Combinatorial Neural Networks Based Model for Identification of Marine Steam Turbine Clustered Parameters

Using classification and approximation neural networks as its basis, a combinatorial model for identifying and simulating certain LNG tanker marine steam turbine plant parameters is presented in this paper. Two basic parts make up the model. The first part uses self-organizing neural networks to classify parameters in adequate clusters. The second part uses static feed-forward neural networks to carry out cluster interrelationships combinatorial identification. The authors then analyze the success of the study results by creating an adequate ranking of all identification-simulation models. Clear insight is provided by this approach into certain cluster interdependencies with the potential to significantly contribute to applications based on lost sensor information estimation and prediction that is not dependent on the cause of loss. Significantly increased sensor information reliability and redundancy directly reflect considerable technical security increases in terms of viewing a whole ship as a floating object, despite the fact that all of the aforementioned is distinctly related to marine propulsion control systems.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/03535320
• Authors:
  • Komadina, Pavao
  • Tomas, Vinko
  • Valcic, Marko
• Publication Date: 2011

Language

• English

Media Info

• Media Type: Print
• Features: Figures; References; Tables;
• Pagination: pp 1-9
• Serial:
  • PROMET-Traffic & Transportation
  • Volume: 23
  • Issue Number: 1
  • Publisher: University of Zagreb
  • ISSN: 0353-5320
  • EISSN: 1848-4069
  • Serial URL: https://traffic.fpz.hr
• Publication flags:

  Open Access (libre)

Subject/Index Terms

• TRT Terms: Approximation (Mathematics); Cluster analysis; Mathematical models; Neural networks; Sensors; Simulation; Steam turbines
• Uncontrolled Terms: Classification systems (Access control); Clustering; Interdependence; Interdependencies; Liquified gas tankers; Numerical rank; Parameter identification; Place rank; Propulsion systems
• Subject Areas: Marine Transportation; Planning and Forecasting; Vehicles and Equipment; I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 01340848
• Record Type: Publication
• Files: TRIS
• Created Date: May 20 2011 5:47PM