Share

IIR document

A variable refrigerant flow (VRF) air-conditioning system refrigerant charge detection method using stacking ensemble learning.

Number: 0936

Author(s) : CHENG H., MU W., CHENG Y., CHEN H., XING L.

Summary

This paper proposes a stacking ensemble learning method to detect the refrigerant charge fault of the variable refrigerant flow (VRF) air conditioning system. Firstly, measurement data is collected from the refrigerant charge fault experiment of a VRF system performed in an enthalpy difference laboratory. Several features are selected based on different criteria after preprocessing the measurement data and importance analysis. To prevent overfitting, no duplicate dataset is created. Then, five classification models have been used in this paper as follows: Support Vector Machine (SVM), Gradient Boosting Machine (GBM), Random Forest (RF), Multi-layer Perceptron (MLP), and K-Nearest Neighbors (KNN). After several parameter optimizations, these models demonstrate accuracy between 84.71% to 91.58%. Finally, different stacking ensemble learning methods integrate these models, achieving the test set's best accuracy rate (96.66%).

Available documents

Format PDF

Pages: 14

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: A variable refrigerant flow (VRF) air-conditioning system refrigerant charge detection method using stacking ensemble learning.
  • Record ID : 30031850
  • Languages: English
  • Subject: Technology
  • Source: Proceedings of the 26th IIR International Congress of Refrigeration: Paris , France, August 21-25, 2023.
  • Publication date: 2023/08/21
  • DOI: http://dx.doi.org/10.18462/iir.icr.2023.0936

Links


See other articles from the proceedings (492)
See the conference proceedings

Indexing