Abstract
Refrigerant leakage is common in the use of dehumidifier, leading to decrease of dehumidification ability. To detect refrigerant capacity effectively and maintain it timely, a neural network with deep learning skills was proposed based on time series. As input of networks, the time series consists of the operating parameters of the dehumidifier at multiple time points. To determine the refrigerant capacity, the proposed method combines the outputs on the neural networks of all the time series examples in a single run of dehumidifier. As the result suggests, the proposed method is capable of detecting the refrigerant capacity with a low missing rate and high accuracy, which improves the maintenance mechanism of the dehumidifier and is of great significance to ensuring a comfortable air environment for user.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ren, H., Liu, S., Gao, Y.: Decompression machine fault diagnosis based on S600 building automation system. Refrig. Air Cond. Electr. Power 28(1), 11–13 (2007)
Gao, Y., Liu, S., Zhang, Z.: An improved fault identification method based on genetic fuzzy rules. Coal Mine Mach. 28(12), 181–183 (2007)
Liang, J.: Study on the real-time fault diagnosis expert system of dehumidifier based on CLIPS. Refrig. Air Cond. 3, 008 (2010)
He, B., Liu, S.: Faults diagnosis for dehumidifier based on genetic fuzzy C-means clustering algorithm. Refrig. Air Cond. Electr. Power Mach. 4, 005 (2009)
Wang, X., Liu, S., Liu, X.: Application of artificial neural network in fault diagnosis of dehumidifier. Mech. Electr. Eng. Technol. 36(7), 62–63 (2007)
Huang, H.: Fault diagnosis of dehumidifier based on RBFNN. Refrig. Air Cond. 4, 019 (2011)
Zhang, Q., Wu, Y., Xu, J.: Fault diagnosis and life prediction of dehumidifier based on genetic neural network. Environ. Eng. 1, 78–83 (2017)
He, W., Gao, Y.: COP prediction of dehumidifier based on GRNN and genetic algorithm and its application in fault diagnosis. Refrig. Air Cond. (Beijing) 9(5), 17–20 (2009)
Huang, Z., Liu, H., Liu, S., et al.: Research on faults diagnosis of dehumidifier based on COP and improved PNN. Refrig. Air Cond. (Sichuan) 24(5), 66–69 (2010)
Gao, Y., Liu, S., Zhang, Z.: Application of ARX model to fault diagnosis of dehumidifier. Comput. Simul. 25(2), 332–335 (2008)
Liu, H., Liu, S., Gao, Y., et al.: Faults diagnosis for dehumidifier based on LS-SVM ARX model. Refrig. Air Cond. Electr. Power 31(5), 47–51 (2010)
Gao, Y., Liu, S., Li, F., et al.: Fault detection and diagnosis method for cooling dehumidifier based on LS-SVM NARX model. Int. J. Refrig 61, 69–81 (2016)
Zhu, D., Yu, S.: A summary of knowledge-based fault diagnosis methods. J. Anhui Univ. Technol. (Nat. Sci.) 19(3), 197–204 (2002)
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167 (2015)
Acknowledgments
This paper was supported by foundation research project No. JCYJ20150730103208405 of Shenzhen Science and Technology Innovation Committee, and open research project of State Key Laboratory of Air-conditioning Equipment and System Energy Conservation, China.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Peng, G., Yang, Z., Wang, M. (2018). Refrigerant Capacity Detection of Dehumidifier Based on Time Series and Neural Networks. In: Qiao, J., et al. Bio-inspired Computing: Theories and Applications. BIC-TA 2018. Communications in Computer and Information Science, vol 952. Springer, Singapore. https://doi.org/10.1007/978-981-13-2829-9_10
Download citation
DOI: https://doi.org/10.1007/978-981-13-2829-9_10
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-2828-2
Online ISBN: 978-981-13-2829-9
eBook Packages: Computer ScienceComputer Science (R0)