“…Each surface is defined in the local coordinates of the part. One ring is male and deprived of six degrees of freedom, the second one is female and movable [7]. The assembly parameters of two cone rings are the coordinate of axis intersection with the top end face of the movable ring.…”
Section: Conjugation Of the Parts Occurs On Surfacesmentioning
The quality of aircraft and rocket engines depends primarily on the geometric accuracy of assembly units and parts. Mathematical models implemented in the form of computer models are used to predict quality indicators (in particular, assembly parameters). Direct modeling of the conjugation process using numerical conjugation and finite-element
models of assemblies requires significant computational resources and is often accompanied by problems convergence of solutions. In order to solve the above problems, it is possible to use neural network models describing the main regularities of the pairing process based on the accumulated results. The work presents a neural network model for predicting assembly parameters of the parts based on the use of actual surfaces of the parts obtained as a result of mathematical modeling. Assembly on conical surfaces is considered. A convolutional neural network was used to predict assembly parameters.
“…Each surface is defined in the local coordinates of the part. One ring is male and deprived of six degrees of freedom, the second one is female and movable [7]. The assembly parameters of two cone rings are the coordinate of axis intersection with the top end face of the movable ring.…”
Section: Conjugation Of the Parts Occurs On Surfacesmentioning
The quality of aircraft and rocket engines depends primarily on the geometric accuracy of assembly units and parts. Mathematical models implemented in the form of computer models are used to predict quality indicators (in particular, assembly parameters). Direct modeling of the conjugation process using numerical conjugation and finite-element
models of assemblies requires significant computational resources and is often accompanied by problems convergence of solutions. In order to solve the above problems, it is possible to use neural network models describing the main regularities of the pairing process based on the accumulated results. The work presents a neural network model for predicting assembly parameters of the parts based on the use of actual surfaces of the parts obtained as a result of mathematical modeling. Assembly on conical surfaces is considered. A convolutional neural network was used to predict assembly parameters.
The quality of aircraft and rocket engines depends primarily on the geometric accuracy of assembly units and parts. Mathematical models implemented in the form of computer models are used to predict quality indicators (in particular, assembly parameters). The work presents a numerical model for predicting assembly parameters of the rotors based on the use of actual surfaces of the parts obtained as a result of measurement on coordinate measuring machines and of mathematical modeling. Aircraft engine turbine rotor parts surfaces are measured on a coordinate measuring machine. Experimental assemblies of three parts were carried out. The measurement error of the parts does not exceed 5 µm, of assembly – 10 µm. Face and radial run-outs were measured for four positions of the disc and retainer in the shaft assembly, with run-outs ranging from 0.08 mm up to 0.1 mm. Calculations of face and radial run-outs of the disc with respect to the base surfaces of the shaft were made using the developed digital model. Relative deviations of the calculated values from the experimental values do not exceed 10%.
Diffractive optical elements (DOEs) with their unique properties allow to form a predetermined beam intensity profile in the focal plane. The use of DOEs in laser material processing technologies reveals new possibilities for controlling the properties and operational characteristics of processed parts. The beam intensity profile formed by DOE is presented in the form of an analytical expression that was used to set a surface heat source for modelling thermal processes in DP 1000 steel. Experimental studies of samples under laser heating were performed. Simulation output results correlate well with the experimental data. The proposed simulation model, based on a precise heat input definition, is an intermediate step to the final goal, which is the prediction of structural changes in the zone of laser beam irradiation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.