全隐LU-SGS算法在高超声速热化学非平衡流刚性问题中的应用

在高超声速热化学非平衡流动计算中,当地气体能量松弛时间、化学反应特征时间与流动时间推进步长量级差异过大会带来严重数值刚性问题,且在高雷诺数条件下,壁面、拐角等强干扰区网格加密使得该问题加剧,导致初始最大CFL数极小,收敛速度缓慢。原始LU-SGS算法仅考虑化学反应源项和对流项的隐式处理,通过推导黏性项Jacobian矩阵谱半径并采用对角近似处理,发展了热化学非平衡FLU-SGS和BLU-SGS两种全隐LU-SGS算法;针对高焓二维圆柱和轴对称返回舱算例,对比改进前后三种算法的收敛特性。结果表明,FLU-SGS及BLU-SGS算法能够快速建立强黏性干扰和大分离流场、解决热化学非平衡复杂流计算中的刚性问题,实现初始最大CFL数3至5个量级的提升,加速收敛效果明显。

压缩机工作过程模拟中数值振荡问题的成因分析及改善方法

为解决数值模拟压缩机工作过程中存在的数值振荡和发散问题,采用四阶后向差分(BDF-4)算法和经典四阶龙格库塔(RK-4)算法分析压缩机基元容积控制方程的稳定性,并对2种算法的计算结果进行对比。结果表明:在吸排气过程中,压力计算的微小偏差会导致质量流量产生较大的误差,从而使基元容积控制方程产生较大的数值刚性,限制了RK-4算法的最大稳定步长;在稳定性上,BDF-4算法的绝对稳定域包含整个负半x轴,可以避免数值刚性引起的振荡和发散,其计算步长不再受数值刚性的限制;对计算效率而言,每步BDF-4算法所消耗时间为同步长RK-4算法的1.7~1.8倍;计算精度上,BDF-4算法与RK-4算法代数精度相同,实际计算时两者的表现也十分接近。BDF-4算法在保证工程精度的前提下提高了压缩机工作过程模拟的稳定性,解决了数值振荡问题,对容积式压缩机的工作过程模拟、新型泄漏通道的研究等有实际意义。

Numerical evaluation of buckling behavior in space structure considering geometrical parameters with joint rigidity

The buckling behavior of single layer space structure is very sensitive. The joint rigidity, moreover, is one of the main factors of stability which may determine the entire failure behavior. Thus, the reasonable stiffness of joint system, which is neither total pin assumption nor perfect fix condition, is very important to apply to the real single layer space one. Therefore, the purpose of this work was to investigate the buckling behavior of single layer space structure, using the development of the upgraded stiffness matrix for the joint rigidity. To derive tangential stiffness matrix, a displacement function was assumed using translational and rotational displacement at the node. The geometrical nonlinear analysis was simulated not only with perfect model but also with imperfect one. As a result, the one and two free nodal numerical models were investigated using derived stiffness matrix. It was figured out that the buckling load increases in proportion to joint rigidity with rise-span ratio. The stability of numerical model is very sensitive with the initial imperfection, responding of bifurcation in the structure.

Stiffness identification of four-point-elastic-support rigid plate

As the stiffness of the elastic support varies with the physical-chemical erosion and mechanical friction, model catastrophe of a single degree-of-freedom(DOF) isolation system may occur. A 3-DOF four-point-elastic-support rigid plate(FERP) structure is presented to describe the catastrophic isolation system. Based on the newly-established structure, theoretical derivation for stiffness matrix calculation by free response(SMCby FR) and the method of stiffness identification by stiffness matrix disassembly(SIby SMD)are proposed. By integrating the SMCby FR and the SIby SMD and defining the stiffness assurance criterion(SAC), the procedures for stiffness identification of a FERP structure(SIFERP) are summarized. Then, a numerical example is adopted for the SIFERP validation, in which the simulated tested free response data are generated by the numerical methods, and operation for filtering noise is conducted to imitate the practical application. Results in the numerical example demonstrate the feasibility and accuracy of the developed SIFERP for stiffness identification.

Development of an Accurate Numerical Methodology for Predicting the First Three Modes of Vibration of an Electric Motor Fixed on a Rigid Base

This paper proposes a numerical methodology for the prediction of the first three modes of vibration of an electric motor fixed on a rigid base. A deep literature review supported the production of four ad hoc prototypes that aided the development of the proposed approach. Tests carried out with the prototypes led to the procurement of the modal parameters be used to calibrate the numerical models, as well as the FRF （frequency response function） curves be used to validate the numerical solution. The validated model allowed structural changes to be then promoted on the prototypes, in order to make them more robust to variations in manufacturing and assembling processes. The mentioned adjustments and structural changes were accomplished by means of a process of structural optimization using Genetic Algorithm. The solution was developed based on the commercial finite element code ANSYS. The practical results obtained in this study show that a numerical model for modal analysis of an electric motor fixed on a rigid base with errors less than 3% for the first three modes of vibration can be achieved, allowing positive structural changes to be performed in the machine design that result in the minimization of manufacturing reworks associated with the dynamic behavior of the studied motor.

Dynamic Behaviors of a Moored Floating Structure with Flexible Skirts

A numerical model was developed by using the dual boundary element method to investigate the dynamic behavior of a moored floating structure with a pair of vertical and flexible skirts attached at its bottom in the linear wave field. Theoretical conception is based on potential theory with linear external forces. The motions of the structure were assumed to be small and linear. The flexible skirts mounted beneath the structure were assumed uniform flexural rigidity and the thickness of the skirts was negligible. Comparison between the present model and Gesraha＇s solution was made to verify the results for a moored floating structure with or without rigid skirts. The influence of the skirt rigidity on the moored floating structure, moored lines and waves is investigated in this study. The results show that, the natural frequencies of structure＇s oscillation, moored force, wave reflection and transmission tend to the region of short-period waves when the flexible rigidity gradually decreases. Positive correlation exists between the aft mooring force and the pitch motion of the floating structure.

Numerical study of ambient pressure for laser-induced bubble near a rigid boundary

The dynamics of the laser-induced bubble at different ambient pressures was numerically studied by Finite Volume Method (FVM). The velocity of the bubble wall, the liquid jet velocity at collapse, and the pressure of the water hammer while the liquid jet impacting onto the boundary are found to increase nonlinearly with increasing ambient pressure. The collapse time and the formation time of the liquid jet are found to decrease nonlinearly with increasing ambient pressure. The ratios of the jet formation time to the collapse time, and the displacement of the bubble center to the maximal radius while the jet formation stay invariant when ambient pressure changes. These ratios are independent of ambient pressure.

Mathematical modeling and numerical simulation of telephone cord buckles of elastic films

An annular sector model for the telephone cord buckles of elastic thin films on rigid substrates is established, in which the von Krman plate equations in polar coordinates are used for the elastic thin film and a discrete version of the Griffith criterion is applied to determine the shape and scale of the parameters. A numerical algorithm combining the Newmark-β scheme and the Chebyshev collocation method is designed to numerically solve the problem in a quasi-dynamic process. Numerical results are presented to show that the numerical method works well and the model agrees well with physical observations, especially successfully simulated for the first time the telephone cord buckles with two humps along the ridge of each section of a buckle.

An element decomposition method with variance strain stabilization

An element decomposition method with variance strain stabilization(EDM-VSS) is proposed. In the present EDM-VSS, the quadrilateral element is first divided into four sub-triangular cells, and the local strains in sub-triangular cells are obtained using linear interpolation function. For each quadrilateral element, the strain of the whole quadrilateral is the weighted average value of the local strains, which means only one integration point is adopted to construct the stiffness matrix. The stabilization item of the stiffness matrix is constructed by variance of the local strains, which can eliminate the instability of the one-point integration formulation and largely increase the accuracy of the element. Compared with conventional full integration quadrilateral element, the EDM-VSS achieves more accurate results and expends much lower computational cost. More importantly, as no mapping or coordinate transformation is involved in the present EDM-VSS, the restriction on the conventional quadrilateral elements can be removed and problem domain can be discretized in more flexible ways. To verify the accuracy and stability of the present formulation, a number of numerical examples are studied to demonstrate the efficiency of the present EDM-VSS.