Mechanical Tool Life Cycle Management System Development And Research

With the proposal of 2025 manufacturing industry development plan.Internet plus and other newly-developed technique become an engine to accelerate digital,networked and intelligent manufacturing development.Information–based manufacturing will greatly assist the development of industry innovation.As an important part of manufacturing,tool-management system's intelligent,automated level greatly attract enterprises' attention.However,there are still lots of problem in achieving this goal.This paper studies part of problem in tool management system development and summarizes some critical technologies to achieve the intelligent,automated goal.In order to implement this management system,this paper mainly studies the following contents:A coding method,which combines group and flexible coding technology,is proposed.A tool component diagram is drew by analyzing the structure of mechanical tool and applying group classification method,which can help to classify machine tools.These classified tools can be labeled using a flexible coding method with high readability and correctness.The whole coding module uses an automated method to transform tool information into letters and numbers.After labeled,A cutter QR code can be easily produced to improve the management efficiency.In this paper,A tool cycle life management platform is developed,which not only provide a scheme to handle tool data in computer but also design a perfect operation management work flow to meet enterprise's customized requirements.The system framework make a tight integration with purchase department,manufacturing shop,the internet of things.To collect workshop data for data mining in production management,an interface routine is also designed.In order to comb the company's tool management operation,like tool checking in and out,purchase,a method of module analysis is used to produce a system function module and work flow chart.The system can run under a safe and efficient circumstance with a well designed database.A lots E-R diagram is designed and a clear relationship between them is set up which reduces data redundancy.Tool life prediction method is a difficult task in tool's management development.Tools life directly influence tool selection strategy,tool exchange and so on.In this paper,a multiple module BP networks prediction method is proposed.The method uses genetic algorithms to optimize weights and bias and generated the best structure by analyzing Gaussian distribution law of prediction error.In order to give a more reasonable estimation after many prediction cells give their results,the paper proposes an valuation evaluation algorithm,which can filter the bad prediction value.It reduces the difficulty of tool life prediction modeling and avoids unreasonable prediction result.At the end of this paper,most of research achievements are applied to this tool management system.With the actual usage in enterprise,the system's rationality and high-efficiency is proved.