Gabrys, B. and Bargiela, A., 1997. Integrated Neural Based System for State Estimation and Confidence Limit Analysis in Water Networks. In: The 8th European Simulation Symposium, Ess 96. Society for Computer Simulation, pp. 398402.
Full text available as:

PDF
Gabrys_Bargiela_1996_Int_SE_CL_Water_Systems.pdf  Accepted Version 54kB 
Abstract
In this paper a simple recurrent neural network (NN) is used as a basis for constructing an integrated system capable of finding the state estimates with corresponding confidence limits for water distribution systems. In the first phase of calculations a neural linear equations solver is combined with a NewtonRaphson iterations to find a solution to an overdetermined set of nonlinear equations describing water networks. The mathematical model of the water system is derived using measurements and pseudomeasurements consisting certain amount of uncertainty. This uncertainty has an impact on the accuracy to which the state estimates can be calculated. The second phase of calculations, using the same NN, is carried out in order to quantify the effect of measurement uncertainty on accuracy of the derived state estimates. Rather than a single deterministic state estimate, the set of all feasible states corresponding to a given level of measurement uncertainty is calculated. The set is presented in the form of upper and lower bounds for the individual variables, and hence provides limits on the potential error of each variable. The simulations have been carried out and results are presented for a realistic 34node water distribution network.
Item Type:  Book Section 

ISBN:  9781565550995 
Number of Pages:  56 
Subjects:  Generalities > Computer Science and Informatics > Artificial Intelligence Generalities > Computer Science and Informatics Technology > Engineering > General Engineering 
Group:  Faculty of Science and Technology 
ID Code:  9644 
Deposited By:  Professor Bogdan Gabrys 
Deposited On:  11 Mar 2009 22:03 
Last Modified:  10 Sep 2014 14:44 
Downloads
Downloads per month over past year
Repository Staff Only  