. In ths reference, TCSC n lne performance and hgh-densty nstallaton of the two models (wthout the data-arng market and blateral market-arng wth mutual agreement) s analyzed. The results ow that prented method can effectvely change n densty wth the mnmum producton cost and mnmum densty, mprove. The results also ndcate that n some cas due to TCSC lne nstallaton cost, cost, densty s ncread. Correspondng Author: Masoud Mohammad Rahm Fard, Department of Electrcal Engneerng, Ardabl Branch, Islamc Azad Unversty, Ardabl, Iran. E-mal: masoud7mrf@gmal.com 744

2 Aust. J. Basc & Appl. Sc., 5(): , Effectve n congeston management, optmal placement FACTS devces n the network s the power of reference (Gtzadeh, M., M. Kalantar, 8) placement optmal TCSC and SC n a IEEE 4 bus standard system evaluated the results ows the lne densty to elmnate condensaton and voltage profle s mproved. Also n reference (Barat, H. et al., 6) the optmal placement UPFC usng genetc algorthm evaluated the results of UPFC placement owed reduced producton cost, cost, densty, mproved densty lnes and Profles s the voltage. The reference genetc algorthm placement UPFC for congeston management market s ared u. In reference (Kazem, A., R. Sharf, 6) the optmal placement TCPS usng crtera bad on real power n the market for the management of ared and mutual densty s taken nto consderaton. In ths reference system as the standard IEEE 3 bus system ud and studed by MATLAB7. Programmng was conducted and the mum socal welfare for lectng the best locaton of TCPS s consdered. In reference (Saravanan, M. et al., 6) one of the applcatons of PSO to dentfy the best locaton of FACTS devces to nstall mnmum cost of FACTS devces has been pad. The temperature lmt reference voltage lnes and lmt bus s consdered. In reference 3 type FACTS devces the SC, TCSC and UPFC ud on the system and IEEE 3 bus, IEEE 8 bus and IEEE 6 bus tested s. The results ow that UPFC capacty s on the system but the hgh nstallaton costs. TCSC nstalled n the mnmum cost recovery system load capacty s better. In reference (Shaheen, H.I. et al., 8) the best locaton of UPFC wth the optmzaton parameters s analyzed. The reference PSO and genetc algorthms for achevng ths goal has been ud. Smulaton system for the IEEE 6 bus and IEEE 4 bus s ud. The results ow that both algorthms for solvng the problem of hgh qualty, specfcatons and performance ntegraton constant calculaton, well have. The results also ow that the PSO technque at the begnnng of GA optmzaton s faster, but wth ncread producton, genetc algorthm, the PSO s better. In terms of computng tme of the PSO technque s faster than GA. The fact that, UPFC the most mportant and most wdely ud FACTS devces consdered goes; ths artcle placement optmal UPFC usng PSO algorthm on the system IEEE 4 bus (Zmmermann, R.D., D. Gan, 5) Transmsson congeston management and Increa socal welfare for power systems restructured s ud. Algorthm Method Brds Communty: Brds n the communty poston algorthm to record the moton of partcles and ther neghbors wll change. Each partcle s a poston that we ow wth x (t) Shows that ths poston P st partcle s the tme t. In ths algorthm each partcle addton to the speed of a locaton s requred: v ( t) x ( t) x ( t ) () Algorthm communty wth a group of brds random answers to start, then answer the problem by optmzaton problem wth space to make deals to arch generatons. Each partcle as multdmensonal wth two values that x d and v respectvely Referrals locaton and speed to the stuaton after d d st of my are partcle s defned. At each stage of the populaton moves every partcle wth two values s the best to date. The frst value, the best answer so far s n terms of competence for each partcle parately obtaned and ths value s called p _ best. Another best value by the algorthm s obtaned socal brds; the best value so far by all partcles n the populaton has obtaned the best overall value and g _ best s called. After p _ best and g _ best by the amount and speed of each partcle accordng to ts new locaton n relatonp to that: v ( t ) wv. ( t) c. rand( p_ best x ) c. rand( g _ best x ) d d d d d d () x d ( t ) xd ( t) vd ( t ) (3) Top w n the relatonp, weght, c and c factors, learnng, and rand a random number n the range (,) s. To prevent algorthm dvergence, the fnal value of each partcle velocty s lmted. v d v, v ] (4) [ mn W, c and c of the algorthm parameters are socal brds. And convergence s dependent value for ths parameter. Here, the procedure s usually equal to c and c and the number.5 s assumed. Convergence strongly depends on the amount of w s better dynamc and must be defned. Thus, the lnear trend n the evoluton of populaton.4 to.9 decreas. Intally ths number must be large as possble by good answers to 745

3 Aust. J. Basc & Appl. Sc., 5(): , provde early and late stages Small w caus a better convergence. Ths reducton n the related form can be defned. w w w w ter mn ter (5) Ths algorthm s the correspondng topology on. In ths algorthm each partcle movement relyng on ts experence and knowledge of all other partcles s performed. Therefore, t s clear that the ntegrty of the communty s many and complete communcaton s establed between the partcles. The algorthm steps are:. Basc populaton s formed randomly.. Determne the competence of partcles wth ther prent poston. 3. Comparson of current competence and the best experences of partcles and ther necessary replacement. 4. Comparson of current competence of each partcle's best prevous result of all partcles. 5. Set the speed vector for each partcle usng the relatonp. 6. Move wth ther new postons related to partcle Return to step and repeat the algorthm to reach convergence. The Propod Method Problem Solvng: FACTS devces nclude nstallaton has two stages. Frst locaton n the network equpment ould be specfed and then control the devce parameters for the desred goal ould be optmal. Method frst propod standard system IEEE 4 bus (Fgure ) to test on t and are the load chon regardless of UPFC on the values and passng power lnes we calculated. Then the effect of UPFC n all dfferent places, UPFC bad on the desred nstvty Replacement and dong calculatons, densty lnes wth and wthout UPFC gan. Calculatons performed wth UPFC Replacement n many dfferent places and algorthm PSO (programmng algorthm n MATLAB) contnued and fnally the most optmal locaton of UPFC nstallaton, we manage specfc densty. In ths paper, the obectve functon, remove the lne congeston and ncread socal welfare and s defned as follows: C p N g. Pg C. P g g (6) Where: C P P P (7) p. g g g And N g Q C. Q Cq. g (8) Where: C Q Q Q q. g g g And (9) TCC Nb Nb D G P P () Relatons 6 and 7, the actve power producton cost functons, Relatons 8 and 9, the reactve power producton cost functons, and about, s related to the total Charge of congeston. Therefore, the obectve functon, the total producton cost can be actve, and reactve and total charge of congeston, would be at least That to acheve mum socal welfare wll Whch s defned as the : (Chanana, S., A. Kumar, 6). SW Mn C ( P ) C ( Q ) TCC () p g q g 746

4 Aust. J. Basc & Appl. Sc., 5(): , In addton to the above relatonps, lmtatons, and the voltage Shn Terms and power producton of the Actve are defned accordng to the relatonp. mn mn and P P P () g g g Terms and lmt the control parameters of UPFC are consdered as follows: mn mn, (3) mn mn, Fg. : Sample system ud (IEEE 4 bus). UPFC Model Ud n the Algorthm Implementaton: The UPFC power necton model own n Fgure has been wdely ud n rearch and studes on the mpact of UPFC system. Therefore ths model for smulaton and mplementaton of the algorthm s ud. Fg. : UPFC power necton model. Relatons ud n the model accordng to Fgure are: P Q to / t sn (4) cos / t (5) sn sn / t P (6) to cos cos / t Q (7) 747

5 Aust. J. Basc & Appl. Sc., 5(): , Any populaton structure n the mplementaton of the PSO algorthm s own n Fgure 3. Bad on the length of a populaton that s 6 Locaton UPFC placement reprent and also ow parameters UOFC. bus Fg. 3: Structure of each populaton. bus mn LMP n the calculaton to ths pont s that the bus that the generator s nstalled f the cost s P PP mn generators operatng n the propod t of LMP but when P P or P P s LMP s determned by the system. Now consderng the above relatons, dstrbuton of samples on the system and results bad on the power lnes crossng between then obtaned by consderng the densty values usng the algorthm lnes PSO, placement optmal UPFC and optmzed for playback on obtanng the best locaton of UPFC n terms of densty delete lnes contnued. Results and Dscussons on the System Studed: Method prented n two cas has been revewed: In the frst ca: the system wthout UPFC and the lne densty sn t lmted. Second ca: In ths mode densty lmt for 8MW lnes there. The UPFC s added to the system. Cost functon coeffcents and propod sale of energy s own n Table. alue of each generator power producton s gven n Table s noted as beng expensve becau of any purcha of generator 6 has been done. LMP values n Table 3 for all bus s gven n Table 4 and the UPFC and lnear parameters whch must be expod s own n Table 5 of the algorthm mplemented general nformaton own s. Table : Cost functon coeffcents and propod energy sales. Bus No Bd ($/MWh) P 33 4 P mn Table : The rate of producton power generators. Bus No. The actve powers generators n ca The actve powers generators n ca Table 3: values of local prce lmt. Bus No. LMP n ca LMP n ca Table 4: parameters UPFC. UPFC nstalled Lne

6 Aust. J. Basc & Appl. Sc., 5(): , Table 5: General Informaton results from algorthm mplementaton. General Informaton Ca Ca Total loss (MW) Total generaton (MW) Total charge congeston ($/h) 57 3 Total cost of actve power ($/h) Total cost of reactve power ($/h) Power flow n lne - (MW) The obectve functon (WS) ($/h) Accordng to Table 5 n the lne densty value of - MW rate densty has reached resolve ths lne ndcates. Can have more lnes are passng normal. Total cost congeston charge mode, 57 Dollars per hour now and n the ca of, 3 Dollars per hour s obtaned on the densty of spendng reducton reprents mode s the total amount of producton cost n ca, 88. Dollars per hour s the amount specfed, but the UPFC mode s ud to value dollars per hour s reduced and The total cost of producton of reactve power n ca, the tme was n 93 dollars, the cond mode, the tme s 3 dollars, whch reprents a decrea whle the amount of total loss rate of 5 MW n ca s reduced. The obectve functon value of about 99 Dollars per hour n the frst ca, the latter s n 885 Dollars per hour, whch reduces the value of the obectve functon, ows an ncrea n socal welfare. Meanwhle accordng to Table 3, LMP n Mode to Mode s reduced and unform that ths also ows an ncrea n socal welfare.the best locaton of UPFC lne -4 has been determned. Concluson: One of the key ssues n the power transmsson network s the congeston of lnes. Restructurng of the systems that open access to transmsson networks s esntal n them, Ths ssue s of double mportance. Becau of the economc explotaton of exstng facltes, and mum u of the nstalled capacty of power transmsson networks, mprove congeston n den lnes, the problem s unavodable. The results prented n the way, that n the best poston to properly Lnear Mode FACTS devces for congeston management provde. Bad on results for the two cas revewed, we can conclude that the u of FACTS devces, especally the u of UPFC and PSO algorthm a good way to reduce charge congeston, producton costs and congeston management and ncrea socal welfare n the electrcty market. REFERENCES Barat, H., M. Ehsan, M. Fotuh-Fruzabad, 6. "Locaton of Unfed Power Flow Controller and ts Parameters ttng for Congeston Management n Pool Market Model usng Genetc Algorthm", Power Electroncs, Drves and Energy Systems, Internatonal Conference on, pp: -7. Chanana, S., A. Kumar, 6. "Effect of Optmally Located FACTS Devces on Actve and Reactve Power Prce n Deregulated Electrcty Markets" Power Inda Conference, pp: Chen, Z.., L.Z. Zhang, J. Shu, 5. "Congeston Management Bad on Partcle Swarm Optmzaton", Power Engneerng Conference, The 7th Internatonal, : 9-3. Gtzadeh, M., M. Kalantar, 8. "A New Approach for Congeston Management va Optmal Locaton of FACTS Devces n Deregulated Power Systems", Electrc Utlty Deregulaton and Restructurng and Power Technologes, Thrd Internatonal Conference on, pp: Kazem, A., R. Sharf, 6. "Optmal Locaton of Thyrstor Controlled Pha Shfter n Restructured Power Systems by Congeston Management", Industral Technology, 6. ICIT 6. IEEE Internatonal Conference on, pp: Meena, T., K. Selv, 5. "Cluster Bad Congeston Management n Deregulated Electrcty Market usng PSO", IEEE Indcon Conference, Chenna, Inda. Nabav, S.M.H., S. Jadd, M.A.S. Masoum, A. Kazem, 6. "Congeston Management n Nodal Prcng wth Genetc Algorthm", Power Electroncs, Drves and Energy Systems, 6. PEDES '6. Internatonal Conference, pp: -5. Reddy, K., S. Reddy, N. Prasad Padhy, R.N. Patel, 6. "Congeston Management n Deregulated Power System usng FACTS Devces", Power Inda Conference, IEEE, pp: 8. Saguan, M., S. Plumel, P. Dessante, J.M. Glachant, P. Bastard, 4. "Genetc Algorthm Assocated to Game Theory n Congeston Management", 8th Internatonal Conference on Probablstc Methods Appled to Power Systems, Iowa State Unversty, Ames, Iowa. Saravanan, M., S. Mary Raa Slochanal, 6. Applcaton of partcle swarm optmzaton technque for optmal locaton of FACTS devces consderng cost of nstallaton and system loadablty, Elver B.., 77:

7 Aust. J. Basc & Appl. Sc., 5(): , Shaheen, H.I., G.I. Raed, S.J. Cheng, 8. Optmal locaton and parameters ttng of UPFC bad on GA and PSO for enhancng power system curty under sngle contngences, IEEE Power and Energy Socety General Meetng-Converson and Delvery of Electrcal Energy n the st Century, pp: -8. Zmmermann, R.D., D. Gan, 5. Matpower a LAB power system smulaton package, Ur s Manual, erson