• 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.

  • TECHNOLOGIES

    ROBUST DESIGN AND MULTIPLE CRITERIA OPTIMIZATION OF ELECTRON BEAM GRAFTING OF CORN STARCH

    Machines. Technologies. Materials., Vol. 11 (2017), Issue 12, pg(s) 583-586

    Electron beam (EB) irradiation has the ability to modify polymer substrates by process of graft copolymerization to synthesize water-soluble copolymers having flocculating potential. Models – depicting the dependencies of the described quality characteristics (their means and variances) from process parameters – are estimated by implementation of the robust engineering methodology for quality improvement. Multiple criteria optimization based on the desirability function approach, involving requirements for economic efficiency, assurance of low toxicity, high copolymer efficiency in flocculation process, good solubility in water, bias, robustness, quality of prediction and the relative importance of responses, is presented.

  • INVESTIGATION AND OPTIMIZATION OF ELECTRON BEAM GRAFTING OF CORN STARCH

    Machines. Technologies. Materials., Vol. 10 (2016), Issue 3, pg(s) 52-55

    Experimental investigation of the modification of starch by grafting acrylamide using electron beam irradiation in order to synthesize water-soluble copolymers having flocculation abilities is performed. The influence of the variation of the parameters acrylamide/starch (AMD/St) weight ratio, electron beam irradiation dose and dose rate, as well as the presence or absence of metallic silver nanoparticles is investigated. The characterization of graft copolymers was carried out by monomer conversion coefficient, residual monomer concentration, intrinsic viscosity and Huggins’ constant. Models, describing the dependencies of the quality characteristics (their means and variances) from the process parameters, are estimated by implementation of the robust engineering approach in the case of qualitative and quantitative factors. Multi-criteria optimization involving requirements for economic efficiency, assurance of low toxicity, high copolymer efficiency in flocculation process and good solubility in water is also presented.

  • 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.