An evolutionary algorithm for synthesizing optical thin-film designs

This paper presents an evolutionary approach, called Family Competition Evolutionary Algorithm (FCEA). to design optical thin-film multilayer systems. FCEA integrates self-adaptive mutations, decreasing-based mutations, and four-layer selections to balance exploration and exploitation. One antireflection coating and one narrow-band rejection filter are presented to demonstrate that our approach is a powerful technique. Our approach consistently performs better than other evolutionary algorithms and other published results on these two problems. From experimental results of antireflection coating, our optimal solutions exhibit a pronounced semiperiodic clustering of layers and these solutions also confirm the theoretical prediction between the optical thickness and the best achievable reflectance.