Abstract

Pumps as turbine (PAT) for small-scale hydropower, pressure relief valves, and hydro-pumping applications have recently gained popularity. In contrast to conventional turbines, pumps do not have movable guiding vanes to accommodate changes in flow rate and head. This limits their use to the single best operating point. As a result, the precise estimation of reverse mode operating parameters is critical. One of the most successful techniques in this regard is the energy loss model. However, it necessitates extremely detailed pump geometry and experimental data. Although it is claimed to be more accurate, the lack of data makes it impractical. This calls for the devising of a pump geometrical parameter prediction model from limited manufacturer information. Using 137 sets of commercial pump operating data, geometry redesign was carried out by MATLAB code. Following that, multivariate linear regression analysis was carried out, and a one-dimensional geometrical prediction model was developed. Then the result was validated with computational analysis and measurement. The model results were 99 %, 100 %, and 95 % correlated for pump inlet, volute, and throat inlet diameters, respectively, with the experimental results. Similarly, the model results were also 99 %, 71 %, and 89 % correlated for pump inlet, outlet, and volute blade width, respectively, with experimental results. The model gave a more accurate geometrical result than the analytical and numerical methods.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call