Optimization of Body Profile Line of Axial Flow Control Valve Based on Surrogate Model

被引：0
|
作者：
Li S. [1 ]
Hu Y. [1 ]
Li C. [1 ]
Wu H. [1 ]
Shen H. [1 ]
机构：
[1] College of Petrochemical Technology, Machinery Industry Pump and Special Valve Engineering Research Center, Lanzhou University of Technology, Gansu, Lanzhou
来源：
Huanan Ligong Daxue Xuebao/Journal of South China University of Technology (Natural Science) | 2023年 / 51卷 / 03期
基金：
中国国家自然科学基金;
关键词：
axial flow control valve; multi-objective optimization; numerical simulation; profile line optimization; surrogate model;
D O I：
10.12141/j.issn.1000-565X.220306
中图分类号：
学科分类号：
摘要：
Aiming at the problems of long optimization cycle and mechanical repetition of the optimization process for large-diameter natural gas pipeline valves at home and abroad, this paper proposed a method for optimizing the profile of axial flow control valves using Kriging surrogate model combined with NSGA-Ⅱ algorithm. Taking the DN600 axial flow regulating valve as the research object, after preliminary evaluation of its performance by Fluent and ANSYS software, it selected the flow value, maximum stress level and maximum deformation as the optimization goals. First, the B-spline curve was used to analyze the valve body type. By changing the coordinates of the control points, the parameterization of the valve body profile could be realized, and the automatic modeling of the valve body profile could be realized. The Latin hypercube sampling method was used to design 100 types of profiles, and the structural optimization sample library was obtained through numerical simulation. Kriging surrogate model for an axial flow control valve was established. Then, the NSGA-Ⅱ algorithm was used to optimize the surrogate model to obtain the Pareto front solution set, and the optimal profile was determined by studying the front solutions of the three profiles and the velocity and pressure variation curves of the horizontal section inside the initial profile. The opening-resistance characteristics, flow-resistance characteristics and internal flow of the pipeline behind the valve were studied before and after optimization. The research results show that the optimized profile flow value increased by 9. 2%, the maximum stress level decreased by 8. 46%, and the maximum deformation is reduced by 6. 2%, the fluid resistance is reduced after optimization, and the influence distance of high-speed flow in the pipeline after optimization is reduced. The performance improvement of the optimized profile proves the effectiveness of the profile optimization method and reveals the potential of the method in the field of valve profile structure optimization. © 2023 South China University of Technology. All rights reserved.
引用
收藏
页码：41 / 52
页数：11
相关论文
共 18 条
  • [1] ZHANG Zhi-chao, GAO Tai-yuan, ZHANG Lei, Aero-heating agent model based on radial basis function neural network ［J］, Acta Aeronautica et Astronautica Sinica, 42, 4, pp. 303-312, (2021)
  • [2] KRISTIN P B, EMILIO P H．, The interplay of optimization and machine learning research ［J］, Journal of Machine Learning Research, 7, 2, pp. 1265-1281, (2006)
  • [3] LI Shu-xun, Zhong LI, ZHOU Ai-min, Optimization and experimental research on the opening profile of the throttle disc of the three-way regulating ball valve ［J］, Journal of Huazhong University of Science and Technology （Natural Science Edition）, 45, 2, pp. 61-66, (2017)
  • [4] RAVINDRAN A C，, KOKJOHN S L．, Combining machine learning with 3D-CFD modeling for optimizing a DISI engine performance during Cold-Start ［J］, Energy and AI, 5, (2021)
  • [5] WANG Y, ZHANG D, Dual-convolutional neural network based aerodynamic prediction and multi-objective optimization of a compact turbine rotor, Aerospace Science and Technology, 116, (2021)
  • [6] Joint use of artificial neural networks and particle swarm optimization to determine optimal performance of an ethanol SI engine operating with negative valve overlap strategy, Energy, 204, (2020)
  • [7] CHENG Hong-liang, YI Wei-lin, JI Lu-cheng, Application of machine learning in the optimal design of centrifugal impeller with high pressure ratio ［J］, Chinese Journal of Engineering Thermophysics, 41, 11, pp. 2734-2741, (2020)
  • [8] MIAO Sen-chun, YANG Jun-hu, WANG Xiao-hui, Optimization of hydraulic turbine blade profile based on neural network-genetic algorithm, J］．Journal of Aerodynamics, 30, 8, pp. 1918-1925, (2015)
  • [9] Optimization of a single-stage double-suction centrifugal pump ［J］, Journal of Physics：Conference Series, 796, 1, pp. 1-10, (2017)
  • [10] DERAKHSHAN S，, TAVAZIANI A，, KASAEIAN N．, Numerical shape optimization of a wind turbine blades using artificial bee colony algorithm ［J］, Journal of Energy Resources Technology, 137, 5, pp. 1-12, (2015)
12