Case Study of a Mult-Infeed HVDC System Paulo Fscher, Jupng Pan, Senor Member, IEEE, Kalash Srvastava, Senor Member, IEEE, Wlbur Wang, and Chao Hong Abstract There are serous concerns relatng to a Mult- Infeed HVDC system when feedng a weak AC network. Typcal ssues concernng mult-nfeed confguratons are: need for coordnaton of the recovery control, need for dfferent DC modulaton strateges to stablze the system, possblty of voltage nstablty of the area recevng large amount of power from multple HVDC lnks and the rsk of mutual commutaton falures. In contrast, f the area recevng electrcal power from multple HVDC transmsson lnks s relatvely strong due to the presence of large amount of generaton unts nearby there are stll some questons that need to be nvestgated such as the ssues underlnng the operaton of such a mult-nfeed system, the proper desgn of the controls of the HVDC systems and the system dynamc performance under extreme contngences. Ths paper nvestgates nto an example of such a mult-nfeed HVDC system. The authors have performed small sgnal analyss of the system to assess nstablty assocated wth the control modes. Electromechancal and voltage stablty analyss were performed for harmful contngences. Dynamc performance analyss was also carred out to analyze the nteracton amongst varous HVDC nverters durng dsturbances. Index Terms HVDC transmsson, mult-nfeed, stablty, dynamc performance, smulaton, control H I. INTRODUCTION VDC systems have tradtonally been operated n relatve electrcal solaton from each other. Wth growng demand for power transmsson over long dstances, the plannng and system operaton have begun to consder the effects of one dc system on the operaton of another dc system through a common AC system where ac mpedance between the two dc systems s small a so called mult-nfeed HVDC system []. Mult-nfeed converters ether share a common ac bus or connected to buses that are electrcally close. The operaton of adjacent dc termnals poses serous concerns for mutual nteracton followng a dsturbance n ether dc systems or n the common ac system. Common mode nteracton through Paulo Fscher de Toledo s wth ABB AB HVDC, SE-77 80 Ludvka and wth the Dvson of Electrcal Machnes and Power Electroncs, Royal Insttute of Technology, Stockholm, S-00 44, Sweden (e-mal: paulo.fscher@se.abb.com or paulo.fscher@ee.kth.se) Kalash Srvastava s wth ABB Corporate Research, 7278 Vasteras, Sweden (e-mal: kalash.srvastava@se.abb.com). Jupng Pan s wth ABB Corporate Research, Ralegh NC 27606 USA (emal: jupng.pan@us.abb.com). Wlbur Wang s wth ABB Chna HVDC, Bejng Chna (emal: wlburweguo.wang@cn.abb.com) Chao Hong s wth Chna Southern Power Grd Co. Ltd., Guangzhou Chna. (emal: hongchao@csg.cn) voltage dstorton, phase mbalance, and ampltude and phase changes has the potental to degrade the performance of each dc system. An example would be the recovery of both dc systems from an ac fault. The nteracton may prolong the combned recovery to pre-fault dc power levels compared wth recovery of one dc system operatng alone [2]. Some studes have found ntmate nteractons between the varous consttuent systems and control coordnaton among the consttuent HVDC lnks were partcular mportant. The man potental problems arsng from Mult-Infeed HVDC systems nclude: Small sgnal nstablty due to control nteractons among the consttuents HVDC lnks Voltage nstablty and collapse Increased commutaton falures n one consttuent HVDC lnk due to ac faults occurrng n the vcnty of a neghborng one Transent ac voltage depresson due to smultaneous recovery of consttuent HVDC lnks after ac faults Ths paper nvestgates nto an example of such a multnfeed HVDC system. The studed system s one of the planned confguratons of the Chna Southern Grd (CSG) system whch wll have fve HVDC transmsson lnks by 200 feedng 5,800 MW of electrcal power nto the load center n Guangdong provnce. The authors have performed small sgnal analyss of the system to assess nstablty assocated wth the control modes. Electromechancal and voltage stablty analyss were performed for harmful contngences. Dynamc performance analyss was also carred out to analyze the nteracton amongst varous HVDC nverters connected to a common AC network durng varous dsturbances. The purpose of the study s to nvestgate f the system under peak load condton would wthstand crtcal contngences and to dentfy the possble problems and countermeasures. II. STUDY SYSTEM OVERVIEW A. Overvew of CSG System The studed mult-nfeed HVDC system s one of the planned confguratons for the CSG system as shown n whch ncludes fve HVDC lnks operatng n a common AC network. These fve HVDC lnks are Gu-Guang I (GUG), Gu-Guang II (GUG2), Tan-Guang (TSQ), 3Gorges-Guang (3GG) and Yunnan-Guang (YUG). The CSG system ncludes several provnces. A large amount of electrcal power s transmtted from west areas to a heavly loaded area n Guangdong provnce by parallel HVDC lnks and 500kV AC transmsson lnes. The dstances between the east and west
南丰电站 毛阳河电站 牛路岭电站 清澜电厂 areas are n the range of 000~500km. Mddle area s the transmsson corrdors. Fgure below shows a geographc dagram of CSG system for the studed confguraton, emphaszng the 5 HVDC transmsson lnks and the 500kV AC transmsson corrdors. These HVDC transmsson lnks wll supply 5,800MW of electrcal power nto the load area n Guangdong. In ths confguraton, the four double crcut AC transmsson lnes wll delvery 6,700MW from west areas to the load area. ( 十二五 ) 云 Yunnan TSQ = 800 MW GUG & 2 = 3000 MW 3GG = 3000 MW YunGuang = 5000 MW ( 十二五 ) YunGuang ( 十二五 ) 南 GUG 2 TSQ 贵 Guzhou GUG 州 South Chna Power Network Plannng for Year 200 200 年南方 500kV 电网规划地理接线图 广 Guangx 西 Hanan 广 Thermal P/P Hydro P/P Pump storage P/P Nuclear P/P Converter staton 500kV S/S 500kV Swtch Stn 500kV SC HVDC lne 500kV AC lne 400kV lne 3GG Guangdong Fgure Geographc dagram of CSG network B. Mult-Infeed Short Crcut Ratos All of the HVDC nverter statons are sttng n a rather strong system. Table presents the calculaton results of Short Crcut Rato (SCR) and Effectve Short Crcut Rato (ESCR) for each converter staton. Table also shows the calculated Mult-Infeed Short Crcut Rato (MSCR) and Mult-Infeed Effectve Short Crcut Rato (MESCR) where the mpact of close connected converters s taken nto consderaton [3]. The results are also graphcally presented n Fgure 2. The calculaton methods are gven n Appendx. Inverter Bus HVDC : 3GG HVDC 2: TSQ HVDC3: GUG I HVDC 4: GUG II HVdc 5: YUG Voltage Ratng Power Ratng Indvdual SCR Mult- Infeed MSCR Indvdual ESCR 东 图例 Mult- Infeed MESCR ± 500 kv 3000 MW 5.5 2.97 4.83 2.49 ± 500 kv 800 MW 9.33 3.99 8.64 3.4 ± 500 kv 3000 MW 7.45 3.98 6.83 3.44 ± 500 kv 3000 MW 9.9 5.86 9.4 5.26 ± 800 kv 5000 MW 7.22 3.84 6.48 3.3 Table Short crcut ratos for the studed confguraton The calculaton shows an SCR value greater than 5 for all nverter statons. Comparatvely, the 3GG nverter staton has the lowest short crcut rato, whle the nverter statons for the TSQ and GUG II have an SCR value above 9. When consderng the nfluence of local reactve power compensaton, the correspondng ESCR for these nverter statons are reduced by 5 to 0 percent respectvely. 2.00 0.00 8.00 6.00 4.00 2.00 0.00 Indvdual SCR Mult Infeed MSCR Indvdual ESCR Mult Infeed MESCR 3000 MW 800 MW 3000 MW 3000 MW 5000 MW ± 500 kv ± 500 kv ± 500 kv ± 500 kv ± 800 kv HVDC : 3GG HVDC 2: TSQ HVDC3: GUG I HVDC 4: GUG II HVdc 5: YUG Fgure 2 Short crcut ratos for the studed confguraton The calculaton shows an SCR value greater than 5 for all nverter statons. Comparatvely, the 3GG nverter staton has the lowest short crcut rato, whle the nverter statons for the TSQ and GUG II have an SCR value above 9. When consderng the nfluence of local reactve power compensaton, the correspondng ESCR for these nverter statons are reduced by 5 to 0 percent respectvely. By comparng the SCR values wth the correspondng MSCR for each of the nverter statons, t can be concluded that all HVDC lnks are electrcally close connected. For example, the 3GG nverter has an SCR=5.5 but the MSCR s 2.97 wth the nfluence from the other nverter statons. The reducton for the TSQ nverter staton s even more dramatc : from SCR=9.33 to MSCR=3.99. One reason s that the nomnal ratng of TSQ s lower as compared to the other converter statons. TSQ s rated 800 MW whle the others are 3000 MW, wth excepton of YUG (5000 MW). The nverter staton of the GUG II s located n the strongest part of the network and also s least affected by the nverters from the other HVDC lnks (the short crcut rato s reduced from SCR=9.90 to MSCR=5.86 when consderng the mpact from the other lnks). Hence, t can be expected that the dynamc performance for ths lnk would not be much nfluenced by the other lnks durng system dsturbances. Smlar analyss and conclusons can be drawn by observng the ESCR and ts correspondng MESCR calculated for each of the converter statons. C. Partcpaton Factor Analyss Wth the Partcpaton Index t s possble to evaluate the ntensty of nteracton between HVDC converter statons. Results of ths calculaton are presented n Table 2. 3GG TSQ GUG I GUG II YUG 3GG 0.207 0.0255 0.0397 0.0237 0.094 TSQ 0.0425 0.55 0.0465 0.089 0.07 GUG I 0.0397 0.0296 0.468 0.073 0.0634 GUG II 0.0237 0.03 0.073 0.064 0.037 YUG 0.0657 0.0252 0.038 0.0222 0.54 Table 2 Partcpaton Index (absolute values)
From the results the followng can be observed: YUG nteracts most wth the other converter statons. Ths s because the nomnal ratng of YUG converter staton s hgher than any other converter staton and YUG converter staton s sttng electrcally close to all the other converter staton, except to GUG II. YUG nteracts more wth 3GG as compared wth the other converter statons. Ths means the dsturbances related to YUG, lke commutaton falures, wll have more domnant effect on the operaton of 3GG as compared to the other converter statons. YUG also receves nfluence from 3GG. The Partcpaton Index receved from 3GG s sgnfcantly hgher as compared wth the others. GUG II s the converter staton that receves the least nfluence from the other converter statons. Snce the nomnal ratng of TSQ s much lower than other converters and hgh short crcut capacty, ts own Partcpaton Index s only 0.55 pu. Because of ths, the contrbuton from the relatvely hgh Partcpaton Indexes, produces the dramatc reducton of short crcut rato, from SCR=9.33 to MSCR=3.99. It can also be noted that, for the TSQ converter staton, the sum of dfferent Partcpaton Indexes of remote converters exceeds the partcpaton Index of ts own converter. III. QUASI-STATIC ANALYSIS OF THE SYSTEM Quas Statc Modal Analyss methodology has been used to study the Voltage Stablty and Power Stablty condtons of the CSG system. Weak elements or areas n the electrcal system can be dentfed by usng the Q-V modal analyss. It s based on a calculaton of the egenvalues, partcpaton factors and V-Q senstvtes. Maxmum Power Curve also ndcates the stablty margn for each ndvdual HVDC lnk. A. Maxmum Power Curves The study of Maxmum Power Curve (MPC curve) s a statc approach for the stablty analyss of an HVDC transmsson system, whch was ntroduced n the earler 980 s and used for a sngle nfeed topology of an HVDC converter, where the power senstvty to small changes n the dc current was studed and power stablty lmts verfed [4]. Ths methodology has been extended to a multple nfeed topology, where dfferent HVDC lnks are connected to dfferent buses [3]. The Maxmum Power Curve and V-I Curve of dfferent HVDC lnks were calculated for the studed CSG system assumng that the dc current n one HVDC lnk was ncreased whle dc currents n the other four HVDC lnks were unchanged. It was also assumed that all HVDC lnks were operatng n Constant Power Control wth a slow response tme. A conservatve assumpton n ths study s that all loads n the system have a constant load characterstc. Ths means that the load remans unchanged when the voltage s changed at the connecton pont. A sample of the results from the calculaton s presented n Fgure 3and Fgure 4. Power at Inverter Termnal(Pd) (p.u.).4.35.3.25.2.5. tsq gug2 gug.05 yug 3gg.05..5.2.25.3.35.4.45.5 Drect Current(Id) (p.u.) Fgure 3 Maxmum power curves for fve HVDC lnks Inverter Termnal AC Voltage(Uac) (p.u.).0 0.99 0.98 0.97 0.96 0.95 0.94 tsq 0.93 gug2 gug 0.92 yug 3gg 0.9.05..5.2.25.3.35.4.45.5 Drect Current(Id) (p.u.) Fgure 4 V-I characterstcs for fve HVDC lnks As can be observed from the curves, the 3GG lnk s operatng very close to the Maxmum Avalable Power (MAP) condton. For ths lnk the crtcal MESCR s 2.49. The low stablty condtons observed n the results can be explaned as follows. The system s qute senstve to reactve power balance n the load area, n partcular close to the 3GG nverter. As shown n Fgure 5, the angle dfference between some areas n the west (generaton area) and areas n the east, close to the load center, exceeds 70 electrcal degrees. Ths ndcates that the AC lnes connectng these areas are heavly loaded. Conservatve assumpton regardng load representaton has been consdered (constant load). A dfferent representaton of the load, havng voltage dependence, would mprove the results. All HVDC transmsson systems are operatng assumng a slow Power Flow Controller. A fast Power Controller would also mprove the stablty condton.
transmsson lnes feedng ths area, are mportant for mantanng system stablty. C. P-V Curves P-V curves of fve AC buses close to the HVDC nverters n the load area were drawn as shown n Fgure 7..3.25 Fgure 5 Power angles n crtcal areas of CSG system B. V-Q Senstvty Analyss From V-Q senstvty analyss t s possble to evaluate the voltage stablty related nformaton from a system-wde perspectve, provdng nformaton regardng the mechansm of nstablty. Ths method can be used to evaluate the securty of the transmsson grd by quantfyng the stablty margns and power transfer lmts. It s also possble to dentfy weak ponts n the system and areas of voltage nstablty, as well as dentfcaton of possble remedal measures to be taken to mprove the system stablty n those dentfed weak ponts. From ths senstvty analyss t has been dentfed that the busses located n the nterface area where the four man transmsson corrdors and some HVDC lnks termnate, are the most senstve busses. Fgure 6 shows the dentfed crtcal area for the CSG system. Weak Area Fgure 6 Identfed crtcal area n CSG system It was also dentfed that most of the long transmsson lnes, belongng to the four man transmsson corrdors and transferrng a large quantty of power from the western generaton area to the eastern load area, have hgh partcpaton relatve to those busses that have hgh V-Q senstvty. The V-Q senstvty of these busses, whch s related to the voltage stablty margn condtons, s sgnfcantly degraded f some of these branches are trpped. Another observaton from ths analyss was that most of the generators located n a load area, or are connected to man AC Voltage at nverter termnals (Vac) (p.u.).2.5..05 Vac-TSQ Vac-GUG2 Vac-GUG 0.95 Vac-YUG Vac-3GG 0.9 0.8 0.85 0.9 0.95.05 Actve Power n load area (P) (p.u.) Fgure 7 P-V curves for the fve AC nverter busses P-V curves relate bus voltage to load wthn the regon, whch gve an approxmate ndcaton of voltage collapse due to excess of load level n a gven area. From the fgure t s possble to observe that the operaton pont of the AC system s close to the nose part of the curves, whch ndcates that the AC system s heavly loaded and ts operatng pont s close to the system stablty lmt. D. Applyng Remedal Measures It has been verfed that the nterface area s the crtcal area n the CSG system. In order to mprove the system performance, a remedal measure would be to add reactve power support devces on some of the busses n that area. Ths means that these devces would support the most crtcal area of the system. Two mportant consderatons should be ponted out. Frst, t s advsable that those devces should have fast response to changes n the system. Ths means that an SVC or STATCOM or fast swtchng shunt devces lke capactor banks and shunt reactors should be used. Second, consderng that ths nterface area s a rather bg area, t s convenent to consder several reactve power support devces located on dfferent busses. IV. FUNDAMENTAL FREQUENCY STABILITY STUDY An ntal nvestgaton of the system regardng the fundamental frequency transent stablty condtons of the studed confguraton for the CSG system was performed wthout consderng the effect of any control actons. The smulaton was performed usng SIMPOW software package. The studes revealed that some of the contngences resulted n nstablty. The harmful contngences were renvestgated consderng varous control actons. A number of control measures have been nvestgated to manage harmful contngences. The general problem wth the network s the mpared power transfer capablty to the man load area n the event of a contngency. Amongst the solutons
dscussed, the best opton was found to be a combnaton of seres and shunt compensaton. Seres compensaton of 30% on three AC transmsson corrdors connectng the generaton area n the west to the man load area n the east and shunt compensaton comprsng of 3x00 swtched shunt capactors n parallel wth 00 MVAR SVC at three locatons n the so called nterface area n the load area was felt adequate n managng the harmful contngences. Smulaton results for sngle pole trp of YUG lnk wthout and wth proper control measures are shown n Fgure 8 and Fgure 9. To manage a bpole trp of YUG, t s necessary to shed some generaton. However, f the amount of seres compensaton s ncreased or seres compensaton s ntroduced on more lnes, t mght be possble to manage a bpole trp of the HVDC transmsson lnk wthout any generaton sheddng. Ether the amount of seres compensaton on the exstng lnes should be ncreased or seres compensaton added to more lnes. The results were qute encouragng and all harmful contngences wth a faulton perod of 00 ms were managed wth ths arrangement. The cases that have been studed are based on the lst of contngences that ncludes: Sngle pole and bpole trp of converters Three phase fault on AC lnes nsde the load area close to nverter bus and subsequent trp of sngle/double crcut lnes after 00 ms. Three phase fault on AC lne feeders to the load area close to nverter bus and subsequent trp of sngle/double crcuts AC lne feeders after 00 ms. Three phase fault at some of the generator termnals and subsequent trp of the generator after 00 ms. Three phase fault at rectfer termnal wth a fault clearng tme of 00 ms. Fgure 8 Sngle pole tro of YUG lnk, base case Fgure 9 Sngle pole trp of YUG lnk, wth proper control measures Power frequency control was employed on all the lnks and t was used n combnaton wth other control measures also n the studes. The Power Frequency Controller s a hgher level controller to the HVDC transmsson lnk whch bascally adds a contrbuton to the set power order based on measured frequency devaton seen at the converter bus. It ncludes a proportonal gan (whch s typcally n the order of 50 000 MW/pu of frequency devaton) and a lead/lag flter to provde a dervatve acton to the controller. A Fast Power Flow Controller (response tme of 50 ms) has shown to be benefcal to the system as compared wth a typcal mplementaton of the controller whch s characterzed by a slow response tme (response tme of second). V. DYNAMIC PERFORMANCE STUDY The man objectve of the dynamc performance study (DPS) s to study the nteracton between the HVDC nverters connected n the Guangdong area. The focus s on the commutaton falure senstvty of dfferent nverters and the recovery performance of dfferent lnks after major dsturbances under normal and contngency system operatng condtons. The purpose of the study s to verfy f, due to these phenomena of concern, t mght be needed to develop coordnated recovery control strategy to acheve acceptable performance of multple HVDC lnks. The smulaton setup was prepared based on PSCAD verson 4... An equvalent of the Chna Southern Grd system was prepared where the major 500 kv buses n the load area n Guangdong are retaned and the major 500 kv ac connectons between the west and east parts of the CSG network are also retaned. The resultant equvalent network conssts of 35 nodes, 274 branches and 22 aggregated dynamc generators. The system steady state and dynamc performance has been verfed n comparson wth the results from the full network smulaton. Further nvestgaton has shown that t s farly accurate to only model 9 aggregated generators as dynamc machnes n the developed CSG equvalent. Another nvestgaton that s presently n progress s to verfy that smlar study results and conclusons could be obtaned by usng a more smplfed equvalent, consderable
smaller than the one that has been bult. Havng obtaned such experence, ths would smplfy future work f a more complex system havng ncreased number of HVDC transmsson lnks would be studed. The overall results of ths DPS study ndcate that satsfactory performance of HVDC transmsson systems can be expected for the analyzed confguraton. The nteracton between the fve HVDC transmsson lnks s mnor manly because of the strength of the AC network. The man fndngs of ths DPS study can be summarzed as follows: Commutaton falure senstvty analyss ndcates that, except for the YUG lnk, commutaton falure of one HVDC lnk caused by mpedance fault or nduced by removal of frng pulse does not spread to the healthy lnks. However, commutaton falure of the YUG lnk may ncur spread commutaton falures of the healthy lnks. For severe AC faults n Guangdong area, the fve HVDC transmsson lnks all suffer from commutaton falure. However, the overall recovery performance of multple HVDC lnks s satsfactory. Fgure 0 shows one case study whch apples a 3 phase 00ms fault close to the YUG nverter staton and trp of the correspondng 500kV AC lne. In the few problematc cases, acceptable performance could be acheved wth proper control parameter settngs for dfferent HVDC lnks. There exsts a rsk of prolonged recovery nvolvng 3GG and YUG lnks f the pump storage power plant n Guangdong area not n servce whch results n reduced dynamc reactve power support to 3GG and YUG nverters. Fgure 0 YUG nverter responses to a close-n 3ph 00ms fault and trp of correspondng 500kV lne VI. CONCLUSIONS Ths paper presented a case study concernng the operaton performance of mult-nfeed HVDC system. The techncal approaches used n ths study nclude power flow analyss, quas-statc modal analyss, transent stablty smulaton analyss and dynamc performance studes. In summary, satsfactory operaton performance can be expected for the studed mult-nfeed HVDC transmsson system. The dentfed potental problems due to extreme contngences can be effectvely mtgated by nstallng adequate seres and shunt reactve power compensaton devces at the crtcal AC transmsson lnes and nterface area. Improved system performance can be acheved by mplementng advanced controls such as fast power flow controllers. The techncal approaches presented n ths paper may be used as a general gudelne of future mult-nfeed HVDC transmsson system studes. However, t should be mentoned that the results obtaned n ths study are consdered as ndcatve of the expected system performance snce addtonal network expanson projects have been proposed. APPENDIX Two basc relatons, Short Crcut Rato (SCR) and Effectve Short Crcut Rato (ESCR) are defned usng the tradtonal formulas as follows: SCR and ESCR SCC = () Pdc SCC Qf = (2) Pdc where, SCC, QF and Pdc are the short crcut MVA avalable on the nverter bus, the bus flter and capactor MVA and rated power of the th lnk, respectvely. These expressons can be generalzed by a complex form, by usng the mpedances Z L or Z E, whch are the Thevenn mpedance and effectve Thevnn mpedance, respectvely, of the converter bus expressed n per unt of the rated power of the HVDC converter staton. Ths wll take the phase angle mpedance nto account. Then, they are defned as SCR = (3) Z L and SCR = (4) Z E However, these two dfferent ways to defne SCR and ESCR gve approxmately smlar ampltude values. Extendng these defntons to Mult-Infeed system, Mult- Infeed Short Crcut Rato (MSCR) and Mult-Infeed Effectve Short Crcut Rato (MESCR) consderng the mpact of the other converters ncluded n the system are gven by MSCR = k (5) Pdc z m=, m where, s the converter bus under consderaton k corresponds the number of HVDC termnal statons m vares from converter # up to converter #k z,m s the th, m th element n the Z BUS matrx n p.u.
If the matrx Z BUS ncludes the shunt compensaton element needed by the converter then the formula express the Mult- Infeed Effectve Short Crcut rato (MESCR). Partcpaton Indexes can be defned as, P = Pdc z, (6) and P m = Pdcm z, m m (7) where P, s the partcpaton ndex of the own converter P,m s the partcpaton ndex of the remote converter that s electrcally connected to the converter bus under consderaton. VII. REFERENCES [] H.P. Lps, "Aspects of Multple Infeed of HVDC Inverter Statons nto a Common AC System," IEEE Transactons on Power Apparatus Systems, vol. 92, pp. 775-779, Mar/Apr. 973. [2] M. Szechtman, et al, "The Behavour of Several HVDC Lnks Termnatng n the Same Area," CIGRE General Sesson, Pars, France, Paper 4-20, 992. [3] Paulo Fscher de Toledo, et al, Multple Infeed Short Crcut Rato aspects related to multple HVDC nto one AC Network; paper presented n Dalan 2005 Conference, Chna [4] IEEE/CIGRE, Gude for Plannng DC Lnks Termnatng at AC Locatons havng Low Short Crcut Capactes, IEEE Standard Department, 993 VIII. BIOGRAPHIES Paulo Fscher de Toledo was born n São Paulo, Brazl. In 978 he receved the M.Sc degree from Mauá Insttute of Technology n São Paulo and n 2007 the PhD degree from Royal Insttute of Technology, Stockholm, all n Electrcal Engneerng. He has been workng n the feld of HVDC (Hgh Voltage Drect Current) snce many years back, frst for Promon Engenhara and then for ASEA/ABB AB. Snce 996 he has been workng wth the system development and system desgn group n Ludvka, Sweden for ABB. He has a vast experence n both classcal-hvdc and VSC-HVDC projects. Jupng Pan (M 97, SM 04) receved hs B.S. and M.S. n Electrc Power Engneerng from Shandong Unversty, Chna and hs Ph.D. n Electrcal Engneerng from Vrgna Tech, USA. He s currently a prncpal consultng R&D engneer wth ABB Corporate Research n USA. Hs expertse ncludes power system modelng and analyss, HVDC transmsson, transmsson plannng studes, energy market modelng and smulaton studes. Kalash Srvastava (SM 996) was born at Fatehpur n the Inda, on October 3, 962. He graduated n Electrcal Engneerng from MMM Engneerng College Gorakhpur (Inda) n 983 He dd hs MTech and PhD n Power Systems from IIT Kanpur (Inda) n 986 and 995 respectvely. He worked n Inda and Italy for dfferent companes before jonng ABB Corporate Research Sweden where he has been workng for past years. Weguo Wang was born at Fujan Provnce, Chna, n 969. He receved hs B.S and M.S. n Industral Automaton from Lanzhou Unversty of Technology and Electrc Power Engneerng from Northeast Chna Power Unversty, and hs Ph.D. n Power Systems from North Chna Electrc Power Unversty (Bejng). He s now workng n HVDC Group ABB Chna. Before that he has been workng n ABB Corporate Research Chna for more than 5 years snce 2003. Hong Chao receved hs B.Sc.(Eng.) and M.Sc.(Eng.) degrees n Electrcal Engneerng from Wuhan Unversty of Hydraulc & Electrc Engneerng n 987 and 990 respectvely, and hs Ph.D. degree n Electrcal Engneerng from the Unversty of Hong Kong n 2000. He s currently a senor engneer wth CSG Technology Research Center. Hs research nterest s n power system smulaton, HVDC and ts controls.