Research In Damage Detection Theory Of K Type Derrick Structure Based On Frequency Domain System Identification

Rig derrick is one of the most important equipment in hoist system. People can install crownblock and hang tackle, hoisting hook and special tools on it. Some derricks in using have been used for a long time in the open, many parts of the derrick have distortion and corruption of different extent, and enlarged screw. Thus, we must detect and evaluate the damage and fully realize the actual status of the derrick structure in using. If we can find the damage immediately, diagnose the location and the extent of the damage, constitute the correct servicing strategy, resume the ability of bearing the weight of load, it is good to postpone the natural life of derrick and to provide the proof of using present derrick economically. Furthermore, it is very important to avoid disaster and to protect the people's lives and estate. The study on damage identification of derrick and making progress in evaluating the ability of load and safety has important realistic value and theoretics significance.Firstly, on the basis of collection and analysis of much literature, this article analyses the technology of structural damage identification of domestic and foreign research present situation, and introduced the basic theory of modal experiment and the basic theory and method of the derrick's modal parameter based on frequency domain system identification.Secondly, the damage identification method based on BP neural networks is discussed and analyzed. All of these such as the basic consept of artificial neural networks, the characteristic, the model, its algorithm are introduced in this process. At the same time, This article elaborates BP neural networks, BP classic algorithm, the Lavender-Marquardt algorithm on the condition of optimization, and BP neural networks that realized in the MATLAB in details.In general, aiming at the problem on the damage location and extent of JJ160/ 41-K type derrick, the method of damage identification based on artificial neural networks is introduced. Vibration modal analysis theory is integrated with BP neural networks so as to detect derrick's damage with the help of ANSYS and MATLAB, at the same time, it has been successful in detecting damaged position and extent likely occurring, the method based on vibration modal analysis theory and BP neural networks is satisfied in this paper. Along with the improvement of precision of the modal analytic apparatus and the technology of identifying parameter, the results is that the neural network applied in structure is feasible and have good application perspective.