An Application of Desulphurization Pretreatment of Molten Iron using Parallel Kernel Regression RBF NN

Kernel regression of RBF NN building on the notion of density estimation is frequently used for modeling prediction. But kernel matrix computation for high dimensional data source demands heavy computing power. To shorten the computing time, the paper designs a parallel algorithm to compute the kernel function matrix of kernel regression of RBF NN. The proposed algorithm has been applied to desulphurization pretreatment of molten iron in metallurgical process to build the prediction of desulphurization pretreatment modeling. The paper then implements the algorithm on a cluster of computing workstations using MPI. Finally, we experiment with the practical data to prove the speedups and accuracy of the algorithm