Immune optimization algorithm in noisy environments solving chance constrained programming

This work investigates a simple immune optimization algorithm in noisy environments for chance constrained programming problems without a priori noisy information. It bases on stochastic simulation and some immune metaphors in the clonal selection principle. The key of the algorithm is to design an adaptive sample allocation scheme and to construct the immune operators of dynamic proliferation and adaptive mutation which strengthen the abilities of noisy compensation and local and global search. Comparative Experiments show that the proposed approach can achieve satisfactory performances including optimized quality, noisy suppression and performance efficiency.