Team from the Universities of Reading and Oxford claim third place in international energy modelled competition

Dr Georgios Giasemedis from the Counting lab at the University of Reading and Steven Haben from department of mathematics at the University of Oxford have beaten off stiff competition to come third in an important international competition for energy demand modelling. Both of the researchers are working on the £30 million Thames Valley Vision project. The competition was previously competed for in 2012 with a University of Reading team claiming first after the disqualification of the top three teams. The competition is organised on a bi-annual basis by one of the world's leading authorities on energy forecasting, Dr Tao Hong from UNC Charlotte. The competition is split into four modelling streams; electrical load forecasting, price forecasting, wind power and solar power forecasting.  

The Reading/Oxford team competed in the electrical load forecasting competition which took place between the 15th August 2014 to the 15th November 2014. Each week a dataset was released and the teams had to give their best probabilistic forecast for the next period. The competition continued for a total of 15 task periods. The Oxford/Reading University team achieved a creditable third place in around 100 initially starting teams. Teams included utility companies who do this as their profession.

The team used a forecasting method based on a hybrid model combining two types of probabilistic forecasting tools, a kernel density estimation and a quantile regression.

The team picked up their third place prize at the PES conference in Denver, Colorado.  

 

Things to do now

Page navigation

See also

 

Search Form

A-Z lists