| Hydraulic Automatic Gauge Control (HAGC) system is an important large-scale core system in strip mill. It is a key to determine the steel band quality and the yield. The characteristics of heavy-load, high degree of accuracy, high response and hydromechatronics coupling are taken on by Hydraulic AGC, which is composed of the hydraulic system, the control system, the breast roller, the servo component and so on. It gets conversion mechanical energy into hydraulic energy with the help of conversion components including force motor, servo valve, cylinders and work interface consisting of rollers, work pieces and dynamic lubricating bearings, forms the cross multiply coupling of the hydromechatronics and complicated internal feedback, it is easy to produce coupling shaking or fault. In 2000, coupling shake of 1450 hot rolling AGC and main transmission system caused the serious out-of-tolerance of steel strip, broke the shaft finally in a steel plant. In 1996, hydraulic AGC of Panzhihua Iron and Steel Co.,Ltd. broke down for a time by the destabilizing fault, and caused enormous losses. So the testing and fault diagnosis of hydraulic AGC are becoming the focus in the iron and steel industry of our country.Based on generalizing the domestic and foreign a large of references, this dissertation systematically analyzes the principle and model of the strip mill hydraulic AGC, combines with the actual project experiences of Panzhihua Iron and Steel Co.,Ltd., ShangHai Smelting Co.,Ltd., and Wuhan Iron and Steel Co.,Ltd. It carries out profound researches on fault diagnosis, system development and off-line testing of the hydraulic system.The full dissertation is divided into seven chapters. The research purpose and meaning are briefly explained in Chapter one, which makes the synthesis narration on present situation and the development tendency of the domestic and foreign related hydraulic testing and the diagnosis technology.The rolling mill hydraulic AGC composition and the control principle are analyzed in Chapter two, which researches on the hydraulic AGC fault types, the mechanism and the failure form. Taking these as the foundation, the fault tree and diagnosis strategies of the hydraulic AGC system are proposed in the dissertation. Italso aims at the particularity of the rolling mill hydraulic AGC control system, and proposes the method of utilizing the on-line monitor, the off-line experiment, the fault imitation to carry on the monitor and the diagnosis, and constructs the overall frame of the hydraulic AGC monitoring and diagnosis system.The dynamic characteristic and the static characteristic experiment and the testing technology of the hydraulic AGC are studied in Chapter three, which develops the integrated testing system of 15000KN the hydraulic AGC synthesis in the Baoshan Iron and Steel Co.,Ltd. and Panzhihua Iron and Steel Co.,Ltd. As it is unable to extract the important characteristic and the fault characteristic of certain hydraulic AGC elements by the ordinary real-time on-line monitoring and the diagnosis technology, so we reconstruct the AGC prototype system test platform of the fault decoupling function, combine with the periodic experiments of the servo elements and the on-line monitoring. It is important to the correct diagnosis and high diagnosis precision of the hydraulic AGC. The dynamic model of the experimental system are established in the dissertation, which proposes the rolling mill hydraulic AGC off-line scene prototype dynamic experiment and the AGC cylinder bidirectional entire journey friction characteristics testing method and the evaluation criteria;it proposes the sequential two-loop control strategy for oil temperature, and solves the question of the system oil temperature controlling low precision of greatly lagged temperature controller. In view of the rolling mill servo valves having the wide range of flow capacity, we develop the flow measurement technology which based on the frequency measurement/ circumferentor, and improves the discharge characteristic testing precision. The engineering project is unified in the dissertation, which develops experimental control software and the rolling mill cylinder bidirectional total excursion friction characteristic testing technology. The inertia force and the test disturbance are eliminated by the strength - speed double stage closed loop and the non- perturbation shift.The extraction of the hydraulic pressure AGC fault characteristic parameters and the disposal of the signal processing technology are studied in Chapter four. The on-line monitoring system which based on the network and the PLC rolling mill hydraulic AGC are established in the chapter, which builds the hydraulic AGC on-linemonitor software system platform with WinCC of Siemens configuration software and develops the user interface. The testing method of basing on the spectrum analytic method of the hydraulic AGC dynamic response is studied in the dissertation, and the method can effectively extract the dynamic characteristic parameters of the hydraulic AGC under the on-line way. The method of the wavelet residual error recover is used to extract the residual error sign of the system weak fault information in the chapter. According to the residual error, we can identify the diagnosis serious degree, start warning the user. The wavelet is used to analyze the position of singular port of a function and degree of singular state and distinguish the sudden fault of the hydraulic AGC system.The questions of the hydraulic AGC fault simulation are mainly researched in Chapter five. Based on AMESim and unit of the graphic physical model method, fault simulation platform is developed in the chapter, which establishes three levels of electricity feedback electro-hydraulic servo valve and the hydraulic AGC system fault simulation model. The dissertation carries on the fault simulation to the common fault of the part and the system.The various intelligent diagnostic methods are analyzed in Chapter six. The BP neural network and the genetic algorithms are used in the chapter, which establishes the genetic neural network model of the complex electro-hydraulic system failure pattern recognition. The fusion technology of the genetic algorithm and the neural network are discussed in the chapter, which uses Visual C++ 6.0 and combines the technology of the database to develop the software of the hydraulic AGC fault intelligent diagnosis. The several key questions are studied in the chapter: the knowledge base establishment, the inference machine structure, the explanation mechanism design as well as the software performance. Taking the electro-hydraulic servo valve as the example, this chapter develops fault pattern recognition system based on the servo valve characteristic curve.The groundwork of the full text and conclusions of the question studied in an all-round way are summarized in Chapter seven, which proposes the tentative plan to the next research work.
|
PDF Full Text Request