Adaptive Sampling Detection Based Immune Optimization Approach and Its Application to Chance Constrained Programming

摘要

This work investigates a bio-inspired adaptive sampling immune optimization algorithm to solve linear or nonlinear chance-constrained optimization problems without any noisy information. In this optimizer, an efficient adaptive sampling detection scheme is developed to detect individual's feasibility, while those high-quality individuals in the current population can be decided based on the reported sample-allocation scheme; a clonal selection-based time-varying evolving mechanism is established to ensure the evolving population strong population diversity and noisy suppression as well as rapidly moving toward the desired region. The comparative experiments show that the proposed algorithm can effectively solve multi-modal chance-constrained programming problems with high efficiency.