Application Of Artificial Immune System In Damage Identification Of Tower Crane Boom

Nowadays,China has become a top construction machinery manufacturing base in the world.During the development of country,machineries played a crucial role in all kinds of construction projects.Though frequently used in construction,large tower cranes are rather accident-prone.According to records,more than 80% of the total mechanical accidents were caused by large tower cranes.In those accidents,tower collapse and boom break caused by overload were two major reasons.Tower cranes accidents threaten people’s safety and property.Therefore,it is timely to perform structural health testing on tower cranes,perform damage identification work,prevent major work equipment from malfunctioning,and eliminate major accidents as well as catastrophic accidents.Based on previous studies,the damage identification method of tower crane boom is studied in this thesis.The main contents are as follows:Firstly,based on finite element theory,the structural simplification principle and unit selection method of tower crane boom were analyzed.ANSYS were applied modelling the tower crane boom.The static stress analysis of the tower crane boom was performed by ANSYS to obtain the unit with maximum stress and large stress variation under typical operating conditions.Then the unit’s damage was identificated later.Dangerous damage location of the boom.Simulate the working condition of the luffing car on the crane boom at a uniform speed,apply a moving load on the crane boom,use the ANSYS finite element software to perform transient dynamic analysis on the tower crane boom,obtain the luffing car The acceleration response signal generated during the movement provides the basis for extracting the damage feature index later.Secondly,the basic theory and method of principal component analysis and wavelet packet analysis are briefly introduced.The principal component analysis is used to analyze and reduce the dimension of the obtained acceleration response signals at multiple measuring points.At the same time,wavelet packet analysis is used to construct energy spectrum as damage identification.The eigenvectors are described,and the implementation steps when using wavelet packet analysis,the selection basis of the wavelet basis,and the determination of the number of wavelet packet decomposition layers are described.Finally,the basic theory of immune algorithm and the implementation process of the main artificial immune algorithm are introduced.The position of the crane boom is judged by the aiNet immune network,and then the immune neural network is used to achieve the tower crane based on the damage location.The degree of damage to the boom is identified.The results show that the aiNet immune network can accurately identify the damage position of the tower crane boom.At the same time,the immune neural network can identify the degree of damage to the tower crane boom.