Effects of K-value in the K-Nearest Neighbors Algorithm on Performace of Chiller Fault Detection and Diagnosis
Corressponding author's email:
nhutlm@hcmute.edu.vnDOI:
https://doi.org/10.54644/jte.76.2023.1309Keywords:
HVAC, FDD, KNN, Energy, CondenserAbstract
Fault detection and diagnosis for water chiller systems help to extend the life of the system, prevent serious damage, and reduce energy consumption. For these reasons, this paper investigates the K value of the KNN algorithm and proposes a fault detection and diagnosis model for the water chiller system based on the K-nearest neighbors algorithm (FDD-KNN). The study results indicated that the FDD-KNN model has an accuracy rate of 99.15% or higher when the value of K is 1. When compared to previous studies for LV3 and LV4 severity faults, the proposed model exhibits high and uniform diagnostic accuracy across all faults and fault levels. In addition, the fault isolation and fluctuating trends of the variables in practice for Normal and fouling of the ConFoul condenser were also checked by the actual database of the water chiller systems, Center Saigon Building, Ho Chi Minh City, Vietnam. The results show that the changing trend of the two variables, TCO and TRC, is consistent with the actual operating conditions of the system. Therefore, the proposed model FDD-KNN with K=1 is completely reliable to apply to fault detection and diagnosis for water chiller systems.
Downloads: 0
References
S. Katipamula and M. Brambley, "Review Article: Methods for Fault Detection, Diagnostics, and Prognostics for Building Systems—A Review, Part I," HVAC&R Research, vol. 11, no. 1, pp. 3-25, 2005.
R. Huang et al., "An effective fault diagnosis method for centrifugal chillers using associative classification," Applied Thermal Engineering, vol. 136, pp. 633-642, 2018.
S. He, Z. Wang, Z. Wang, X. Gu, and Z. Yan, "Fault detection and diagnosis of chiller using Bayesian network classifier with probabilistic boundary," Applied Thermal Engineering, vol. 107, pp. 37-47, 2016.
J. Gao, H. Han, Z. Ren, and Y. Fan, "Fault diagnosis for building chillers based on data self-production and deep convolutional neural network," Journal of Building Engineering, vol. 34, 2021, doi: 10.1016/j.jobe.2020.102043.
K. Chen, Z. Wang, X. Gu, and Z. Wang, "Multicondition operation fault detection for chillers based on global density-weighted support vector data description," Applied Soft Computing, vol. 112, 2021, doi: 10.1016/j.asoc.2021.107795.
J. Liu et al., "Data-driven and association rule mining-based fault diagnosis and action mechanism analysis for building chillers," Energy and Buildings, vol. 216, 2020, doi: 10.1016/j.enbuild.2020.109957.
Y. Wang, Z. Wang, S. He, and Z. Wang, "A practical chiller fault diagnosis method based on discrete Bayesian network," International Journal of Refrigeration, vol. 102, pp. 159-167, 2019.
N. Settouti, M. E. A. Bechar, and M. A. Chikh, "Statistical Comparisons of the Top 10 Algorithms in Data Mining for Classication Task," International Journal of Interactive Multimedia and Artificial Intelligence, vol. 4, no. 1, 2016.
Z. Zhang, H. Han, X. Cui, and Y. Fan, "Novel application of multi-model ensemble learning for fault diagnosis in refrigeration systems," Applied Thermal Engineering, vol. 164, 2020, doi: 10.1016/j.applthermaleng.2019.114516.
K. Yan, A. Chong, and Y. Mo, "Generative adversarial network for fault detection diagnosis of chillers," Building and Environment, vol. 172, 2020, doi: 10.1016/j.buildenv.2020.106698.
A. S. Glass, P. Gruber, M. Roos, and J. Todtli, "Qualitative model-based fault detection in air-handling units," IEEE Control Systems, vol. 15, no. 4, pp. 11–22, 1995.
J. E. B. M. C. Comstock, "Development of Analysis Tools for the Evaluation of Fault Detection and Diagnostics in Chillers ASHRAE Research Project RP-1043," Purdue University, Ray W. Herrick Laboratories, West Lafayette, 1999.
Downloads
Published
How to Cite
Issue
Section
Categories
License
Copyright (c) 2023 Journal of Technical Education Science
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright © JTE.
