病态半参数回归模型的岭估计求解

针对半参数回归模型求解过程可能出现的法方程病态问题,提出了用岭估计原则改进半参数模型的求解。通过模拟算例将岭估计解法和其他方法进行了比较,结果表明,岭估计解法能较好地解决半参数回归模型求解过程中的病态问题。     

基于谱修正迭代的Bursa模型

利用Bursa模型进行空间直角坐标转换时,求解的过程中如果法方程的系数矩阵是病态,则求得的转换参数并不可靠.在谱修正迭代法的基础上,推导了基于谱修正迭代的Bursa模型公式,通过算例可知该方法可有效改善病态矩阵对坐标转换的影响,其坐标转换精度比Bursa模型高.     

正则化方法在结构模型修正中的应用研究

大型结构的模型修正求解问题多呈现不同程度的病态,实测数据的微小误差都有可能造成求解的失效。该文研究了测量噪声影响下模型修正病态系统的求解问题。首先,介绍了结构模型修正的求解问题以及数学上常用的正则化方法;然后,通过一悬臂梁模型数值算例探讨了截断奇异值分解正则化方法和"L"曲线法在结构模型修正求解中的适用性。结果显示,适当的正则化可以有效的解决模型修正病态系统的求解问题;另外,该方法对于部分满足离散Picard条件的模型修正方程同样适用。最后,通过一导管架平台物理模型试验对该正则化方法的实用性进行了验证。     

利用双h类岭估计求解板块运动的欧拉参数

在求解板块运动参数的过程中,选用的台站分布越合理,参加解算的测站数据越多,求解的参数精度也就越高,但随之带来的就是系数矩阵的病态性问题,为了解决这个问题,采用了双h类岭估计方法对北美板块的欧拉参数进行估计,求出欧拉参数及其中误差,以及板块运动水平速度的残差百分比,并将求得的欧拉参数与其他模型进行对比,结果显示模型间具有较好的一致性,但仍有一定的差异;此外,测站速度残差0~1mm的百分比在80%以上,表明利用此方法求解精度较好。     

泥床上的波浪衰减：—伪塑性体模型

由于不同的淤泥其流变特性不同，部分淤泥的本构关系可以用伪塑性模型近似，因此本文提出上层为理想流体，下层为伪塑性体的双层流体模型模拟淤泥床的波浪衰减。文中利用线性化方法求解伪塑性体这种非线性模型的边界值问题，并根据色散方程的数值解讨论了模型参数对波浪衰减的影响。文中最后指出，模拟泥床上波良衰减的其它非线性模型也可以通过本文提出的等效功原理求解。     

利用空间投影理论求解RPC模型

利用遥感影像成像时的空间投影模型并顾及全球DEM建立了一组虚拟的三维控制点,用于求解RPC模型,解决了控制点数量严重不足时RPC模型的求解问题。试验证明该方法理论严密、精度较高。     

采用两步式模型修正过程识别海上风电支撑结构的损伤

针对海上风电支撑结构损伤识别的难点,将一种新的模型修正方法应用到了三脚架式海上风电支撑结构的损伤识别中。采用数学上常用的正则化方法,解决了实测模态含噪声条件下损伤识别"病态"方程组的求解问题;结合Guyan扩阶技术,实现了基于低阶不完备模态的损伤识别。另外,为提高损伤识别精度,结合算例提出一种两步式模型修正过程:第一步,初步定位损伤位置;第二步,精确识别损伤位置和损伤程度。数值研究结果表明,采用两步式模型修正过程的方法对于海上风电支撑结构的损伤识别是有效的。     

黄海北部海域三维潮流数值模型

基于湍流闭合的三维斜压流体动力学方程组,建立了河口及附近浅海域的三维环流数值模型,模型采用ＡＤＩ差分方法妥动量方各,采用佞步方法求解标量方程,采用总变差减小格式（ＴＶＤ）来求解各方程中的平 流项。模型提高了计算精度,同时使计算量大幅度缩小。利用本文所建立的数值模型对鸭绿江口及附近海域的潮流场进行了数值模拟,结果表明,模型具有良好的精度。     

基于空间投影理论的RPC模型求解方法研究

利用遥感影像成像时的空间投影模型并顾及全球DEM建立了一组虚拟的三维控制点用于求解RPC模型,解决了严格成像模型未知或过于复杂难以建立,同时控制点数量严重不足时RPC模型的求解问题。试验证明该方法理论严密,精度较高。     

无人机航空像片绝对定向非迭代解算方法研究

利用无人机航空像片的模型点重心化坐标和地面重心化坐标构成单位四元数,通过最优化求解绝对定向旋转矩阵,并基于旋转矩阵解算平移参数和比例参数,实现绝对定向的非迭代解算。与传统方法相比,这种方法具有无需初始值、可靠性高等优点。实验结果表明,方法可以得到较高的解算精度,能够满足无人机航空像片定位的需要。     

基于地理加权回归模型的水深反演算法研究

利用卫星多光谱数据反演浅海水深是水深测量的一种重要手段。已有水深反演方法是在研究区建立统一的数学参数的反演模型,未考虑由于海底底质和水质变化导致的空间非平稳性问题。本文使用地理加权回归模型(Geographically Weighted Regression, GWR)对回归参数在空间上进行估计,针对GWR模型的带宽对反演精度的影响,使用了交叉验证(Cross Validation,CV)的方法来确定最佳带宽,并以南海永兴岛和甘泉岛海域为实验区,基于WordiVew-2多光谱数据对使用GWR模型的可行性和精度进行了验证。实验结果表明:永兴岛研究区GWR模型精度较线性回归模型提高了36.05%,在0~5,5~10,10~15和15~20 m区间,精度分别提高了49.46%,39.97%,12.36%和49.68%;甘泉岛研究区GWR模型精度较线性回归模型提高了8.08%,在0~5,5~10,10~15和15~20 m区间,精度分别提高了12.05%,16.23%,4.49%和12.23%,表明GWR模型具有更好的水深反演效果。     

变分同化方法反演海气耦合模型参数的研究

采用变分资料同化技术,结合最优控制思想,对一个海气耦合模型的模式参数和强迫项进行了反演,结果表明,采用该方法对模式进行优化,既可以补偿模式参数不准确性给预报带来的误差,又可以对模式参数本身进行修正和估计,为将来在实际应用中改善更复杂的预报模式、提高预报准确率提供了一个可借鉴的思路。     

基于PSO-LSSVM的局部海洋三维地磁场建模方法

局部海洋地磁场模型可以为水下地磁匹配导航与磁性目标探测等工作提供高精度的海域地磁背景场数据。针对目前区域地磁场模型构建方法中未考虑到高度或深度变化的影响,截断阶数难以确定及存在边界效应等问题。根据海洋磁力测量的特点,顾及磁测点高度数据,引入粒子群优化的最小二乘支持向量机法(PSO-LSSVM)构建局部海洋三维地磁场模型。在仿真与实测数据建模试验中,与Taylor多项式法、BP神经网络法及曲面样条函数法比较,结果表明,无论是在地磁变化平缓区还是复杂区,PSO-LSSVM法的建模精度均是最高的,建议在局部海洋地磁场模型构建中采用。     

Use of Lade's Single Surface Work-Hardening Model to Investigate Mechanical Behavior of Decomposed Soils

In this study, it was attempted to assess soil parameters necessary for Lade's single surface work-hardening model that reviewed the physical and mechanical properties of granite soil located in Korea based on the results of triaxial compression tests. In addition, finite element analyses coupled with the determined soil parameters as inputs were conducted based on Lade's single surface work-hardening model and the results were compared with element test results. It could be seen that, in predicting undrained mechanical behavior, the single surface model was reproducing the stress-strain relation obtained through element tests at high accuracy. It is worthwhile to inform that these differences in the initial loading stage and the impossibility to predict swelling behavior are caused by the fact that there is no prediction model for changes in shear properties, especially in dilatancy properties due to particle crushing occurring while element tests are conducted. Hence, it is concluded that, to expand the applicability of Lade's single surface work-hardening constitutive model to practical problems, the model should be modified in relation to the dilatancy of soils.     

基于大气参数拟合的GF-1多光谱影像大气校正

针对大气校正常使用的基于查找表的方法建表速度缓慢、查找精度低等问题,提出在6S大气校正模型的基础上,以气溶胶厚度为自变量对大气参数进行拟合的算法,使用该算法对GF-1多光谱影像进行大气校正研究。对半球反射率、路径辐射项等效反射率、大气透过率3个大气参数使用三次多项式进行拟合,拟合后的决定系数都达到了0. 999 8;根据拟合结果使用MODIS气溶胶产品作为输入,逐像元计算大气参数,并对影像进行大气校正,将校正结果与FLAASH模型校正结果进行对比。结果表明,该算法能有效校正大气对GF-1多光谱影像的影响,并与FLAASH校正结果具有很好的一致性。     

Determination of Ocean Tide Loading Displacement by GPS PPP with Priori Information Constraint of NAO99b Global Ocean Tide Model

The published global ocean tide models show good agreement in deep oceans and exhibit differences in complex coastal areas, along with subsequent Ocean Tide Loading Displacement (OTLD) modeling differences. Meanwhile, OTLD parameters (amplitudes and phase lags) derived by Global Positioning System (GPS) Precise Point Positioning (PPP) approach need long time to converge to a stable state and show poor precision of S2, K1, and K2 constituents. Based on the fact that no constraint is imposed in the current kinematic solution, a new method is put forward, in which global ocean tide model predictions are taken as the priori information constraints to speed up the convergence rate and improve the accuracy of the GPS-derived OTLD parameters. First, the data of tide gauge from 01 January 2014 to 31 December 2016 are used to generate the harmonic parameters to evaluate the accuracy of six global ocean tide models and a regional ocean tide model (osu.chinesea.2010). Osu.chinesea.2010 model shows good agreement with the tide gauge results, while NAO99b model presents relatively large difference. The predictions from osu.chinesea.2010 and NAO99b model are employed as reference and the prior information, respectively. Second, continuous observations of 12 GPS sites during 2006–2013 in Hong Kong are collected to generate three dimensional OTLD amplitudes and phase lags of eight constituents using PPP with prior information constraints approach and harmonic analysis. Third, comparing the convergence time of eight constituents from PPP without and with priori information constraints approaches, the results show that the new method can significantly improve the convergence rate of OTLD amplitude estimates which obtain a certain level of stability seven years earlier than that derived by the PPP without priori information constraints. Precision of OTLD parameters derived by the new method is about 1 mm. By comparing with the precision of single PPP approach, the accuracy of eight constituents has been improved, especially for S2, K1, and K2 constituents. Finally, through comparing the different correction effects of OTLD estimates on the coordinates and their time series of the ground GPS stations, the results show that OTLD estimates derived by the new approach have similar influence as the osu.chinasea.2010 ocean tide model. The new method provides an effective means to improve the convergence and precision of the GPS-derived OTLD parameters, and achieve a similar correction as the high precision ocean tide model.     

Parameter identification of a compliant nonlinear SDOF system in random ocean waves by reverse MISO method

The determination of the drag and inertia coefficients, which enter into the wave force model given by Morison's equation, is particularly uncertain and difficult when a linear spectral model is used for ocean waves, and the structure is compliant and has nonlinear dynamic response. In this paper, a nonlinear System Identification method, called Reverse Multiple Inputs–Single Output (R–MISO) is applied to identify the hydrodynamic coefficients as well as the nonlinear stiffness parameter for a compliant single-degree-of-freedom system. Four different types of problems have been identified for use in various situations and the R–MISO has been applied to all of them. One of the problems requires iterative solution strategy to identify the parameters. The method has been found to be efficient in predicting the parameters with reasonable accuracy and has the potential for use in the laboratory experiments on compliant nonlinear offshore systems.     

波浪与抛石潜堤相互作用的MPS法数值模型研究

为了研究波浪与抛石潜堤相互作用过程中大自由表面变形和堤内渗流等强非线性紊流运动问题,利用改进的MPS法,建立了模拟波浪与抛石潜堤相互作用的MPS法数值计算模型。模型将抛石潜堤假定为均质多孔介质,采用Drew的二相流运动方程描述多孔介质内外的流体运动;通过在动量方程中增加非线性阻力项,并引入亚粒子尺度紊流模型,模拟波浪与可渗结构物相互作用过程中的紊流运动。选取“U”型管中多孔介质内渗流过程和孤立波与可渗潜堤相互作用两个典型的渗流问题,通过将数值计算结果与理论解和实测值的对比分析,对所提出的MPS法紊流渗流模型的模拟精度进行验证。结果表明:基于改进的MPS法构建的垂向二维紊流渗流模型可以很好地再现“U”型管中多孔介质内渗流以及波浪作用下可渗潜堤内外的复杂流场,显著缓解流-固界面处的压力震荡与粒子分布不均匀问题,实现了较高的模拟精度。     

复合模型系统的控制策略

复合模型中的物理模型和数学模型是依靠计算机控制系统实时耦合起来的。复合模型对控制系统的精度要求较高 ,但是流量的控制中存在着水流的滞后性以及水位控制中的超调现象 ,这些问题都将影响控制的精度。本文介绍采用改进的PID与Smith控制算法相结合的控制策略实现流量控制 ,以解决水流的滞后性 ;采用变参数PID控制策略实现水位控制。试验证明两者都取得了较好的控制效果