Research On Thermal Wave Testing For Adhesive Bonded Structure Damage Of Hidden Frame Glass Curtain Wall

Hidden frame glass curtain wall is widely used in large-scale buildings,especially the ones in central business districts.Hidden frame glass curtain wall safety accidents were reported in recent years,thus detection of adhesive bonded structure damage of hidden frame glass curtain wall not only guarantees production and living environment,but also keeps public safety.To improve the efficiency and simplify operation against present methods of detection,this dissertation presents thermal wave testing.In the method,heating resource heats the subjects and then thermal imager captures the thermal images containing temperature differences between damaged areas and normal areas.Therefore,damage evaluation can be processed reliably as well as adhesive bonded structure damage can be detected intuitively and rapidly.The research on thermal wave testing for adhesive bonded structure damage of hidden frame glass curtain wall is significant to its detection,evaluation and early warning.The work is supported by Guangdong Technology Plan Project(2016A040403013).The main work of this dissertation contains:(1)Typical forms of adhesive bonded structure damage were discussed.Then heat conduction differential equations and heat conduction system were established to analyze characteristics of thermal wave propagation.Based on the characteristics,heating function and radiator were selected to build up thermal wave testing platform.(2)Mid-and-far infrared heating source based on fuzzy self-tuning PID control was developed after the accomplishment of electronic circuit design.Steady time of the source is40s without overshoot and steady-state error.Including sizes,material and craftworks,reflector of the resource was designed in consideration of the parameters of thermal imager.Radiant illumination of the source was optimized by simulated annealing algorithm and its result was simulated.The simulated result shows uniformity of the source is 1155W/m~2.(3)Thermal sequence enhancement algorithm was proposed.In the algorithm,optimum thermal images were selected according to weighted information entropy.Then laplacian pyramids were built up to separate features and details of the thermal images and fused through different rules.Additionally,upper computer with two interfaces was designed using multithreading technology.(4)Thermal wave testing platform was built up and glass curtain wall specimens were manufactured.Method of heating time was proposed after the relationship between heating time and uniformity was found out.The result is heating time should not less than 6s.Then,experiments on thermal wave testing with different heater in the daytime and nighttime were carried out.The result shows that infrared heating resource reaches best effect in the nighttime with performance increasing 44.36%under depth no more than 3mm.Thermal sequence enhancement algorithm and gray-scale linear enhancement algorithm were compared.The result shows that fusion enhancement algorithm increases the visibility of damage especially the deep damage(not less than 4mm)as well as the amount of information.