Modelling Room Cooling Capacity With Fuzzy Logic Procedure

Abstract

The primary aim of this study is to develop a model for estimation of the cooling requirement of residential rooms. Fuzzy logic was employed to model four input variables (window area (m 2), roof area (m 2), external wall area (m 2) and internal load (Watt). The algorithm of the inference engine applied sets of 81 linguistic rules to generate the output variable in Cooling Load rating. A paired t-test was carried out using SPSS version 20 package, with the results of human professionals' calculations for each assessed rooms compared with the model generated Cooling Load capacities for the same set of variables. The human calculation and model results were observed to be strongly correlated (r=0.880, p<0.001) with no significant difference between the two sets of variables (t19=-1.697, p>0.001). On the average, human calculation values were 221.5 points lower than model calculated values (at 95% confidence interval [-447.55, 4.50]). The study proposed a model to size the cooling capacity of residential rooms. The model is capable of providing results comparable to that of human professionals in the application area. It is simple and can find its usefulness among building consultants/professionals and home owners.

Modelling oil and gas flow rate through chokes: A critical review of extant models

“…The fuzzy logic system has the ability to capture the non-linear relationship of an input-output model without an exact mathematical formula (Liu and Li, 2005). Modelling with fuzzy logic entails utilizing a linguistic approach (descriptive language) established on fuzzy logic with fuzzy propositions (Adeyemi et al, 2016). The operational mechanism of fuzzy logic is to map an input space (universe of discourse) to an output space, using a list of "if then" statements referred to as rules (Castillo and Melin, 2001).…”