Research On Cutting Tool Life Prediction And Management Technology Of Milling In Titanium Alloys Based On Data-driven

Recently,the development of traditional manufacturing has promoted in the direction of intelligence,automation,and digitization with the rapid growth of artificial intelligence,robotics,and the Internet of Things.As a significant tool in the manufacturing process,the efficient use and management of cutting tools will have a significant impact on the efficiency and the cost of the entire production process.The accuracy of tool life prediction and the effectiveness of management are two main problems in this field.The remaining useful life(RUL)prognostication model in the cutting tool processing process has been established based on the data-driven method together with a tool management database system developed in this paper,the main work is as follows:(1)An integrated life prediction model based on the trajectory similarity prediction algorithm and the differential evolution support vector machine algorithm is established.The former can predict the RUL of the cutting tool according to the variation law of the historical signal feature quantity.The latter can realize the optimal solution based on the limited data amount and improve the model prediction accuracy.In addition,the complicated parameter setting problem is solved by using the differential evolution algorithm.(2)A number of tools samples are selected for life prediction experiments.The force signal which collected in the cutting process is analyzed in time domain,frequency domain,and wavelet analysis,the relationship between the signal feature quantity and the cutting tool wear amount is explored,five eigenvectors such as the root mean square value and energy of the signal are determined as the input values of the prognostic model.(3)The prediction accuracy of the integrated life prediction model is verified by experiments and the respectively accuracy in stable cutting stage of the sample tool was 88.5%,87.5% and 90.5%.The result shows that the integrated prediction model proposed in this paper has higher precision in RUL prediction of cutting tools compared with other models.(4)A tool management database system based on C/S architecture is developed whose development platform is MATLAB and database technology is SQL Server 2008 R2,integrating life management,basic information management and procurement management functions,and it can predict the RUL of cutting tools by extracting the force signal.