SENSITIVITY ANALYSIS OF THE TEST PARAMETERS OF A SOLAR FLAT PLATE COLLECTOR FOR PERFORMANCE STUDIES Subhra Das 1, 2 * , Bibek Bandyopadhyay 2 and Samir Kr. Saha 1 1 Department of Mechanical Engineering, Jadavpur University, Kolkata 2 Solar Energy Centre, Institutional Area, Gurgaon-Faridabad Road (19 th Milestone), Gurgaon *Corresponding author: Phone: 0124-4144524, E-mail address: nips.subhra@gmail.com Abstract The BIS test procedure for performance of solar flat plate collectors allows a certain limit of deviation for the control parameters to define the steady state condition. During actual performance studies, sometimes these limits are exceeded. In this paper an attempt has been made to find out the effect on the performance curve of the collector as a result of exceeding this prescribed limit by a certain small amount. Regression analysis has been used to study the sensitivity of the collector design to the response parameters. It has been concluded that a variation in the range ±0.01 to ±0.04 over and above the prescribed deviation does not significantly affect the response parameters (i.e, thermal efficiency, F R U L and F R (ȗα)). Keywords: Flat Plate Collector, Thermal performance test, Sensitivity analysis, Regression analysis. 1. Introduction The Indian standard procedure for testing solar flat plate collectors and reporting the performance was proposed by Bureau of Indian Standards (IS: 12933: 2003, Part 5, Second Revision). The standard sets limits for environmental conditions, specifies test procedures and calls for the thermal performance to be reported in terms of efficiency of the collector (Ș) which is expressed as a linear function of (T i -T a )/G, where T i is the fluid inlet temperature , T a is the ambient temperature and G is the Solar irradiance on the collector plane. The thermal efficiency test is conducted under steady state conditions. To ensure that the collector is operating in steady state conditions over a given test period, it is proposed that none of the experimental parameters deviate from the mean values over the test period by more that the limits prescribed by BIS. It is found that the limits set by the BIS to ensure steady state condition is very difficult to achieve in practice. Our aim in this paper is to study the effect of a small variation in the design parameters on the efficiency of the collector. Various methods have been employed in the literature for modeling the variations of design parameters (or variables). The methods are classified into five major category: fuzzy set theory (Wood and Antonsson, 1987), interval methods (Chen and Ward, 1995), sensitivity analysis methods (Sandgren et al., 1985, Whiting et al., 1999), probabilistic-based methods (Clarke et al., 2001) such as robust design (Sundaresan, et al., 1993; Parkinson, et al., 1993, Xiaoping Du, et al., 1998) and statistical method ( Paul A. Funk, 2000, N.S.Thakur et al., 2003, S.D.Sharma et al.,2005). The difference between the fuzzy set theory and probabilistic methods for dealing with different types of uncertainties are discussed by Wood and Antonsson (1990). While sensitivity analysis is focused on reducing the local rate of change of the design performance, the robust design or Taguchi method (Phadke, 1989; Taguchi, 1993) goes one step further by introducing uncertainty or noise in the system and generating optimal solutions that could reduce the impact of the uncertainty in a global scale. To measure the goodness of fluctuating performance on the variation of the design parameters, a metric needs to be formulated. In this paper we propose to use the “Sensitivity Analysis” to compare the performance of the collector by varying the design parameters with the performance of the collector based on BIS test procedure. In our study we have considered the effect of variation on two design parameters viz., “inlet fluid temperature” and “temperature difference between collector inlet and outlet temperature”