离心泵智能优化设计体系研究现状与展望

Research Status and Prospect of Intelligent Optimization Design System for Centrifugal Pumps

摘要: 离心泵作为关键流体输送设备，其优化设计对提升能效、降低运行成本具有重要意义。本文从性能影响因素、主流优化方法及优化目标三个维度系统梳理了当前研究现状。研究表明，叶轮与蜗壳的几何参数(如叶片数、出口安放角、包角等)对泵的性能具有显著影响，是优化的基础。在方法层面，基于CFD的数值模拟已成为核心分析工具；以正交设计、拉丁超立方抽样为代表的试验设计方法能高效构建样本空间；以响应面、Kriging、神经网络等为代表的代理模型技术有效替代了耗时仿真；遗传算法、粒子群算法等智能优化方法则实现了多目标高效寻优。当前研究已呈现出从“经验驱动”向“数据与模型协同驱动”的转变，并在提升效率、扬程、空化性能及降低振动等方面取得显著成效。未来，随着智能算法与多物理场仿真的进一步融合，离心泵优化设计将朝着更高效、更精准、更系统的方向发展，为实现泵装备的绿色化与智能化提供坚实支撑。

Abstract: Centrifugal pumps, as key fluid transfer equipment, have an optimized design that is of great significance for improving energy efficiency and reducing operating costs. This paper systematically reviews the current research status from three dimensions: performance influencing factors, mainstream optimization methods, and optimization objectives. Research shows that the geometric parameters of the impeller and volute (such as the number of blades, outlet installation angle, wrap angle, etc.) have a significant impact on the performance of the pump and are the basis for optimization. At the method level, CFD-based numerical simulation has become a core analysis tool; experimental design methods represented by orthogonal design and Latin hypercube sampling can efficiently construct sample spaces; surrogate model techniques represented by response surface, Kriging, and neural networks effectively replace time-consuming simulations; intelligent optimization methods such as genetic algorithms and particle swarm optimization achieve efficient multi-objective optimization. Current research has shown a shift from “experience-driven” to “data and model collaborative-driven”, and has achieved significant results in improving efficiency, head, cavitation performance, and reducing vibration. In the future, with the further integration of intelligent algorithms and multi-physics field simulation, the optimization design of centrifugal pumps will develop towards more efficient, more accurate, and more systematic directions, providing solid support for the greening and intelligence of pump equipment.