• TECHNOLOGIES

    GRAPHICAL USER INTERFACE FOR OPTIMIZATION OF ELECTRON BEAM WELDING BY NEURAL AND REGRESSION MODELS FOR OBTAINING DEFECTFREE WELDS

    Machines. Technologies. Materials., Vol. 12 (2018), Issue 2, pg(s) 76-79

    This paper considers the process electron beam welding of stainless steel type 1H18NT in vacuum. Based on experimental data, the influence of the variations of the following process parameters: electron beam power, welding velocity, the distances from the magnetic lens of the electron gun to the beam focus and to the surface of the treated sample is investigated.

    Neural and regression models for the geometry characteristics of the welded joints: surface of the weld cross-sections, weld depths and mean weld widths of the samples are estimated, as well as models for defining the areas of the process parameters, where the appearance of defects is or is not expected. The obtained models are used for developing the graphical user interface aiming investigation and prediction of the electron beam welding characteristics and process parameter optimization. This software can be implemented for supporting the operator’s choice of appropriate work regimes, obtaining the required welds quality standards, for education and investigations.

  • THE SCANING NOZZLE HOT AIR SYSTEM FOR THERMOGRAPHIC DETECTION OF THE SURFACE INCORPORATED HIDDEN DEFECTS

    Machines. Technologies. Materials., Vol. 8 (2014), Issue 11, pg(s) 49-51

    The scanning nozzle hot air system for thermographic detection of the surface incorporated hidden defects is proposed. Subsurface defects in the sample are detected using the high resolution thermal imaging camera FLIR SC7000. To introduce additional energy in are searched sample, a scanning hot air (about 110°C) nozzle is applied (a patent application P.403346). The hidden defect causes a temperature increase in comparison with the remaining area what is a result of changes in emissivity. The results are compared with the pulse thermography method using the xenon lamp for excitation.

  • THE TERMOGRAPHIC ANALYSIS OF THE WELDING BY TIG

    Machines. Technologies. Materials., Vol. 8 (2014), Issue 11, pg(s) 46-48

    Thermography measurements allow to detect the defects that may appear on a joint at welding of components. Energy pulse generated by a xenon lamp with adequate power in a short period of time is sufficient for thermal excitation and enables to register the temperature distribution using the thermography high resolution camera FLIR SC7000. The impulse with 6kJ energy and 6ms time generate sufficient power to measure the temperature distribution on the surface of the weld tested. During cooling the temperature of the area with defect changes more slowly than in the areas without defects, because of to the less intense heat dissipation. This allows the registration of defects in welds "on-line" at the production process. Material used for analysis detection of defects in the welded joints is Inconel 718, stainless steel 410 and stainless steel 321. The peak energy which flow throw the samples with defects in the welded joints its completely or partially blocked. It cause different temperature distribution on the surface in the places where the connection discontinuity take place.

  • TECHNOLOGICAL BASIS OF “INDUSTRY 4.0”

    NEURAL NETWORKS FOR DEFECTIVENESS MODELING AT ELECTRON BEAM WELDING

    Industry 4.0, Vol. 2 (2017), Issue 1, pg(s) 5-8

    This paper considers the process electron beam welding in vacuum of stainless steel 1H18NT. Neural network based models are developed and used for the description of the defectiveness, depending on the process parameters – electron beam power, welding velocity, the distance between the main surface of the magnetic lens of the electron gun and the beam focusing plane and the distance between the main surface of the magnetic lens of the electron gun and the sample surface. Neural network (NN) models, based on a multi-layered feedforward neural network, trained with Levenberg-Marquardt error backpropagation algorithm are compared with NN models, based on Pattern recording neural network, trained with Conjugate Gradient Algorithm. The neural networks are trained, verified and tested using a set of experimental data. The obtained models are implemented to predict areas of process parameters, where the appearance of defects is most probable and the location of welding regimes that should be avoided.