Uncertainty evaluation of efficiency measurement in laboratory conditions

Abstract

Uncertainty is a parameter associated with the result of a measurement, which characterizes the dispersion of the values that could reasonably be attributed to the measurand. All measurements are subject to uncertainty and a measurement result is complete only when it is accompanied by a statement of the associated uncertainty, such as standard deviation or uncertainty band.The identification of uncertainty of individual sensors and standardization in calibration process is under progress in Turbine Testing Lab (TTL) at Kathmandu University (KU). The lab aims to establish a Francis turbine test rig and procedures that meet the standards set by International Electrotechnical Commission (IEC). This needs a major investment on calibration systems and development of procedures to estimate the uncertainty in measurement more precisely than ever before.This study aims to identify the sources of uncertainty and their quantification for efficiency measurement systems at laboratory condition. The study is done on the 14 kW Francis turbine project at TTL, KU. Based on different published papers regarding uncertainties, type A and type B uncertainties were evaluated separately and then combined to get combined uncertainty of each components. The uncertainty in efficiency was then calculated using the rules of uncertainty calculation.In this paper, the standard method is illustrated to evaluate the uncertainties in sensor-based measurement system. The same method is used in investigation of uncertainty to the extent it was possible and feasible in the lab i.e. study could be mainly focused only on torque transducer. The value of efficiency at 1300 rpm is calculated to be 66.7% ± 0.35%, where ± 0.35% represents the uncertainty in efficiency for the same rpm value. The results indicate uncertainty level unforeseen close to meet the IEC standards which could be result of including only parts of uncertainties. All uncertainties should be evaluated for future research papers.