Research On Drag Reduction Optimization Of CFD Flow Channel Profile Of Axial Flow Control Valve

The government work report of 2021 pointed out: We will do a solid job in the comprehensive green and low-carbon transition such as carbon peaking and carbon neutrality,as soon as possible to achieve sustainable economic development.Since the "12th Five-Year Plan" energy structure adjustment,natural gas has gradually become the main force in the energy transition.With the construction of the national "West-East Gas Pipeline" project,natural gas has been mainl y transported by pipelines,among which axial flow control valve has become one of the important control components in the pressure adjustment and distribution process of long-distance pipelines to urban receiving stations.The profile design of large-caliber axial-flow valve structure relies on the surveying and mapping of small and medium caliber axial-flow valve to scale up and so on.The lack of relevant basic research results in the larger conveying resistance of large-caliber axial-flow regulating valve,and the overall product is not "low-carbon".This paper optimizes the flow path profile of the axial flow control valve to reduce the local resistance loss caused by the valve during the conveying process,so as to achieve the purpose of saving energy and reducing consumption,and improving the efficiency of pipeline transportation efficiency.On the basis of investigating relevant domestic and foreign research,the optimal Latin hypercube experimental design was adopted,and the axial flow control valve profile is optimized based on methods such as comput ational fluid dynamics(CFD),BP neural network and NSGA-II genetic algorithm.The following research works have been carried out:(1)Based on the CFD method,considering the influence of the boundary layer and the surface roughness of different parts on the calculation results,a numerical simulation method that can accurately calculate the internal flow field information of the axial flow control valve is obtained,and the axial flow control under di fferent opening degrees is studied.With the internal flow field information of the valve,the flow characteristic curve of the axial flow control valve is obtained.(2)Sample points are obtained based on the optimal Latin hypercube experimental design method,and the target parameter values are automaticall y calculated using Isight software integrated with 3D software Solid Works and CFD software,and a database between design variables and target parameters is established.Based on the surrogate model prediction theory,the sample database is trained to obtain the surrogate model between the target parameter value of the axial flow control valve and the design variable parameter,and the optimal parameter solution is obtained by combining with the NSGA-II genetic algorithm.(3)Based on the prediction theory of different agent models,four test functions are used to compare the prediction accuracy of different agent models for different problems.Use the axial flow control valve profile optimization sample database to compare and analyze the prediction accurac y of different proxy models,and find the proxy model with higher prediction accuracy.(4)The Pareto solution set is obtained by using the BP neural network model combined with the NSGA-II algorithm,and the design parameter values corresponding to the optimal individual with the flow resistance coefficient and the basin volume are used for numerical simulation calculation,and the structure parameters and internal parameters of the original axial flow control valve are calculated.Compare and analyze the flow field information.