本项目针对在现实世界中普遍存在的且具有高度复杂特征的动态优化问题开展智能生物地理学优化算法(Biogeography-based Optimization, BBO)参数自适应及变结构模式研究。通过比较性数值方法及数学分析手段探究BBO算法在典型动态优化问题中的参数及结构的性能特征，引导设计具有参数自适应及结构自调整能力的智能BBO算法。参数自适应的目的在于使得算法具有环境引导能力的智能优化特征，变结构功能使算法结构的自适应具有引导算法收敛方向及保持种群多样性的能力，对种群多样性进行合理的监控与评估，适时实现种群多样性的保持与恢复，从而保证算法具有高效求解动态优化问题的能力。项目将基于典型动态优化问题开展算法性能的验证，有助于完善算法的优化能力。该设计将为BBO算法及其它进化计算方法的进一步发展提供借鉴和参考。

This research focuses on the design of parameter self-adaption and variable structure for Biogeography-based Optimization (BBO) to solve dynamic optimization problems which are complicated and common in real world. In this research, parameters and structures in BBO are to be investigated to reveal their effects to performance by numerical comparison and mathematical analysis. To enhance BBO’s performance in dynamic optimization, parameter self-adaption strategy and a variable structure are proposed. Design of adaption strategy is to enhance BBO’s ability to deal with changes in dynamical environment, while variable structure makes BBO play corresponding roles in different steps of evolutionary process, which take into account both convergence direction and population diversity. To avoid stagnation and premature, population diversity will be evaluated and maintained during the whole optimization process. By employing benchmarks and several practical problems, the proposed algorithm will be tested and the results will be analyzed by statistics. The comparisons between proposed algorithm and other classical evolutionary algorithms will be conducted to improve the designs. The ideas in this research will also provide experiences and inspirations for other evolutionary algorithm in dealing with dynamic optimization problems.