混流泵能量性能多参数优化研究

发布时间：2018-01-18 06:40

本文关键词：混流泵能量性能多参数优化研究　出处：《江苏大学》2017年硕士论文　论文类型：学位论文

【摘要】：混流泵在国民经济中应用广泛,高效混流泵水力模型的研发一直是水力机械领域重要和基础的研究方向之一。本文以一台比转速为336的导叶式混流泵为研究对象,通过模型泵测试、数值计算和理论分析相结合的方法,提出了一种不改变叶轮出口名义直径的混流泵能量性能多参数优化方法,并进行了试验验证。本文的主要工作及成果:1.在系统地归纳国内外叶片泵内流特性研究方法、优化设计方法的基础上,总结了常用的叶片泵水力优化方法,并对神经网络和遗传算法的研究现状作了介绍。2.设计加工了模型泵并进行了性能测试,包括外特性、压力脉动及振动特性试验,测试结果表明:(1)与设计要求相比,模型泵效率低了3.92个百分点,扬程低了1.9 m,性能需要优化;(2)与设计工况相比,非设计工况的脉动情况较为紊乱,各监测点在一定的频率范围内均有不同程度的振荡;(3)叶轮进出口处振动均以X方向为主,Y、Z方向的振动情况随工况增加变化不大,叶轮出口振动幅值较大,X、Y、Z三个方向分别比叶轮进口的振动速度幅值高6.78%、5.86%、7.73%。3.提出了一种不改变叶轮出口名义直径的混流泵性能多参数优化方法:(1)选取叶轮的叶片出口边倾斜角θ、叶片后盖板包角φ、叶片中间流线进口安放角β1、叶片中间流线出口安放角β2、叶片出口宽度l、前盖板倾角T1、后盖板倾角T2、叶片数z等8个参数作为优化变量;(2)采用RBF神经网络模型建立混流泵性能预测模型,其中训练及检验的样本点由正交试验和CFD数值计算方法确定;(3)应用多岛遗传算法对预测模型进行求解,从而建立了一种不改变叶轮出口名义直径的混流泵性能多参数优化方法。采用该方法对模型泵能量性能进行了优化,数值计算结果表明:优化后模型泵性能得到显著提升,在设计流量点,效率提高了4.39个百分点,达到了85.24%,扬程也提升了14.6%。4.对优化后模型泵的外特性、压力脉动及振动特性进行了试验测试,并基于CFD结果对比了优化前后模型泵的内流变化,结果表明:(1)在设计点,优化后泵扬程和效率分别提高了11.5%和4.32个百分点;(2)1.0 Qd工况下,优化后泵进口、叶轮出口、导叶出口、泵出口的压力脉动峰峰值分别降低了62.9%、37.5%、28.57%、62.5%;(3)设计工况下,优化后叶轮进口法兰T1测点X、Y、Z方向的振动速度幅值分别减小了12.2%、14.28%、25%;叶轮出口法兰T2测点X、Y、Z方向的振动速度幅值分别减小了4.71%、9.52%、35.29%;(4)优化后导叶内扬程损失降低了1.23%,在原模型导叶中间截面发生回流及低速区域有所改善且优化后叶轮叶片背面的低压区域压力有所提高,有效抑制了压差引起的不稳定流动。
[Abstract]:Mixed flow pump is widely used in national economy. The research and development of hydraulic model of high efficiency mixed flow pump has been one of the important and basic research directions in the field of hydraulic machinery. In this paper, a guide vane mixed flow pump with specific speed of 336 is taken as the research object, and the model pump is tested. Based on the combination of numerical calculation and theoretical analysis, a multi-parameter optimization method for the energy performance of a mixed-flow pump without changing the nominal diameter of the impeller outlet is proposed. The main work and result of this paper are as follows: 1. On the basis of systematically summarizing the research methods of internal flow characteristics and optimization design of vane pumps at home and abroad, the commonly used hydraulic optimization methods of vane pumps are summarized. The research status of neural network and genetic algorithm is introduced. 2. The model pump is designed and manufactured, and its performance is tested, including external characteristics, pressure pulsation and vibration characteristics. The test results show that compared with the design requirements, the efficiency of the model pump is 3.92% lower, the lift is 1.9 m lower, and the performance of the model pump needs to be optimized. (2) compared with the design condition, the pulsation of the off-design condition is more disordered, and each monitoring point has different degrees of oscillation in a certain frequency range; 3) the vibration at the inlet and outlet of the impeller is mainly in the direction of X direction and the vibration in the direction of YJ Z has little change with the increase of working conditions, and the amplitude of vibration at the outlet of the impeller is larger than that at the outlet of the impeller. The amplitude of vibration velocity in the three directions of Z is 6.78% higher than that in the inlet of impeller, respectively, and 5.86% higher than that of impeller inlet. 7.73 / 3.A multi-parameter optimization method of mixed flow pump performance, which does not change the nominal diameter of impeller outlet, is proposed. The blade exit angle 胃 and the back cover angle 蠁 are selected. Eight parameters, such as inlet angle 尾 1, outlet angle 尾 2, outlet width l, front cover dip angle T 1, back cover plate inclination angle T 2 and blade number z, were used as optimization variables. (2) the performance prediction model of mixed flow pump is established by using RBF neural network model, in which the sample points of training and inspection are determined by orthogonal test and CFD numerical calculation method. 3) using multi-island genetic algorithm to solve the prediction model. Thus, a multi-parameter optimization method for the performance of the mixed flow pump without changing the nominal diameter of the impeller outlet is established, and the energy performance of the model pump is optimized by this method. The results of numerical calculation show that the performance of the model pump has been improved significantly after the optimization, and the efficiency of the model pump has been improved by 4.39 percentage points to 85.24% percentage points at the design flow point. The external characteristics, pressure pulsation and vibration characteristics of the optimized model pump are tested, and the internal flow changes of the model pump before and after optimization are compared based on the CFD results. The results show that at the design point, the pump head and efficiency are increased by 11.5% and 4.32 percentage points, respectively. Under the condition of 1.0 Qd, the peak value of pressure pulsation at the pump inlet, impeller outlet, guide vane outlet and pump outlet were reduced by 62.9% and 37.5% respectively. 62.5; (3) under the design condition, the vibration velocity amplitude in the direction of the T1 measuring point of the inlet flange of the impeller has been reduced by 12.2and 14.28, and the amplitude of the vibration velocity in the direction of XY _ (Z) has been reduced respectively. The amplitude of vibration velocities in the direction of X _ (Y) Z at the T _ 2 measuring point of impeller outlet flange decreases 4.71% and 9.52% 35.29% respectively; (4) after optimization, the loss of the inner head of the guide vane is reduced by 1.23, and the pressure of the low pressure area on the back of the impeller blade is improved in the middle section and the low speed area of the original model. The unstable flow caused by pressure difference is effectively restrained.
【学位授予单位】：江苏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH313

【参考文献】