基于分级免疫萤火虫算法的桥梁振动传感器优化布置研究

针对结构健康监测系统设计的振动传感器优化布置问题，在基本萤火虫算法的基础上引入等级划分策略和免疫机制，提出了一种分级免疫萤火虫算法。采用二重结构编码，弥补了基本萤火虫算法只能用于连续优化问题的不足；建立等级划分制度，使不同等级种群内部形成独立的搜索空间，维持了种群多样性，并让优质个体得以保留；引进免疫机制，进行萤火虫的选择、记忆、交叉和变异，增强了算法的全局搜索能力和局部寻优能力；文末利用足尺Benchmark桥梁模型，对算法参数进行了敏感性分析，并开展了振动传感器优化布置方案的选择。结果表明，与基本离散型萤火虫算法相比，分级免疫萤火虫算法的计算效率和寻优结果均有显著提升，能够很好地解决振动传感器优化布置问题。

OPTIMAL VIBRATION SENSOR PLACEMENT FOR BRIDGES USING GRADATION-IMMUNE FIREFLY ALGORITHM

To find the optimal vibration sensor placement(OVSP) during designing a structural health monitoring system, the gradation-immune firefly algorithm(GIFA) was proposed by introducing the gradation strategy and the immune pattern to improve the original firefly algorithm. The dual-structure coding method was employed to overcome the shortage that the original firefly algorithm can only be applied to optimal problems with continuous variables. The gradation strategy was developed to limit individuals with different gradations in their respective search space. As a result, the diversity of population is ensured and the individuals with good performance are inherited. Furthermore, the immune pattern is utilized to perform selecting, memorizing, crossing and mutating for fireflies and enhance the capability of global searching and local optimization of the GIFA. A full-scale benchmark cable-stayed bridge was employed as a case study. The parametric sensitivity of the proposed GIFA was discussed and the optimal sensor configurations were offered. The results indicate that the computational efficiency of the GIFA and the quality of optimal solutions provided by the GIFA are dramatically improved when comparing with the simple discrete firefly algorithm. The GIFA is an excellent approach to solve OVSP problems.