Dynamic state Estimation & Voltage Stability

The DEVS project Dynamic State Estimation And Voltage Stability Of Transmission And Distribution Grids With A Large Share Of Distributed Generation Capacity

Klaas Visscher, Project Manager Energy research Center of the Netherlands, [email protected]

Financially supported by SenterNovem (EOS-LT)

10-11-2009

http://www.devs-project.nl

Overview • The DEVS project • Problem definition – Future scenario – Objectives

• Problem approach – – – –

Classical State Estimation DEVS Dynamics State Estimator DEVS Control Strategy Outlook on distributed Agent based approach

• Future grid extension scenario’s • Current status of the project

10-11-2009

2

The DEVS project Particip ant

Organisation name

SAG member

Commitme nt

1

ECN

No

Financial

2

TUD

No

Financial

3

Alliander (Continuon Netbeheer)

Yes

Financial

4

KEMA

Yes

Financial

5

ENECO Netbeheer

Yes

In-kind

6

DELTA N.V.

Yes

In-kind

10-11-2009

• +/- 1,6 MEuro (real Euros) • 4 years + 0,5 year • Started March 2007 Aim: • control strategies as a solution to identified voltage stability problems

3

Problem definition •

Trend to increase distributed generation – Each user node can behave as both sink and source for power.

•

Unpredictable and large power fluctuations of decentralized generation. Would result in: – Frequent blackouts in parts of the grid, – economic loss and social chaos in the areas concerned would be the result. (A risk that should be avoided)

•

Conventional approach of state estimation insufficient – voltage and current transformers – from the point of accuracy as well as time resolution.

•

New voltage control strategies are needed – Much more accurate determination of power exchange necessary at the nodes – Specifically designed to cope with decentralized generation.

10-11-2009

4

Future scenario (NL) • 6 GW offshore windpower

Decentralised generation:

• 2 GW onshore windpower • 7 GW micro CHP & CHP • HVDC interconnections Also expected: • Storage systems (vehicles)

10-11-2009

“All electricity generation facilities that generate an amount of power that does not depend on control actions of distribution and transmission network operators.” 5

Objectives Future grid with a large share of decentralised generation: – several thousands Megawatt of wind energy (onshore and offshore) and – several thousands Megawatt of micro CHP units in households:

1. Identification of the main voltage stability problems in the network 2. Detection of voltage stability problems by means of Dynamic State Estimation (DSE), 3. To device and test control strategies as a solution to the identified voltage stability problems.

10-11-2009

6

Problem approach • Dynamic State Estimation and Voltage Control – in distribution networks

• Determination of the dynamic system state from on-line phasor measurements – by calculating real and reactive power flows and their evolution in time with the proposed DSE.

A scalable solution: • Phasor measurements 10 kV/20kV – take place at the 10 kV or 20 kV side of the 10 kV / 400 V transformers.

• Going up to the higher voltage levels similar measurements are done at 50 kV / 10 kV, 150 kV/ 50 kV etc. 10-11-2009

7

Classical State Estimation Topology

State

Status of Switches Quality

Measurements & Accuracy

Faults in Measurements & Topology

Pseudomeasurements State Estimator

Determine state (voltage magnitudes and angles) from a redundant set of noisy and erroneous measurements. • SE spreads errors over all states. • Bad data detection and identification. Static-state estimation: Based on static model. Dynamic-state estimation: Based on dynamic model.

10-11-2009

8

DEVS Dynamic State Estimator

(Dynamic) State Estimator

Voltage Instability Detector and Predictor DEVS Dynamic State Estimator

• Classical state estimator extended with instability detection and prediction. • State estimation based on accurate phasor measurements.

10-11-2009

9

DEVS Control Strategy

10-11-2009

10

Outlook on hierarchical distributed approach

10-11-2009

11

Future grid extension scenario’s North Holland grid • HVDC to Great Britain

• Offshore WIND power sevralk GW – Beverwijk connection

• CHP several GW • microCHP several GW Tests: • SC-tests at HV • Voltage and reactive power behaviour. 10-11-2009

12

Current status of the project • NH grid modeled – ECN: Matlab-Simulink (SimPowerSystems) – TUD: RTDS (Real Time Digital Simulator)

• Aggregated Wind farm models – (ECN)

• Agent based loadflow – Proof of Concept

• Survey of modern State Estimation • Hierarchical Agent based voltage instability prevention – TU Delft / Proof of Concept

• Future grid extension scenario’s – Test Cases established 10-11-2009

13

Thank you for your attention [email protected]

10-11-2009

14