Prediction Model for Complete Characteristic of Pump as Turbine by Hybrid Approach Using Ensemble Regression Model and Polynomial Regression

Abstract Centrifugal pumps operated in reverse mode (popularly called pump as turbine (PAT)) are well-established as green energy converters, mainly used in micro-hydro applications due to their lower capital cost. However, the appropriate selection of the pump for reverse-mode application is a critical issue due to not availability of its characteristic curve. Therefore, various researchers have proposed models to prediction PAT parameters from pump characteristics based on diverse approaches. Still, there is hardly any model which can predict the head and flow rates for the PAT with less than 20 % deviation in both of them. This work proposes a novel approach to predict the characteristics of PAT with the integration of ensemble regression modelling for Best Efficiency Point (BEP) and from it, by polynomial equations complete characteristics curve. The data from in-house experiments and open literatures were used to develop the proposed model excluding the data of four-pumps, which are typically used for validation of this proposed model. The predicted parameters of PAT from pump characteristics show good agreement for four selected PATs with less than 10% maximum deviation. Compared to the literature prediction models, it shows a lower deviation for same 4 PATs. A case study based on the field data is carried out to justify the applicability. It reveals that by using this model selected pump when operated in PAT will produce higher power and generate more revenue compared to literature models. Though the difference is lower, for a large number of PATs it will be significantly higher.