一种高度并行的Schwarz型混乱松弛法及其收敛性证明

文[1—2]把混乱松弛思想引入到Schwarz交替法中,构造了一种Schwarz型混乱松弛法。但这个方法在进行第n+1步迭代时,在拟边界上必须要用到第n步迭代的值,从而影响了算法的并行性,得不到相应的同步或异步MIMD并行算法、为此,本文给出一种高度并行的Schwarz型混乱松弛法,这个方法包括了Schwarz交替法及其相应的同步和异步MIMD并行算法。对于二阶线性与非线性微分方程Dirichlet问题,本文应用微分方程极值原理证明了该方法的收敛性。     

演化策略用于高维故障样本集最优统计聚类分析

针对液体火箭发动机推进系统超高维故障样本数据的聚类问题 ,提出基于演化策略的最优统计聚类算法。为预防算法过早收敛 ,演化策略采用了父本适应值的动态调整值与共享函数 ,并针对超高维数据聚类提出了控制参数的适应性调整技术 ;为使算法能最终跳出局部最优死区 ,提出算法的局部调整策略。该算法用于液体火箭发动机典型故障仿真数据集分析 ,并取得了最优聚类结果。此外 ,还基于IRIS数据集比较了该算法与FKCN模糊自主聚类算法。仿真分析表明了算法在高维数据聚类分析中的优点。     

涡轮发动机复杂区域的结构化网格生成技术

通过对二维泊松方程增加伪时间项，采用时间推进法求解生成二维网格，利用拉格朗日插值生成三维网格，给出了一种在涡轮发动机复杂区域中生成结构化网格的计算方法。该方法很好地保证了网格的质量和正交性。大量的计算算例表明该网格的生成方法和区域分解算法相结合可以对复杂区域流场的流动细节进行很好的模拟。     

模糊自组织神经网络在航空发动机故障诊断中的应用

介绍了模糊自组织神经网络在航空发动机故障诊断中的应用方法,并通过实例验证了该方法在发动机故障分类中的实用性。该方法具有结构算法简单、无监督自学习和侧向联想等功能。它有很好的应用前景,可以广泛应用于发动机的故障诊断。      

非线性混合回归演化算法研究

基于最小二乘法的常规非线性回归算法的缺点,提出了方程结构与系数的混合回归演化策略方法。在指定最大回归项数的前提下,该混合回归演化策略先找到一些基本满足拟合要求（回归项数与精度）的初始种子回归方程,再以回归项数逐步减少、拟合精度逐步提高为准则,对初始种子回归方程进一步混合回归演化,最后得到回归项数最少、拟合精度最高的最佳回归方程。此外,利用该混合回归演化策略对液体火箭发动机常见的几个经验关系式进行了非线性回归拟合实例分析,拟合残差分布表明该混合回归演化策略方法是有效的。     

斜切反喷管粘性流场数值模

为了在级间分离期间提供反推力，许多固体火箭发动机前端都装有一组斜切反喷管，由于反喷管气动型面具有尖点，并在超音速区有台阶，喷管内存在一系列激波，并伴有流动分离现象。从雷诺平均的非定常Ｎａｖｉｅｒ－Ｓｔｏｋｅｓ方程出发，利用时间相关法，采用ＭａｃＣｏｒｍａｃｋ两步显格式，结合Ｂａｌｄｗｉｎ－Ｌｏｍａｘ代数湍流模型，数值模拟了斜切反喷管流场。计算得到的壁面压强分布与风洞吹风实验测得的压强分市相当一致。     

平行混合的变比热准确解与次准确解

在变比热平行混合计算中，一般多以近似等比热比ｋ处理。变比热气动函数式的推导，提出了解此问题的有力工具。本文较详细地介绍了五ｋ准确解及三ｋ次准确解，证明次准确解足够近似，而定比热解则误差很大。     

跨音松弛法在叶栅绕流中的应用

文章介绍跨音松弛法应用于叶栅绕流的定常与非定常流场计算。在定常流中,采用弦线法向小扰动假定;在非定常流中,采用线化非定常分量小扰动假定,从而简化了数值求解过程。由于松弛法大大地节省机时和内存,便于工程上推广应用。     

Review of shock wave detection method in CFD post-processing

In the present computational fluid dynamics (CFD) community, post-processing is regarded as a procedure to view parameter distribution, detect characteristic structure and reveal physical mechanism of fluid flow based on computational or experimental results. Field plots by contours, iso-surfaces, streamlines, vectors and others are traditional post-processing techniques. While the shock wave, as one important and critical flow structure in many aerodynamic problems, can hardly be detected or distinguished in a direct way using these traditional methods, due to possible confusions with other similar discontinuous flow structures like slip line, contact discontinuity, etc. Therefore, method for automatic detection of shock wave in post-processing is of great importance for both academic research and engineering applications. In this paper, the current status of methodologies developed for shock wave detection and implementations in post-processing platform are reviewed, as well as discussions on advantages and limitations of the existing methods and proposals for further studies of shock wave detection method. We also develop an advanced post-processing software, with improved shock detection.     

l1-based calibration of POD-Galerkin models of two-dimensional unsteady flows

This paper discusses a physics-informed methodology aimed at reconstructing efficiently the fluid state of a system. Herein, the generation of an accurate reduced order model of two-dimensional unsteady flows from data leverages on sparsity-promoting statistical learning techniques. The cornerstone of the approach is l₁ regularised regression, resulting in sparsely-connected models where only the important quadratic interactions between modes are retained. The original dynamical behaviour is reproduced at low computational costs, as few quadratic interactions need to be evaluated. The approach has two key features. First, interactions are selected systematically as a solution of a convex optimisation problem and no a priori assumptions on the physics of the flow are required. Second, the presence of a regularisation term improves the predictive performance of the original model, generally affected by noise and poor data quality. Test cases are for two-dimensional lid-driven cavity flows, at three values of the Reynolds number for which the motion is chaotic and energy interactions are scattered across the spectrum. It is found that: (A) the sparsification generates models maintaining the original accuracy level but with a lower number of active coefficients; this becomes more pronounced for increasing Reynolds numbers suggesting that extension of these techniques to real-life flow configurations is possible; (B) sparse models maintain a good temporal stability for predictions. The methodology is ready for more complex applications without modifications of the underlying theory, and the integration into a cyber-physical model is feasible.