Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: soft modelling

Sortuj według:

Ogranicz wyniki do:

Soft modelling of the shaping of metal profiles in rapid tube hydroforming technology

Sadłowska Hanna, Kochański Andrzej

Computer Methods in Materials Science

2022

Vol. 22, No. 4

201--210

The paper presents an approach to the impact of process parameters in innovative RTH (Rapid Tube Hydroforming) technology for shaping closed metal profiles in flexible and deformable dies. In order to implement the assumed deformation of the deformed profile, the RTH technology requires the monitoring and control of numerous technological parameters, including geometric, material, and technological variables. The paper proposes a two-stage research procedure considering hard modelling (constitutive) and soft modelling (data-driven). Due to the complexity of the technological process, it was required to develop a numerical finite element method FEM model focused on obtaining the adequate profile deformation measured by the ellipsoidality of the cylindrical profile. Based on the results of the numerical experiments, a preliminary soft mathematical model using ANN was developed. Analysing the soft model results, several statistical hypotheses were made and verified to investigate the significance of selected process parameters. Thanks to this, it was possible to select the most important process parameters, i.e., the properties of moulding sands used for RTH dies: the angle of internal friction and cohesion.

Application of instance-based learning for cast iron casting defects prediction

Sika Robert, Szajewski Damian, Hajkowski Jakub, Popielarski Paweł

Management and Production Engineering Review

2019

Vol. 10, No. 4

101--107

The paper presents an example of Instance-Based Learning using a supervised classiﬁcation method of predicting selected ductile cast iron castings defects. The test used the algorithm of k-nearest neighbours, which was implemented in the authors’ computer application. To ensure its proper work it is necessary to have historical data of casting parameter values registered during casting processes in a foundry (mould sand, pouring process, chemical composition) as well as the percentage share of defective castings (unrepairable casting defects). The result of an algorithm is a report with ﬁve most possible scenarios in terms of occurrence of a cast iron casting defects and their quantity and occurrence percentage in the casts series. During the algorithm testing, weights were adjusted for independent variables involved in the dependent variables learning process. The algorithms used to process numerous data sets should be characterized by high eﬃciency, which should be a priority when designing applications to be implemented in industry. As it turns out in the presented mathematical instance-based learning, the best quality of ﬁt occurs for speciﬁc values of accepted weights (set #5) for number k = 5 nearest neighbours and taking into account the search criterion according to “product index”.