Faculty of Business and Economics, Chair of Energy Economics, Prof. Dr. Möst The Influence of Voltage Stability on Congestion Management Cost in a Changing Electricity System www.ee2.biz Fabian Hinz 15th IAEE European Conference Vienna, September 2017
1 2 3 4 Motivation Model Development Staus Quo 2014 Future scenario 2025 TU Dresden, Chair of Energy Economics, Fabian Hinz 2
Load vs. generation Loop flows Congestion mgmt. cost [mio. EUR] Congestion management causes high cost Challenge Current / Real power Development of congestion mgmt. cost, causes 750 Curtailment Redispatch & Countertrading 0 30 45 32 58 164 198 159 269 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Direct flow from North to South Load distribution Wind distribution Load concentrated in the South and West Wind concentrated in the North Power flows from North to South cause loop flows via Eastern Europe Phase shifting transformers being installed Loop flow via PL, CZ and AT Source: BNetzA Monitoring Reports 2007-2016 TU Dresden, Chair of Energy Economics, Fabian Hinz 3
Availability of reactive power in the transmission grid declines Challenge Voltage / Reactive power Reactive power supply: conventional and future scenario Conventional supply Future supply Transmission grid 110 kv grid Medium / low voltage grid Reactive power consumption Reactive power consumption Electricity feed-in 62% 8% 42% 13% 28% 38% 51% 59% 2014 2025 2035 Reactive power supply Conventional supply through large power plants Availability in the transmission grid decreases Supply can be replaced by RES in the distribution grid Source: Kraftwerksliste BNetA 2015, Netzentwicklungsplan 2015 Controllable reactive power Offshore TSO DSO TU Dresden, Chair of Energy Economics, Fabian Hinz 4
Redispatch measures conducted in order to solve current and voltage problems Current- and voltage-induced redispatch Real power Current-induced redispatch Reactive power Voltage-induced redispatch Situation expansive Current too high cheap Voltage too low expansive Voltage okay Q cheap Power plant 1 not dispatched Power plant 2 fully dispatched Power plant 1 not dispatched Power plant 2 Reactive power Redispatch expansive Current okay cheap Voltage okay Q expansive Voltage okay Q cheap Power plant 1 ramped up Power plant 2 ramped down Power plant 1 Reactive power Power plant 2 Reactive power More expansive power plant ramped up in order to alleviate transmission line More expansive power plant ramped up in order to provide reactive power Redispatch cost TU Dresden, Chair of Energy Economics, Fabian Hinz 5
1 2 3 4 Motivation Model Development Staus Quo 2014 Future scenario 2025 TU Dresden, Chair of Energy Economics, Fabian Hinz 7
Redispatch cost calculated in a 3-step approach Reactive power [Mvar] Model approach Step 1 Market model Electricity market model (copper plate) for Germany and neighboring countries to generate power plant dispatch NTC-based trade between market zones Only real power (P) dispatch Reactive power behavior of 380 KV line 600 500 400 Step 2 Real power: current-induced redispatch Step 3 Reactive power: voltage-induced redispatch Estimation of current-induced redispatch based on a transmission & 110 kv distribution grid model Usage of ELMOD to calculate load flows, overloads and least-cost redispatch Penalty cost for international redispatch Estimation of reactive power dispatch and voltageinduced redispatch Usage of ELMOD LinAC, a linearized AC model to account for voltage stability and reactive power flows Iterative approach to account for quadratic reactive power behavior of electricity lines 300 200 100 0-100 0 500 1000 1500 Q_cap Q_tot Line load [MVA] Q_ind Iterative calculation of quadratic inductive reactive power behavior TU Dresden, Chair of Energy Economics, Fabian Hinz 8
Redispatch models use linearized real and reactive power flow calculations Simplified model formulation of redispatch models Market model Current-induced Redispatch Voltage-induced Redispatch Target function marg Min cost n P P,market Genn gen n n N Thermal limit: LineCurrent l Thermallimit l Restrictions Voltage TS: 0, 97 p. u. U n 1, 03 p. u. Voltage DS: 0, 94 p. u. U n 1, 06 p. u Gen P Gen n P, Gen n Q Gen Q Grid balance Real power: Gen n P Dem n P = σ m N g n,m U n U m b n,m (θ n θ m ) Iterative calculation Reactive power: Gen n Q Dem n Q Loss n Q = σ m N b n,m U n U m g n,m (θ n θ m ) 1) U n / U m... Voltage magnitude at node n / m Θ n / Θ m... Voltage angle at node n / m g n,m / b n,m... Conductance / susceptance between node n and m TU Dresden, Chair of Energy Economics, Fabian Hinz 9
Current and voltage are represented reasonably well by the redispatch model Model quality of ELMOD AC and ELMOD LinAC Current [A] Voltage [p.u.] LinAC MAE RSME amape 1) I [A] 22.9 39.6 0.69% Good fit for current LinAC MAE RSME amape 1) U [kv] 2) 2.0 2.5 0.53% Reasonable fit for voltage Comparison between redispatch model (ELMOD LinAC) and AC load flow model (ELMOD AC), Germany, 16 grid situations 1) Adjusted Mean Absolute Percentage Error: adjusted in relation to nominal voltage / thermal limit 2) On 380 kv level TU Dresden, Chair of Energy Economics, Fabian Hinz 10
110 kv grid set developed based on OSM data and other public sources Data set for grid model OSM data Substations 380 / 220 / 110 kv Electricity lines 380 / 220 / 110 kv Nodes with generation and demand Auxiliary nodes Lines start / end, technical parameters updated with TSO static grid models Transformers 380 / 110 kv 220 / 110 kv 380 kv 220 kv 110 kv Power plants / RES Attribution to nodes Plants: based on addresses and coordinates RES: based on OSM data / RES database Load Attribution based on GDP and population of surrounding area Nodes: ~5700 Lines: ~6500 Substations: ~370 TU Dresden, Chair of Energy Economics, Fabian Hinz 11
1 2 3 4 Motivation Model Development Staus Quo 2014 Future scenario 2025 TU Dresden, Chair of Energy Economics, Fabian Hinz 12
Good fit between congestions in model and reality Congested grid elements: Model results vs. reality Model results 2014 Monitoring report 2014 Frequency of congested grid elements Good fit between for border areas to Poland, Czech Republic and Denmark Fit for Remptendorf- Redwitz line Congestions in the North West and Center not reliably recognized Distribution grid congestions in the North fit local curtailment compensation Source: BNetzA Monitoring Report 2015 TU Dresden, Chair of Energy Economics, Fabian Hinz 13
Results Taking into account voltage stability, redispatch patterns change High load and high wind feed-in situation: current- and voltage-induced redispatch Current-induced Current- and voltage induced Ramp-down of power plants in the North Curtailment mainly in Schleswig-Holstein Ramp-up in the South and Austria Additional redispatch in the South to cover reactive power requirements Additional ramp-downs in the North TU Dresden, Chair of Energy Economics, Fabian Hinz 14
Cost p.a. [mio. EUR] Reactive power from the 110 kv grid decreases voltage-induced redispatch cost Redispatch costs 2014 in Germany 180 160 140 120 113.4 174.7 17.1 44.1-13.4 (-8%) 161.3 13.3 34.6 Redispatch cost Germany 2014 Comparison of voltage- / current- induced redispatch Cost reduction potential from 110 kv grid reactive power sources 100 80 60 40 20 78.1 35.4 78.1 35.4 78.1 35.3 Redispatch and curtailment cost is mainly currentinduced 8% reduction possible through reactive power from the distribution grid 0 Current-induced Current- / Voltage-induced with 110 kv sources Curtailment U Redispatch U Curtailment I Redispatch I TU Dresden, Chair of Energy Economics, Fabian Hinz 15
1 2 3 4 Motivation Model Development Staus Quo 2014 Future scenario 2025 TU Dresden, Chair of Energy Economics, Fabian Hinz 16
Market zone split decreases redispatch cost more than 110 kv reactive power Redispatch costs 2025 under full grid extension, combined and split DE/AT market zone Redispatch cost 2025 Comparison of DE/AT market zone and split Cost reduction potential from 110 kv grid reactive power sources -23 Combined zone Status Quo With 110kV sources 4 4 106 105 244 241 293 278 304 308 933 956 Only redispatch cost! Additionally welfare effects on wholesale markets have to be considered! -179 Zone split Status Quo With 110kV sources 4 4 103 102 347 322 346 307 Cost reduction potential through 110 kv sources increases Overall reduction of redispatch cost through splitting of DE/AT market zone -18 777 759 CZ PL AT DE Int. redispatch TU Dresden, Chair of Energy Economics, Fabian Hinz 17
Considerably higher cost under grid extension delay savings potential increases Redispatch costs 2025 under full and delayed grid extension Redispatch cost 2025 Comparison of full and delayed grid extension Cost reduction potential from 110 kv grid reactive power sources -23 Full grid extension Status Quo 703 106 With 110kV sources 703 4 4 Annuity @ 4% WACC + 2% O&M cost 105 244 241 293 278 308 304 1,659 1,636 Grid extension CZ PL AT DE Int. redispatch +1,166 5 year grid extension delay Status Quo With 110kV sources 179 14 178 14 260 249 1,930 1,894 Considerably higher cost under grid extension delay Under grid extension delay, the cost reduction potential from 110 kv sources increases 443 445-46 2,825 2,779 TU Dresden, Chair of Energy Economics, Fabian Hinz 18
Total cost [bn. EUR] Which degree of grid extension is economically reasonable? Relationship between grid extension and redispatch cost 2.8 2.70 Redispatch cost 2025 Alteration of grid extension level (# of HVDC links) Comparison of total grid extension cost 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 2.46 2.09 1.86 1.87 1.60 1.56 1.31 Grid extension cost Total cost - Market zone combined Total cost - Market zone split 1.50 1.29 1.49 1.31 1.55 1.37 1.61 1.44 1.68 1.53 1.79 1.65 0.8 0.6 0.4 0.2 0.0 Delayed 0.19 0 HVDC 0.34 0.43 1 HVDC 0.53 2 HVDC 3 HVDC 0.61 4 HVDC 0.70 5 HVDC Full 0.79 6 HVDC 0.89 7 HVDC 1.03 8 HVDC TU Dresden, Chair of Energy Economics, Fabian Hinz 19
Conclusions Key take-aways Current- and voltage induced redispatch will play an important role in future electricity systems Usage of 110 kv reactive power sources can slightly limit redispatch costs Market zone layout has a much higher impact Grid extensions required to impede extreme cost increases number of HVDC links in grid development plan seems reasonable TU Dresden, Chair of Energy Economics, Fabian Hinz 20
Faculty of Business and Economics, Chair of Energy Economics, Prof. Dr. Möst Thank you for your attention! Dipl. Wi.-Ing. Fabian Hinz Chair of Energy Economics Faculty of Business and Economics TU Dresden Email: fabian.hinz@tu-dresden.de Phone: +49 351 463 39896 www.ee2.biz