National Repository of Grey Literature 2 records found  Search took 0.02 seconds. 
Classification of Steel Connections in IDEA StatiCa Application
Nekut, Tomáš ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
The goal of this thesis was to design and implement application which would be able to predict the most suitable connection type of two steel members. This information could be used as a clue by a structural engineer and so make design of steel construction easier for him. Implemented application is able to automatically process finished project files that were created by structural engineers and prepare a data set by extracting features from them. The application is then able to train a neural network on this data set and using it predict suitable connection type of unconnected steel members. The precision of model finally reached 81 %. Prediction of connection types using artificial intelligence is not widely used yet but could work and could be possibly usable as is shown also in this thesis.
Classification of Steel Connections in IDEA StatiCa Application
Nekut, Tomáš ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
The goal of this thesis was to design and implement application which would be able to predict the most suitable connection type of two steel members. This information could be used as a clue by a structural engineer and so make design of steel construction easier for him. Implemented application is able to automatically process finished project files that were created by structural engineers and prepare a data set by extracting features from them. The application is then able to train a neural network on this data set and using it predict suitable connection type of unconnected steel members. The precision of model finally reached 81 %. Prediction of connection types using artificial intelligence is not widely used yet but could work and could be possibly usable as is shown also in this thesis.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.