Policy-based reserves for power systems


METADATA ONLY
Loading...

Author / Producer

Date

2012

Publication Type

Presentation

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

We present a new approach to providing backup for large, fluctuating renewable infeeds in power systems, based on the concept of reserve policies. The approach can be viewed as a generalization of existing reserve schemes, and uses robust optimization with recourse to determine operating rules for power system entities such as generators and storage units. These rules, or policies, set in advance the responses of these entities to errors in the prediction of loads and renewables once they have been measured. We show how affine policies can be specified such that the power network constraints, namely matching supply and demand, respecting transmission line ratings, and the operating limits of power system entities, are satisfied for all possible realizations of the prediction error. The error is assumed to be bounded and may arise from any number of spatially and/ or temporally correlated sources. Affine policies are compared with existing reserve operation modes under standard modelling assumptions, and consistent cost reductions are reported for multi-day benchmark studies featuring a poorly-predicted wind infeed. Efficient prices for such “policy-based reserves” are derived, and we propose new reserve products that could be bought and sold on electricity markets. We also report some early success in solving larger-scale versions of the problem using ADMM.

Permanent link

Publication status

published

External links

Editor

Book title

Journal / series

Volume

Pages / Article No.

Publisher

IfA Internal Seminar Series

Event

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Organisational unit

03416 - Morari, Manfred (emeritus) check_circle

Notes

Lecture at IfA Internal Seminar Series in Zürich in November 2012.

Funding

Related publications and datasets