Lightweight Design And Research On The Gearbox Of Pumping Unit

With the rapid development of China’s economy,the demand for petroleum resources in China is increasing.However,due to the greater technical difficulties in the exploitation of oil and gas reservoirs in China,the exploitation costs of our petroleum enterprises are relatively high,seriously affecting the economic benefits of our petroleum enterprises.Therefore,it is urgent to take corresponding measures to conserve energy and reduce consumption.As a key component of the pumping power transmission system,the reducer requires lightweight design,and lightweight design of the reducer box is an important component.On the premise of ensuring that the bearing capacity and service life of the box meet the requirements,making the box lighter is also of great significance for improving the competitiveness of the product.This paper proposes a method for integrating proxy model technology and combinatorial optimization algorithms to carry out lightweight design work on the gearbox box of beam pumping units.The specific research content and main research results are as follows:1.Build test functions and real polynomial response surface proxy models,Kriging proxy models,and sparse grid proxy models,respectively.The prediction accuracy of these three proxy models is compared from four error indicators: determination coefficient,root mean square error,average of absolute error values,and average error rate.The results show that the prediction accuracy of the sparse grid proxy model is the highest among the three.Therefore,the sparse grid proxy model technology is selected to conduct lightweight design research on the gearbox box of beam pumping units.2.Comparative analysis of the efficiency of optimization algorithms in lightweight box design was conducted.A typical reducer box was selected as the research object.In the single objective optimization,11 schemes were designed with the minimum mass as a single objective.It was found that the efficiency of the single optimization scheme and the combined optimization scheme in the single objective optimization study was not significantly different.The comparison found that the optimization efficiency of the multi island genetic algorithm combined with the direct search method was the best,reducing 18.1% of the initial design weight.In multi-objective optimization,11 schemes were designed with the minimum mass and the minimum first order vibration frequency as the optimization objectives.By comparison,it was found that the archived microgenetic algorithm combined with the direct search method had the highest optimization efficiency,with a weight reduction of 18.0% compared to the initial design.3.Based on the above research,a lightweight design method for the gearbox box of a beam pumping unit is proposed using a sparse mesh proxy model combined with a multi-objective combinatorial optimization algorithm(archived micro genetic algorithm combined with direct search method).After verifying the optimized structure,it was found that the strength and stiffness of the box met the design requirements,and resonance would not occur.Finally,the quality of the box was reduced by 7.3%,achieving the goal of energy conservation and consumption reduction.