WASPAS based Taguchi approach to reduce the energy consumption of a typical office building

Better insulation or material with low thermal conductivity is a crucial factor in new construction and retrofitting existing structures where energy efficiency is the objective. Effective energy savings need the careful selection of construction materials to accomplish this. This research work takes into account a typical office building for this purpose. This paper presents the WASPAS-based Taguchi optimization approach, which is used to determine the optimal air-conditioning load of a typical office building. To model a typical office building, AutoCAD software has been utilized, and building energy simulation using HAP software has been conducted. The optimization of the air-conditioning load involved selecting eight factors, namely the type of HVAC system (A), wall material (B), window glass U-value (C), window glass shading coefficient (D), window-to-wall ratio (E), insulation material for the roof (F), thickness of roof insulation (G), and air infiltration flow rate (H) at mixed levels. An orthogonal L18 arrangement is used for the experimental tests, and in each trial, the cooling and heating load as well as the annual energy consumption of the typical office building are calculated. The WASPAS method is then used to convert the multiple-response problem into a single-response problem after a decision matrix is generated by the S/N ratios of all three response variables. The optimal levels for each of the eight factors are determined after reviewing the response table for Taguchi tests. By calculating the percentage contributions of each of the eight factors, the analysis of variance method is also used to pinpoint the primary factors that had a significant impact on the results. The results showed that the A2B3C1D2E1F2G3H1 combination is the most effective set of all eight factors, and the wall material of the office building has the largest impact, contributing 38.23%. Finally, the prediction model has been developed for all three response variables.