高斯混合模型结合加权似然的目标跟踪算法

鉴于高斯混合模型对背景变化快时无法精确检测出目标和目标跟踪的适应性差等瑕疵，提出了基于加权似然跟踪器来改进高斯混合模型实现运动目标跟踪算法。主要引入了自适应高斯混合模型来实时检测运动目标，然后空间加权似然来进行视频中的目标定位，引入加权似然期望值来改进高斯混合模型处理视频中的多尺度、多角度变化的目标跟踪不精准问题。通过VOT 2014 dataset对比实验结果表明提出的基于加权似然跟踪（Weighted Likelihood Tracking，WLT）和改进高斯混合模型（Improved Gaussian Mixture Model，IGMM）的目标跟踪算法较传统高斯混合模型跟踪算法在跟踪的精度有较大提高。在应对多尺度、多角度变化的目标跟踪表现出了较大的优势。     

Fermentation of table cream by Lactobacillus plantarum strains: effect on fungal growth,aflatoxin M1 and ochratoxin A

Aflatoxin M1 (AFM₁) and ochratoxin A (OTA) are two main mycotoxins in milk and dairy products. In the present work, the ability of four Lactobacillus strains (L. plantarum PTCC 1058, L. plantarum LP3, L. plantarum AF1 and L. plantarum LU5) to remove AFM1 and OTA in fermented cream was studied during 24 h fermentation. The antifungal activity of the mentioned lactobacilli against the defined fungi (Aspergillus flavus PTCC 5004, Aspergillus parasiticus PTCC 5018, Aspergillus nidulans PTCC 5014, Aspergillus ochraceus PTCC 5060) was also evaluated. The results showed that the cell counts of all strains were increased by 64–70% during fermentation. All Lactobacillus strains decreased the amount of AFM₁ significantly (P ≤ 0.05) in the range of 26–52%, which the highest AFM₁-reducing effect was related to L. plantarum LU5 (from 0.5 to 0.24 μg kg⁻¹). The mean OTA removal by Lactobacillus strains in fermented cream also ranged from 32 to 58%. Amongst Lactobacillus strains, the cell-free culture supernatants of L. plantarum LU5 showed the highest (inhibition zone of 26.7 ± 1.2 mm) and L. plantarum LP3 and L. plantarum PTCC 1058 the lowest antifungal activities. The fermented creams contained Lactobacillus strains exhibited the highest and lowest antifungal activities against A. ochraceus and A. parasiticus, respectively. L. plantarum LU5, with the inhibition zone of 27.6 ± 0.9 mm, was the most effective fungal inhibitor, while L. plantarum PTCC 1058 had the lowest antifungal activity.     

基于改进飞蛾扑火优化算法的MIMO雷达相位编码信号集设计

提出一种MIMO雷达相位编码信号集设计方法，该方法对飞蛾扑火优化算法进行改进，利用改进的飞蛾扑火算法优化相位编码信号集，提高其正交性.在改进方法中，增加对火焰的更新过程，使火焰不仅能够保存最优飞蛾位置还具有搜索最优解的功能.优化过程以自相关函数旁瓣峰值和、互相关函数峰值和，以及自相关函数旁瓣能量、互相关函数能量作为算法适应度函数.仿真实验结果表明，改进方法明显提高了飞蛾扑火优化算法的收敛速度和收敛精度，且优化得到的信号集具有较低的自相关旁瓣值和互相关值.     

平面并联机构工作空间的三维螺旋扫描CAD求解

针对平面并联机构无奇异位置工作空间求解困难、过程繁琐、计算量大等问题，提出了基于CAD求解平面并联机构工作空间的三维螺旋扫描方法。将[n]自由度平面并联机构分解成[n]条支链进行独立分析，得到每条支链下末端执行器的可达区域，再将所有支链可达区域取交集即为平面并联机构工作空间。应用SolidWorks软件建立平面并联机构模型，进行几何特征处理，通过自动求解器求解，将求解过程图形化，快速得到同轴布局5R机构和平面3-RPR并联机构的无奇异位置工作空间。通过同轴布局5R机构的运动学实验，验证了该求解方法的可行性。     

基于过程免疫力的动态电压恢复器优化配置

优质电力园区以最低成本满足用户对电能质量的特殊需求,成为解决电压暂降问题的新途径,已在国内外开展了广泛的试点应用。针对中国优质电力园区规划缺少经济高效的优化配置方法的实际情况,提出了一种基于过程免疫力的动态电压恢复器优化配置方法。基于工业过程电压暂降免疫力,以满足用户用电需求为目标,以治理设备购置费用最低为约束,采用遗传算法对园区供电质量分级制定动态电压恢复器的优化配置方案,实现工业园区电压暂降治理的经济性。对大型精密温控系统的电压暂降治理装置的优化配置方法进行仿真计算,验证了所提方法的有效性与经济性。     

基于克里金理论的辐射剂量场插值重构与可视化研究

使用基于克里金理论的插值计算方法对均匀网格与随机排列等不同分布类型的辐射剂量场进行了数据重构与可视化研究,并将重构结果与常用的反距离加权插值法进行了对比,分析了不同插值方法对辐射剂量场重构精度与可视化显示效果的影响。结果表明:对于多数情况,克里金插值法在辐射剂量场重构的精度与可视化显示的效果上均明显优于反距离加权插值法,算例中克里金插值法的结果与原始数据之间的偏差在10%以内,而反距离加权插值法的结果与原始数据之间的偏差基本在15%以上;但在辐射剂量场梯度变化较大且已知数据极少的区域,克里金插值法的计算结果偏差会增大,此时可使用反距离加权插值作为其补充,综合使用。本文的研究工作,验证了克里金插值法在辐射剂量场重构与可视化方面的应用潜力,可为推动辐射剂量场可视化仿真技术的发展提供技术支持。     

Generalised multivariable supertwisting algorithm

The purpose of this paper is to present an extension of the generalised supertwisting algorithm (STA) to the multivariable framework. We begin by introducing an algorithm that may be deemed as a linear, quasicontinuous, or discontinuous multivariable system, depending on the functions that define them. For the class represented by such an algorithm we prove the robust, Lyapunov stability of the origin and characterise the perturbations that preserve its stability. In particular, when its vector field is discontinuous or quasicontinuous our algorithm is endowed with finite‐time stability. Due to its resemblance to the scalar case, we denote such finite‐time stable systems as generalised multivariable STA. Furthermore, the class of finite‐time stable systems comprise the currently available versions of STAs. To finalise, by means of simulation examples, we show that our proposed finite‐time stable algorithms are well suited for signals online differentiation and highlight their dynamical traits.     

基于博弈分析的流域水资源配置

针对传统规划方法难以体现水资源配置过程中多种复杂因素的问题,基于博弈理论建立宏观调控和市场经济相结合的水资源配置动态博弈模型,求解其子博弈精炼纳什均衡下各博弈参与人的策略,并运用改进量子遗传算法求解该博弈模型。对A河流域水资源配置结果表明,在保证流域公共生态用水2.0×10~8 m~3的基础上,各用水子区自发约束自身的取水量,达到均衡时,社会总收益为11.27亿元,相比自由取水模型,社会总收益增加了0.95亿元,增幅为9.2%。     

Hierarchical Learning Water Cycle Algorithm

In order to improve the global searching ability of Water Cycle Algorithm (WCA), the hierarchical learning concept is introduced and the Hierarchical Learning WCA (HLWCA) is proposed in this paper. The underlying idea of HLWCA is to divide the solutions into collections and give these collections with hierarchy differences. One of the collections has a higher hierarchy than others and utilizes an exploration-inclined updating mechanism. The solutions in this high hierarchy collection are the exemplars of other collections. The other collections are sorted according to the exemplars’ function value and the solutions in these collections actively choose whether to follow their own exemplar or not. Through different updating mechanisms of collections, the global searching ability is improved while the fast convergence and strong local search ability of WCA are retained. The proposed HLWCA is firstly experimented on IEEE CEC 2017 benchmark suite to testify its performance on complex numerical optimization tasks. Then, it is tested on four practical design benchmark problems to verify its ability of solving real-world problems. The experimental results illustrate the efficiency of the proposed algorithm.     

Probabilistic tree-based representation for solving minimum cost integer flow problems with nonlinear non-convex cost functions

The minimum cost flow problem (MCFP) is the most generic variation of the network flow problem which aims to transfer a commodity throughout the network to satisfy demands. The problem size (in terms of the number of nodes and arcs) and the shape of the cost function are the most critical factors when considering MCFPs. Existing mathematical programming techniques often assume the cost functions to be linear or convex. Unfortunately, the linearity and convexity assumptions are too restrictive for modelling many real-world scenarios. In addition, many real-world MCFPs are large-scale, with networks having a large number of nodes and arcs. In this paper, we propose a probabilistic tree-based genetic algorithm (PTbGA) for solving large-scale minimum cost integer flow problems with nonlinear non-convex cost functions. We first compare this probabilistic tree-based representation scheme with the priority-based representation scheme, which is the most commonly-used representation for solving MCFPs. We then compare the performance of PTbGA with that of the priority-based genetic algorithm (PrGA), and two state-of-the-art mathematical solvers on a set of MCFP instances. Our experimental results demonstrate the superiority and efficiency of PTbGA in dealing with large-sized MCFPs, as compared to the PrGA method and the mathematical solvers.