多模态遗传算法研究及其进展

在对多模态遗传算法（MGA）的研究历史和经典算法进行综述的基础上,展望了MGA的研究进展,重点对笔者提出的几种适于多峰值求解的遗传算子及其构成的算法特性进行了分析和比较,指出了多模态遗传算法在工程实践中的重要作用。     

Bi-/multi-modal pore formation of PLGA/hydroxyapatite composite scaffolds by heterogeneous nucleation in supercritical CO2 foaming

Scaffolds with multimodal pore structure are essential to cells differentiation and proliferation in bone tissue engineering.Bi-/multi-modal porous PLGA/hydroxyapatite composite scaffolds were prepared by supercritical CO_2 foaming in which hydroxyapatite acted as heterogeneous nucleation agent.Bimodal porous scaffolds were prepared under certain conditions,i.e.hydroxyapatite addition of 5%,depressurization rate of 0.3 MPa·min~(-1),soaking temperature of 55℃,and pressure of 9 MPa.And scaffolds presented specific structure of small pores(122 μm±66 μm)in the cellular walls of large pores(552μm±127μm).Furthermore,multimodal porous PLGA scaffolds with micro-pores(37 μm±11 μm)were obtained at low soaking pressure of 7.5 MPa.The interconnected porosity of scaffolds ranged from(52.53±2.69)% to(83.08±2.42)%by adjusting depressurization rate,while compression modulus satisfied the requirement of bone tissue engineering.Solvent-free CO_2 foaming method is promising to fabricate bi-/multi-modal porous scaffolds in one step,and bioactive particles for osteogenesis could serve as nucleation agents.     

Multimodal Luminescent Yb3+/Er3+/Bi3+-Doped Perovskite Single Crystals for X-ray Detection and Anti-Counterfeiting

Anti-counterfeiting techniques have become a global topic since they is correlated to the information and data safety, in which multimodal luminescence is one of the most desirable candidates for practical applications. However, it is a long-standing challenge to actualize robust multimodal luminescence with high thermal stability and humid resistance. Conventionally, the multimodal luminescence is usually achieved by the combination of upconversion and downshifting luminescence, which only responds to the electromagnetic waves in a limited range. Herein, the Yb³⁺/Er³⁺/Bi³⁺ co-doped Cs₂Ag_0.6Na_0.4InCl₆ perovskite material is reported as an efficient multimodal luminescence material. Beyond the excitation of ultraviolet light and near-infrared laser (980 nm), this work extends multimodal luminescence to the excitation of X-ray. Besides the flexible excitation sources, this material also shows the exceptional luminescence performance, in which the X-ray detection limit reaches the level of nGy s⁻¹, indicating a great potential for further application as a colorless pigment in the anti-counterfeiting field. More importantly, the obtained double perovskite features high stability against both humidity and temperature up to 400 °C. This integrated multifunctional luminescent material provides a new directional solution for the development of multifunctional optical materials and devices.     

基于朴素贝叶斯模型的新闻故事分割方法

提出一种基于朴素贝叶斯模型的新闻视频故事分割方法。通过对新闻视频进行镜头检测,获得候选故事边界点,从候选边界点周围镜头提取多模态中级特征,形成属性集合作为输入,应用朴素贝叶斯模型对候选边界点进行分类后对结果进行后处理,得到新闻故事。实验结果表明,该方法获得了较高的查准率和查全率,对不同类型的新闻节目有良好的适应性。     

多模噪声中粒子滤波算法研究

粒子滤波是一种解决非高斯滤波问题的有效方法,受到许多领域的研究人员的重视。在扩展卡尔曼滤波(EKF)的基础上,提出一种基于多层感知器(MLP)的扩展卡尔曼滤波算法。利用扩展卡尔曼粒子滤波器和MLP对当前时刻状态重要性采样,引入MLP对样本进行重采样。该算法能有效利用测量值的最新信息,对状态估计的误差更小。在实验中,对于多模噪声非线性系统,该算法与另外算法进行比较。结果证明,所提算法性能优异于其他算法。     

A memetic particle swarm optimisation algorithm for dynamic multi-modal optimisation problems

Many real-world optimisation problems are both dynamic and multi-modal, which require an optimisation algorithm not only to find as many optima under a specific environment as possible, but also to track their moving trajectory over dynamic environments. To address this requirement, this article investigates a memetic computing approach based on particle swarm optimisation for dynamic multi-modal optimisation problems (DMMOPs). Within the framework of the proposed algorithm, a new speciation method is employed to locate and track multiple peaks and an adaptive local search method is also hybridised to accelerate the exploitation of species generated by the speciation method. In addition, a memory-based re-initialisation scheme is introduced into the proposed algorithm in order to further enhance its performance in dynamic multi-modal environments. Based on the moving peaks benchmark problems, experiments are carried out to investigate the performance of the proposed algorithm in comparison with several state-of-the-art algorithms taken from the literature. The experimental results show the efficiency of the proposed algorithm for DMMOPs.     

Research on Modified Shifting Balance Genetic Algorithms

The increasing overlap of core and colony populations during the anaphase of evolution may limit the performance of shifting balance genetic algorithms. To decrease such overlapping,so as to increase the local search capability of the core population,the sub-space method was used to generate uniformly distributed initial colony populations over the decision variable space. The core population was also dynamically divided,making simultaneous searching in several local spaces possible. The algorithm proposed in this paper was compared to the original one by searching for the optimum of a complicated multi-modal function. The results indicate that the solutions obtained by the modified algorithm are better than those of the original algorithm.     

一种新的混合免疫算法

免疫算法是在免疫系统识别多样性的启发下所设计出的一种新的多峰值函数的寻优算法。文章针对目前人工免疫算法在收敛于全局最优方面存在的缺陷,提出了一种模拟退火免疫算法,引入Boltzmann退火选择算子,利用模拟退火算法可选择性接受恶化解的特点克服传统免疫算法的缺陷,并通过实验证明了该混合免疫算法的优越性能。     

一种改进的演化算法

将GT算法和粒子群优化（PSO）算法结合并加以改进,采用non uniform变异算子提高局部搜索能力和算法的稳定性,同时引入种群划分等策略,构建了一种新的演化算法(记为GT POPDM PSO)。该算法比上述两种算法具有更好的性能,特别是对多峰函数优化等问题计算效果更好。