Time disaggregation of water consumption readings by means of stochastic methods

Residential water demand is highly unsteady and stochastic, motivating the development of methods that model it as a series of rectangular pulses following a Poisson process (PRP methods). In order to generate the demand series, these methods require parameters of the instantaneous water demand, such as frequency of water use, and intensity and duration of stochastic demand pulses, each of them defined by its mean, variance and probability distribution. Procedures for obtaining those parameters are generally based on direct observation of instantaneous water demand by registering it every second at selected homes for several days. That direct technique is impractical, because of the enormous amount of data generated and processed, and the need of special equipment. A method for estimating the necessary parameters for simulating the instantaneous water demand from meter readings of >ls (e.g. Imin) is presented. The proposed method considers some principles from the Neyman-Scott (N-S) process, such as the disaggregation of the accumulated water volume, based on a comparison between statistical moments of the observed larger interval demand series and theoretical moments of the instantaneous water demand. An objective function express ing the relation between theoretical and observed moments is formulated and minimized by non linear programming. The intensity, duration and frequency or arrival rate of the instantaneous demand pulses are thus obtained. Using these results, instantaneous water demand series, or demand series with any averaging interval, can be generated. The method is validated by comparing the generated demand series with observed demand series.