Inverse methodology for calibration of numerical models of water supply systems

• Main Author: Bertelo, António Pedro Tão Nunes
• Publication Date: 2018
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10773/24612
• topic: Water supply networks; Calibration; Inverse model; Nodal demand; Optimization

Since the costs of the water management companies cover the costs of the operations of the water distribution system, it is, therefore, necessary to adapt the water consumption of the consumers with the best price negotiated by the companies. The calibration of nodal demands of a water distribution system is a process of approximation of the predicted values calculated through a hydraulic model to the observed values. Two methods for calibration of nodal demands were analyzed: i) Classical Inverse method and ii) Inverse model method. The Inverse model method incorporates the Gradient method developed by Todini. This method is known for solving the system of mass and energy balance equations. The inverse model presents itself as more efficient and is constituted by the separation of known and unknown variables from the continuity and energy equations of the water supply network. Nodal demands with similar characteristics are aggregated to make the model determined. The Gauss-Newton method is then applied to solve the model. Both methods were analyzed through a case study of a simple hydraulic network for a transient and permanent regime.