Group technology using neural nets

Gwiazda, Aleksander; Knosala, Ryszard

doi:10.1016/s0924-0136(96)02566-6

Cited by 21 publications

(9 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5,6 Such structure is characteristic for both types of joints. It should state that the other investigations area should be search in the idea of linking the designs process of a structural form with the design process of the material form of a given technical mean [10,11].…”

Section: Discussionmentioning

confidence: 99%

Hardness analysis of welded joints of austenitic and duplex stainless steels

Topolska

2016

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

Abstract. Stainless steels are widely used in the modern world. The continuous increase in the use of stainless steels is caused by getting greater requirements relating the corrosion resistance of all types of devices. The main property of these steels is the ability to overlap a passive layer of an oxide on their surface. This layer causes that they become resistant to oxidation. One of types of corrosion-resistant steels is ferritic-austenitic steel of the duplex type, which has good strength properties. It is easily formable and weldable as well as resistant to erosion and abrasive wear. It has a low susceptibility to stress-corrosion cracking, to stress corrosion, to intercrystalline one, to pitting one and to crevice one. For these reasons they are used, among others, in the construction of devices and facilities designed for chemicals transportation and for petroleum and natural gas extraction. The paper presents the results which shows that the particular specimens of the ][ joint representing both heat affected zones (from the side of the 2205 steel and the 316L one) and the weld are characterized by higher hardness values than in the case of the same specimens for the 2Y joint. Probably this is caused by machining of edges of the sections of metal sheets before the welding process, which came to better mixing of native materials and the filler metal. After submerged arc welding the 2205 steel still retains the diphase, austenitic-ferritic structure and the 316L steel retains the austenitic structure with sparse bands of ferrite σ. IntroductionDuplex steels are the result of combining the microstructure of ferritic and austenitic stainless steels. In the case of ferritic steel it is primarily a Nickel one and in the case of austenitic steel it is a Chromium-Nickel one. Their advantages results from properties of both mentioned groups of steels. The next advantage is that higher properties could be obtained with lower level of alloying elements, mainly expensive Nickel. What results in lower price of duplex steels. It cause that nowadays the maritime construction are more and more often design using the different duplex steels. It is also related with good weldability of this group of steels. There are analysed other areas of application [1,2]. Among others duplex stainless steel offers also excellent resistant to corrosion and is characterised by the very high mechanical strength parameters. The high corrosion resistance of this steels ensures significantly more uptime than carbon steels and conventional stainless steels. It is particularly visible for the super duplex and hyper duplex steels. On the other hand the light duplex steels characterises by the same level of mechanical parameters like conventional steels.The mechanical strength allows designing lighter constructions, which are more compact and which need less welding operations. The high mechanical properties are the result of combination both

show abstract

Section: Discussionmentioning

confidence: 99%

Hardness analysis of welded joints of austenitic and duplex stainless steels

Topolska

2016

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

show abstract

“…The reason why the ANNs are used in science and technique is their ability to treat dynamic, complex processes with often complicated nature or imprecisely determined processes as "black boxes". In other words, an ANN makes it possible to find a relation between numerous input variables and the output variable from the process with no necessity to build complicated mathematical equations [6][7][8]. Fig.…”

Section: Application Of Ann Models In Decision-making Related To Prod...mentioning

confidence: 99%

“…In industry, nearly 70% of all applications are recognitions of images, including also their classification, grouping and processing. In management and operation of production systems, the ANNs are more and more often used for [1,3,6,9]:  control of production processes and robots,  analysis of production problems,  diagnostics of electronic systems of machines,  selection of workers and input materials,  optimization of commercial activity, utilization of wastes, movements of the robot,  planning of overhauls of machines,  forecasting.…”

Section: Application Of Ann Models In Decision-making Related To Prod...mentioning

confidence: 99%

Efficiency forecasting of electric bundle assembly with use of ANN model

Burduk

Dybała

Machado

2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Management of production systems requires making immediate decisions based on the data generated in bulk by IT systems. In this case, it can be helpful to use models of artificial neural networks (ANN) that, on the grounds of accessible data, will determine results of the made decision. One of the key problems in production companies is determination of execution time and cost of a production order. The problem is especially important in a company manufacturing a variable product line with a big part of manual operations. In the article, the way of building an ANN model for efficiency forecasting of the assembly process of electric bundles is presented. With regard to the very wide and variable product line, the products with different complexity degree are manufactured on three types of assembly lines. The assembly processes are performed on the assembly lines manually by groups of workers, so efficiency of the process is influenced mostly by skills and experience of these workers. Therefore, numbers of new assembly workers assigned to individual assembly lines and quantities of new products in the production schedule are selected as explanatory variables in the ANN model. The explained variable in the ANN model is production volume of the manufactured electric bundles.

show abstract

“…However, the development of this technology that affects the achievement of better accuracy of printouts caused that 3D printing has become a method of receiving ready-made objects which are characterized by very good properties [6]. The possibilities of using spatial printers mainly depend on production methods [7], materials, which are available on the market and the volume of the printer. In practice, all 3D printing methods are similar because they form the object layer by layer in order to create complex shapes.…”

Section: Introductionmentioning

confidence: 99%

Algorithms of control parameters selection for automation of FDM 3D printing process

et al. 2017

View full text Add to dashboard Cite

Abstract. The paper presents algorithms of control parameters selection of the Fused Deposition Modelling (FDM) technology in case of an open printing solutions environment and 3DGence ONE printer. The following parameters were distinguished: model mesh density, material flow speed, cooling performance, retraction and printing speeds. These parameters are independent in principle printing system, but in fact to a certain degree that results from the selected printing equipment features. This is the first step for automation of the 3D printing process in FDM technology.

show abstract

Group technology using neural nets

Cited by 21 publications

References 1 publication

Hardness analysis of welded joints of austenitic and duplex stainless steels

Hardness analysis of welded joints of austenitic and duplex stainless steels

Efficiency forecasting of electric bundle assembly with use of ANN model

Algorithms of control parameters selection for automation of FDM 3D printing process

Contact Info

Product

Resources

About