Alberta Electric System Operator (AESO)

Oscillatory dynamics and corridor stress in the Alberta electric system

North American SynchroPhasor Initiative March 2015

2

Agenda

• Alberta Electrical System Overview

• Operational Challenges

• WISP Participation

• HVDC integration

• Synchrophaser monitoring software

• Integration into control room/EMS

• Data Mining Overview

• Results of Data Mining Project

Alberta Grid - Topology

3

11,169

15,526

6513 1674

6553

514 272

22,322 km of transmission

• Intertie to Montana (230KV) MT

Geographic map of Alberta

4

Voltage Levels:

• 500KV

• 240KV

• 138KV

• 69KV

Load vs Generation Locations

• Fort McMurray Oil Sands in NE of province (BTF Generation)

• Major load centers in Calgary(South) and Edmonton (Center)

• Major Generation in Center West of province and FT Mc Murray areas

• 1.5GW of wind in south of province

• 500KV intertie to BCH located east of Calgary (South)

• 230KV intertie to Montana located at southern most point of province

5

Generation Locations

6

Operational Challenges

• 240KV backbone south to central and central to north

• Wind ramp in the province can incur issues (WPRM)

• Intertie 500KV from south and 240KV to Montana – WECC Loop Model to reduce contingencies from loop flows

• Large volume of North to south power transfers

• Industrial load

7

Load Schedule - Weekly

• AIES Weekly Winter Load Schedule

8

WISP Participation

• The AESO as part of the Wisp Program installed SEL Phasor Measurement Units at 5 location in Alberta

– Bennett ( 500KV tie to BCH)

– Picture Butte (240KV tie to Montana)

– Langdon (240KV SVC and (4) 240KV Lines)

– Sundance (Generator and 240KV Lines)

– Genesee (Generator)

• Future Locations (HVDC)

– Sunnybrook (Northwest converter station)

– Crossings (Southwest converter station)

– Newell (Southeast converter station)

– Heathfield (Northeast converter station)

9

PMU Locations

10

Purpose

• Drivers – Eastern Alberta Transmission Line (485KM - 500KV DC) will

extend from the southeast of Alberta to the northeast.

– Western Alberta Transmission Line (347KM – 500KV DC) will extend from the south of Alberta to the central west

– Expected in service date in 2015

• As part of the HVDC integration we will be installing PMU’s at both ends of the HVDC lines

• AESO pursued an integrated real time toolset to provide real time monitoring and early warning capability of transient stability issues

11

ALSTOM Phasor Point/OpenPDC

• The ALSTOM Phasor Point was selected for the Real time monitoring tool to be integrated as a stand-alone application

• OpenPDC (GPA) was selected as the Data Concentrator for Phasor data

• Data connections with external entities – WECC

– Northwest Energy (Montana)

– BC Hydro

– Bonneville Power Administration

12

Control Room/EMS Integration

• Phasorpoint Sandbox was delivered from ALSTOM in 2013 – Used for application training and proof of concept

• Production standalone Phasorpoint integration – Integrated with the OpenPDC to gather real time PMU data

from remote PMU sites

– Engineering support staff used to gain knowledge of the product

– Available outside the control room to operators to gain familiarity with

• No Control Room consoles have been deployed with the Phasorpoint software at this time, expected to be completed by HVDC commissioning.

13

Problem

• Determine ‘How To Operate with Phasor data in Real Time”

• Determine what are acceptable limits for the PMU fluctuation vs. what is considered an alarm and what is considered an actionable event for the SC

14

Data Mining with Alstom

Dynamic Performance Baselining • Identify oscillatory modes

– Estimate where the modes are “observable”.

– Typical amplitude and damping levels.

• Suggest alarm/alert limits in e-terraPhasorPoint

Angle Baselining • Identify distribution of angle differences across critical transfer paths

– Under different operational conditions

– Correlate MW transfers

– Change in angular separation for every 100MW change in power

AESO PhasorPoint System

NWE region

BPA region

Angle Difference Alerts

Overview Status Indicators

System Level Map

Event History Viewer

Individual stations

Oscillatory Stability Management in PhasorPoint

1/F MODE FREQUENCY

MODE AMPLITUDE

A

MODE DECAY TIME

EXP(-t/ ? )

MODE PHASE

Mode Frequency

Mode Amplitude

Mode decay time

Mode Phase

Exp(-t/τ)

Measured P / f / δ

Simultaneous multi-oscillation detection and characterisation direct from measurements

For each oscillation detected, alarm on:

mode damping and/or

mode amplitude for

Operations

Planning & Analysis, Plant Performance

Unlimited oscillation frequency sub-bands

Individual alarm profiles for each sub-band

In operational use since 1995 Does not use system model

Post-event analysis

Dynamic model validation

Damping controller performance assessment

Dynamic performance baselining

Early warning of poor damping (two level alarms)

Fast Modal Analysis: Alarms

Trend Modal Analysis: Analysis

Wide area mode alarms Mode locus plot with alarm thresholds

So ……

• Needs to be information for the Operator

• Great data does not mean anything if we cant take action

• Napsi challenge

• Without RT just theoretical

• Huge value in decision making, financial and reliable

18

Dominant Modes – aggregated view

A very low frequency mode

WECC North-South Mode A

WECC North-South Mode B

Inter-area Dynamical Behavior

Higher frequency Inter-area dynamics

in the power flow

Sub-bands or mode boundaries

MW

1 month duration

Inter-area Dynamical Behavior

Little activity in higher bands

System Frequency

1 month duration

Modes and Sub-band Boundaries

Mode (Hz) Lower Bound

Higher Bound

0.05 SB1 0.00 0.11

0.23 SB2 0.11 0.29

0.42 SB3 0.29 0.46

0.62 SB4 0.46 0.70

0.83 SB5 0.70 0.89

-NA- SB6 0.89 1.10

-NA- SB7 1.10 1.49

Summary – Damping Ratio by Sub-bands

0

2

4

6

8

10

12

14

DR (%)99.9%

DR (%)99.5%

MW99.9%

MW99.5%

mHz99.9%

mHz99.5%

SB 1SB 2SB 3SB 4SB 5SB 6

Dominant modes

Example – Sub-band 2 [0.11 – 0.29Hz]

US – Canada flow

Alberta North-South Flow

3 MW

Participation varies from signal to signal

Event triggered oscillation Locus plot MW vs. Damping Ratio

Example – Sub-band 2 [0.11 – 0.29Hz]

4.5 mHz

Locus plot mHz vs. Damping Ratio

Participation about the same in all frequency signals

Locus plot mHz vs. Damping Ratio

MW angle Correlation

18 20 22 24 26 28 30 32800

900

1000

1100

1200

1300

1400

1500

1600

θarea(deg) →

P SoK

(MW

)

Syste Co d t o o So o

v_genesee_330p_34415 - v_langdon_102s_34484 linear

∆ θ / ∆ P (100MW) = 2.25 deg

∠North - South

Summary

• Recommended Limits – For oscillation monitoring

– For monitoring angle-pairs

• Derived from data statistics and field expertise • Minimizes false alarms • Much lower hysteresis to be able to detect events! • No significant difference in angle base-lines for

– Weekdays and week-ends

– Net Import vs. Net Export

• Should be repeated with HVDC system in-service.

Thank you

“Importing” vs. “Exporting”

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 230

5

10

15

20

25

30

35

Ang

le (d

egre

e)

Never exported during day-time

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 230

5

10

15

20

25

30

35

Hour →

Ang

le (d

egre

e)

Typical Angle behavior

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 2314

15

16

17

18

19

20

21

22

23

24

Hour

Ang

le (d

egre

e)

ExportingImporting

Export condition angle is always lower than “import” condition

Sharp increase in angle in

morning hours

Never exported during day-time