Energy Management of A Smart Railway Station Considering Regenerative Braking and Stochastic Behaviour of ESS and PV Generation

Transcription

1 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 1 Energy Managemen of A Smar Railway Saion Considering Regeneraive Braking and Sochasic Behaviour of ESS and PV Generaion İbrahim Şengör, Suden Member, IEEE, Hasan Can Kılıçkıran, Suden Member, IEEE, Hüseyin Akdemir, Suden Member, IEEE, Bedri Kekezoğlu, Ozan Erdinç, Senior Member, IEEE, and João P S Caalão, Senior Member, IEEE Absrac The smar grid paradigm has provided grea opporuniies o decrease energy consumpion and elecriciy bills of end-users Among a wide variey of end-users, elecrical railway sysems (ERSs) wih huge insalled power should be considered as a vial opion in order o avoid wased energy provided ha an energy managemen sysem is uilized In his sudy, a mixedineger linear programming (MILP) model of a railway saion energy managemen (RSEM) sysem is formulaed by a sochasic approach, aiming o uilize he emerged regeneraive braking energy (RBE) during he braking mode in order o supply saion loads Furhermore, he proposed RSEM model is composed of an energy sorage sysem (ESS), RBE uilizaion, phoovolaic (PV) generaion unis, and an exernal grid in his paper The passengers impac on RBE as well as he sochasic behaviour of he iniial sae-of-energy (SOE) of ESS along wih uncerainy of PV generaion by he RSEM model are also evaluaed The model is esed under a bunch of case sudies formed considering several combinaions of he cases ha an ESS or PV are available or no and using RBE is possible or no Index Terms Energy sorage sysems, mixed ineger linear programming, railway energy managemen sysem, regeneraive braking energy, sochasic programming NOMENCLATURE The main nomenclaure used in his paper is expressed below Oher symbols and abbreviaions are defined where hey firs appear Abbreviaions ERS ESS Elecrical railway sysem Energy sorage sysem This work was suppored by FEDER funds hrough COMPETE 22 and by Poruguese funds hrough FCT, under Projecs SAICT- PAC/4/215 - POCI FEDER-16434, POCI FEDER- 6961, UID/EEA/514/213, UID/CEC/521/213, UID/EMS/151/213 and SFRH/BPD/13744/214 Also, he research leading o hese resuls has received funding from he EU Sevenh Framework Programme FP7/ under gran agreemen no 3948 (projec SiNGULAR) I Sengor, HC Kilickiran, H Akdemir, B Kekezoglu, and O Erdinc are wih Yildiz Technical Universiy, Isanbul 3422, Turkey ( isengor@yildizedur; hckiran@yildizedur, hakdemir@yildizedur, bkekez@yildizedur, and oerdinc@yildizedur) O Erdinc is also wih INESC-ID, Insiuo Superior Técnico, Universiy of Lisbon, Lisbon 149-1, Porugal ( ozanerdinc@gmailcom) João P S Caalão is wih INESC TEC and he Faculy of Engineering, Universiy of Poro, Poro , Porugal, also wih he C-MAST, Universiy of Beira Inerior, Covilhã 621-1, Porugal, and also wih INESC- ID, Insiuo Superior Técnico, Universiy of Lisbon, Lisbon 149-1, Porugal ( caalao@ubip) MILP PV RSEM RB RBE SOE Indices and Ses s w Parameers CE ESS CR ESS DE ESS DR ESS N 1 N 2 P load P P V,s P RBE SOE ESS,ini w SOE ESS,min SOE ESS,max T λ buy λ sell π s π w Variables P ESS,ch Mixed-ineger linear programming Phoovolaic Railway saion energy managemen Regeneraive braking Regeneraive braking energy Sae-of-energy Scenario ses of iniial PV generaion Period of he day index in ime unis [min] Scenario ses of iniial SOE of ESS Charging efficiency of he ESS Charging rae of he ESS [kw per min] Discharging efficiency of he ESS Discharging rae of he ESS [kw per min] Maximum power ha can be drawn from he grid [kw] Maximum power ha can be sold back o he grid [kw] Railway saion power demand during period [kw] Power generaed by PV during period for scenario s [kw] Power obained from braking energy of rain during period [kw] Iniial SOE of he ESS for scenario w [kwh] Minimum SOE limi of he ESS [kwh] Maximum SOE limi of he ESS [kwh] Number of ime inervals in one hour Price of energy bough from he grid [e/kwh] Price of energy sold o he grid [e/kwh] Probabiliy value of relaed scenario for PV generaion Probabiliy value of relaed scenario for iniial SOE of ESS ESS charging power during period for scenarios s and w [kw] (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

2 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 2 P ESS,disch P grid P ESS,us P RBE,us P sell SOE ESS u ESS u grid ESS discharging power during period for scenarios s and w [kw] Power supplied from he grid during period for scenarios s and w [kw] Power used from ESS during period for scenarios s and w [kw] Power used from RBE during period for scenarios s and w [kw] Power sold o he grid during period for scenarios s and w [kw] SOE of he ESS during period for scenarios s and w [kwh] Binary variable: 1 if during charging period for scenarios s and w, else Binary variable: 1 if during charging period for scenarios s and w, else I INTRODUCTION A Moivaion and Background Pracical evidence suggess ha he energy efficiency is among he mos crucial facors for decreasing carbon emissions [1] Moreover, depleion of fossil fuel reserves and progressively increasing elecriciy demand have suppored he rising concerns abou efficien use of energy [2] Therefore, policy makers pay paricular aenion o consumers wih high level of energy demand due o he vas poenial of energy recovery lying behind [3] Elecrical railway sysems (ERSs) are one of he examples for large-scale consumers ha are regarded as effecive resources boh in order o achieve carbon emission arges and o decrease consumpion of energy I has been saed ha, he amoun of greenhouse gas emissions (GHG) of ransporaion is presened as 14% of oal emission by 21, which is prediced o be doubled by 25 in [4] Ye anoher repor revealed ha 234% of he world s carbon emissions originaed from ransporaion in 213 Also, emission rae from rail ransporaion accouns for 35% of oal ransporaion [5] Regeneraive braking (RB) is a he hear of our undersanding of energy recovery depending on he huge amoun of energy consumpion and generaion paerns in ERSs RB is roughly defined as he process of ransforming braking energy of rain during he deceleraion ino elecrical energy using racion moors and effecively use of his regeneraed energy [6] More han one opion is available for using he regeneraed energy such as giving back o caenary line o supply energy for oher rains, or soring in an energy sorage sysem (ESS) in order o uilize or sell in anoher period in he fuure, or direcly injecing o he grid via a reversible subsaion [7] I is possible ha he oal energy consumpion of ERSs can be reduced beween 1% and 45% by means of RB sysems [8] The smar grid concep presens more reliable, efficien, safe, and modernized power sysems and offers a new perspecive o he energy managemen philosophy of ERSs Besides, using advanced compuaional feaures and bidirecional communicaion equipmen, he smar grid paradigm provides a chance o communicae beween demand side and he uiliy [9] Furhermore, soring and selling energy become more possible for consumers via proliferaed ESSs hanks o he smar grid concep [1] Anoher beneficial aspec of smar grid concep is o inegrae renewable energy sources such as phoovolaic (PV) and wind As a consequence, aforemenioned circumsances arise he smar energy managemen approach which can be implemened in railway ransporaion applicaions by designing a railway saion energy managemen (RSEM) concep [11], [12] B Lieraure Overview The opic of increasing energy efficiency based on soring regeneraive braking energy (RBE) in ERSs has drawn he aenion of various researchers around he world Ciceralli e al [13] presened an energy managemen sraegy for wayside ESS o ake advanage of maximum RBE during braking periods The model considered he acual volage and curren value of ESS, and power sysem losses by forecasing of rain moion parameers such as ineria forces and acceleraion Noneheless, i was saed ha RBE was used for acceleraion of oher rain a he saion and changes in passengers number were no noiced while he rain was operaed in braking and mooring mode, addiionally ha paper negleced he unknown iniial sae-of-energy (SOE) of ESS Khayyam e al [14] proposed a railway energy managemen sysem archiecure considering he smar grid vision Train loads, on-board and wayside ESS as well as disribued generaion unis were considered joinly for dynamic opimal energy uilizaion using he presened managemen scheme However, i should be noed ha sored energy was no used for he saion loads and passengers effec on RBE was no considered, also uncerainy of iniial SOE of ESS was ignored in ha sudy A hierarchical energy managemen sraegy ha has an ESS and a microgrid was suggesed for unidirecionally supplied power o railway saion in [15] Furhermore, disinc scenarios were developed as wheher hey include microgrid or no, in order o evaluae he proposed managemen sysem from he perspecive of he economic benefis Moreover, i was claimed ha he conrol problem of energy consumpion level saed in [16], [17] was solved On he oher hand, line opology such as curves and slopes as well as availabiliy of uncerain behaviour PV generaion and sochasic characerisic of iniial SOE of ESS were no considered in [15] [17] Lu e al [18] suggesed a power managemen sraegy o enhance he energy saving of a diesel muliple-uni rain by using dynamic programming and nonlinear programming framework According o he presened resuls in ha paper, fuel consumpion cos was reduced by 7% Moreover, his managemen sraegy ook ino accoun line opology and changes in passengers number affecing energy recovery However, he opion of leveraging from RBE in order o mee he saion demand was ignored Furhermore, he uncerainy of iniial SOE of ESS was negleced in he menioned paper Nasr e al [19] and Pankovis e al [2] invesigaed he benefis of using RBE so as o decrease he wased energy in ERSs I was aimed ha he sored energy from RBE was reused specifically for he saion loads such as elevaors, escalaors, lighing ec, no for railway applicaions in [19] The wind and (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

3 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 3 PV based renewables, RBE, and ESS were aken ino accoun in order o enhance he energy efficiency in [2] However, boh [19] and [2] ignored he impac of passengers number, line opology, las bu no leas variable iniial SOE of ESS, while only he [2] evaluaed he PV, bu, wihou sochasiciy of renewables Hernandez and Suil [21] proposed a DC microgrid including renewables based on PV and RBE ogeher wih ESS o charge elecric vehicles nex o he rain saion In his paper, he sraegy of power managemen, converer conrol and he impac of size of he ESS componens saed as a challenging problem by he prior sudies were sored ou so as o maximize he usage of renewables Neverheless, renewables were only used o charge elecric vehicles, no for supplying saion loads and passengers changes were no aken ino accoun in [21] I should be noed ha in his ype of sudies a reliable infrasrucure is required o ensure ha power exchange can be carried ou smoohly For his reason, he elecrical proecion has a criical role The relaed requiremens abou inerconnecion was well-examined in [22], [23], however, his opic is no considered in his paper for he sake of he clariy Aguado e al [24] developed a mehodology for opimal operaion of ERSs o evaluae he poenial of renewable energy resources ogeher wih RBE and maximize savings in he operaional coss Even hough i was declared ha he uncerainies relaed o renewables were considered hrough a sochasic approach, he uncerainy due o iniial SOE of ESS and passengers number were no invesigaed in [24] These sudies ogeher wih many oher sudies no referred here considered he opic from differen poins of view in order o enhance he efficien use of energy based on smar grid vision wihin ERSs C Conribuions In his sudy, a Mixed Ineger Linear Programming (MILP) model of RSEM concep covering ESS, RBE, PV, and differen pricing schemes in order o evaluae he operaion of a railway saion is propounded Regarding he iniial SOE of ESS and PV generaion as uncerain parameers, operaional assessmen of he railway saion is carried ou using sochasic programming approach Keeping in mind he valuable conribuions made by prior sudies, his paper inends o make he conribuions saed below: The effecs of uncerain iniial SOE of ESS and PV generaion as well as differen pricing schemes on RSEM are evaluaed considering several case sudies RBE obained from rains is uilized in order o parially mee inernal demand of railway saion The impac of he number of passengers varying wih inensiy during he day on calculaed RBE is considered D Organizaion The res of he paper is organized as follows: Secion II provides he necessary background informaion for he operaion of railway vehicle and presens he mahemaical Regeneraive Braking Energy PV Generaion Energy Sorage Sysem RSEM Elecrical Grid Fig 1 The block diagram of RSEM Smar Meer Power Flow Communicaion Flow Saion Loads formulaion of energy managemen model Hereafer, Secion III describes he evaluaed case sudies and relaed resuls Secion IV finalize he paper wih concluding remarks and makes suggesions abou possible fuure sudies II METHODOLOGY The block diagram of RSEM sraegy is demonsraed in Fig 1 The RSEM sysem manages he operaion of a smar railway saion in a subway line aking ino accoun RBE, pricing signal received from he uiliy, and ESS The erm used as smar railway saion in his paper implies he saion infrasrucure is able for bidirecional power and informaion flow which is compaible wih well-known smar grid concep In RSEM sysem, only he inernal demand of a saion is considered and energy consumpion of he rain is assumed o be supplied via racion ransformers The res of his secion gives informaion abou he mahemaical model of rain moion and he proposed energy managemen model A Mahemaical Model of Train Moion In order o deermine he poenial of RBE, he mahemaical model of rain moion is used This subsecion presens he model of rain moion The rain moion is based on Newon s one dimensional moion laws and direcly affeced by no only he line opology bu also he characerisics of racion devices: n F i = m a (1) i=1 In (1), F i represens he oal forces ha affec he rain moion, m is he roaing rain mass, and a is he acceleraion of he rain The forces affecing he rain moion are divided (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

4 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 4 Fag ϕ Fig 2 The forces acing on rain moion V ino wo main caegories as F r and F ag, and are illusraed in Fig 2 F r : Force generaed by racion moors I is considered as posiive in racion mode, while negaive in braking mode F ag : Toal forces ha play negaive role agains rain moion I consiss of line gradien, line curve, and resisance caused by own rain moion F r F ag = m a (2) Equaion (1) can be rearranged by subsiuion of oal forces ha ac on rain moion wih F r and F ag (2) F ag = F r + F gr + F c (3) The oal forces ha have a negaive effec on rain moion are obained by sum of F r, F gr, and F c which symbolize resisance caused by own rain moion, gradien of line, and curve of line, respecively, in (3) Herein, F r is modelled by using well-known Davis formula [25] P = (m a + F ag ) v η g η m η i 3, 6 + P a (4) I = P V l (5) In (4), P indicaes he insananeous power of rain and i is assumed as posiive while rain acceleraes Conversely, when rain brakes, P is considered as negaive due o he generaed power in racion moors on rain Also, P a represens auxiliary loads of rain while η g, η m, and η i are efficiency of gear, racion moors, and inverers respecively Lasly, v symbolizes rain speed a ime I should also be noed ha I represens he insananeous curren and V l indicaes he line volage in (5) B Energy Managemen Model In his paper, minimizing he oal daily cos of railway saion elecriciy consumpion is deermined as he objecive of he energy managemen model The objecive funcion is composed of he probabiliy value of relaed PV generaion and iniial SOE of ESS scenarios (π s ) and (π w ), power bough from he grid (P) grid and power sold o he grid (P) sell which are considered as variables during he period for scenarios s and w In addiion, he ime dependen pricing signals (λ buy ) Fr and (λ sell ) are used in energy managemen model ( P grid min π w π s s w λ buy P sell λ sell T The main focus of his sudy is merely minimizing he daily operaional cos of he railway saion Therefore, he oher possible coss apar from he operaional coss such as he invesmen coss of he necessary communicaion sysem infrasrucure, invesmen of ESS, along wih he wear and ear cos of ESS or any oher sysem componens are no aken ino accoun Moreover, decision opions for ime granulariy T is no limied as considered in his paper and can be exended according o preferences of he relaed designer of he model, such as 1h, 3 min, 15 min, ec 1) Power Balance: The mos crucial equaion ha forms he basis of he model is given in (7) This equaion saes ha grid (P), grid PV (P,s P V ) and ESS (P ESS,us ) can be used ogeher in a combined form or independenly o supply inernal power demand of railway saion loads (P load ) and ESS (P ESS,ch,s ) or o sell available power o he grid (P) sell I is worhy o noe ha inheren consrains due o he naure of power exchange of grid and ESS will be explained in furher subsecions [26] P grid + P,s P V + P ESS,us = P load + P ESS,ch + P, sell, s, w (7) 2) ESS Modelling: In order o evaluae he uncerain characerisic of he iniial SOE of ESS along wih he PV generaion, he ESS model in [26] is revised and made compaible for sochasic programming approach Equaion (8) enforces ha he discharging efficiency of ESS (DE ESS ) affecs he amoun of available power (P ESS,us ) o supply he inernal loads of he railway saion, which is obained from discharging power of ESS (P ESS,disch ) Furhermore, in inequaliies (9) and (1), a binary variable is used o model he physical naure of ESS depending on he fac ha an ESS canno be charged and discharged a he same ime Toal charging power of he ESS composed of he effecive usable power from RBE (P RBE,us ) and he power aken from grid in order o charge he ESS (P ESS,ch ), is limied by ESS charging rae (CR ESS ) in (9) due o modelling purpose of he limied charging naure of he ESS Correspondingly, in (1), discharging rae of ESS (DR ESS ) draws an upper limi for he uilizable power provided from ESS (P ESS,disch ) Equaion (11) demonsraes he mahemaical relaionship beween he remaining SOE of ESS from he previous ime inerval (SOE 1,s,w), ESS charging energy supplied from RB and/or elecrical grid, and discharging energy used for meeing inernal railway saion loads in order o obain he SOE of ESS (SOE) ESS for every ime inernal As far as he SOE value of ESS a he beginning is concerned, he iniial SOE of ESS (SOEs,w ESS,ini ) is assigned as SOE of ESS using (12) I is worhy o noe ha he iniial SOE of ESS is assumed o change in a sochasic manner according o he relaed scenarios Las bu no leas, SOE of ESS is resriced in he range of allowed maximum (SOE ESS,max ) and minimum ) (6) (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

5 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 5 (SOE ESS,min ) values by using he consrains (13) and (14) P RBE,us P ESS,us + P ESS,ch P ESS,disch SOE ESS = P ESS,disch DE ESS,, s, w (8) CR ESS u ESS,, s, w (9) DR ESS (1 u ESS ),, s, w (1) = SOE 1,s,w ESS + CE ESS ( P ESS,ch + P RBE,us T P ESS,disch T, 1 s, w (11) SOE ESS = SOEw ESS,ini, if = 1 s, w (12) SOE ESS SOE ESS,max,, s, w (13) SOE ESS SOE ESS,min,, s, w (14) 3) RBE Modelling: The oal amoun of available RB power ha can be used for charging of he ESS is indicaed as P RBE Alhough he main arge is o use his energy as much as possible, some amoun of P RBE can ineviably be wased due o he he maximum allowed charging capaciy of ESS Regarding his variable naure of he uilized energy from RB over he ime, a variable called P RBE,us is defined in order o model he energy used for charging purposes of ESS Equaion (15) helps o avoid his variable o ake higher values han RBE obained from rain braking P RBE P RBE,us, s, w (15) 4) Power Exchange Consrains: I is worhy o noe ha power canno be bough from he grid and sold o he grid during he same ime inerval In order o model aforemenioned consrain a binary variable (u grid ) is used As can be seen from (16) and (17), he saion is able o draw power from he grid when u grid is 1, and sell back power o he grid when u grid is P grid N 1 u grid, s, w (16) P sell N 2 (1 u grid ), s, w (17) III TEST AND RESULTS Wih he aim of evaluaing he RBE and ESS effec along wih he sochasic behaviour of he iniial SOE of he ESS and PV generaion on he daily cos of a railway saion, he proposed MILP model is esed in GAMS v2413 sofware wih CPLEX v12 solver [27] I should be noed ha RBE calculaion is performed by modelling he M1A ligh mero line in RAILSIM 8 sofware [28] The M1A ligh mero line wih a lengh of 197 kilomeer and 18 saions, is one of he busies mero line in Isanbul The roue map of M1 ligh mero line can be seen from Fig 3 Concerning o obain closer resuls o real case, he acual daa relaed o racion moor sizes ogeher wih he opological feaures of line such as gradien and curve, which are supplied ) Fig 3 The roue of Isanbul M1 ligh mero line Daily passengers Time of he day Fig 4 The daily passenger profile of M1 ligh mero line Power [kw] Time of he day M1B Fig 5 The daily load demand profile of Bahcelievler railway saion M1A from Mero Isanbul Co, are used I is also considered in his sudy ha passengers number in he evaluaed saion dynamically changes The dynamic number of passengers indirecly affecs he RBE amoun by leading o an aleraion in he oal mass of he rain To obain more accurae resuls, he acual passenger profile of he saion aken from Isanbul Mero Co is considered, which is given in Fig 4 Anoher imporan parameer used in his sudy is he saion load demand Assessmens are carried ou for only Bahcelievler Saion Figure 5 picures he power demand profile of Bahcelievler Saion which is recalculaed based on he acual energy measuremens of he saion The load specrum of saion is assumed as consising of escalaors, elevaors, lighing, heaing, venilaion and air condiioning Moreover, he sampling ime of he recorded daa is reorganized o be 1 minue due o very shor RB ime In he RSEM srucure, i is assumed ha railway saion loads can be supplied by grid, PV or ESS while ESS can be charged by eiher uilizing RBE, PV or grid Figure 6 illusraes RB power profile of he relaed saion I is worhy o underline (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

6 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE s4 s8 5 Power [kw] Time of he day Fig 6 The daily RB power profile for Bahcelievler Saion Irradiion [W/m 2 ] s1 s3 s5 s7 s9 s2 s4 s6 s8 s1 Price [ /kwh],12,1,8,6,4,2 Time of he day Fig 8 The examined scenarios for PV generaion 111 ESS1 ESS2 Fig 7 Dynamic pricing signal Time of he day ha he whole RBE is consumed wih he purpose of meeing inernal saion demand via ESS Also, he opion ha he power supplied by ESS and PV o he grid in a reverse way, is evaluaed in his paper Las bu no leas, RBE is considered as i is indirecly sold o he grid over ESS Considering he evaluaion of daily operaion cos of he saion, i is more likely ha pricing signal plays an imporan role The communicaion infrasrucure beween uiliy and saion required for he RSEM o ake dynamic acions is assumed o be provided owing o a smar meer a he saion In his sudy, hree differen pricing schemes are considered, namely dynamic pricing signal, ime of use, and fixed price The fla price is aken as 84 e/kwh and i is assumed ha i does no change during he whole day Apar from he fla price, energy price in ime of use scheme is considered as 5 e/kwh, 81 e/kwh, and 127 e/kwh for he ime period of 23:-7:, 7:-18:, and 18:-23:, respecively Las bu no leas, Fig 7 illusraes he ime varying pricing signal used in his sudy I is worhy o underline ha he average value of he scheme given in [29] is manipulaed o obain a value approximae o he fixed price signal in order o creae opporuniy for a more realisic comparison I should be noed ha he selling price is assumed as equal o he buying price for every single scheme One of he specificaions of he proposed ESS model is ha ESS has a oal capaciy of 1 kwh Addiionally, i is assumed ha charging and discharging raes of he ESS are limied o 1 kw per minue ogeher wih he charging and Roo Fig 9 Scenario ree composed of PV generaion and iniial SOE of ESS scenarios PV1 PV4 discharging efficiencies of 95 Lasly, i is no allowed for ESS o be discharged below 2 kwh I should be reminded ha he daily operaional cos of saion is regarded as independen from invesmen cos of he ESS Alhough assessmen of iniial SOE of he ESS can be realized assigning an exac value o he iniial SOE of he ESS, i canno be precisely known under some condiions due o he various reasons, which means he problem needs o be sochasically programmed and evaluaed Therefore, in his sudy iniial SOE of he ESS is examined no PV8 PV ESS3 ESS9 ESS1 ESS2 ESS3 ESS9 ESS1 ESS2 ESS3 ESS (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

7 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE s4w1 Grid s4w1 PV s4w1 ESS s4w9 Grid s4w9 PV s4w9 ESS s8w9 Grid s8w9 PV s8w9 ESS Power [kw] : 1:1 1:2 1:3 1:4 1:5 1:6 1:7 1:8 1:9 1:1 1:11 1:12 1:13 1:14 1:15 Time of he day Fig 1 Decomposiion of used power in order o mee he load demand TABLE I COMPARISON OF DIFFERENT CASE STUDIES Descripion of Cases Dynamic Pricing Signal Time of Use Signal Fla Pricing Signal Toal Operaional Cos Cos Reducion Toal Operaional Cos Cos Reducion Toal Operaional Cos Cos Reducion Base Case (None of RBE, ESS, or PV) Base Case Base Case Base Case Case 2 (Considering only ESS) % % % Case 3 (Considering only PV) % % % Case 4 (Considering boh ESS and RBE) % % % Case 5 (Considering boh ESS and PV) % % % Case 6 (Considering all of ESS, RBE, and PV) % % % TABLE II COMPARISON OF CASE STUDIES FOR DIFFERENT PV SIZES Descripion of Cases Dynamic Pricing Signal Time of Use Signal Fla Pricing Signal Toal Operaional Cos Cos Reducion Toal Operaional Cos Cos Reducion Toal Operaional Cos Cos Reducion Base Case (None of RBE, ESS, or PV) Base Case Base Case Base Case Case 2 (Considering all of ESS, RBE, and 1 kw-pv) % % % Case 3 (Considering all of ESS, RBE, and 9 kw-pv) % % % Case 4 (Considering all of ESS, RBE, and 8 kw-pv) % % % Case 5 (Considering all of ESS, RBE, and 7 kw-pv) % % % only deerminisically bu also sochasically for he scenarios of having 2, 3, 4, 5, 6, 7, 8, 9 and 1 kwh iniial SOE of he ESS Furhermore, probabiliy of each scenario of iniial SOE of ESS is chosen as equal In his sudy, i is assumed ha he railway saion is able o be supplied by PV generaion uni Regarding he uncerain behaviour of PV power generaion uni, he problem is modelled as i reflecs he sochasic naure of PV Therefore, 1 differen scenarios are considered so as o properly model he problem in a sochasic manner Included irradiaion and emperaure daa aken from [3] for 1 differen days is used o calculae generaion power profiles which are assumed as 1 differen scenarios for PV generaion I should be noed ha he specificaions of PV panels given in [31] are used while compuing power generaion profiles by using irradiaion and emperaure daa Figure 8 illusraes he evaluaed scenarios for PV generaion Upper of his figure also explains he paern difference beween cloudy and sunny day scenarios using wo (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

8 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 8 Power [kw] s4w1 s4w9 s8w :3 11:31 11:32 11:33 11:34 11:35 11:36 11:37 11:38 11:39 11:4 11:41 11:42 11:43 11:44 11:45 Time of he day Fig 11 Power sold back o he grid in seleced scenarios scenarios as an example, he PV generaion scenarios roughly follow he same paern a he res of he day Assessmens are also carried ou abou effec of he PV generaion uni size on daily operaional cos creaing 4 cases such as PV sizes are 7, 8, 9 and 1 kw Combining 1 scenarios for PV generaion wih 9 scenarios for iniial SOE of ESS, a scenario ree is consruced, which includes 9 differen scenarios Graphical demonsraion of scenario ree is given in Fig 9 For he sake of clariy, he evaluaions of he graphical resuls are invesigaed based on resuls of seleced 3 scenarios while he resuls in Table I and II cover he whole 9 scenarios While selecing he scenarios used in graphical resuls, i was considered ha he resuls provide opporuniy for comparing sunny and cloudy days as well as differen SOE levels The decomposiion of used power so as o supply he saion loads for 3 seleced scenarios is given in Fig 1 for a very shor ime inerval Each column represens he insananeous power drawn boh from he grid and he ESS ogeher wih power used from PV generaion for 3 seleced scenarios, namely scenario s4w1, s4w9, and s8w9 As can be seen in Fig 1, he power drawn from he grid is severely affeced by he iniial SOE of he ESS and PV generaion, even in he lae hours of he day Owing o he wo way power exchange infrasrucure of he smar railway saion, he power sold back o he grid is shown for a shor ime inerval in Fig 11 Sold power is observed as more sable for he sunny day scenario compared o cloudy one I should be underlined ha when uncerain behaviour of iniial SOE of ESS is inroduced, he resuls of ESS scenarios are similar excep for minor irregulariy, which are observed a ESS scenario wih full SOE level I is obvious ha charging and discharging saes of he ESS are direcly affeced by eiher RBE usage, PV generaion or iniial SOE of he ESS, as seen in Fig 12 I is worhy o underline ha 3 seleced scenarios relaed o he iniial SOE of he ESS PV generaion are presened in Fig 12 for he case ha includes RBE and PV generaion wih 1 kw under he dynamic price scheme Neverheless, i can be also seen from he menioned figure ha afer RBE is inroduced abou a 6:, he firs rain arrives o he saion, SOE of he ESS aleraion increases for all scenarios Table I encapsulaes he base case ogeher wih five differen cases assessed in his sudy considering he differen pricing schemes I can be deduced from he able ha uilizaion of only ESS, only PV, or combinaions of he ESS, PV and RBE Energy [kwh] s4w1 s4w9 s8w9 Time of he day Fig 12 The variaions in SOE of ESS during a day for each scenario in case RSEM includes RBE and ESS and is operaed under dynamic price scheme have significan impac on reducing oal daily operaional cos of he smar railway saion for he sochasic approach The case including none of RBE, ESS, or PV is assumed as he base case, while i is he wors case evaluaed considering he fla pricing signal Dynamic pricing and ime-of-use signals are also considered so as o highligh he impac of smar grid applicaions by using smar meering feaures even hough no in mos of acual railway saions I can be seen ha differen pricing schemes provide a grea opporuniy for minimizing oal daily operaional cos of he railway saion Using RSEM, nearly 2-3% drop in cos is ensured for he sochasic approach even if he railway saion is equipped wih only ESS Furhermore, reusing of RBE ogeher wih he ESS and PV is anoher and he mos efficien opion, which provides more han 35% cos reducion for he sochasic approach I is worhy o sae ha he aforemenioned evaluaions are conduced in case of RSEM operaed under dynamic pricing signal The resuls belong o he cases ha is creaed for evaluaing he PV size impac on daily operaional cos of smar railway saion, are given in Table II I can be said ha he increase in PV size resuls in a significan decrease in daily operaional cos of he smar railway saion Similar o he resuls given in Table I, he mos severe decrease in operaional cos is obained under he cases wih dynamic price scheme, which emphasize he imporance of smar grid concep Neverheless, he insallaion cos should be considered while deciding he required PV size, which is assumed as ou of scope for his paper IV CONCLUSION Aiming o reduce dependence on fossil fuels and o relieve he public anxieies on global climae change by decreasing he GHGs, efficien use of energy has become an imporan opic Due o he high reusable energy poenial lying behind he ERSs, railway operaion can be considered as a key facor for reaching he goals in energy efficiency (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

9 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 9 This sudy presened a MILP model of RSEM for evaluaing dynamic variaions of passengers, differen pricing schemes, and sochasic naure of he iniial SOE of he ESS along wih he uncerainies in PV generaion while aiming o minimize daily operaional cos of a railway saion by using is own insrumens such as ESS, PV, RBE and exernal one ie grid I was assumed ha ESS, PV, RBE and a smar meer allowing o operae Bahcelievler railway saion under differen pricing signals, form he RSEM srucure The calculaion of RBE was carried ou using RAILSIM sofware In addiion, niney differen scenarios were evaluaed in order o explore he impac of iniial SOE of he ESS and PV generaion, which were considered as a parameer ha canno be precisely known by RSEM I should be noed ha wo-way power flow beween grid and saion was considered in his paper Therefore, RSEM sysem managed he power flow in saion regarding he opions ha buying from and selling o he uiliy In order o evaluae he impac of RBE, differen pricing schemes and sochasic naure of he ESS along wih he uncerainies in PV generaion on daily cos of railway saion, six differen cases were creaed in his sudy The case ha railway saion has no ESS and PV or is no able o uilize RBE was seleced as base case for all kind of pricing signals The resuls showed ha he reducion in cos of daily elecriciy consumpion of railway saion is possible using he ESS, neverheless, using ESS ogeher wih RBE had a remendous effec on he daily cos and decreased i by nearly 16% in sochasic approach One unanicipaed finding was ha, he cases ha uilizes only he PV, and ESS along wih PV, resuled in same decrease rae as 19% in he cos of daily operaional The mos significan reducion was observed when he saion was able o use all of RBE, ESS, and PV, which was calculaed as 35% I should be emphasized ha due o he sochasic behaviour of iniial SOE of he ESS and PV generaion, RSEM response in erms of saion power flows during he day changed according o he relaed scenario In his paper, all he examinaions were carried ou hrough a MILP model of RSEM considering i is comprised of ESS, PV, RBE and smar meer Regarding his, auhors would like o indicae ha his research can be exended o inegrae demand response sraegies o RSEM, which is considered as a fuure sudy ACKNOWLEDGMENT The auhors would like o hank Mero Isanbul Co for heir valuable cooperaion as sharing he real daa belong M1A and allowing o use RAILSIM sofware REFERENCES [1] H Kanchev, F Colas, V Lazarov, and B Francois, Emission reducion and economical opimizaion of an urban microgrid operaion including dispached pv-based acive generaors, IEEE Transacions on Susainable Energy, vol 5, no 4, pp , 214 [2] H Saboori and R Hemmai, Considering carbon capure and sorage in elecriciy generaion expansion planning, IEEE Transacions on Susainable Energy, vol 7, no 4, pp , 216 [3] V Bukarica and Z Tomsic, Design and evaluaion of policy insrumens for energy efficiency marke, IEEE, vol 8, no 1, pp , 217 [4] R K Pachauri, M R Allen, V R Barros, J Broome, W Cramer, R Chris, J A Church, L Clarke, Q Dahe, P Dasgupa e al, Climae change 214: synhesis repor Conribuion of Working Groups I, II and III o he fifh assessmen repor of he Inergovernmenal Panel on Climae Change IPCC, 214 [5] The Inernaional Energy Agency (IEA) and he Inernaional Union of Railways (UIC), Railway Handbook, hps://wwwieaorg/opics/ranspor/publicaions/railwayhandbook216/, 216, Online; accessed 1 Augus 217 [6] A Adinolfi, R Lamedica, C Modeso, A Prudenzi, and S Vimercai, Experimenal assessmen of energy saving due o rains regeneraive braking in an elecrified subway line, IEEE Transacions on Power Delivery, vol 13, no 4, pp , 1998 [7] A González-Gil, R Palacin, P Bay, and J Powell, A sysems approach o reduce urban rail energy consumpion, Energy Conversion and Managemen, vol 8, pp , 214 [8] A González-Gil, R Palacin, and P Bay, Susainable urban rail sysems: Sraegies and echnologies for opimal managemen of regeneraive braking energy, Energy conversion and managemen, vol 75, pp , 213 [9] M Colloa and G Pau, An innovaive approach for forecasing of energy requiremens o improve a smar home managemen sysem based on ble, IEEE Transacions on Green Communicaions and Neworking, vol 1, no 1, pp , 217 [1] K Harwig and I Kockar, Impac of sraegic behavior and ownership of energy sorage on provision of flexibiliy, IEEE Transacions on Susainable Energy, vol 7, no 2, pp , 216 [11] S Boudoudouh and M Maâroufi, Smar conrol in a dc railway by muli agen sysem (mas), in Elecrical Sysems for Aircraf, Railway, Ship Propulsion and Road Vehicles & Inernaional Transporaion Elecrificaion Conference (ESARS-ITEC), Inernaional Conference on IEEE, 216, pp 1 6 [12] H Moghbeli, H Hajisadeghian, and M Asadi, Design and simulaion of hybrid elecrical energy sorage (hees) for esfahan urban railway o sore regeneraive braking energy, in Power Elecronics and Drive Sysems Technologies Conference (PEDSTC), 216 7h IEEE, 216, pp [13] F Ciccarelli, A Del Pizzo, and D Iannuzzi, Improvemen of energy efficiency in ligh railway vehicles based on power managemen conrol of wayside lihium-ion capacior sorage, IEEE Transacions on Power Elecronics, vol 29, no 1, pp , 214 [14] S Khayyam, F Ponci, J Goikoexea, V Recagno, V Bagliano, and A Moni, Railway energy managemen sysem: Cenralized decenralized auomaion archiecure, IEEE Transacions on Smar Grid, vol 7, no 2, pp , 216 [15] H Novak, M Vašak, and V Lešić, Hierarchical energy managemen of muli-rain railway ranspor sysem wih energy sorages, in Inelligen Rail Transporaion (ICIRT), 216 IEEE Inernaional Conference on IEEE, 216, pp [16] M Vašak, M Baoić, N Perić, and M Bago, Opimal rail roue energy managemen under consrains and fixed arrival ime, in Conrol Conference (ECC), 29 European IEEE, 29, pp [17] M Vašak, M Baoić, M Bago, and N Perić, Consrained energy opimal conrol of rains: Concepual implemenaion, in 3h Conference on Transporaion Sysems wih Inernaional Paricipaion Auomaion in Transporaion, 21, pp [18] S Lu, S Hillmansen, and C Robers, A power-managemen sraegy for muliple-uni railroad vehicles, IEEE ransacions on vehicular echnology, vol 6, no 2, pp 46 42, 211 [19] S Nasr, M Iordache, and M Pei, Smar micro-grid inegraion in dc railway sysems, in Innovaive Smar Grid Technologies Conference Europe (ISGT-Europe), 214 IEEE PES IEEE, 214, pp 1 6 [2] P Pankovis, J Pouge, B Robyns, F Delhaye, and S Brisse, Towards railway-smargrid: Energy managemen opimizaion for hybrid railway power subsaions, in Innovaive Smar Grid Technologies Conference Europe (ISGT-Europe), 214 IEEE PES IEEE, 214, pp 1 6 [21] J Hernandez and F Suil, Elecric vehicle charging saions feeded by renewable: Pv and rain regeneraive braking, IEEE Lain America Transacions, vol 14, no 7, pp , 216 [22] F Sanchez-Suil, J Hernández, and C Tobajas, Overview of elecrical proecion requiremens for inegraion of a smar dc node wih bidirecional elecric vehicle charging saions ino exising ac and dc railway grids, Elecric Power Sysems Research, vol 122, pp , 215 [23] J C HERNANDEZ, F S SUTIL, and P G VIDAL, Proecion of a mulierminal dc compac node feeding elecric vehicles on elecric (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion

10 This aricle has been acceped for publicaion in a fuure issue of his journal, bu has no been fully edied Conen may change prior o final publicaion Ciaion informaion: DOI 1119/TSTE , IEEE 1 railway sysems, secondary disribuion neworks, and pv sysems, Turkish Journal of Elecrical Engineering & Compuer Sciences, vol 24, no 4, pp , 216 [24] J Aguado, A J Sanchez-Racero, and S de la Torre, Opimal operaion of elecric railways wih renewable energy and elecric sorage sysems, IEEE Transacions on Smar Grid, 216 [25] D Seimbille, Design of power supply sysem in dc elecrified ransi railways-influence of he high volage nework, 214 [26] O Erdinc, N G Paerakis, T D Mendes, A G Bakirzis, and J P Caalão, Smar household operaion considering bi-direcional ev and ess uilizaion by real-ime pricing-based dr, IEEE Transacions on Smar Grid, vol 6, no 3, pp , 215 [27] GAMS, CPLEX 12 Solver Descripion, hps://gamscom/laes/docs/solvers/cplex/indexhml, 212, online; accessed 8 March 217 [28] SYSTRA Consuling, Inc, RAILSIM 8 User Manual,, 28, online; accessed 11 March 217 [29] K M Tsui and S-C Chan, Demand response opimizaion for smar home scheduling under real-ime pricing, IEEE Transacions on Smar Grid, vol 3, no 4, pp , 212 [3] NREL, hps://midcdmznrelgov/, 217, Online; accessed 9 Augus 217 [31] SUNPREME, Daashee of GxB 5 Panel, hp://sunpremecom/wp- conen/uploads/217/5/ sunpreme-daashee-gxb-5w- Rev-1-1pdf, 217, Online; accessed 9 Augus 217 İbrahim Şengör (S 15) received he BSc degree from İsanbul Technical Universiy, and MSc degree from Yıldız Technical Universiy, İsanbul, Turkey in 213 and 216, respecively He is currenly working as a Research Assisan a he Elecrical Engineering Deparmen of Yıldız Technical Universiy, Turkey while pursuing his PhD sudies His research ineress include elecrificaion of railway sysems, renewable energy sysems and smar grid echnologies Hasan Can Kılıçkıran (S 17) received he BSc and MSc degrees from Yıldız Technical Universiy, İsanbul, in 211 and 214, respecively He is currenly working as a Research Assisan a he Elecrical Engineering Deparmen of Yıldız Technical Universiy, Turkey while pursuing his PhD sudies Proecion of Power Sysems, Wind Energy Inegraion o Power Sysems, and Smar Railway Energy Managemen Sysems are main research ineress of him Hüseyin Akdemir (S 17) was born in Konya, Turkey He compleed BSc and MSc a he Deparmen of Elecrical Engineering, Yıldız Technical Universiy in 214 and 216, respecively He is currenly working as a Research Assisan a he Elecrical Engineering Deparmen of Yıldız Technical Universiy, Turkey while pursuing his PhD sudies His research ineress include Renewable Energy Sysems, Proecion of Power Syems and Lighing Technology Bedri Kekezoğlu received he BSc, MSc, and PhD degrees from Yıldız Technical Universiy, Isanbul, Turkey, in 25, 27, and 213, respecively He is currenly working as an Assisan Professor a he Elecrical Engineering Deparmen of Yıldız Technical Universiy His research ineress include Power Sysems, Power Qualiy and Wind Power Energy Ozan Erdinç (M 14-SM 16) received he BSc, MSc, and PhD degrees from Yildiz Technical Universiy (YTU), Isanbul, Turkey, in 27, 29, and 212, respecively Unil May 213, he worked in he privae secor in differen posiions including elecrical insallaions, renewable energy invesmens and as procuremen exper In June 213, he became a Posdocoral Fellow wih he Universiy of Beira Inerior, Covilhã, Porugal, under he EU-FP7 funded Projec Smar and Susainable Insular Elecriciy Grids Under Large-Scale Renewable Inegraion Laer, he joined he Deparmen of Elecrical Engineering, YTU, Isanbul, where in April 216 he obained he ile of Doçen Dr (Associae Prof Dr) He is currenly also a Researcher wih he INESC-ID, Lisbon, Porugal His research ineress are hybrid renewable energy sysems, elecric vehicles, power sysem operaion, and smar grid echnologies João P S Caalão (M 4-SM 12) received he MSc degree from he Insiuo Superior Técnico, Lisbon, Porugal, in 23, and he PhD degree and Habiliaion for Full Professor ( Agregação ) from he Universiy of Beira Inerior (UBI), Covilhã, Porugal, in 27 and 213, respecively Currenly, he is a Professor a he Faculy of Engineering of he Universiy of Poro (FEUP), Poro, Porugal, and Researcher a INESC TEC, INESC-ID/IST-UL, and C-MAST/UBI He was he Primary Coordinaor of he EU-funded FP7 projec SiNGULAR ( Smar and Susainable Insular Elecriciy Grids Under Large- Scale Renewable Inegraion ), a 52-million-euro projec involving 11 indusry parners He has auhored or coauhored more han 55 publicaions, including 185 journal papers (more han 5 IEEE Transacions/Journal papers), 325 conference proceedings papers, 31 book chapers, and 14 echnical repors, wih an h-index of 34 and over 45 ciaions (according o Google Scholar), having supervised more han 5 pos-docs, PhD and MSc sudens He is he Edior of he books eniled Elecric Power Sysems: Advanced Forecasing Techniques and Opimal Generaion Scheduling and Smar and Susainable Power Sysems: Operaions, Planning and Economics of Insular Elecriciy Grids (Boca Raon, FL, USA: CRC Press, 212 and 215, respecively) His research ineress include power sysem operaions and planning, hydro and hermal scheduling, wind and price forecasing, disribued renewable generaion, demand response and smar grids Prof Caalão is an Edior of he IEEE TRANSACTIONS ON SMART GRID, an Edior of he IEEE TRANSACTIONS ON SUSTAINABLE ENERGY, an Edior of he IEEE TRANSACTIONS ON POWER SYSTEMS, and an Associae Edior of he IET Renewable Power Generaion He was he Gues Edior-in-Chief for he Special Secion on Real-Time Demand Response of he IEEE TRANSACTIONS ON SMART GRID, published in December 212, and he Gues Edior-in-Chief for he Special Secion on Reserve and Flexibiliy for Handling Variabiliy and Uncerainy of Renewable Generaion of he IEEE TRANSACTIONS ON SUSTAINABLE ENERGY, published in April 216 Since May 217, he is he Corresponding Gues Edior for he Special Secion on Indusrial and Commercial Demand Response of he IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS He was he recipien of he 211 Scienific Meri Award UBI-FE/Sanander Universiies and he 212 Scienific Award UTL/Sanander Toa, in addiion o an Honorable Menion in he 217 Scienific Awards ULisboa/Sanander Universiies Moreover, he has won 4 Bes Paper Awards a IEEE Conferences (c) 217 IEEE Personal use is permied, bu republicaion/redisribuion requires IEEE permission See hp://wwwieeeorg/publicaions_sandards/publicaions/righs/indexhml for more informaion