Joint state and parameter estimation under bounded measurement error for water networks

Considers problems of joint state and parameter estimation in general water network based on its nonlinear mathematical model and bounds on measurement inaccuracies. It is well known that the network model is given originally in a form of nonlinear implicit equations and its explicit analytical form is not available. The explicit input-output mapping is available only numerically and its advantage is that so called intermediate variables are eliminated from the model. A drawback is that numerical properties of iterative algorithms in a reduced space are usually worse. The author derives estimation algorithms by using an implicit model of a network although the problem dimensionality will then increase