.png)
WDNetGIS exploits GIS interface that allows great flexibility to users working on the analysis, design and management of water networks. WDNetGIS, is fully integrated with WDNetXL, can import data from standard * .inp EPANET files, generable by all commercial software. It is equipped with a 3D viewer with data query function, and integrates completely with WDNetXL by means of shapefiles.
The WDNetXL is an integrated system for water distribution network (WDN) analysis, planning and management distributed as MS-Excel® add-ins. It integrates advanced and robust WDN hydraulic simulation with topological analysis and optimization strategies to support technicians for complex WDN analysis, design and management problems. The WDNetXL system permits to realize just-in-time technology transfer from technical research to WDN management sector through a holistic platform, which is ready for possible extensions to the latest innovations, thus providing an upgradable support to match current and future technical needs.
The availability of large and detailed databases together with the increased computational capabilities has motivated researchers to propose innovative techniques and methodologies to mine information from data. The Evolutionary Polynomial Regression (EPR) [Giustolisi and Savic (2006)] has been introduced in the hydroinformatics community as a hybrid data-driven technique, which combines the effectiveness of genetic algorithms with numerical regression for developing simple and easily interpretable mathematical model expressions. The multi-objective search paradigm has been introduced [Giustolisi and Savic (2009)] for developing multiple models by simultaneously optimizing fitness to training data and parsimony of resulting mathematical expressions. Such improvement allows for a sudden understanding of existing patterns in data
ANN MOGA (ANNs by Multi-Objective Genetic Algorithm) is a tool developed on the homonymous modelling methodology based on the ANNs paradigm [Giustolisi and Simeone (2006)]. The tool employs a particular structure of ANN named the Input-Output Neural Network (IONN). In particular, it is based on a MOGA approach for construction of IONN models, which prevents potential overfitting troubles caused by poor generalization capabilities of the identified ANNs [Giustolisi and Laucelli (2005)]. This can be obtained by minimizing the model’s input dimension and the number of hidden neurons (flexibility) while preserving fitness properties.