Research On Layout Of Monitoring Points On Structure Surface Based On Element Radiation

Real-time and rapid prediction of the radiated noise field of the underwater platform is helpful to analyze the sound field characteristics of the underwater platform’s own radiated noise and evaluate its own acoustic concealment performance,and provide a priori information for active noise reduction.In order to ensure the comprehensive and accurate forecast results,it is necessary to place a large number of monitoring points on the surface of the platform,but due to the limitations of the size of the underwater platform,implementation difficulty,cost control and other factors,it is impossible to place enough monitoring points on the surface of the underwater platform.It will bring about the problem that the prediction accuracy of radiated noise sound field decreases,so it is of great significance to study how to effectively improve the prediction accuracy of radiated noise.In this paper,two prediction accuracy compensation schemes are provided for the problem of forecast accuracy decline caused by insufficient monitoring points.The main research contents of this paper are as follows:1.Starting from the rapid prediction method of the radiated sound field of the structure,the advantages and disadvantages of the numerical algorithms such as the boundary element method,the equivalent source method,the kirchhoff approximation method and the element radiation superposition method are compared,and then the element radiation superposition method is used to establish the structure radiation.The sound field model was used,and the accuracy of the model was verified by simulation.At the same time,the prediction model was used as the underlying logic of compensation scheme evaluation.2.Starting with the method of evenly distributing monitoring points,the influence of the number of axial and circumferential monitoring points on the prediction accuracy of the radiated sound field of the structure is studied.The study found that there is a threshold of the number of monitoring points that varies with frequency in both the axial and circumferential directions.If this threshold is exceeded,the change in the number of monitoring points has less impact on the forecast accuracy,and vice versa,and the threshold of the number of monitoring points will increase with the frequency.increase.In addition,in order to solve the problem of insufficient monitoring points in the process of uniform distribution,two supplementary methods of virtual monitoring points are studied—the interpolation algorithm based on thin shell theory and the matrix pre-storage algorithm based on equivalent source.Both methods can improve the prediction accuracy.The interpolation algorithm based on thin shell theory introduces less error in the process of supplementing virtual monitoring points,but it is only suitable for regular shells,while the matrix pre-stored method based on the equivalent source method is not limited to shells.The shape of the volume is limited by the selection of the equivalent source surface,the precision of the equivalent source method,and the size of the storage space.3.Based on the idea of non-uniform distribution method,an optimal monitoring point distribution method is researched—the modal guarantee criterion distribution method based on uniform candidate points.Based on this method,the monitoring point distribution plan is obtained,and the analysis The influence of factors such as the number of regions,regional distribution,number of monitoring points,eigenvalue matrix,and areas that cannot be deployed on the accuracy of the forecast results is analyzed.The conclusion shows that the number of regions has an optimal value,and the forecast accuracy is the highest at this time;the more monitoring points,the higher the forecast accuracy,but limited by the actual situation and the threshold of the number of monitoring points,the number of monitoring points has an upper limit;the eigenvalue matrix should be consistent with The judgment method is matched,and the prediction accuracy is the highest in the matching case;the influence of the unplaceable area on the accuracy is related to its location.If there are monitoring points in the unplaceable area before removal,the forecast accuracy will be reduced,otherwise it will have no effect on the forecast accuracy.