Methodology of a Data Validation and Reconstruction Tool to Improve the Reliability of the Water Network Supervision

The drinking water network of Aigües del Ter-Llobregat (ATLL) has applied a methodology to obtain a reliable data. The methodology is based on a first phase to validate the data with six levels and a second phase that they analysis and reconstruct the non-validate data. INTRODUCTION Every day, the telecontrol system supervised the whole network status. This process needs accurate and reliable sensors data. If we work with a wrong data, we do not able to control efficiently the network and we can not detect the anomalous situations as faults or leakages. So, we need to analyzes the received data and validate. If the data is non-validate, we have to replace by well forecast. The most important information, in a drinking water network, is captured by flowmeters. The flowmeters measures the water consumption of the several towns and the introduced water in the network. The flow analysis is used to differentiate the correct and the wrong data. Moreover, the analysis is used to estimate the wrong data and their confidence intervals. Anyway, the method needs to know the other sensors values of the network, as valve position, pumps status and levels tank, to correlate with the flow. This methodology is been applied in the drinking water of Aigües del Ter-Llobregat (ATLL)