Дослідження можливостей покращення якості виробів 3D-принтерної технології
Date
2017-03-28
Journal Title
Journal ISSN
Volume Title
Publisher
Видавництво Львівської політехніки
Abstract
Проаналізовано методи контролю якості виробництва на 3D-принтері для технології селективного
лазерного плавлення та лазерного припікання.
The analysis methods of quality control on production 3D- printer for selective laser sintering technology of special products via the sizes of which we can estimate the geometric parameters and via an additional control system we can high up their quality in real time. Increasing demands for precision measurement raises new problems of optimization of mathematical models of measuring transformations, and adequate processing of experimental data. This problem in modern MIs is particularly relevant due to the capabilities of inexpensive hardware implementation in basis of modern microelectronic components that opens the possibility of computing realization directly in the measuring path. Obtaining the necessary precision in many cases is only possible in the case of optimal mathematical models that provide in a certain sense the best approach of MI general transformation function. Rational choice of mathematical model of transformation function in many cases improves the measuring accuracy or expands the measurement range of preset accuracy. In this regard, becomes important choice of suitable criterion during processing the experimental data. Normal criteria mostly applicable for analyzing experimental data, is the most common method of mean squared errors, which consists in minimizing the sum of squares of the errors and computing the average of these squares. Unfortunately, the rootmean- square approximation does not provide achievement of the lowest difference between the estimator and function that is estimated at all points of observation, which is desirable during the precision processing of experimental data. Therefore, for solving the calibration tasks should be used the minimax criterion which ensures the minimum possible errors of reproducing the experimental calibration characteristics. Physical modeling is an experimental method of scientific research, which implies the substitution of the studied physical process by other similar to it of the same physical nature – by model. Physical model is a smaller or larger physical copy of an object. The geometries of model and object are often similar in the sense that one is a rescaling of the other; in such cases the scale is an important characteristic. Geometrically similar to the original the model can be both reduced and increased in the comparison with original sizes, and the model of process or phenomenon may differ from the real process by the quantitative physical characteristics such as power, energy, process pressure etc. In a broad sense, any physical experiment conducted in laboratory, including an experiment with natural object or part of it, is a physical modeling. The latter is based on the similarity theory and dimensional analysis, establishing the similarities criteria. The identity of the latter for a nature and the model provides the ability to transfer the experimental results obtained by physical modeling, in natural conditions. With the implementation of relevant conditions of physical modeling, i.e. the identity of similarity criteria, the values of variables that characterize a real phenomenon of proportionality of the similar points in space and at similar moments of time, become to be proportional to values of the same variables for the model. Presence of such proportionality allows perform recalculation of experimental results that were obtained on a model by multiplying the value of each of the identified variables on a constant for all values a given dimension set factor – the similarity factor.
Проанализированы методы контроля качества производства на 3D-принтере для технологий селективного лазерного плавления и припекания.
The analysis methods of quality control on production 3D- printer for selective laser sintering technology of special products via the sizes of which we can estimate the geometric parameters and via an additional control system we can high up their quality in real time. Increasing demands for precision measurement raises new problems of optimization of mathematical models of measuring transformations, and adequate processing of experimental data. This problem in modern MIs is particularly relevant due to the capabilities of inexpensive hardware implementation in basis of modern microelectronic components that opens the possibility of computing realization directly in the measuring path. Obtaining the necessary precision in many cases is only possible in the case of optimal mathematical models that provide in a certain sense the best approach of MI general transformation function. Rational choice of mathematical model of transformation function in many cases improves the measuring accuracy or expands the measurement range of preset accuracy. In this regard, becomes important choice of suitable criterion during processing the experimental data. Normal criteria mostly applicable for analyzing experimental data, is the most common method of mean squared errors, which consists in minimizing the sum of squares of the errors and computing the average of these squares. Unfortunately, the rootmean- square approximation does not provide achievement of the lowest difference between the estimator and function that is estimated at all points of observation, which is desirable during the precision processing of experimental data. Therefore, for solving the calibration tasks should be used the minimax criterion which ensures the minimum possible errors of reproducing the experimental calibration characteristics. Physical modeling is an experimental method of scientific research, which implies the substitution of the studied physical process by other similar to it of the same physical nature – by model. Physical model is a smaller or larger physical copy of an object. The geometries of model and object are often similar in the sense that one is a rescaling of the other; in such cases the scale is an important characteristic. Geometrically similar to the original the model can be both reduced and increased in the comparison with original sizes, and the model of process or phenomenon may differ from the real process by the quantitative physical characteristics such as power, energy, process pressure etc. In a broad sense, any physical experiment conducted in laboratory, including an experiment with natural object or part of it, is a physical modeling. The latter is based on the similarity theory and dimensional analysis, establishing the similarities criteria. The identity of the latter for a nature and the model provides the ability to transfer the experimental results obtained by physical modeling, in natural conditions. With the implementation of relevant conditions of physical modeling, i.e. the identity of similarity criteria, the values of variables that characterize a real phenomenon of proportionality of the similar points in space and at similar moments of time, become to be proportional to values of the same variables for the model. Presence of such proportionality allows perform recalculation of experimental results that were obtained on a model by multiplying the value of each of the identified variables on a constant for all values a given dimension set factor – the similarity factor.
Проанализированы методы контроля качества производства на 3D-принтере для технологий селективного лазерного плавления и припекания.
Description
Keywords
метрологія, контроль якості, моніторинг, аналіз, 3D-принтер, додаткове виробництво, metrology, quality control, monitoring, analysis, 3D-printer, additional production, метрология, контроль качества, мониторинг, анализ, 3D-принтер, дополнительное производство
Citation
Яцишин С. Дослідження можливостей покращення якості виробів 3D-принтерної технології / Святослав Яцишин, Ігор Полянський // Вимірювальна техніка та метрологія : міжвідомчий науково-технічний збірник. — Львів : Видавництво Львівської політехніки, 2017. — Том 78. — С. 74–79.