Extraction of Knowledge from Artificial Neural Networks modeled to produce Optimized Operational Rules of Water Supply Systems

• Grant number: 12/25413-3
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: July 01, 2013
• End date: March 31, 2014
• Field of knowledge: Engineering - Civil Engineering - Hydraulic Engineering
• Principal Investigator: Luisa Fernanda Ribeiro Reis
• Grantee: Frederico Keizo Odan
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The present research project proposes the creation of a new methodology to optimize the real time operation of a water supply system in order to reduce operational costs. It will be used a modeled Artificial Neural Network (ANN) to directly reproduce the optimized operation using data obtained from a supply system.For a better understanding of the factors that influence the operation, a tool will be used to extract the knowledge captured by the ANN, which is like a black box, i.e., the relationship between input and output data is complex and is not easily comprehensible.It is expected that the proposed methodology will accelerate the optimizing process to obtain the optimized operation rules in time to use it on the real time operation, and will also give a better understanding of the factors that contributes to operation with reduced costs.