Smart water supply system: a quasi intelligent diagnostic method for a distribution network

Abstract

Constantly developing monitoring systems provide a large amount of raw data. In many cases, the operators of water supply systems (WSS) have already reached their perception limit for analysing information. Therefore, the managing process of the WSS requires quasi-intelligent informatics systems, the main purpose of which is to minimise the WSS operating costs in addition to maintaining the proper water delivery to customers. It can be achieved by the detection of abnormal functioning of WSS operations (leakages, water outages). The standard SCADA monitoring systems, in most cases, are not able to distinguish a significant water leakage and water tank filling process. Such cases occur relatively often in complex water supply systems with many water tanks. The aim of this paper is to present the quasi intelligent method of detecting abnormal WSS functioning, including its concepts and effects after a 3 month period operation. The essence of the detection method is the integration of numerical model (built-in Bentley WaterGEMS software) and SCADA monitoring system. The monitoring data are constantly compared to the simulation results and when accepted accordance limits are exceeded, the appropriate alerts are generated. Such solution cause the WSS operator does not need to analyse SCADA system indications constantly. The additional application of the method enables the detection of essential water leakages.