摘要
针对传统优化方法效率低的不足,应用响应面模型方法和多目标遗传算法对某型钻井泵泵体进行了轻量化设计。首先,利用中心复合试验的设计方法,通过在设计空间抽取样本点进行数值模拟,建立了由3个设计参数所决定的泵体的响应面模型。其次,利用多目标遗传算法的循环逼近优化方法对响应面模型进行优化,结合实际从Pareto最优解集中选择1组作为最终优化参数。最后,利用优化参数进行仿真结果验证。仿真结果表明:在泵体前板变形和曲轴轴承座周围变形基本不变的条件下,优化解可使泵体质量减小9%左右。该优化设计方法比传统优化方法更高效,可以将其应用于类似机械结构的优化设计中。
To address the low efficiency of the traditional optimization method,the response surface model method and the multi-objective genetic algorithm have been used to conduct lightweight design of a certain type of drilling pump body.Firstly,using the design method of central composite test,numerical simulation of sampling points selected in the design space has been conducted to establish the response surface model of the pump body detennined by three design parameters.Then,the response surface model has been optimized using the cyclic approximation optimization method of multi-objective genetic algorithm.A set of parameters is selected as the final optimization parameters from the Pareto optimal solution set taking real work scenario into consideration.Finally,the simulation results are verified by the optimization parameters.The simulation results show that the optimal solution can reduce the pump mass by about9%under the condition that the deformation of the front plate and around the crankshaft bearing seat of the pump body basically remain the same.The optimized design method is more efficient than the traditional optimization method and could be applied to the optimization design of similar mechanical structure.
作者
刘军
张宏
常平
黄华
Liu Jun;Zhang Hong;Chang Ping;Huang Hua(School of Mechanical and Electronical Engineering, Lanzhou University of Technology;Lanzhou LS Petroleum Equipment Engineering Co. , Ltd.)
出处
《石油机械》
2017年第10期22-26,共5页
China Petroleum Machinery
基金
国家工信部智能制造综合标准化与新模式应用示范项目"石油装备关键零部件数字化铸造及智能制造新模式"(2016125)
关键词
钻井泵
响应面模型
遗传算法
轻量化
优化设计
drilling pump
response surface model
genetic algorithm
lightweight
design optimization
