Probability-based optimum design of refrigerated warehouses

Abstract A basic criterion in the design of refrigerated commercial warehouse is to minimize the operational energy. Although the input parameters like outside temperature, solar radiation, material properties and the dimensions of walls and roof are random in nature, very little fluctuations are desired in the inside conditions. A design produced by considering the randomness of the input variables and the fixed inside conditions would be more rational and realistic than the one produced according to deterministic procedures. In addition to the thickness of insulation for the walls and roof, the thickness of walls, orientation of the building and the building envelope parameters also influence the energy requirements of a warehouse. In this work the probability based optimum design of refrigerated warehouses, using interior penalty function method of optimization, is considered. For the calculation of heat gain, the design day based on the maximum value of mean plus three times the standard deviation of solar-air temperature, computed from the hourly meteorological data, is chosen. The optimum point is checked with Kuhn-Tucker conditions. A sensitivity analysis of the optimum results is also performed.