(159f) Efficient Optimization of Particulate Systems Described by Monte Carlo Method: Determination of Nucleation Parameters from Kinetic Data

In this paper the inverse problem is solved to determine silver and gold nucleation parameters from kinetic data. The kinetic data is obtained using a stopped flow reactor and the change of size with time is determined using surface plasmon theory. The dynamic of the metal formation is simulated using MC simulation. Solving the inverse problem of this system is very difficult because the model is given in a form of a MC simulation, which accurate solutions are time consuming. To make optimization feasible two approaches are presented here: (a) Use of coarse-grained methods (PEMC and t-PEMC) recently introduced in the literature ([1], [2]) and (b) Modify the stochastic algorithm to optimize systems with noise (make rough MC simulations). The algorithm used to solve the optimization problems is based on artificial chemical plant paradigm [3]. Comparison of the results, in terms of CPU time and parameters obtained are presented using both approaches.

[1] R. Irizarry, Fast Monte Carlo Methodology for Multivariate Particulate Systems-I: Point Ensemble Monte Carlo (2008) Chemical Engineering Science 63, 95-110.

[2] R. Irizarry, Fast Monte Carlo Methodology for Multivariate Particulate Systems-II: t-PEMC (2008) Chemical Engineering Science 63, 111-121.

[3] R. Irizarry, LARES: An Artificial Chemical Process Approach for Optimization (2004) Evolutionary Computation Journal, 12 (4), 435-460.