Dimensionamento de tanques de armazenamento de água cinza incorporando estocasticidade e riscos

Abstract

This work presents a methodology for stochastic sizing of greywater tanks (reservoirs) used to supply toilet demands. Data from three surveys that specified the hourly flow rates for each hydro-sanitary equipment (shower, washbasin and toilet) were used. A procedure to convert daily statistical standard deviations per device to hourly deviations was proposed in order to incorporate uncertainties in the data. The sizing was performed by means of an optimization model capable of finding the best storage capacity within predetermined limits of minimum reliability and maximum risk. Simulations were conducted for a total of 25 reliability and risk values in each of the three case studies. In all tests, it was found that the reliability within the risk is greater than the initial reliability proposed, so that the real risk is lower than the fixed risk and the capacity calculation procedure was valid. The results for the three data series showed similarity with regard to the two input variables (reliability and risk) in relation to capacity, that is, to the extent that more reliability and less risk are desired, more storage capacity will be needed. The largest volume for the reservoir among the data from the three case studies was approximately 112 liters, which would correspond to a tank with 1.0 m2 of base area and 11.2 cm of height. A system was proposed where the greywater from an upper floor would be directed to a tank located on the ceiling of the bathroom on the lower floor that would serve to meet the demand of the toilet and, if at times this input was higher than consumption, overflow would occur to a specific greywater downpipe.