离心水泵的优化设计及其仿真

发布时间：2018-01-12 15:13

本文关键词：离心水泵的优化设计及其仿真　出处：《西北农林科技大学》2011年硕士论文　论文类型：学位论文

【摘要】：离心泵广泛应用于国民经济和社会生活的各个领域,在石油、化工、冶金、城市给排水、农业灌溉等领域发挥着重要的作用。随着计算机技术和流体动力学(CFD)技术的发展,人们对离心泵的性能要求也越来越高。现阶段离心泵的优化设计大多数是单目标优化且没有考虑叶轮和蜗壳的匹配关系,流场模拟忽略了叶轮和蜗壳间的耦合特性,导致离心泵的优化设计结果不甚理想。本文的研究方法和内容对于改善离心泵性能和完善离心泵优化设计理论具有借鉴意义。本文以低比转速离心泵为研究对象,建立了离心泵的损失、汽蚀性能和稳定性的多目标优化数学模型,采用Matlab遗传算法工具箱进行优化计算。根据优化后的几何参数建立离心泵的流场模型并进行了网格划分和边界条件设定。数值模拟离心泵全流场以及叶轮和蜗壳耦合特性并预测了其扬程和效率等性能指标。考虑离心泵叶轮和蜗壳间的匹配关系,确定最佳工况点下的理论扬程;建立了以离心泵的能量损失、汽蚀性能和H -Q曲线无驼峰的多目标优化数学模型,并根据分目标函数的重要性不同,利用线性加权法统一目标函数;确定了适合本研究问题的优化变量范围。采用Matlab遗传算法工具箱进行优化计算,得出离心泵几何参数的最优组合解,使离心泵的能量损失减少了1130.7w,汽蚀余量降低了0.4258m,离心泵的H -Q曲线稳定性也有很大的改善,理论分析优化结果的可行性。采用Pro/ENGINEER软件进行离心泵的叶轮和蜗壳三维流场的建模,使用CFD前处理软件Gambit对离心泵整机三维流场模型进行网格划分,确定了边界条件和流体类型。选择Fluent软件对离心泵的整体三维流场以及叶轮和蜗壳间的耦合特性进行数值模拟。选择Fluent的后处理功能得到了离心泵内部全流场压力云图、速度矢量以及叶轮和蜗壳耦合面上的速度矢量图;分析了不同工况下离心泵内部全流场的压力和速度分布的变化;从数值模拟数据中导出离心泵进出口处的压力和叶轮绕Z轴的力矩,对离心泵的理论扬程和效率进行性能预测,进一步验证了数值模拟和优化设计的正确性。本文将多目标优化设计、遗传算法和CFD流场数值模拟结合起来,对改善离心泵的整机性能和研究全流场的复杂运动规律具有重要的借鉴意义。
[Abstract]:Centrifugal pumps are widely used in various fields of national economy and social life, in petroleum, chemical, metallurgy, urban water supply and drainage. Agricultural irrigation and other fields play an important role. With the development of computer technology and fluid dynamics (CFDs) technology. At this stage, the optimization design of centrifugal pump is mostly single-objective optimization and does not consider the matching relationship between impeller and volute. The coupling between the impeller and the volute is neglected in the flow field simulation. The research methods and contents of this paper are useful for improving the performance of centrifugal pump and improving the theory of optimal design of centrifugal pump. In this paper, a multi-objective optimization mathematical model of centrifugal pump's loss, cavitation performance and stability is established. Matlab genetic algorithm toolbox is used for optimization calculation. According to the optimized geometric parameters, the flow field model of centrifugal pump is established, and the grid and boundary conditions are set up. The full flow field of centrifugal pump is simulated numerically. The coupling characteristics of impeller and volute are predicted and the performance indexes such as lift and efficiency are predicted. Considering the matching relationship between impeller and volute of centrifugal pump, the theoretical head at the best working condition is determined. A multi-objective optimization mathematical model with energy loss, cavitation performance and H-Q curve without hump of centrifugal pump is established. According to the importance of sub-objective function, linear weighting method is used to unify the objective function. The range of optimization variables suitable for this problem is determined. By using Matlab genetic algorithm toolbox, the optimal combination solution of geometric parameters of centrifugal pump is obtained, and the energy loss of centrifugal pump is reduced by 1130.7w. The cavitation margin is reduced by 0.4258m, and the stability of H-Q curve of centrifugal pump is improved greatly. The three-dimensional flow field of impeller and volute of centrifugal pump is modeled by Pro/ENGINEER software. The three-dimensional flow field model of centrifugal pump was meshed with CFD pre-processing software Gambit. The boundary conditions and fluid types are determined, and the whole three-dimensional flow field of centrifugal pump and the coupling characteristics between impeller and volute are numerically simulated with Fluent software. The post-processing function of Fluent is selected to obtain the pressure cloud diagram, velocity vector and velocity vector diagram on the coupling surface of impeller and volute in the whole flow field of centrifugal pump. The variation of pressure and velocity distribution in the whole flow field of centrifugal pump under different working conditions is analyzed. The pressure at the inlet and outlet of the centrifugal pump and the torque of the impeller around the Z axis are derived from the numerical simulation data. The theoretical head and efficiency of the centrifugal pump are predicted, which further verifies the correctness of the numerical simulation and the optimization design. In this paper, the combination of multi-objective optimization design, genetic algorithm and numerical simulation of CFD flow field is of great significance for improving the performance of centrifugal pump and studying the complex motion law of the whole flow field.
【学位授予单位】：西北农林科技大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH311

【引证文献】