Share

IIR document

Multi-input multi-output (MIMO) artificial neural network (ANN) models applied to economized [s]croll compressors.

Number: pap. n. 1321

Author(s) : ZIVIANI D., BAHMAN A., GROLL E.

Summary

Predicting the compressor performance is an essential part in the design and optimization of HVAC&R equipment. The AHRI standard polynomial equations are well suited to map the performance of single-stage fixed-speed positive displacement machines, but present a number of well documented shortcomings especially in the case of compression enhancements (e.g. oilflooding, injection, variable-speed). To this end, a Multi-Input Multi-Output Artificial Neural Network (ANN) model has been developed to map the performance of single-phase and two-phase injected scroll compressors. The ANN models are based on a multi-layer structure whose number of neurons has been optimized to obtain high-accuracy predictions. The models have been
developed by using the open-source Keras package. The trained ANN models have been compared with the current state-of-the-art correlations available and they outperformed the existing correlations in terms of accuracy.

Available documents

Format PDF

Pages: 9

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: Multi-input multi-output (MIMO) artificial neural network (ANN) models applied to economized [s]croll compressors.
  • Record ID : 30026751
  • Languages: English
  • Source: Proceedings of the 25th IIR International Congress of Refrigeration: Montréal , Canada, August 24-30, 2019.
  • Publication date: 2019/08/24
  • DOI: http://dx.doi.org/10.18462/iir.icr.2019.1321
  • Notes:

    Keynote


Links


See other articles from the proceedings (632)
See the conference proceedings

Indexing