Share

IIR document

Frosting weight and refrigerating capacity prediction of fin evaporator based on random finite element method and ridgelet neural network.

Author(s) : ZHAO B., YI R., GAO D., et al.

Type of article: Article, IJR article

Summary

In order to confirm the proper defrosting measurement, the Frosting weight and refrigerating capacity of fin evaporator should be predicted correctly, therefore the new prediction model is proposed by combining the ridgelet neural network and random finite element method. Firstly, the ridgelet neural network is constructed through using ridgelet function as excitation function of hidden layer, and the basic structure of the ridgelet neural network is designed. Secondly, the training algorithm of ridgelet neural network is put forward based on improved genetic algorithm, the intervals of weights and thresholds can be changed dynamically for the improved genetic algorithm. Thirdly, the frosting random finite element model of fin evaporator is constructed, the frosting parameters of fin evaporator are considered as random variables, and the simulation results under different working condition can be used as testing samples and training samples. Finally, the Frosting weight and refrigerating capacity of fin evaporator are predicted based on new prediction model, simulation analysis shows that the new prediction model has highest prediction precision and efficiency.

Available documents

Format PDF

Pages: 37-46

Available

  • Public price

    20 €

  • Member price*

    Free

* Best rate depending on membership category (see the detailed benefits of individual and corporate memberships).

Details

  • Original title: Frosting weight and refrigerating capacity prediction of fin evaporator based on random finite element method and ridgelet neural network.
  • Record ID : 30025485
  • Languages: English
  • Source: International Journal of Refrigeration - Revue Internationale du Froid - vol. 99
  • Publication date: 2019/03
  • DOI: http://dx.doi.org/10.1016/j.ijrefrig.2018.11.046

Links


See other articles in this issue (45)
See the source

Indexing