Algorithm 3 Pseudo-code for constructing a local radial basis function model.

1:if update the local surrogate model then   2:      The information of the local surrogate model at this time is kept for the add-point strategy;   3:end if   4:for each ${\mathrm{x}}_{\mathrm{i}}^{\mathrm{t}}$ do   5:      Use Euclidean distance to find the n nearest neighbors of the individual from the database and form the set ${\mathrm{NP}}_{\mathrm{i}}$;   6:end for   7:Merge the sets generated by each individual to form the training set $\mathrm{NP}$;   8:Constructing a new local surrogate model with the training set $\mathrm{NP}$;