Nondestructive Detection Of Highway Grid Structure Canopy

The work of collection of expressway charge becomes more and more important following the great development of expressway. And the effects of charge for canopy which can keep out wind and rain from work of charge become more important as well. Because of the development of society and the progress of science, the charge for canopy has changed from brickwork to brick structure and frame structure and to grid structure which is used widely nowadays. The maintenance of charge for canopy of grid structure becomes more important because it is increasing. The specialness and complexity of grid structure make the repair more difficult and cost higher. This thesis put forward a method—nondestructive testing to the charge for canopy of grid structure. This method can make prevention decision to the damage. It fixes up test points (sensors) in the grid structure so as to get the frequency of vibration between the rods. And then calculate the cross-correlation function to locate the damage. It is the preliminary positioning method mentioned in this thesis. Then I choose extreme learning machine and support vector machine to get the accurate detection, and to find the accurate truss damage rod parts further. The method offered by this thesis will give more reference and help to the detection of grid structure, avoiding the damage of charge for canopy of grid structure timely, and to reduce the hidden danger of its safety.