A Method for Improved Efficiency in the Dynamic Modelling of Hydraulic Networks.pdfVIP

A Method for Improved Efficiency in the Dynamic Modelling of Hydraulic Networks JE van Zyl, DA Savic, GA Walters Centre for Water Systems, University of Exeter, UK www.ex.ac.uk/ESE/WaterSystems/ j.e.van-zyl@exeter.ac.uk Abstract The authors report on work in which the efficiency and accuracy of dynamic analysis of water supply systems may be improved by utilizing a new, two pronged approach: Firstly, an assumption is made which results in the general reservoir equation reducing to a nonhomogeneous linear differential equation, for which an analytical solution may be found. Secondly, a general polynomial description of th e known demand behaviour is incorporated into the method. The resulting set of equations is applied in conjunction with static analyses to describe implicitly the system’s hydraulic behaviour. The method is applied to an example problem to show that significant improvements in accuracy and speed of dynamic analysis may be achieved. Introduction The static hydraulic behaviour of water distribution systems is highly nonlinear, requiring analysis by a computationally expensive iterative process. The dynamic behaviour of these systems is described by a nonlinear nonhomogeneous differential equation for which no analytical solution is available, requiring another iterative scheme and further increasing the computational burden. In cases where large numbers of dynamic simulations are required and/or computational efficiency is a principal requirement, such as operational optimization and real-time control, the computational cost can severely limit the efficacy of the method. To minimize the number of static simulations, dynamic hydraulic analysis is conventionally done by employing a simple explicit Euler integration