Sensitivity analysis for CMU type heat-transfer monitor for OTEC

Abstract

Measurement of the fouling resistance, R/sub f/, by a Heat Transfer Monitor (HTM) has some inevitable uncertainty. To better understand the origins of such uncertainty, a sensitivity analysis of the data-reduction algorithm of the improved Agronne National Laboratory (ANL) HTM was performed. Nine parameters were chosen for analysis: tau, the time constant for the cooling curve; the seawater temperature; the flow velocity; the Wilson Plot slope and intercept; the inside diameter of the tube; the thermal conductivity of the tube; the correction factor for the thermal entrance effect; and the natural convection heat transfer coefficient for air. The first step in the analysis was to generate a set of uncertainty equations based on the partial derivatives of the equations in the data-reduction algorithm. Then, after all other uncertainties were set to zero, the uncertainty in the parameter being examined was allowed to vary. Each time, the corresponding uncertainty in the nominal fouling resistance was calculated. Also, a further study was undertaken to examine the effects of the uncertainty of the fouling factor caused by different values of the fouling resistance. As a result of this study, it was determined that the uncertainty in the fouling factor was highly sensitive tomore » five parameters: tau, the flow velocity, the Wilson Plot slope, the correction for the entrance effect, and the water temperature. Small uncertainties in these values will result in relatively high uncertainties in the fouling resistance. Thus, the major causes for any high level uncertainties can be directly attributable to the uncertainties in a few specific parameters.« less