Scholarlink

This article considers nonlinear autoregressive neural networks (NARNET) for multistep ahead forecast of Steam Table properties. Experimental information has been gathered and considered for learning of the NARNET model using Levenberg–Marquardt algorithm. The accurateness is analysed in terms of different error indicators. The outcomes show that the maximum RMSE values obtained for output attribute ‘specific enthalpy of liquid’ and ‘specific enthalpy of vapour’ were 0.939 and 1.986 correspondingly for a maximum step size of 30 estimations. Moreover, excellent BCC values in order of 0.99 were obtained for both the attributes. The model can be effectively used in Academia as well as Industries for precise prediction of steam thermodynamic properties particularly of interpolated data saving time and labour.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Steam Table Data Using Nonlinear Autoregressive Neural Network (NARNET)

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Tarun Kumar Dhiman,
  • Ashwani Kharola,
  • Vishwjeet Choudhary,
  • Rahul,
  • Priyanshu Thapliyal,
  • Sankula Madhava

摘要

This article considers nonlinear autoregressive neural networks (NARNET) for multistep ahead forecast of Steam Table properties. Experimental information has been gathered and considered for learning of the NARNET model using Levenberg–Marquardt algorithm. The accurateness is analysed in terms of different error indicators. The outcomes show that the maximum RMSE values obtained for output attribute ‘specific enthalpy of liquid’ and ‘specific enthalpy of vapour’ were 0.939 and 1.986 correspondingly for a maximum step size of 30 estimations. Moreover, excellent BCC values in order of 0.99 were obtained for both the attributes. The model can be effectively used in Academia as well as Industries for precise prediction of steam thermodynamic properties particularly of interpolated data saving time and labour.