Statistical Modelling of Power Grids
April 4, 2019
When: 20190523
Where: Dogpatch Labs - The CHQ Building
Venue Capacity: 80 people The capacity of the venue is about 80 people but because of no-shows, we have a policy of first-come, first-served on the night in the event of the room filling up. Managing no-shows is a problem all Meetups have to deal with and this is the fairest method we can think of. While we have never been in a position to turn anyone away for capacity reasons, it is always a possibility, so please arrive early to avoid disappointment on the night. Speaker: Joe Wheatley Power grids are a popular example of complex infrastructure and are a type of oscillator network. In some situations complex infrastructure can show unexpected behaviour e.g. Braess paradox, instabilities or even collapse. Most modern power grids contain grid-scale renewable generation that is not under the control of the grid operator. Modelling this using standard statistical techniques or from first principles is problematic. Here I demonstrate a successful statistical modelling approach that describes the output of individual thermal power plant as probabilistic (Markov) processes controlled by grid-scale covariates. A detailed quantitative model of the Australian power grid (NEM) was obtained using this method. Hopefully some general data-science lessons will be learnt along the way.