Forging Machine Tool Running State Detection And Fault Diagnosis Method Study

With the rapid development of intelligent manufacturing equipment,forging machine as an important based manufacturing industry,many countries take effective measures to promote the development of intelligent manufacturing field,the annual investment of major special funds for industrial manufacturing machine tool industry up to 100 million.The technical development of forging machine tool is directly related to the development of many basic industries,as automobile,machinery,aviation,shipping and others.Therefore,the intelligent state detection and fault diagnosis level of forging machine tool has become an important symbol of a country's industrial modernization level.Industry 4.0 is one of the top ten future projects proposed by Germany.The project is co-financed by the German Federal Bureau of Education and Research and the Federal Ministry of Economy and Technology.The investment is expected to be as high as 200 million euros.It aims to upgrade the intelligent level of industrial manufacturing and establish intelligent factories with adaptability,resource efficiency and genetic engineering.Combining with the goal of Made in China 2025 to build a strong manufacturing country,the state detection and fault diagnosis of forging machine tools is of great significance in the new era of Internet in the field of intelligent manufacturing,the following are the contents of the paper.Firstly,the structure and function of JB39-630/3 closed four point forging machine are studied.For the purpose of dividing the state detection of the research object into several subsystems,according to the requirement of on-site state detection,various types of sensors are installed to collect the temperature,vibration and pressure data of the machine tool,and a detection system composed of PLC,switch,router,camera and other equipment is proposed.At the same time,the corresponding solutions are proposed for the three engineering problems.Then,the multi-source diagnosis system of forging machine tool is designed.At the same time,the system is composed of three parts: sensor press state acquisition,multi-source information fusion state detection hardware device and human-computer interaction software.The first two are responsible for data extraction in forging process.The latter has responsible for data display and alarm of forging machine,and has database storage function for subsequent data analysis.The PC configuration software adopts the Internet remote monitoring mode based on B/S framework.The whole system can be accessed remotely by Internet IP address through TCP/IP protocol.Then,the hardware and the software design of the state detection system of the forging machine is completed.Siemens S7-300 PLC is selected as the controller for the hardware device.With the power supply module,expansion module and I/O acquisition module,the state detection cabinet with specifications of 1m *2m and composed of 18 modules is designed.The cabinet is applied in the manufacturing plant.STEP7 software is used to control PLC ladder diagram and TCP/IP information communication.Meanwhile,according to the development goal of the detection system,we develop the whole system by the way of system function partition.Web Access configuration software is used to design forging machine state detection interface and realize Internet publishing;SQL Server 2008 database is used for real-time storage and follow-up analysis of forging machine data.Then,research on intelligent fault tree method for multi-source information fusion of forging machine tools is completed.Firstly,Using the vibration,temperature and lubrication times signals of the three-stage drive wheel controlled by the main motor,a 12-dimensional high-low frequency characteristic space is constructed.The vibration signal is in a complex environment,which should be denoised by wavelet packet filter,normalized energy extraction and PCA preprocessing.Secondly,a multi-source information fusion structure with complex attributes is designed independently,and an intelligent fault tree diagnosis system based on multi-source information fusion for forging machine tools is proposed.Then,the fault tree model of forging machine is established,and the minimum segmentation set is obtained by qualitative analysis.Finally,the improved recursive neural network is used to solve the weight of fault tree branches under 12-dimensional high input.The result is taken as local evidence.Combined with expert experience,the improved D-S evidence is used to fuse the information of output layer,so as to improve the diagnostic accuracy and solve the impact of single point data mutation.Finally,the experimental verification of the state detection system of the forging machine is carried out,and the test results are displayed.The user account management module,status detection and alarm module,SQL Server 2008 database module are tested and verified.The test results prove the stability of the detection system.At the same time,the intelligent fault tree based on multi-source information fusion is tested and verified.The experimental results prove the feasibility of the diagnosis method.