The Engineering Problem. Calculating GIC Flow through the EHV System

Transcription

1 The Engineering Problem Calculating GIC Flow through the EHV System 1

2 Creating the GIC System Model Since the EHV system is a three-phase balanced network, it is only necessary to model a single-phase equivalent DC network for GIC North 500 kv calculations: (1) R1= Ω Ie1 V1-3 Re=0.5 Ω Ie2 R2 = R1-3= Ω (2) V1-8 Central 500 kv (3) R1-8= Ω R1= Ω RS= Ω Ie3 Ie4 Re= Re=0.5 Ω Ie5 R2 = Rc= Ω (5) V3-8 R3-8= Ω V4-9 Central 230 kv (4) V4-6 R4-6=3.4182Ω East 230 kv (6) R1= Ω R1= Ω RS= Ω RS= Ω South 500 kv (8) R4-9= Ω Ie6 Re=1.0 Ω Ie7 Rc= Ω Re= Rc= Ω Re= R2= R2= East 138 kv (7) R1= Ω Ie10 R2 = South 230 kv (9) Ie9 Re=0.5 Ω (10) Three-Phase Model Single Phase DC model 2

3 Data Requirements To perform a GIC Analysis the following data items are required for the equipment listed below, at voltage levels greater than 200 kv and up for a given system area: Feeder Data (extracted from load flow or short circuit databases) From & To Bus 60 Hz AC Resistance or DC resistance Transformer Data (extracted from transformer test reports) From & To Bus DC winding resistivity Construction: Auto, Y-Y, Y-Delta, etc. Substation Data (extracted from GPS and test data) Latitude & Longitude Substation Ground Grid Resistance 3

4 GIC/DC Equivalent Models Component 60 Hz Model DC/GIC Model Transmission Line R = R DC Series Capacitor Shunt Capacitor or or R = inf. R = inf. Series Reactor R = R DC Shunt Reactor R = R DC 4

5 GIC/DC Equivalent Models (Cont.) Component 60 Hz Model DC/GIC Model Auto Transformer R s = R DC R c = R DC Y Y Transformer / PAR Y Y R P = R DC R S = R DC Y Delta Transformer / PAR Y D R P = R DC R S = inf. Ground Grid R= 3*R G 5

6 Transformer Data Data Quality Recommendations When modeling transformers for GIC Analysis it is best practice to use DC winding resistances found in transformer test reports. Note that it is common for transformer test reports to list the DC winding resistance of three-phase units as three (3) times the single-phase value. It is important to account for this (divide by a factor of three (3)), when applicable. Modeling transformer winding resistances using series resistances (R 12 ) extracted from load flow data bases can produce inconsistent results. Actual transformer DC winding resistance is a direct function of winding length, which can vary significantly dependent on transformer construction type (core form, shell form). Modeling Substation Ground Grid Resistivity Using accurate ground grid resistivity is important. When modeled lower than actual values more GIC will flow to ground (GIC flows are locally over conservative and less conservative elsewhere). When modeled higher than actual values GIC will be redirected elsewhere in the system. 6

7 Sample Transformer Test Report 7

8 GMD Model Sizing When building a GIC model it is important to account for the impact of neighboring systems. However, there is presently no consensus on how far outside of a study area GIC system models should be built. The following can be considered: Build a GIC model that encompasses the entire system of interest as well as all transmission lines extending into neighboring busses. Use a conservative earthing resistance (transformer resistances plus ground grid resistance) at each neighboring substation location (~ ohms) to increase GIC flow into system of interest and allow for conservative results. 8

9 Calculating the Induced E on Transmission Lines When using the 1-D planewave approach exact transmission line routing is irrelevant when calculating the induced E (V im ). All that is necessary is the vector distance between the transmission line origin and destination and the angular difference between the direction of the E and the transmission line vector. V im =E T= E T cos θ E θ T T x T y V T y E 9

10 Sample Problem 1 The system shown below has a transmission line vector of T km and is subjected to a GMD with an associated E V/km, calculate the GIC Flow. R DC = 1.4 Ω Substation B R G = 0.05 Ω Y D Substation A R G = 0.05 Ω R DC = 0.1 Ω D Y R DC = 0.2 Ω 10

11 Sample Problem 1 Solution Step 1: GIC System Model 1.4 Ω V im 0.2 Ω 0.1 Ω Step 2: Calculate Induced E V im = (100 km)(1 V/km)cos(60 ) V im = 50 V 3*0.05 Ω I e 3*0.05 Ω I e = Step 3: Calculate Substation Neutral Current V im R T0TAL = 50 V ( ) Ω = 50 V 2.0 Ω = 25 A 11

12 The Engineering Problem Quantifying the Impact of GIC Flows 12

13 The System-Wide Impact Ref: 2012 Special Reliability Assessment, Interim Report: Effects of Geomagnetic Disturbances on the Bulk Power System, NERC 13

14 How Do You Solve the Problem? GIC Analysis Finite Element Analysis Electromagnetic Transients Analysis Voltage Stability Analyses Ref: 2012 Special Reliability Assessment, Interim Report: Effects of Geomagnetic Disturbances on the Bulk Power System, NERC 14

15 Asymmetrical Transformer Saturation Dependent on Transformer Core Type and Construction Reluctance of the DC flux path dictates the degree of saturation. Three phase, three limb, core form transformers are typically most resistant to saturation from GIC (to a certain threshold). All other transformers types are less resistant and approximately equally susceptible. Three Major Impacts Reactive Power Loss Difficult to Harmonic Generation Estimate (Unit Specific) Loss of Life Ref: 2012 Special Reliability Assessment, Interim Report: Effects of Geomagnetic Disturbances on the Bulk Power System, NERC 15

16 Asymmetrical Transformer Saturation Saturated Core (Field Intensity Plot) Saturated Core (Flux Plot)

17 Asymmetrical Transformer Saturation Magnetizing Current Primary Current FFT Phase A FFT Phase C FFT Phase B July A/Phase GIC Excitation

18 Transformer Manufacturer Analyses and Testing Detailed analyses can be performed by manufacturers to provide more accurate harmonic and reactive power loss evaluations. Manufacturer analysis and testing is required to establish thermal models of transformers.

19 Estimating Reactive Power Losses Linear relationships between transformer reactive power losses and GIC have been published. Formulas based on test data and measured values. Estimated reactive losses are imperfect but can be used for voltage stability analyses EPRI / Kappenman Loss Approximation Core Design Single Phase Q = V (I ex (I Eff )) Dong / Liu / Kappenman approximation Q = k1*i EFF + V(I ex ) Three Phase Shell Form Three Phase 3 Legged Core Form Three phase 5 Legged Core Form k I ex = Transformer exciting current (without GIC) I Eff = Effective GIC in the transformer windings For Y-Y/Y-Delta Transformers: i P i For Autotransformers: ai s ic a 1 a N s N c V H V L 1

20 Estimating Reactive Power Losses Ref:TPL Technical Conference, July 17, 2014 NERC

21 Transformer Thermal Models Ref:TPL Technical Conference, July 17, 2014 NERC

22 Transformer Thermal Assessment Technique Outlined in TPL Supporting Documents Ref:TPL Technical Conference, July 17, 2014 NERC

23 Transformer Thermal Assessment Technique Outlined in TPL Supporting Documents Ref: NERC TPL Screening Criterion for Transformer Thermal Impact Assessment

24 Estimating Harmonics Although linear relationships between Harmonics and GIC magnitude have been suggested, there presently isn t significant confidence with this work. Variation in transformer DC flux path reluctance can have significant influence on harmonic magnitude and profile. Linear vs. GIC? 100% Harmonic Content (% of net harmonic current) 90% 80% 70% 60% 50% 40% 30% 20% 10% H(n) 1 H(n) 2 H(n) 3 H(n) 4 H(n) 5 H(n) 6 H(n) 7 H(n) 8 H(n) 9 H(n) 10 0% GIC Current Ieff (A/Phase) Ref: 2012 Special Reliability Assessment, Interim Report: Effects of Geomagnetic Disturbances on the Bulk Power System, NERC 24

25 Steady State Voltage Stability Analysis for GIC A systems ability to withstand voltage collapse from the Mvar losses from GIC can be evaluated using methods similar to traditional voltage stability analyses (P-V and V-Q) and the results of a GIC flow analysis. Sample Voltage Stability Analysis Approach for GIC 1. Calculate GIC flows for system. 2. Estimate Mvar losses associated with each transformer modeled. 3. Model loads at each transformer location (equal to 1% of Mvar losses to start). 4. Increase loads linearly until voltage collapse or maximum Mvar loss levels are reached. a) TPL specifies studies for both System On-Peak load and Off- Peak Load Near-Term Planning Horizon cases. b) Must consider several storm directions (0-180 Deg. each 15 Deg.?) 5. Repeat Steps 1-4 until no voltage violations identified, adding mitigation if necessary (cap banks, operational measures, neutral blocking, etc.). 25

26 NERC TPL Assessment Process Ref:TPL Technical Conference, July 17, 2014 NERC

27 Why is Geomagnetically Induced Current Important? 27

28 NERC TPL Compliance Requirements and Timetable Requirement R1 R2 R5 R6 R3 R4 R7 Description Identify the indiviual(s) responsible for managing TPL compliance. Complete GIC System Model Construction Complete GIC Flow Analyses Complete Transformer Thermal Assessment Establish Steady State Voltage Criteria for system during GMD events Perform System Planning Assessment for GMD events Submit Corrective Action Procedure addressing how system performace will be met. NERC Enforcement Dates 07/01/ /01/ /01/ /01/ /01/ /01/ /01/2022

29 NERC TPL Contents Currently Under for Revision at per FERC Mandate for TPL Modify the benchmark GMD event definition used for GMD Vulnerability Assessments; Make related modifications to requirements pertaining to transformer thermal impact assessments; Require collection of GMD-related data. NERC is directed to make data available; and Require deadlines for Corrective Action Plans (CAPs) and GMD mitigating actions.

30 FERC Vision of GMD Standard Implementation

31 Localized Extreme Events vs. Spatial Averaging for the Benchmark Event Ref: Benchmark Geomagnetic Disturbance Event Description, May 12, 2016 NERC