The Development Of The Mechanical Press Fault Diagnosis System

As a major division of forging press machine, mechanical presses are characterized by high production efficiency, high material utilization, and an ability to improve internal organization and mechanical properties of processed parts. They are widely used in the industries of aerospace, automobile, household appliances, instrumentation, defense industry, chemical containers, electronics and others. However, it is precisely because they are widely applicable, and are often used as key devices in production line that their faults will directly reduce production efficiency, increase production costs, and lower product quality, and even cause serious casualties and adverse social consequences. Therefore, it is necessary to conduct condition monitoring and fault diagnosis of mechanical presses, and to develop a system for it. which plays a very significant role in production efficiency improvement, equipment enhancement, and safety precautions and environment improvement in press shops. However, at present, it is still hard to see research and development on a system for condition monitoring and intelligent fault diagnosis.Based on this, this study aims to develop a system for condition monitoring and intelligent fault diagnosis of a mechanical press by employing digital signal processing technology and artificial intelligence technology, and vibration mechanism of typical parts’faults of a mechanical press. The system consists of hardware and software. Hardware includes vibration sensors to measure and capture vibration signals, data acquisition cards and a computer. Software is realized by compiling programs with Microsoft’s C#language, which is primarily used for signal analysis and processing, and identification and diagnosis of faults. The software is the core of the system, including9functional modules as follows:user registration&login module, user management module, data management module, maintenance management module, data acquisition module, signal preprocessing module, classic signal analysis module, modern signal processing and analysis module, fault diagnosis module, in which signal preprocessing module, classic signal analysis module, and modern signal processing analysis module are developed based on the principle of digital signal processing, and are mainly used to pick up characteristic signal. Fault diagnosis module is developed based on back-propagation neural network, and is mainly used for intelligent identification and diagnosis of the characteristics of faults.Lastly, in this paper, the actual faults of JB23-6.3open-type tiltable press are taken as an example to verify the functionality of the system. The experimental results show that the system does well in identification and diagnosis of common faults of a mechanical press. But the system also has some defects, for example, there are only a few rules of fault diagnosis because of fewer types of mechanical press components for diagnosis and insufficient research on failure mechanisms. So,further study should be carried out in order to develop a versatile system for intelligent fault diagnosis.