Determination of the parameters of the model of an ion-exchange sorption model (Q1323942)

The authors study the determination of the parameters of a model of ion exchange in a filter filled with a porous sorbent. The model is widely used in problems of physical chemistry and, in particular, in mathematical modelling of the extraction of elements from sea water by means of ion-exchange sorption. The observation data comprises the values of the component concentrations at the outlet of the filters as functions of time \(u_ n(t)\). Suppose that \(u_ n(t)\) are known (measured). We denote the set of parameters of the model by \(p\), and consider the functional \[ J(p)=\sum^ N_{n=1} \alpha_ n\int^ T_ 0 \{c_ n(\ell,t)-u_ n(t)\}^ 2dt, \] where \(c_ n(\ell,t)\) is the solution of the direct problem with the chosen set of parameters \(p\) and \(\alpha_ n\) are given weighting factors. The inverse problem of finding the parameters of the system from experimental data is thus formulated and consists in determining \(\bar p\) such that \(J(\bar p)=\min J(p)\). A difference scheme of first-order accuracy for solving the direct problem is proposed. A gradient method turns out to be very efficient for solving the problem of minimizing the functional \(J(p)\), because of the possibility of an analytic representation of the gradient of the functional \(J(p)\) using the solution of the conjugate problem. An algorithm for solving the conjugate problem is also presented. In the last section of the article the efficiency of the proposed algorithm is illustrated by several numerical examples.