Algorithm for Controlling Mechanical Properties of Hot Rolled Steels Using Bayesian Network Model

2012 ◽  
Vol 706-709 ◽  
pp. 1444-1447
Author(s):  
Oleg S. Khlybov ◽  
Igor V. Dubinin
Keyword(s):  
Mechanical Properties ◽  
Chemical Composition ◽  
Probability Distribution ◽  
Bayesian Network ◽  
Control Method ◽  
Steel Production ◽  
Steel Grade ◽  
Rolling Temperature ◽  
Joint Probability Distribution ◽  
Coiling Temperature

The work presents a control method for on-line adjustment of mechanical properties of as rolled steels produced at hot strip mills. The key idea of the method is a probabilistic causal (Bayesian) network which represents in a form of a directed acyclic graph the joint probability distribution of mechanical properties, chemical composition and temperature–strain parameters acting during hot rolling. As a slab moves along the mill the distribution is used for continuous recalculating the posterior probability of all mechanical properties conditioned by chemical composition and all other process parameters which become known to the moment of recalculating. Finally, when a strip is just before the finishing group we evaluate the probability distribution of finishing rolling temperature and coil temperature given the strip has the target mechanical properties. It generates new setups for these temperatures A pilot version of the method has been just implemented at CSP–line at Vyksa, Russia, United Metallurgical Company’s steel production site The adjustment is realized through appropriate correction of finish rolling temperature or/and coiling temperature setups of the mill automatic control system after the last chemical analysis of the current heat is made at the start of casting. Only “cautious” corrections of the temperatures are permissible so far (deviation from predefined level not more than ±30 degrees for each temperature) and the main aim of them is to set off the influence of chemistry variations on mechanical properties scatter of a given steel grade. The results of using the algorithm show that even these limited but interconnected actions reduce approximately twice the standard deviation of the mechanical properties inside a steel grade.

Optimization of Chemical Composition for Pressure Vessel Steel Grade P460NL2 - Industrial Trial Research

2017 ◽  
Vol 1143 ◽  
pp. 45-51 ◽  
Author(s):  
Costel Durduc-Roibu ◽  
Elena Drugescu
Keyword(s):  
Mechanical Properties ◽  
Chemical Composition ◽  
Steel Grade ◽  
Chemical Compositions ◽  
Pressure Vessel Steel ◽  
Vessel Steel ◽  
Industrial Trial ◽  
Tension Tests ◽  
Rolling Mode ◽  
Strength And Toughness

The research examines the improvements of mechanical properties of yield strength and toughness for an optimized chemical composition B2 used in industrial trial in comparison with usual chemical composition A1 used for pressure steel grade with higher strength and toughness. For both chemical compositions we rolled three plates with thickness 8, 10 and 12 mm. Rolling mode was a control rolling followed by normalizing heat treatment. Samples from each plate from opposites corner in as rolled and normalized state was taken and tested: spectral analysis, mechanical properties: tension tests, Charpy-V notch impact test. Differences between A1 and B2 chemical compositions are given by the micro alloing elements used and the overall results showed increasing YS and toughness values in the range of euronorm requirements.

Learning Bayesian Networks

2011 ◽  
pp. 315-321
Author(s):  
Marco F. Ramoni ◽  
Paola Sebastiani
Keyword(s):  
Probability Distribution ◽  
Bayesian Networks ◽  
Bayesian Network ◽  
Graphical Models ◽  
Graphical Model ◽  
Joint Probability ◽  
Joint Probability Distribution ◽  
Probabilistic Graphical Model ◽  
Conditional Probability Distribution ◽  
Statistics And Probability

Born at the intersection of artificial intelligence, statistics, and probability, Bayesian networks (Pearl, 1988) are a representation formalism at the cutting edge of knowledge discovery and data mining (Heckerman, 1997). Bayesian networks belong to a more general class of models called probabilistic graphical models (Whittaker, 1990; Lauritzen, 1996) that arise from the combination of graph theory and probability theory, and their success rests on their ability to handle complex probabilistic models by decomposing them into smaller, amenable components. A probabilistic graphical model is defined by a graph, where nodes represent stochastic variables and arcs represent dependencies among such variables. These arcs are annotated by probability distribution shaping the interaction between the linked variables. A probabilistic graphical model is called a Bayesian network, when the graph connecting its variables is a directed acyclic graph (DAG). This graph represents conditional independence assumptions that are used to factorize the joint probability distribution of the network variables, thus making the process of learning from a large database amenable to computations. A Bayesian network induced from data can be used to investigate distant relationships between variables, as well as making prediction and explanation, by computing the conditional probability distribution of one variable, given the values of some others.

scholarly journals Penerapan Metode Bayesian Network Model Untuk Menghitung Probabilitas Penyakit Sesak Nafas Bayi

2018 ◽  
Vol 2 (1) ◽  
pp. 62
Author(s):  
Hasniati Hasniati ◽  
Arianti Arianti ◽  
William Philip
Keyword(s):  
Probability Distribution ◽  
Bayesian Network ◽  
Conditional Probability ◽  
Network Model ◽  
Posterior Probability ◽  
Prior Probability ◽  
Joint Probability ◽  
Joint Probability Distribution ◽  
Bayesian Network Model

Bayesian Network dapat digunakan untuk menghitung probabilitas dari kehadiran berbagai gejala penyakit. Dalam tulisan ini, penulis menerapkan bayesian network model untuk menghitung probabilitas penyakit sesak nafas pada bayi. Bayesian network diterapkan berdasar pada data yang diperoleh melalui wawancara kepada dokter spesialis anak yaitu data nama penyakit, penyebab, dan gejala penyakit sesak nafas pada bayi. Struktur Bayesian Network penyakit sesak nafas bayi dibuat berdasarkan ada tidaknya keterkaitan antara gejala terhadap penyakit sesak nafas. Untuk setiap gejala yang direpresentasikan pada struktur bayesian network mempunyai estimasi parameter yang didapat dari data yang telah ada atau pengetahuan dari dokter spesialis. Data estimasi ini disebut nilai prior probaility atau nilai kepercayaan dari gejala penyakit sesak nafas bayi. Setelah diketahui prior probability, langkah berikutnya adalah menentukan Conditional probability (peluang bersyarat) antara jenis penyakit sesak nafas dengan masing-masing gejalanya. Pada langkah akhir, nilai posterior probability dihitung dengan mengambil nilai hasil joint probability distribution (JPD) yang telah diperoleh, kemudian nilai inilah yang digunakan untuk menghitung probabilitas kemunculan suatu gejala. Dengan mengambil satu contoh kasus bahwa bayi memiliki gejala sesak, lemah, gelisah dan demam, disimpulkan bahwa bayi menderita penyakit sesak nafas Pneumoni Neonatal sebesar 0,1688812743.

Learning Bayesian Networks

2011 ◽  
pp. 674-677
Author(s):  
Marco F. Ramoni ◽  
Paola Sebastiani
Keyword(s):  
Probability Distribution ◽  
Bayesian Networks ◽  
Bayesian Network ◽  
Graphical Models ◽  
Graphical Model ◽  
Joint Probability ◽  
Joint Probability Distribution ◽  
Probabilistic Graphical Model ◽  
Conditional Probability Distribution ◽  
Statistics And Probability

Born at the intersection of artificial intelligence, statistics, and probability, Bayesian networks (Pearl, 1988) are a representation formalism at the cutting edge of knowledge discovery and data mining (Heckerman, 1997). Bayesian networks belong to a more general class of models called probabilistic graphical models (Whittaker, 1990; Lauritzen, 1996) that arise from the combination of graph theory and probability theory, and their success rests on their ability to handle complex probabilistic models by decomposing them into smaller, amenable components. A probabilistic graphical model is defined by a graph, where nodes represent stochastic variables and arcs represent dependencies among such variables. These arcs are annotated by probability distribution shaping the interaction between the linked variables. A probabilistic graphical model is called a Bayesian network, when the graph connecting its variables is a directed acyclic graph (DAG). This graph represents conditional independence assumptions that are used to factorize the joint probability distribution of the network variables, thus making the process of learning from a large database amenable to computations. A Bayesian network induced from data can be used to investigate distant relationships between variables, as well as making prediction and explanation, by computing the conditional probability distribution of one variable, given the values of some others.

Learning Bayesian Networks

2011 ◽  
pp. 1124-1128
Author(s):  
Marco F. Ramoni ◽  
Paola Sebastiani
Keyword(s):  
Probability Distribution ◽  
Bayesian Networks ◽  
Bayesian Network ◽  
Graphical Models ◽  
Graphical Model ◽  
Joint Probability ◽  
Joint Probability Distribution ◽  
Probabilistic Graphical Model ◽  
Conditional Probability Distribution ◽  
Statistics And Probability

Born at the intersection of artificial intelligence, statistics, and probability, Bayesian networks (Pearl, 1988) are a representation formalism at the cutting edge of knowledge discovery and data mining (Heckerman, 1997). Bayesian networks belong to a more general class of models called probabilistic graphical models (Whittaker, 1990; Lauritzen, 1996) that arise from the combination of graph theory and probability theory, and their success rests on their ability to handle complex probabilistic models by decomposing them into smaller, amenable components. A probabilistic graphical model is defined by a graph, where nodes represent stochastic variables and arcs represent dependencies among such variables. These arcs are annotated by probability distribution shaping the interaction between the linked variables. A probabilistic graphical model is called a Bayesian network, when the graph connecting its variables is a directed acyclic graph (DAG). This graph represents conditional independence assumptions that are used to factorize the joint probability distribution of the network variables, thus making the process of learning from a large database amenable to computations. A Bayesian network induced from data can be used to investigate distant relationships between variables, as well as making prediction and explanation, by computing the conditional probability distribution of one variable, given the values of some others.

Development of Modeling Algorithm of the Characteristics of the Steels in the Subsystem of Management of the Production Resources of Steel Production

2019 ◽  
Vol 20 (10) ◽  
pp. 615-622
Author(s):  
S. S. Kochkovskaya
Keyword(s):  
Mechanical Properties ◽  
Mathematical Model ◽  
Chemical Composition ◽  
Control Mechanism ◽  
Chemical Elements ◽  
Functional Model ◽  
Steel Production ◽  
Model Management ◽  
Roll Steel ◽  
Software Product

The article is devoted to the development of an algorithm for modeling the characteristics of steels in accordance with customer requirements. A mathematical model is presented that takes into account the optimal levels of the main factors and their interaction, providing the required values of the characteristics. The next step in the study of the mathematical model was modeling by means of functional modeling methodology IDEF0. Input and output data, as well as normative documents of model management and mechanisms of this management for building a functional model are defined. The control mechanism was the software product OptimalSostav, developed using the object-oriented programming language Delphi. The software product is designed to simulate the characteristics of steels using the specified limits of permissible minimum and maximum values of mechanical properties. The algorithm of realization of the control mechanism, which is based on the fractional factor analysis, is described. The presented algorithm allows to identify the influence of chemical composition on the mechanical properties of steels in the form of mathematical and graphical dependencies and to determine the specified mechanical properties that meet the requirements of the customer. The main possibilities and the scope of the software product, which allows to solve the problem of predicting the optimal chemical composition, providing the required mechanical characteristics, as well as to adjust the process of steel melting within a given chemical composition to achieve the desired set of properties. The application of the software product on the example of the analysis of the influence of chemical elements on the mechanical properties of 75HMF roll steel is shown. The results of modeling in the form of mathematical and graphic dependences are given and the estimation of efficiency of application of the software product is given. The results of solving the problem of approximation of the obtained graphic dependences of the influence of chemical elements on the mechanical properties of the studied steel grade by means of Microsoft Excel are presented. It is established that the developed mathematical and algorithmic software of the software product allows to study the percentage of chemical elements in relation to the total composition of the alloy, based on the obtained pie charts and dependency graphs. The adequacy of the model is confirmed by experimental results.

Modelling of Mechanical Properties of Pearlitic Rail Steel

2019 ◽  
Vol 798 ◽  
pp. 3-8
Author(s):  
Patiphan Boonsukachote ◽  
Saranya Kingklang ◽  
Vitoon Uthaisangsuk
Keyword(s):  
Mechanical Properties ◽  
Flow Stress ◽  
Chemical Composition ◽  
Impact Toughness ◽  
Rail Steel ◽  
Steel Grade ◽  
High Yield ◽  
The Individual ◽  
Microstructure Characteristic ◽  
Pearlitic Rail Steel

Railway has become more essential for both mass and goods transportation so that the rails are required to carry higher loads and exhibit longer lifetime. Thus, mechanical properties, especially strength and toughness of rail steel must be continuously increased. In the present work, microstructure, tensile properties and impact toughness of a pearlitic rail steel grade 900A were firstly characterized. It was found that the investigated steel showed high yield and tensile strengths, but moderate elongation. Subsequently, representative volume elements (RVE) model was employed to investigate the effects of bainitic phase on mechanical properties of pearlitic rail steels. The flow stress curves of the individual phases were defined with regard to the chemical composition. As a result, the relationships between predicted yield strengths and tensile strengths in dependence on the phase fraction of bainite were provided. The model can be used to identify the proper microstructure characteristic of rail steel.

Study of the character and causes of destruction of the cardan shaft of the propeller engine

2019 ◽  
Vol 85 (12) ◽  
pp. 43-50
Author(s):  
D. A. Movenko ◽  
L. V. Morozova ◽  
S. V. Shurtakov
Keyword(s):  
Mechanical Properties ◽  
Chemical Composition ◽  
Yield Point ◽  
Protective Coating ◽  
Fracture Pattern ◽  
Corrosion Cracking ◽  
Tempered Martensite ◽  
Neck Formation ◽  
Static Mechanism ◽  
Shaft Surface

The results of studying operational destruction of a high-loaded cardan shaft of the propeller engine made of steel 38KhN3MFA are presented to elucidate the cause of damage and develop a set of recommendations and measures aimed at elimination of adverse factors. Methods of scanning electron and optical microscopy, as well as X-ray spectral microanalysis are used to determine the mechanical properties, chemical composition, microstructure, and fracture pattern of cardan shaft fragments. It is shown that the mechanical properties and chemical composition of the material correspond to the requirements of the regulatory documentation, defects of metallurgical origin both in the shaft metal and in the fractures are absent. The microstructure of the studied shaft fragments is tempered martensite. Fractographic analysis revealed that the destruction of cardan shaft occurred by a static mechanism. The fracture surface is coated with corrosion products. The revealed cracks developed by the mechanism of corrosion cracking due to violation of the protective coating on the shaft. The results of the study showed that the destruction of the cardan shaft of a propeller engine made of steel 38Kh3MFA occurred due to formation and development of spiral cracks by the mechanism of stress corrosion cracking under loads below the yield point of steel. The reason for «neck» formation upon destruction of the shaft fragment is attributed to the yield point of steel attained during operation. Regular preventive inspections are recommended to assess the safety of the protective coating on the shaft surface to exclude formation and development of corrosion cracks.

Structural features and strength behavior of TRIP/TWIP steels

2020 ◽  
pp. 5-18
Author(s):  
D. V. Prosvirnin ◽  
◽  
M. S. Larionov ◽  
S. V. Pivovarchik ◽  
A. G. Kolmakov ◽  
...  
Keyword(s):  
Mechanical Properties ◽  
Chemical Composition ◽  
Thermomechanical Processing ◽  
Maximum Load ◽  
Structural Features ◽  
Stacking Fault ◽  
Stacking Fault Energy ◽  
Structural And Functional Properties ◽  
Twip Steels ◽  
The Creation

A review of the literature data on the structural features of TRIP / TWIP steels, their relationship with mechanical properties and the relationship of strength parameters under static and cyclic loading was carried out. It is shown that the level of mechanical properties of such steels is determined by the chemical composition and processing technology (thermal and thermomechanical processing, hot and cold pressure treatment), aimed at achieving a favorable phase composition. At the atomic level, the most important factor is stacking fault energy, the level of which will be decisive in the formation of austenite twins and / or the formation of strain martensite. By selecting the chemical composition, it is possible to set the stacking fault energy corresponding to the necessary mechanical characteristics. In the case of cyclic loads, an important role is played by the strain rate and the maximum load during testing. So at high loading rates and a load approaching the yield strength under tension, the intensity of the twinning processes and the formation of martensite increases. It is shown that one of the relevant ways to further increase of the structural and functional properties of TRIP and TWIP steels is the creation of composite materials on their basis. At present, surface modification and coating, especially by ion-vacuum methods, can be considered the most promising direction for the creation of such composites.

Sign in / Sign up

Export Citation Format

Share Document