Fuzzy Mode Identification Method For Propagation Of 3-d Straight Through Crack Of Ear Plate Of Mast Structure

With the development of industry, high strength materials such as metal have been widely used in construction industry as component of steel structure buildings, bridge, transmission tower and mast etc. But it also has some limitations. It is arouse local fatigue such that structure damage could easily happen by alternating load because steel structure's dead weight is light and its highly span is wider than general structure. Crack is a good manifestation of local fatigue.Accidents caused by crack have occurred among all kinds of industries, such as pressure vessel, drilling platform, petroleum pipeline, atomic energy reactor structure, motor case and aircraft landing gear etc. Investigation also indicted that surface crack propagation is the main cause of most accidents and component surface crack has extreme devastating effect on production and utilization. Hence it is crucial to focus the research on initiation and propagation of structure and component crack.Guyed mast structure is a high flexibility structure system with small stiffness and is comprised of vertical shaft and cordelles laid along high direction, what has been applied to communication, broadcast and data transmission service widely and played an important role in national economical development. Collapse of guyed mast has been caused by local fatigue of ear plate all the time. This paper concentrated on identification of crack's propagation which is one of two stage of fatigue, namely initiation and propagation.This paper introduces a finite element solid modeling to construct 3-D straight through crack of ear plate of mast structure, which is different from conventional bottom-up per-node direct modeling. Given the fact that the latter modeling method have to define coordinates of every single node precisely, then constructing model from dot to line, from line to surface,it is generally extremely complicated and not very efficient and flexible. Considering the need of mesh generation and result output, we create an isolated body near crack zone to relieve the conflict of mesh generation and crack parameterized modeling and to accelerate computing times simultaneously. Fuzzy mode identification method for crack identification is built based on fuzzy mode library and a membership function to define membership degree between test mode and library mode. Hence it is paramount to build an effective fuzzy mode identification library for the chosen identification method. This paper define sensitive degree of ear plate surface's different node under distinct crack parameters and choose ten most sensitive node as sample to construct a mapping relation of fuzzy set between crack length of different conditions and nodes'stress field. Finally I choose stress field under two different crack's length as test data set to valid the proposed fuzzy mode identification model, and the experimental results have demonstrated that the proposed model is very effective.