ANÁLISE DE PERFIS DE CONSUMO DE ÁGUA POR TELEMEDIÇÃO PARA IDENTIFICAÇÃO DE UNIDADES COM RESERVATÓRIOS INDEPENDENTES DE PRESSÃO

Abstract

The objective of this work was to identify consumption patterns that reveal the presence of reservoirs using telemetry and time series correlation analysis. To this end, a methodology was developed in the R language to process large volumes of hourly consumption and instantaneous pressure data, which were synchronized following a robust pre-processing stage to ensure temporal and spatial quality. The statistical analysis was based on calculating the Pearson Correlation Coefficient (r) individually for each consumer unit, under the premise that the storage infrastructure acts as a hydraulic decoupling element. Strong positive correlations (r>0.55) indicated Direct Supply, while weak or null correlations (r<0.30) indicated Hydraulic Decoupling (Cisterns). The application of the methodology to a sample of 24 consumer units revealed a notable dichotomy, with 92% of the sample showing hydraulic decoupling (reservoirs), contrasting with only 8% that demonstrated direct pressure dependency. The results validate the effectiveness of the Pearson Correlation Coefficient as an efficient tool for segmenting the consumption profile. This predominance of demand inelastic to pressure points to a significant opportunity for loss management, allowing the utility to implement pressure management strategies to reduce the network's average hydraulic load and real leakages without compromising the volume supplied to customers. The main limitation of the method lies in its high sensitivity to the quality and consistency of input data.