机械式回转式拧瓶机的设计及工程分析【说明书+CAD+UG】
机械式回转式拧瓶机的设计及工程分析【说明书+CAD+UG】,说明书+CAD+UG,机械式,回转,式拧瓶机,设计,工程,分析,说明书,仿单,cad,ug
英文原文Applications4.1 IntroductionThis chapter demonstrates the scope of the method developed for the three-dimensional analysis of a screw compressor. The CFD package used in this case was COMET developed by ICCM GmbH Hamburg, today a part of CD-Adapco. The analysis of the flow and performance characteristics of a number of types of screw machines is performed to demonstrate a variety of parameters used for grid generation and calculation.The first example is concerned with a dry air screw compressor. A common compressor casing is used with two alternative pairs of rotors. The rotors have identical overall geometric properties but different lobe profiles. The application of the adaptation technique enables convenient grid generation for geometrically different rotors. The results obtained by three dimensional modelling are compared with those derived from a one-dimensional model, previously verified by comparison with experimental data.The relative advantages of each rotor profile are demonstrated.The second example shows the application of three dimensional flow analysis to the simulation of an oil injected air compressor. The results, thus obtained, are compared with test results obtained by the authors from a compressor and test rig, designed and built at City University. They are presented in the form of both integral parameters and a p-indicator diagram. Calculations based on the assumptions of the laminar flow are compared to those of turbulent flow. The effect of grid size on the results is also considered and shown here.The third example gives the analysis of an oil injected compressor in an ammonia refrigeration plant.This utilises the real fluid property subroutines in the process calculations and demonstrates the blow hole area and the leakage flow through the compressor clearances.The fourth example presents two cases, one of a dry screw compressor to show the influence of thermal expansion of the rotor on screw compressor performance and one of a high pressure oil-flooded screw compressor to show the influence of high pressure loads upon the compressor performance. 4.2 Flow in a Dry Screw CompressorDry screw compressors are commonly used to produce pressurised air, free of any oil. A typical example of such a machine, similar in configuration to the compressor modelled, is shown in Figure 4-1. This is a single stage machine with 4 male and 6 female rotor lobes. The male and female rotor outer diameters are 142.380 mm and 135.820 mm respectively, while their centre lines are 108.4 mm apart. The rotor length to main diameter ratio l/d=1.77. Thus, the rotor length is 252.0 mm. The male rotor with wrap angle =248.40 is driven at a speed of 6000 rpm by an electric motor through a gearbox. The male and female rotors are synchronised through timing gears with the same ratio as that of the compressor rotor lobes i.e. 1.5. The female rotor speed is therefore 4000 rpm. The male rotor tip speed is then 44.7m/s, which is a relatively low value for a dry air compressor. The working chamber is sealed from its bearings by a combination of lip and labyrinth seals.Each rotor is supported by one radial and one axial bearing, on the discharge end, and one radial bearing on the suction end of the compressor. The bearings are loaded by a high frequency force, which varies due to the pressure change within the working chamber. Both radial and axial forces, as well as the torque change with a frequency of 4 times the rotational speed. This corresponds to 400Hz and coincides with the number of working cycles that occur within the compressor per unit time.Figure 4-1 Cross section of a dry screw compressorThe compressor takes in air from the atmosphere and discharges it to a receiver at a constant output pressure of 3 bar. Although the pressure rise is moderate, leakage through radial gaps of 150 m is substantial. In many studies and modelling ,procedures, volumetric losses are assumed to be a linear function of the cross sectional area and the square root of pressure difference, assuming that the interlobe clearance is kept more or less constant by the synchronising gears. The leakage through the clearances is then proportional to the clearance gap and the length of the leakage line. However, a large clearance gap is needed to prevent contact with the housing caused by rotor deformation due to the pressure and temperature changes within the working chamber. Hence, the only way to reduce leakage is to minimise the length of the sealing line. This can be achieved by careful design of the screw rotor profile. Although minimising,leakage is an important means of improving a screw compressor efficiency, it is not the only one. Another is to increase the flow area between the lobes and thereby increase the compressor flow capacity, thereby reducing the relative effect of leakage. Modern profile generation methods take these various effects into account by means of optimisation procedures which lead to enlargement of the male rotor interlobes and reduction in the female rotor lobes. The female rotor lobes are thereby strengthened and their deformation thus reduced. To demonstrate the improvements possible from rotor profile optimisation, a three dimensional flow analysis has been carried out for two different rotor profiles within the same compressor casing, as shown in Figure 4-2. Both rotors are of the “N” type and rack generated.Figure 4-2N Rotors, Case-1 upper, Case-2 lowerCase 1 is an older design, similar in shape to SRM “D” rotors. Its features imply that there is a large torque on the female rotor, the sealing line is relatively long and the female lobes are relatively weak.Case 2, shown on the bottom of Figure 4-2, has rotors optimised for operating on dry air. The female rotor is stronger and the male rotor is weaker. This results in higher delivery, a relatively shorter sealing line and less torque on the female rotor. All these features help to improve screw compressor performance.The results of these two analyses are presented in the form of velocity distributions in the planes defined by cross-sections A-A and B-B, shown in Figure 4-1.In the case of this study, the effect of rotor profile changes on compressor integral performance parameters can be predicted fairly accurately with one-dimensional models, even if some of the detailed assumptions made in such analytical models are inaccurate. Hence the integral results obtained from the three-dimensional analysis are compared with those from a one-dimensional model.4.2.1 Grid Generation for a Dry Screw CompressorIn Case-1, the rotors are mapped with 52 numerical cells along the interlobe on the male rotor and 36 cells along each interlobe on the female rotor in the circumferential direction. This gives 208 and 216 numerical cells respectively in the circumferential direction for the male and female rotors. A total of 6 cells in the radial direction and 97 cells in the axial direction is specified for both rotors. This arrangement results in a numerical mesh with 327090 cells for the entire machine. The cross section for the Case-1 rotors is shown in Figure 4-3. The female rotor is relatively thin and has a large radius on the lobe tip. Therefore, it is more easily mapped than in Case-2 where the tip radius is smaller, as shown in Figure 4-4.Figure 4-3 Cross section through the numerical mesh for Case-1 rotorsThe rotors in Case 2 are mapped with 60 cells along the male rotor lobe and 40 cells along the female lobe, which gives 240 cells along both rotors in the circumferential direction. In the radial direction, the rotors are mapped with 6 cells while 111 cells are selected for mapping along the rotor axis. Thus, the entire working chamber for this compressor has 406570 cells. In this case, different mesh sizes are applied and different criteria are chosen for the boundary adaptation of these rotors. The main adaptation criterion selected for the rotors is the local radius curvature with a grid point ratio of 0.3 to obtain the desired quality of distribution along the rotor boundaries. By this means, the more curved rotors are mapped with only a slight increase in the grid size to obtain a reasonable value of the grid aspect ratio. To obtain a similar grid aspect ratio without adaptation, 85 cells would have been required instead of 60 along one interlobe on the female rotor. This would give 510 cells in the circumferential direction on each of rotors. If the number of cells in the radial direction is also increased to be 8 instead of 6 but the number of cells along axis is kept constant, the entire grid would contain more then a million cells which would, in turn, result in a significantly longer calculation time and an increased requirement for computer memory. Figure 4-4 Cross section through the numerical mesh for Case-2 rotors4.2.2 Mathematical Model for a Dry Screw CompressorThe mathematical model used is based on the momentum, energy and mass conservation equations as given in Chapter 2. The equation for space law conservation is calculated in the model in order to obtain cell face velocities caused by the mesh movement. The system of equations is closed by Stokes, Fouriers and Ficks laws and the equation of state for an ideal gas. This defines all the properties needed for the solution of the governing equations.4.2.3 Comparison of the Two Different Rotor Profiles The results obtained for both Case 1 and Case 2 compressors are presented here. To establish the full range of working conditions and to obtain an increase of pressure from 1 to 3 bars between the compressor suction and discharge, 15 time steps were required. A further 25 time steps were then needed to complete the full compressor cycle. Each time step needed about 30 minutes running time on an 800 MHz AMD Athlon processor. The computer memory required was about 400 MB.In Figure 4-5 the velocity vectors in the cross and axial sections are compared. The top diagram is given for Case-1 rotors and the bottom one for Case-2. As may be seen, the Case 2 rotors realised a smoother velocity distribution than the Case 1 rotors. This may have some advantage and could have increased the compressor adiabatic efficiency by reduction in flow drag losses. In both cases, recirculation within the entrapped working chamber occurs as consequence of the drag forces in the air as shown in the figure. On the other hand, different fluid flow patterns can be observed in the suction port. The velocities within the working chambers and the suction and discharge ports are kept relatively low while the flow through the clearance gaps changes rapidly and easily reaches sonic velocity. Figure 4-5 Velocity field in the compressor cross section for Case1 and Case2 rotorsFigure 4-6 Velocity field in the compressor axial section for Case1 and Case2 rotorsThese differences are confirmed in the view of the vertical compressor section through the female rotor axis, shown in Figure 4-6. In Case 2, lower velocities are achieved not only in the working chamber but also in the suction and discharge ports. In the suction port, this is significant because of the fluid recirculation which appears at the end of the port. This recirculation causes losses which cannot be recovered later in the compression process. Therefore, many compressors are designed with only an axial port instead of both, radial and axial ports. Such a situation reduces suction dynamic losses caused by recirculation but, on the other hand, increases the velocity in the suction chamber which in turn decreases efficiency. Some of these problems can be avoided only by the design of screw compressor rotors with larger lobes and a bigger swept volume and a shape which allows the suction process to be completed more easily. However, rotor profile design based on existing one-dimensional procedures neglects flow variations in the ports and hence is inferior for this purpose. In such cases, only a full three dimensional approach such as this, will be effective. 中文译文应用4.1简介本章介绍了对螺杆压缩机的三维分析开发的方法的范围。在这种情况下,采用由ICCM GmbH Hamburg开发的CFD软件,现在是CD-Adapco的一部分。对一定数量的螺杆机器的类型的流程和性能特性的分析是用来展示用于栅格一代和演算的各种各样的参量。 第一个例子是关于一个干螺杆空气压缩机。一个常见的压缩机外壳是使用两个可选双转子。转子具有相同的整体几何性质但是有不同的叶剖面。适应技术的应用可以方便使网格生成几何不同的转子。三维模型得到的结果与从一个一维模型获得的那些比较,以前被核实与实验数据相比。演示了每个转子配置文件的相对优势。第二个例子显示了三维流动分析模拟注入油空气压缩机的应用。如此得到的结果与从压缩机的作者和试验台,设计和建造城市大学通过以下方式获得的测试结果进行了比较。他们提出了两个积分的形式参数和一个p-示意图。计算基于的假设是层流与湍流流动的那些进行比较。网格尺寸对计算结果的影响也被认为是在这里。第三个例子给出了油中注入的制冷压缩机的分析。这利用了现实的流体属性的过程中计算的子程序,并演示吹孔区域和通过压缩机的间隙泄漏流。第四个例子呈现两种情况,一是显示的干式螺杆压缩机的转子的螺杆式压缩机的性能,热膨胀的影响和高压油没螺杆式压缩机中的一个,以显示的影响高压负荷时压缩机的性能。4.2干燥螺丝压缩机的流程 干燥螺丝压缩机是常用的生产被加压的空气,不需要任何油。这样机器的一个典型的例子,在配置与被塑造的压缩机相似,在表4-1显示。这是一个有4个阳性和6个阴性转子叶单级机。阳性和阴性的转子外直径分别为142.380毫米和135.820毫米,而他们的中心线108.4毫米。转子长度的主直径比L / D = 1.77。因此,转子长度252毫米。阳转子与包角= 248.40在每分钟6000转的速度驱动,通过齿轮箱由一个电动马达。阳性和阴性的转子通过定时齿轮同步与压缩机转子裂片即1.5的相同比率。因此,阴性的转子转速为每分钟4000转。阳转子叶尖速度然后44.7米/ s,这是相对低的值,为干燥的空气压缩机。工作腔密封从它的轴承,由唇,迷宫式密封的组合。每个转子是由一个径向和轴向轴承和一个径向轴承在放电结束后吸入端的压缩机。轴承是由一个高频力加载,它会因在工作腔的压力变化而变化。径向和轴向的力,以及频率的旋转速度的4倍的转矩变化。这对应于400Hz和发生在压缩机内的每单位时间的工作周期数一致。 压缩机以空气从大气排到一个接收器3个恒定的输出压力。虽然压力上升是温和的,经过150径向间隙泄漏是巨大的。在许多研究和建模过程中,容积损失被认为是一个线性函数的横截面积和压差的平方根假设叶片间间隙保持或多或少不变的同步齿轮。然后,通过该间隙的泄漏间隙和泄漏管路的长度成比例。然而,一个大的间隙是必要的,以防止转子变形,由于工作腔内的压力和温度的变化所造成的与壳体接触。因此,减少泄漏的唯一方法是将密封线长度。这可以通过仔细的螺杆转子型线设计实现。尽管最小化泄漏是一个重要的手 图4-1 干式螺杆压缩机的截面段,提高了螺杆压缩机效率,却不是唯一的一个。另一个是提高叶流之间的区域,从而提高压气机叶流量,从而减少了相对效应的泄漏。现代配置生成方法把这些不同的影响考虑通过优化程序,导致扩大阳转子叶片和减少阴性转子叶。阴性的转子叶是加强及其变形从而降低。为了证明可能从转子齿形优化,改善已进行了三维流场计算在两个不同的转子型线在同一个压缩机壳体,如图4-2所示。生成两个转子的“N ”型和机架。例1是一个比较老的设计,形状类似SRM “D”的转子。它的特点意味着阴转子上,有一个大的转矩,密封线是比较长的相对较弱阴性叶。显示在图4-2的底部,例2的转子的优化操作在干燥的空气。阴性的转子是强大而阳性的转子是较弱的。这结果在较高的输送,一个相对较短的密封线和扭矩少阴转子上。所有这些特点有助于提高螺杆压缩机的性能。 这两个分析结果中的横截面定义的平面A-A、B-B速度分布的形式出现,如图4-1所示。 在本研究的情况下,转子型线的变化对压缩机的整体性能参数的影响可以相当准确地预测的一维模型,即使在这样的分析模型作了详细的假设是不正确的。因此,从三维分析得到的积分结果与一维模型的比较。4.2.1用于干式螺杆压缩机的网格生成。 在例1中,转子被映射52个数值细胞沿叶片间的阳转子和36个细胞沿着每个叶片间的阴转子的圆周方向。这给出了分别在圆周方向上的208和216的数值的单元格的阳性和阴性的转子。总共有6个细胞在径向方向上,并在轴向方向上的97个细胞被指定为两个转子。这种安排导致整个机器327090细胞的数值啮合。例1转子的截面如图4-3所示。阴性转子比较薄,在叶顶大半径上。因此,它是更容易比分析映射在尖端半径越小,如图4-4所示。图42 转子N ,例1上,例2下图4-3 通过案例1转子截面数值网格在例2中的转子被映射60细胞沿凸转子突齿40细胞沿阴性叶瓣这给沿两个转子在圆周方向上的240个细胞。在径向方向上,转子被映射到与6个细胞,111细胞被选择为沿转子轴的映射。因此,该压缩机的整个工作腔有406570个细胞。在这种情况下,不同的大小被应用,并且这些转子的边界适应不同标准的选择。转子的主要适应选择的标准是与某个网格点分布,以获得所需的质量比为0.3,沿转子的边界的局部曲率半径。通过这种方式,更多的弯曲转子映射只有一个轻微增加网格大小来获得一个合理的价值网格的长宽比。为了获得一个类似85个细胞所需要的网格长宽比,而不是沿着一个叶片间的60阴转子的。这将给510细胞在圆周方向上每个转子。如果细胞的数量在径向方向也增加到8代替6但数量的细胞沿轴不变,则整个网格将包含更多,然后一百万细胞,从而反过来导致计算时间大大延长,增加计算机内存要求。通图4 -4 过数值网格横截面为例2转子 4.2.2干式螺杆压缩机的数学模型 所用的数学模型是基于在2章给出了动量,能量和质量守恒方程。这个方程计算空间的保护模型是为了获得细胞面速度引起的啮合运动。方程系统是封闭的斯托克城,傅立叶和菲克的法律和理想气体状态方程。这是定义控制方程解决方案所需的所有属性。4.2.3两个不同的转子的比较 获得的结果对两例1和例2压缩机介绍如下。建立完整的范围的工作条件和获得增加压力从1到3酒吧的压缩机吸、排之间,15次步骤是必需的。一个进一步的25次步骤然后需要完成完整的压缩机循环。每个时间步需要大约30分钟,运行时间在一个800 MHz的AMD Athlon处理器。计算机内存要求约400 MB。 图4-5的速度矢量在十字架和轴向部分进行比较。前图给出案例1和转子底部一个案例2。如可以看到的,在第2种情况的转子实现更平滑的速度分布比第1种情况的转子。可以增加压缩机绝热效率,减少流动阻力损失。在这两种情况下,再循环在裹入工作腔发生的后果在空中拖曳力如图。 另一方面,不同的流体流动模式可观测到吸入口。工作腔和吸入阀和排出端口的速度范围内保持相对低的,而流过的间隙间隙的变化迅速,方便达到声速。图4-5 压缩机截面的例1和案例2转子速度场图4-6 压缩机轴向部分案例1和案例2转子速度场 这些区别根据垂直的压缩机部分被证实 通过阴性电动子轴,显示在图4-6上。在例2中,低速度达到不仅在工作室还在入口及出口的港口。在吸入口,这是很重要的,因为液体再循环,结束时出现端口。这个循环造成损失是无法恢复后的压缩过程。因此,许多压缩机的设计只有一个轴向港口而不是两个港口,径向和轴向港口。这种情况下减少吸入动态而造成损失的再循环,但另一方面,增加的速度,吸入腔从而降低效率。这些问题中的某些问题,可避免仅由螺杆压缩机的转子的设计中具有较大的叶和更大的扫过容积的形状,这使得更容易地完成吸入过程。然而,电动子根据现有的一维做法的外形设计忽略在口岸上的流程变化并且为此下等。在这种情况下,只有一个完整的三维方法如此,这将是有效的。
收藏