Online-Simulation of water distribution networks allows for estimating the current state of the entire network in near real-time. Measurement data coming from sensors at selected positions in the real network are used for driving a mathematical simulation model. Therefore the information gained from the measurements is extended and covers the whole system. As part of online monitoring or decision support systems online-simulations have multiple applications in operations and control of water supply networks. Although sensors and techniques for data transfer as well as mathematical simulation techniques are highly developed the practical applicability of online-simulations for decision support in large networks still suffers from the high time requirements for the whole cycle of measurement data updates, simulation and post-processing. One common approach to tackle this problem is the aggregation of the underlying models. However, for some applications like contaminant source identification the information that is lost can be crucial for the reliability of the decisions that are made based on the online-simulation results. In order to enhance online calculations and to improve the practical applicability of large online simulation models an adaptive calculation framework has been developed that allows for running the model with different levels of accuracy but using all one and the same data base. For each problem an adequate level of accuracy is chosen. Higher functions like source identification or optimisation algorithms that require a number of extra simulations can be focused on selected regions of the network. For the subnetwork in question detailed data are used whereas the rest of the system is omitted or, if necessary, considered with a lower level of accuracy. The framework is based on topological analysis and decomposition of the network graph. The paper describes the basic concepts and demonstrates its applicability by means of contaminant source identification.
Deuerlein, Jochen W.; Montalvo Arango, Idel; Piller, Olivier A.; Ung, Hervé; and Gilbert, Denis, "Adaptive Modeling Of Water Supply Networks For Improved Practical Applicability Of Hydraulic Online-Simulation" (2014). CUNY Academic Works.