Applications of artificial neural network for the prediction of flow boiling curves

Guanghui Su, Kenji Fukuda, Koji Morita, Mark Pidduck, Dounan Jia, Tatsuya Matsumoto, Ryo Akasaka

Research output: Contribution to journalArticlepeer-review

25 Citations (Scopus)


An artificial neural network (ANN) was applied successfully to predict flow boiling curves. The databases used in the analysis are from the 1960’s, including 1,305 data points which cover these parameter ranges: pressureP=100–1,000 kPa, mass flow rateG=40–500 kg/m2-s, inlet subcoolingΔTsub =0–35°C, wall superheatΔTw = 10–300°C and heat fluxQ=20–8,000kW/m2. The proposed methodology allows us to achieve accurate results, thus it is suitable for the processing of the boiling curve data. The effects of the main parameters on flow boiling curves were analyzed using the ANN. The heat flux increases with increasing inlet subcooling for all heat transfer modes. Mass flow rate has no significant effects on nucleate boiling curves. The transition boiling and film boiling heat fluxes will increase with an increase in the mass flow rate. Pressure plays a predominant role and improves heat transfer in all boiling regions except the film boiling region. There are slight differences between the steady and the transient boiling curves in all boiling regions except the nucleate region. The transient boiling curve lies below the corresponding steady boiling curve.

Original languageEnglish
Pages (from-to)1190-1198
Number of pages9
Journaljournal of nuclear science and technology
Issue number11
Publication statusPublished - Nov 2002

All Science Journal Classification (ASJC) codes

  • Nuclear and High Energy Physics
  • Nuclear Energy and Engineering


Dive into the research topics of 'Applications of artificial neural network for the prediction of flow boiling curves'. Together they form a unique fingerprint.

Cite this