Evaluation of hybrid forecasting methods for organic Rankine cycle: Unsupervised learning-based outlier removal and partial mutual information-based feature selection

•The nonlinear characteristics of real organic Rankine cycle (ORC) data are analysed.•An unsupervised learning-based algorithm is proposed for outlier removal.•A partial mutual information-based feature selection is performed.•Our hybrid method has superior performance in ORC forecasting. The constr...

Full description