约束非负矩阵分解框架下高维自适应粒子群端元提取

传统基于约束非负矩阵分解NMF(Nonnegative Matrix Factorization)的高光谱端元提取算法一般存在两个问题:一方面,以固定惩罚系数方法处理端元提取的约束优化问题,难以较好权衡目标项与约束项间的关系,影响提取效果;另一方面,求解过程通常基于梯度算法,依赖于初始值和步长的设定,容易陷入局部最优。针对这些问题本文提出约束NMF框架下高维自适应粒子群端元提取算法HAPSO(High-dimension Adaptive Particle Swarm Optimization)。该算法在端元距离最小约束的NMF框架下,利用粒子群算法PSO替代原梯度算法以增强全局搜索能力;采用高维PSO方法解决了多波段高维问题,并结合种群信息构建自适应惩罚机制以实现端元提取中目标与约束的合理权衡。通过模拟影像和真实影像的实验,证实该算法与传统的NMF端元提取算法相比能够更合理地权衡约束和避免局部最优,具有较好的端元提取效果。     

Hydraulic geometry and maximum flow efficiency as products of the principle of least action

Basic flow relationships have previously been seen to be insufficient to explain the self‐adjusting mechanism of alluvial channels and as a consequence extremal hypotheses have been incorporated into the analyses. In contrast, this study finds that by introducing a channel form factor (width/depth ratio), the self‐adjusting mechanism of alluvial channels can be illustrated directly with the basic flow relations of continuity, resistance and sediment transport. Natural channel flow is able to reach an optimum state (Maximum Flow Efficiency (MFE), defined as the maximum sediment transporting capacity per unit available stream power) with regard to the adjustment of channel form such that rivers exhibit regular hydraulic geometry relations at dominant or bankfull stage. Within the context of MFE, this study offers support for the use of the concepts of maximum sediment transporting capacity (MSTC) and minimum stream power (MSP). Furthermore, this study indicates that the principle of least action is able to provide a physical explanation for the existence of MFE, MSTC and MSP. Potential energy is minimized and consequently sediment transport is maximized in alluvial channels. Copyright © 2000 John Wiley & Sons, Ltd.     

基于Surpac的铜陵新桥矿床露天开采境界优化

矿床的境界优化就是要设计一个最为接近理想状态的模型,使矿山企业利润最大化,在满足经济和边坡角几何约束的条件下,设计一个最终开采境界.基于三维地质矿业软件,可以将三维可视化技术与矿床的后期开采、生产进度编排进行紧密的结合,完成矿体形态与工程实体的衔接,其不同于传统境界优化的方法,可以在三维空间内进行境界优化的应用和研究.本研究选取SURPAC三维地质矿业软件系统,结合铜陵新桥矿床的实际需求,基于三维数字矿山模型,使用L-G算法规范的数学逻辑结构对露天采场进行境界优化.其结果可以为扩大经济效益提供详实的数据基础,并为新桥矿床开展更为合理的开采生产安排提出合理化建议.     

On vertical dynamic stability and cross‐section optimization of closed‐ended hollow pile by considering soil compaction effects

This paper conducts a comprehensive study on the effects of expansion force after pile driving on the vertical vibration of the hollow pile. The initial radially inhomogeneous strain field of soil in disturbed soil region and dynamic shear modulus of remolded soil are constructed by applying the cylindrical cavity expansion method. The equation governing the incremental motion of the soil is consequently deduced on the basis of incremental deformations superposed on an underlying finite deformation. The longitudinal impedance of the top of the pile and the velocity response in frequency and time domains are also numerically studied. The relations between the expansion force after pile driving and the velocity response of the pile with different wall thickness are discussed accordingly. The results suggest that a pile has a better dynamical stability when the characteristics of the section are optimized and interacting force with soil medium gets smaller.     

地球物理抗差估计和广义逆方法

为进行观测数据误差较大的地球物理资料的反演,引人抗差估计,称作地球物理抗差估计;为使病态方程组的反演解更可靠,又结合广义逆方法进行了算法改进．首先介绍抗差最小二乘（ＲＬＳ）的基本原理,然后推寻出适于广义逆反演方法的改进算式,最后举例加以讨论．分析表明,抗差估计可以有效地抑制地球物理观测异常的影响,得出正常模式下的较好估计值;而用改进后的算式和广义逆反演可以使反演解更加改善,不仅如此,改进后的方法还能直接给出解估计的可靠性评价．     

Raingauge network optimization in a tropical urban area by coupling cross-validation with the geostatistical technique

An adequate and reliable raingauge network is essential for observing rainfall data in hydrology and water resource applications. A raingauge network developed for a catchment area is commonly extended periodically to increase data accuracy. Due to financial constraints, the network is reviewed for the optimal number of stations. A new optimization approach is developed in this study by coupling a cross-validation technique with a geostatistical method for raingauge network optimization to prioritize raingauge stations. The spatial interpolation error of the spatial rainfall distribution, measured as the root mean square error (E_rms) optimization criterion is applied to a raingauge network in a tropical urban area. The results indicate that this method can successfully optimize the number of rainfall stations in an existing raingauge network, as the stations are prioritized based on their importance in the network.     

Intelligent fuzzy weighted input estimation method for the multi-story buildings with unknown ground motion acceleration

This paper presents a novel intelligent fuzzy weighted input estimation method which effiviently and robustly estimates the unknown ground motion accelerations. The new input estimation method includes the Kalman Filter (KF) and the recursive least square estimator (RLSE), which is weighted by the fuzzy weighting factor proposed based on the fuzzy logic inference system. By directly synthesizing the Kalman filter with the estimator, this work presents an efficient robust forgetting zone, which is capable of providing a reasonable compromise between the tracking capability and the flexibility against noises. The excellent performace of this inverse method is demonstrated by solving the earthquake-excitation estimation problem, and the proposed algorithm is compared by alternating between the constant and adaptive weighting factors. The results reveal that this method has the properties of better target tracking capability and more effective noise reduction.     

亚毫米波天线有限元模型参数修正复合优化方法

天线有限元模型与实际天线往往存在结构参数误差和材料参数误差等,使有限元模型不能真实反映实际天线结构在各种载荷下的变形特性.针对亚毫米波天线,提出了亚毫米波天线有限元模型参数修正复合优化方法.该方法包括3步,第1步建立天线参数化有限元模型,并进行参数灵敏度分析;第2步,以第1步中得到的高灵敏度参数为优化变量建立优化函数,以初始模型与假想实际天线变形特性的差值最小为目标进行优化;第3步,基于零阶优化法和一阶优化法对天线有限元模型进行复合优化,使天线有限元模型逐渐逼近实际天线.为验证所提出参数修正优化方法的可行性,针对一台1.2 m亚毫米波天线,将对其进行假想实验获得的变形特性作为优化目标,分别对多个材料参数在不同的载荷工况下进行优化,结果表明天线有限元模型经过参数优化后与假想实际天线一致,证明了所提出的亚毫米波天线有限元模型参数优化修正方法可行,并在重力载荷下对修正优化后的天线模型的变形规律进行分析.这种优化方法可以推广应用到大口径亚毫米波天线的有限元模型修正中.     

滑动最小二乘法求解地震波波动方程

本文首次将固体力学领域中的滑动最小二乘拟合法用于求解地震波波动方程.与有限差分法相比,该方法所采用的拟合思想使得解在全空间域更加连续,同时仍然保持高精度的特性;与有限元法相比,该方法绕过变分原理,经过算法的有限差分化之后,其所需的计算量及数据存储量均可大大降低.薄膜震动的数值算例表明了该方法的上述特点.最后,通过对几个模型的合成地震模拟试验,进一步证明了滑动最小二乘法用于波动方程数值模拟的可行性.