PL
 |
EN
Ten serwis zostanie wyłączony 2025-02-11.
Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl
Preferencje help
Polish version English version Język
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 16

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Content available A New Neural Networks Model for Calculating the Continuous Cooling Transformation Diagrams
100%
Trzaska J.
|
|
tom Vol. 63, iss. 4
2009--2015
EN
The article shows a new model of Continuous Cooling Transformation (CCT) diagrams of structural steels and engineering steels. The modelling used artificial neural networks and a set of experimental data prepared based on 550 CCT diagrams published in the literature. The model of CCT diagrams forms 17 artificial neural networks which solve classification and regression tasks. Neural model is implemented in a computer software that enables calculation of a CCT diagram based on chemical composition of steel and its austenitizing temperature.
2
Content available Empirical Formulae for The Calculation of Austenite Supercooled Transformation Temperatures
100%
Trzaska J.
|
2015
|
tom Vol. 60, iss. 1
181--185
EN
The paper presents empirical formulae for the calculation of austenite supercooled transformation temperatures, basing on the chemical composition, austenitising temperature and cooling rate. The multiple regression method was used. Four equations were established allowing to calculate temperature of the start area of ferrite, perlite, bainite and martensite at the given cooling rate. The calculation results obtained do not allow to determine the cooling rate range of ferritic, pearlitic, bainitic and martensite transformations. Classifiers based on logistic regression or neural network were established to solve this problem.
PL
W pracy przedstawiono zależności empiryczne do obliczania temperatury przemian austenitu przechłodzonego na podstawie składu chemicznego, temperatury austenityzowania i szybkości chłodzenia. Zastosowano metodę regresji wielorakiej. Opracowano cztery równania, które umożliwiają obliczenie temperatury początku przemiany ferrytycznej, perlitycznej, baini-tycznej i martenzytycznej. Wyniki obliczeń nie pozwalają na wyznaczenie zakresu szybkości chłodzenia, dla których występują przemiany ferrytyczna, perlityczna, bainityczna i martenzytyczna. Do rozwiązania problemu opracowano klasyfikatory stosując regresję logistyczną lub sztuczne sieci neuronowe.
3
Content available remote Calculation of the steel hardness after continuous cooling
100%
Trzaska J.
Archives of Materials Science and Engineering
|
2013
|
tom Vol. 61, nr 2
87--92
EN
Purpose: The paper presents method in predicting hardness of steel cooled continuously from the austenitizing temperature, basing on the chemical composition, austenitizing temperature and cooling rate. Design/methodology/approach: In the paper it has ©been applied a hybrid approach that combined application of various mathematical tools including logistic regression and multiple regression to solve selected tasks from the area of materials science. Findings: Modelling make improvement of engineering materials properties possible, as well as prediction of their properties, even before the materials are fabricated, with the significant reduction of expenditures and time necessary for their investigation and application. Practical implications: The worked out relationships may be used in computer systems of steels’ designing for the heat-treated machine parts. Originality/value: The paper presents the method for calculating hardness of the structural steels, depending on their chemical composition, austenitizing temperature and cooling rate.
4
Content available Examples of Simulation of the Alloying Elements Effect on Austenite Transformations During Continuous Cooling
89%
Trzaska J. , Trzaska J.
|
|
nr 1
5
Employment of neural networks for analysis of chemical composition and cooling rate effect on CCT diagrams shape
63%
Dobrzański L. A. , Trzaska J.
Inżynieria Materiałowa
|
2004
|
tom R. XXV, nr 3
472-475
EN
The paper presents possibility of employment of the original supercooled austenite transformation anisothermic diagrams forecasting method for analysis of the chemical composition effect on the CCT diagrams shape. The developed model makes it possible to substitute computer simulation for the costly and time consuming experiments. The information derived from calculations make it possible to plot diagrams illustrating the effects of the particular elements or pairs of elements, as well as cooling rate and/or austenitizing temperature, on any temperature or time describing transformations in steel during its continuous cooling. Evaluation is also possible of the effect of the aforementioned factors on hardness and fractions of the particular structural constituents.
PL
W pracy przedstawiono możliwość zastosowania oryginalnej metody prognozowania anizotermicznych wykresów przemian austenitu przechłodzonego do analizy wpływu składu chemicznego na postać krzywych CTPc. Opracowany model pozwala na zastąpienie kosztownych i czasochłonnych eksperymentów przez symulację komputerową. Uzyskane w wyniku obliczeń informacje pozwalają na sporządzenie wykresów ilustrujących wpływ pojedynczych pierwiastków lub par pierwiastków stopowych a także szybkości chłodzenia i/lub temperatury austenityzowania na dowolną temperaturę lub czas opisujące przemiany w stali podczas chłodzenia ciągłego. Możliwa jest również ocena wpływu wspomnianych czynników na twardość i udział poszczególnych składników strukturalnych.
6
Content available remote Numerical simulation of the alloying elements effect on steels’ properties
63%
Sitek W. , Trzaska J.
|
2011
|
tom Vol. 45, nr 1
71--78
EN
Purpose: The goal of the research carried out was evaluation of alloying elements effect on high-speed steels hardness and fracture toughness and austenite transformations during continuous cooling of structural steels. Design/methodology/approach: Multi-layer feedforward neural networks with learning rule based on the error backpropagation algorithm were employed for modelling the steels properties. Then the neural networks worked out were employed for the computer simulation of the effect of particular alloying elements on the steels’ properties. Findings: Obtained results show that neural network are useful in evaluation of synergic effect of alloying elements on selected materials properties when classical investigations’ results do not provide evaluation of the effect of two or more alloying elements. Practical implications: Numerical simulation presented in the work, based on using the adequate material models may feature an alternative for classical investigations on effect of alloying elements on steels’ properties. Originality/value: The use of the neural networks as an tool for evaluation of the chemical composition effect on steels’ properties.
7
Zastosowanie sieci neuronowych do prognozowania wykresów CTPc
63%
Dobrzański L. A. , Trzaska J.
|
|
tom R. XXIV, nr 6
291-294
PL
W pracy przedstawiono metodykę modelowania zależności między składem chemicznym i temperaturą austenityzowania, a kinetyką przemian przechłodzonego austenitu podczas chłodzenia ciągłego, z wykorzystaniem sieci neuronowych. Opracowany model umożliwia obliczenie kompletnego wykresu CTPc dla stali o znanym składzie chemicznym i analizę oddziaływania poszczególnych pierwiastków na charakterystyczne punkty oraz krzywe przemian austenitu przechłodzonego, a także twardość uzyskaną w wyniku chłodzenia. Pozwala również na prognozowanie struktury uzyskanej w stali w wyniku chłodzenia z określoną szybkością z temperatury austenityzowania, przez ilościowy opis udziałów procentowych ferrytu, perlitu, bainitu oraz martenzytu z austenitem szczątkowym.
EN
The paper presents the methodology of modelling using the neural networks of the relationship between the chemical composition and austenitizing temperature, and the supercooled austenite transformation kinetics during the continuous cooling. The model worked out makes it possible to calculate a complete CCT diagram for the steel with a known chemical composition and analysis of the influence of particular elements on the characteristic points and transformation curves of the supercooled austenite, and also the hardness resulting from cooling. It makes also possible forecasting of the structure developed in steel as a result of cooling at a particular rate, by the quantitative description of the percentages of ferrite, pearlite, bainite, and martensite with the retained austenite.
8
Zmodyfikowana metoda Tartagli obliczania hartowności stali
63%
Sitek W. , Trzaska J.
Inżynieria Materiałowa
|
2006
|
tom Vol. 27, nr 3
272-275
PL
W pracy przedstawiono zmodyfikowaną metodę obliczania krzywych hartowności opracowaną przez Tartagliego, Eldisa oraz Geisslera i rozszerzoną przez T. Inoue. Metoda ta wykorzystuje podobieństwo krzywej Jominy do funkcji sekans hiperboliczny. Zaproponowane przez autorów równania empiryczne umożliwiają obliczenie krzywej hartowności na podstawie składu chemicznego stali. Wymagana jest jednak znajomość współczynników regresji charakterystycznych dla danego gatunku stali. W pracy zaproponowano zastąpienie części równań przez modele sieci neuronowych. Do obliczeń wykorzystano obszerny zbiór danych eksperymentalnych obejmujących składy chemiczne stali maszynowych oraz wyniki pomiarów twardości uzyskanych w próbie Jominy. Opracowane modele sieci neuronowych umożliwiają obliczenie krzywej Jominy na podstawie składu chemicznego w analizowanym zakresie stężeń masowych pierwiastków.
EN
The modified hardenability curves calculation method is presented in the paper, initially developed by Tartaglia, Eldis, and Geissler, later extended by T. Inoue. The method makes use of the similarity of the Jominy curve to the hyperbolic secant function. The empirical formulae proposed by the authors make calculation of the hardenability curve possible basing on the chemical composition of the steel. However, regression coefficients characteristic for the particular steel grade must be known. Replacing some formulae by the neural network models is proposed in the paper. The extensive set of the experimental data was used for calculations, encompassing the chemical compositions of machine steels and the hardness test results obtained in the Jominy tests. The developed neural network models make calculation of the Jominy curve possible basing on the chemical composition within the analysed range of the mass fractions of elements.
9
Content available Empirical Formulae for The Calculation of Austenite Supercooled Transformation Temperatures
63%
Trzaska J.
|
|
nr 1
10
Employment of neural networks for modelling of corrosive wear of hard magnetic composite materials
51%
Dobrzański L. A. , Drak M. , Trzaska J.
Inżynieria Materiałowa
|
2004
|
tom R. XXV, nr 3
709-712
EN
The paper presents a neural network model for evaluation of the rate of corrosive wear of the polymer matrix hard magnetic composite materials with particles of the powdered rapid quenched Nd-Fe-B strip with addition of metallic powder: iron, aluminium, CuSn10 type cast copper-tin alloy and X2CrNiMo17-12-2 high-alloy steel. A neural network model was established based on the research results from the investigations carried out in two corrosive environments. Three types of input data were used in the investigation: the contribution of the added powder, the nominal variable that defined the corrosive environment and the time duration of the test. The percentage corrosive wear of the surface was the output produced from such input data.
PL
W pracy przedstawiono model sieci neuronowej wyznaczania stopnia zużycia korozyjnego materiałów kompozytowych magnetycznie twardych o osnowie polimerowej wzmacnianych cząstkami magnetycznie twardymi Nd-Fe-B z dodatkiem proszku metalowego: żelaza, aluminium, odlewniczego stopu miedzi z cyną CuSn10, stali wysokostopowej X2CrNiMo17-12-2. Na podstawie wyników badań wykonanych w 2 ośrodkach korozyjnych ustalono model sieci neuronowej. Na wejściu przyjęto udział proszku dodatku, zmienną nominalną określającą rodzaj środowiska korozyjnego i czas trwania testu, natomiast na wyjściu wyrażony w procentach stopień zużycia korozyjnego powierzchni.
11
Content available remote An artificial intelligence approach in designing new materials
51%
Sitek W. , Trzaska J. , Dobrzański L. A.
|
|
tom Vol. 17, nr 1-2
277--280
EN
Purpose: The paper presents the computer aided method of chemical composition designing the metallic materials with a required property. Design/methodology/approach: The purpose has been achieved in two stages. In the first stage a neural network model for calculating the Jominy curve on the basis of the chemical composition has been worked out. This model made possible to prepare, in the second stage, a representative set of data and to work out the neural classifier that would aid the selection of steel grade with the required hardenability. Findings: Obtained results show that AI tools used are effective and very useful in designing new metallic materials. Research limitations/implications: The presented models may be used in the ranges of mass concentrations of alloying elements presented in the paper. The methodology presented in the paper makes it possible to add new grades of steel to the models. Practical implications: The worked out models may be used in computer systems of steel selection and designing for the heat-treated machine parts. Originality/value: The use of the artificial intelligence method, particularly the neural networks as a tool for designing the chemical composition of steels with the required properties.
12
Content available remote Neural network application in simulations of composites Al-Al2O3 tribological properties
51%
Dobrzański L. , Kremzer M. , Trzaska J. , Włodarczyk-Fligier A.
|
2008
|
tom Vol. 30, nr 1
37-40
EN
Purpose: The purpose of this paper is application of neural networks in tribological properties simulation of composite materials based on porous ceramic preforms infiltrated by liquid aluminium alloy. Design/methodology/approach: The material for investigations was manufactured by pressure infiltration method of ceramic porous preforms. The eutectic aluminium alloy EN AC – AlSi12 was use as a matrix while as reinforcement were used ceramic preforms manufactured by sintering of Al2O3 Alcoa CL 2500 powder with addition of pore forming agents as carbon fibres Sigrafil C10 M250 UNS manufactured by SGL Carbon Group Company. The wear resistance was measured by the use of device designed in the Institute of Engineering Materials and Biomaterials. The device realize dry friction wear mechanism of reciprocating movement condition. The simulation of load and number of cycles influence on tribological properties was made by the use of neural networks. Findings: The received results show the possibility of obtaining the new composite materials with required tribological properties moreover those properties can by simulated by the use of neural networks. Practical implications: The composite materials made by the developed method can find application among the others in automotive industry as the alternative material for elements fabricated from light metal matrix composite material reinforced with ceramic fibers. Originality/value: Worked out model of neural network can be used as helpful tool to prediction the wear of aluminium matrix composite materials In condition of dry friction.
13
Content available remote Application of the artificial neural networks for prediction of magnetic saturation of metallic amorphous alloys
51%
Konieczny J. , Dobrzański L. , Tomiczek B. , Trzaska J.
|
2008
|
tom Vol. 30, nr 2
105-108
EN
Purpose: The aim of the work is to employ the artificial neural networks for prediction of magnetic saturation of the amorphous alloys with the iron and cobalt matrix. Design/methodology/approach: It has been assumed that the artificial neural networks can be used to assign the relationship between the chemical compositions of amorphous alloys, temperature of heat treatment and magnetic saturation. In order to determine the relationship it has been necessary to work out a suitable calculation model. It has been proved that employment of genetic algorithm to selection of input neurons can be very useful tool to improve artificial neural network calculation results. The attempt to use the artificial neural networks for predicting the effect of the chemical composition and temperature of heat treatment on the magnetic saturation BS succeeded, as the level of the obtained results was acceptable. Findings: Artificial neural networks, can be applied for predicting the effect of the chemical composition and temperature of heat treatment on the magnetic saturation. Research limitations/implications: Worked out model should be used for prediction of magnetic saturation only in particular groups of amorphous alloys, mostly because of the discontinuous character of input data. Practical implications: The results of research make it possible to calculate with a certain admissible error the magnetic saturation Bs value basing on combinations of concentrations of the particular elements and heat treatment temperature. Originality/value: In this paper it has been presented an original trial of prediction of the required magnetic properties of the iron and cobalt amorphous alloys.
14
Content available A New Neural Networks Model for Calculating the Continuous Cooling Transformation Diagrams
50%
Trzaska J.
|
15
Content available remote Neural networks aided future events scenarios presented on the example of laser surface treatment
45%
Dobrzańska-Danikiewicz A. , Trzaska J. , Jagiełło A. , Jonda E. , Labisz K.
|
2011
|
tom Vol. 51, nr 2
69-96
EN
Purpose: The purpose of the paper is to present a methodological concept allowing to demonstrate the development directions of materials surface engineering according to the level of generality and the intensity of the phenomena analysed on other phenomena. Design/methodology/approach: A set of analytical methods and tools was used to present the development directions of materials surface engineering at the three levels analysed, i.e.: a macro-, meso- and microlevel. The analytical methods and tools comprise the scenario method, artificial neural networks, Monte Carlo method, e-Dephix method, statistical lists as bar charts, foresight matrices together with technology development tracks, technology roadmaps, technology information sheets and the classical materials science methods. Findings: A research methodology allowing to combine a presentation and description of the forecast future events having a varied level of generality and capturing the cause and effect relationships existing between the events. Research limitations/implications: The methodological concept discussed, implemented with reference to materials surface engineering, has a much broader meaning, and can be successfully applied in other technology foresights, and also in industrial and thematic foresights after minor modifications. Practical implications: The outcomes of the research conducted may be and should be used in the process of creating and managing the future of materials surface engineering and, within the time horizon of 20 years, may and should influence positively the development of the economy based on knowledge and innovation, sustainable development and the statistical level of the technologies used in industry, especially in small- and medium-sized enterprises. Originality/value: An own methodological concept constitutes an original way of presenting the development directions of the investigated field of knowledge. The use of neural networks
16
Content available Empirical Formulas for the Calculations of the Hardness of Steels Cooled From the Austenitizing Temperature
38%
Trzaska J.
|
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.