|
Mechanical equipment is material foundation of the enterprise production, and is an important part of the productivity. In industrial production, as time goes on, it would be have various forms of wear, and lead to the equipment efficiency and precision reduce. Thus we make our products quality decreased, even a serious accident will result in equipment. So carring out the subject research on the equipment failure and performance diagnostic technology is of great significance.Fault and performance diagnosis is a rapidly developing interdisciplinary, It combines testing technology, software engineering, computer technology, signal processing, pattern recognition, artificial intelligence, decision sciences, information science, and many other modern scientific technology, and become not only focus on theoretical research, but also to the practical application of modern engineering science, and gradually form a complete system, theoretical rigor and significance of great works of the new discipline. see from most our country enterprise on the equipment malfunction disposal system, they all basically adopts habits”regular maintenance”and”afterwards maintenance”. Though”regular maintenance”can find some early fault, and prevent the occurrence of parts of the accidents, but it will be forming some unnecessary waste, because they are too frequently replacement parts for not need maintenance equipment that caused excess maintenance. For”afterwards maintenance”, any abnormal condition and abrupt failure which causes downtime repair and production beats pause, All production processes cause necessarily the backlog, serious influence on the smooth completion of the production plan. Therefore it need early carry out fault diagnosis for equipment in the process. Do“ancestors step found fault, eliminate hidden dangers before the failure”.This topic research content is the condition monitoring and fault diagnosis that depended on the 5# Blast-Furnace Top pressure Recovery Turbine Unit of the Laiwu Steel Corporation Energy Utility Plant.The ultimate goal of the research is finally get the diagnostic conclusion which has important guiding role to repair fault, through analysis comprehensively and recognize the equipment working status, fault type and the severity of the renovation, according to different conditions to TRT, which shows in the vibration, noise, temperature, hydraulic, rotor, speed, odor, leakage and other characteristics of all rules to the comprehen- sive analysis of the information. Based on the analysis to the common fault mechanism basis of TRT, A deep research on the principle of neural network technology method and its application technical features, constructs the neural network model of TRT fault diagnosis system through combining malfunction features to find corresponding characteristic. Mean- while do some neural network training in the light of TRT turbine rotor fault samples, it realizes the TRT system fault diagnosis of interface development that based on VB software. This paper applied fault diagnosis technology to achieve the protection to the unit, avoids unnecessary parking which caused by the blast furnace top pressure increases instantaneous. It provides strong support to ensure long-term safe and stable, smooth running and increase economic benefits to TRT.
|