Robust regression of the the output power of a gas-and-steam combined cycle power plant

(Prof. Dr. Thorsten Dickhaus)

This project deals with the application of robust regression methods (see e.g. the methods lmRob, glmRob in robustbase) to data on the output power of a gas and steam combined cycle power plant, and compares these with results from the prediction of a full load electrical power output of a base load operated combined cycle power plant using machine learning methods. The associated dataset is available at the following link.

Elementary programming skills (preferably in R) and basic statistical knowledge (ideally: successful completion of an introductory statistics course) are required. Basic knowledge of calculus and linear algebra is also recommended. The duration of the work within the research group should be at least four weeks. In addition, a written paper and a final presentation in the seminar are expected. Further information and requirements can be found in the module and event catalog of the Department of Mathematics.