| In the process of machining,the tool assumes a significant part and its condition largely determines the quality of the product to a large extent.The tool will be disturbed by many factors such as friction,chemical corrosion and material extrusion in the process of machining,which makes the life of tool machining shorten.If the tool life can be well predicted,it will help a lot in the efficiency and accuracy of machining,etc.The traditional tool life prediction methods are poorly applicable and the prediction accuracy is not high.Therefore,the establishment of a tool life prediction model that can meet the actual needs and has good prediction performance is a good guide to the growth of manufacturing efficiency and product quality assurance.According to the characteristics of the object under study in this paper,first,a tool life prediction model isestablished using RBF neural network,then the model is optimized by an improved PSO algorithm,at last,a tool life prediction management system is developed by combining the study contents and relevant computer technology.Themain research of this paper is as follows:(1)A tool life prediction model based on RBF neural network is proposed.For the problem that the traditional prediction methods have limited application scenarios and cannot predict accurately,a model is established using RBF neural network,which is selected according to the actual situation by analyzing many factors that affect the tool life,which is used as the input of the model and the tool life as the output,and the RBF neural network is trained to learn to achieve the life prediction,and finally the effectiveness of the proposed model is verified through experiments.(2)An optimized model using the improved PSO algorithm is proposed.The model optimization is carried out using the PSO algorithm for the problem of insufficient prediction accuracy of the established tool life prediction model.Firstly,the shortcomings of the PSO algorithm are improved by adjusting the learning factor,adaptive adjustment of inertia weights and high-speed convergence based on chaos theory,and then the improved PSO algorithm is used to determine the values of three key parameters,namely the center,width and connection weights of the RBF neural network in the model,so as to achieve model optimization.The average relative error of the model prediction is 6.16%,which is 17.14% lower than the original model,which improves the prediction accuracy of the model and further verifies the feasibility of the optimization method,at the same time,it also shows that the tool life prediction model of PSO-RBF neural network has better prediction effect,and can better meet the practical needs.(3)Design and implementation of a tool life prediction management system.By analyzing the functional requirements and development conditions of the system in conjunction with the research on tool life prediction,a system that can predict tool life and also manage tool information,inventory,procurement,and scheduling functions was implemented using SSM framework,Vue framework,Java development technology,and My SQL database. |
