Approximate branch-and-bound global optimization using B-spline hypervolumes

This paper presents a B-spline-based branch-and-bound algorithm for unconstrained global optimization. The key components of the branch-and-bound, a well-known algorithm paradigm for global optimization, are a subdivision scheme and a bound calculation scheme. For these schemes, we first introduce a B-spline hypervolume to approximate an objective function defined in a design space, where the approximation is based on Latin-hypercube sampling points. We then describe a proposed algorithm for finding global solutions approximately within a prescribed tolerance. The algorithm includes two procedures that are performed iteratively until all stopping conditions are satisfied. One involves subdivision into mutually disjoint subspaces and computation of their bound information, both of which are accomplished by using B-spline hypervolumes. The other updates a search tree that represents a hierarchical structure of subdivided subspaces during the solution process. Finally, we examine the computational performance of the proposed algorithm on various test problems that cover most of the difficulties encountered in global optimization. The results show that the proposed algorithm is complete without using heuristics and has good potential for application in large-scale NP-hard optimization.     

Two phase flow 1D turbulence model for poly-disperse upward flow in a vertical pipe

A 1D test-solver was developed in recent years for modeling of two phase bubbly flows in pipe geometry. The solver considers a number of bubble classes and calculates bubble-size resolved void fraction profiles in the radial direction. A successful implementation was achieved regarding bubble forces models (non-drag forces). Discrepancies appeared when coalescence and breakup rates were significant. These rates depend upon local turbulence quantities, which are possible reason for discrepancies. Originally the test-solver is equipped by Sato model (Sato, Y., Sadatomi, M., Sekoguchi, K., 1981. Momentum and heat transfer in two-phase bubble flow. I. International Journal Multiphase Flow 7, 167–177 .) which accounts for turbulence via shear- and bubble-induced viscosities calculated out of empirical correlations. One equation for the turbulent kinetic energy was solved, while the dissipation rate was calculated out of a correlation. In order to improve calculation of the local turbulence parameters, a two-phase k– turbulence model was adopted instead. The account for the bubble-induced turbulence was made via a source term taken out of literature. Comparisons between new and old turbulence modeling against experimental data showed better agreement for the new model. The experiments covered a wide range of water and air superficial velocities for upward bubbly flow in two pipe's diameters: 50 and 200 mm. The main feature of the new model is providing more reliable values of turbulence parameters for application in coalescence and breakup models. A comparison with CFX 5.7 calculations in a 50 mm pipe showed better calculation results when the source term was considered in the k– equations. An implementation into CFX is planned.     

基于随机模型的光伏发电置信容量评估方法

随着光伏发电在电网中穿透功率水平的不断提高,光伏发电的置信容量评估问题逐渐成为电网规划中关注的新热点。针对光伏发电与电网负荷的不确定性,文中在序贯小时确定性模型的基础上建立了光伏出力与电网负荷的随机模型,利用序贯小时确定性模型反映光伏出力与电网负荷以及光伏电站之间在长时间尺度上的相关性,采用秩相关系数来描述短时间尺度上的相关性,提出了一种可处理随机变量相关性的基于拉丁超立方采样的光伏发电置信容量评估方法。通过对IEEE-RTS 79系统进行评估,验证了所提出方法的有效性和准确性。     

PBX炸药粘弹性损伤本构模型参数敏感度分析

建立了基于拉丁超立方抽样和Spearman秩相关系数的本构模型参数敏感度分析方法,实现对PBX炸药粘弹性损伤本构模型参数敏感度的整体性分析。以PBX炸药柱体压缩的分析为例,采用该方法确定了各本构参数对端面载荷值的敏感度及排序。并分析了抽样样本量,以及载荷加载速率对本构参数的敏感度的影响。同相关文献结果进行比较,证实了该方法的可行性。该文的方法可在参数识别中提前预测各参数对实验结果的敏感度,为本构模型参数识别中缩减待识别参数和选择合适的模拟实验提供理论基础。     

LH8000配电站（房）智能测控系统主要功能

针对配电设备点多面广、相对分散和不易管理的问题,在遵循模块化设计的前提下成功的研制出了新一代的LH8000配电站（房）智能监控系统,推动了我国电网管理自动化、信息化、集中化和智能化的发展。本文对LH8000配电站（房）智能测控系统主要功能进行介绍并做出解释。     

改进黑猩猩优化算法的测试数据生成研究

自动生成测试数据的关键在于能否生成覆盖率高、纠错能力强的数据。针对目前测试数据生成效率低及黑猩猩优化算法仍存在易陷入局部最优、收敛精度低等问题,提出一种正余弦扰动策略黑猩猩优化算法（chimp optimization algorithm for sine-cosine perturbation strategy,SC-ChOA）。使用拉丁超立方策略初始化种群,增强种群的多样化;引入非线性衰减收敛因子来平衡算法的全局和局部勘探能力;在位置更新时添加正余弦扰动因子,避免群体陷入局部范围搜索而导致的算法停滞现象。使用测试函数与标准黑猩猩优化算法及常用的遗传算法进行对比实验,验证算法的有效性;将改进算法应用到测试数据生成领域,通过在桩中插入分支函数来建立适应度函数,以促进测试数据的优化。为验证改进算法在测试数据生成方面的有效性,使用多个基准程序进行算法对比实验,结果表明SC-ChOA在测试数据生成的覆盖率、平均迭代次数和运行时间上均有明显优势。     

一种改进拉丁方抽样免疫遗传算法*

针对遗传算法求解问题中保持群体多样性能力不足、早熟、耗时长以及求解成功率低等缺点,依据拉丁方抽样方法对遗传算法中的交叉算子进行重新设计;结合免疫机理定义染色体浓度、设计克隆选择策略,提出了一种改进拉丁方抽样免疫遗传算法。利用旅行商问题以及最大子团问题为实例对新算法进行了验证,实验结果表明新算法在解的质量、收敛速度等各项指标上均好于经典遗传算法和佳点集遗传算法,说明了新算法的优越性与可行性。     

基于非子采样Contourlet变换的遥感图像融合算法

针对人类视觉特性, 以及全色高分辨图像和多光谱遥感自身的特点, 提出一种非下采样 Contourlet (NSCT) 域的图像融合新策略. NSCT 具有好的多分辨、移不变和多方向等特性, 能对图像中的边缘和围线信息给出渐近最优表示. 为了更好地保持空间分辨率和颜色分量, 引入基于 LHS 变换的亮度成分叠加策略. 实验结果表明: 本文提出的融合方法在提高空间分辨率的同时较好地保持了光谱信息. 与传统的 PCA 方法、基于 IHS 的融合方法、基于小波加权的融合方法, 以及同样采用本文的融合策略、分别基于小波变换和基于 Contourlet 变换的融合策略相比较, 本文方法在视觉效果和客观衡量指标两方面都有所改善.     

基于改进LHS模式的可靠性设计优化

在设计优化中,确定性优化由于没有考虑输入量的不确定性,其优化结果可能不可靠(不安全),因此基于可靠性的设计优化(reliability-based design optimization,RBDO)得到关注.然而可靠性设计优化计算量大,尤其对于高维问题.基于此,提出一种新方法--改进拉丁超立方体取样(Latin hyp...     

软包锂电池电芯封装铝塑膜外壳拉深工艺

通过拉深试验对铝塑膜材料特性进行分析,在DYNAFORM软件中定义铝塑膜材料属性并进行仿真试验。结合单因素试验和正交试验对影响铝塑膜拉深成形性能的各工艺参数进行显著性分析,采用响应曲面法、拉丁超立方试验设计和多目标粒子群优化算法相结合的方法对影响铝塑膜成形性能显著的参数（如压边力、模具圆角半径、摩擦因数和拉深速度）进行优化,优化后的铝塑膜拉深成形时,其壳体最薄处厚度为55 μm。试验验证了铝塑膜拉深成形工艺优化的结果可行。