| The de velopment of automated machinery m akes t he steel m esh w elding production e fficiency gr eatly i mproved,however t here has be en s ome w elding problems appeared, such as f alse welding and lack of weld. These problems often could not been de tected by a rtificial m anipulation e ffectively, b ecause o f i ts time-consuming and the negative impacts of human factors. So there need a advanced equipments a nd c onvenience instruments be introduced to w elding pr ocess f or quantitative an alysis t o minimize the f actors, as t o ach ieve t he p urpose o f fa st reflection of the welding quality.Welding quality monitoring process uses the voltage and current signals in the welding process, directly or indirectly. By analyzed the signals to get information of the w elding qua lity. I n t his pa per, one monitoring s ystem for s creen pr ojection welding was put forward against the false welding and lack of weld in the screen welding, the hardware equipments were also selected to the industrial environment.In this paper, t he a dvantages of L abVIEW and MATLAB are used. I n LabVIEW,by using the neural network tool of MATLAB to compile the part of choice of welding parameters. Using BP neural network training can be required to reference welding pa rameters, s aving opt ions, a djust parameters of t he t ime. T he m ain application of LabVIEW graphical programming language software components in monitoring system t o a nalyze s ignals in time do main a nd f requency domain, a nd could obt ain e ffective qua ntitative da ta a bout w elding s pot qua lity f rom t he spectrogram. T hen cal culated the dynamic resistance welding process curve, b y calculating the maximum of curve to reflect the size of nugget, and then judge the quality of welding. And through a large number of tests to verify the accuracy of the system, and prove the system is able to achieve the functions of real-time quality monitoring in the welding process.
|
PDF Full Text Request