A Novel Hybrid Deep Neural Network Model to Predict the Refrigerant Charge Amount of Heat Pumps

Improper refrigerant charge amount (RCA) is a recurring fault in electric heat pump (EHP) systems. Because EHP systems show their best performance at optimum charge, predicting the RCA is important. There has been considerable development of data-driven techniques for predicting RCA; however, the current data-driven approaches for estimating RCA suffer from poor generalization and overfitting. This study presents a hybrid deep neural network (DNN) model that combines both a basic DNN model and a thermodynamic model to counter the abovementioned challenges of existing data-driven approaches. The data for designing models were collected from two EHP systems with different specifications, which were used for the training and testing of models. In addition to the data obtained using the basic DNN model, the hybrid DNN model uses the thermodynamic properties as a thermodynamic model. The testing results show that the hybrid DNN model has a prediction performance of 93%, which is 21% higher than that of the basic DNN model. Furthermore, for model training and model testing, the hybrid DNN model has a 6% prediction performance difference, indicating its reliable generalization capabilities. To summarize, the hybrid DNN model improves data-driven approaches and can be used for designing efficient and energy-saving EHP systems.

[1]  Lourdes F. Vega,et al.  Transport properties of HFC and HFO based refrigerants using an excess entropy scaling approach , 2018 .

[2]  Huanxin Chen,et al.  Sensitivity analysis for PCA-based chiller sensor fault detection , 2016 .

[3]  Michael I. Jordan,et al.  Learning with Mixtures of Trees , 2001, J. Mach. Learn. Res..

[4]  James E. Braun,et al.  Virtual Refrigerant Pressure Sensors for Use in Monitoring and Fault Diagnosis of Vapor-Compression Equipment , 2009 .

[5]  Haitao Hu,et al.  Influence of oil on flow condensation heat transfer of R410A inside 4.18 mm and 1.6 mm inner diameter horizontal smooth tubes , 2010 .

[6]  Yoshua Bengio,et al.  Random Search for Hyper-Parameter Optimization , 2012, J. Mach. Learn. Res..

[7]  Huanxin Chen,et al.  Optimized neural network-based fault diagnosis strategy for VRF system in heating mode using data mining , 2017 .

[8]  Bo Fan,et al.  A hybrid FDD strategy for local system of AHU based on artificial neural network and wavelet analysis , 2010 .

[9]  Shanlin Yang,et al.  A deep learning model for short-term power load and probability density forecasting , 2018, Energy.

[10]  Reinhard Radermacher,et al.  Two-stage heat pump system with vapor-injected scroll compressor using R410A as a refrigerant , 2009 .

[11]  Yong Chan Kim,et al.  The effects of improper refrigerant charge on the performance of a heat pump with an electronic expansion valve and capillary tube , 2002 .

[12]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[13]  Davy Geysen,et al.  Thermal load forecasting in district heating networks using deep learning and advanced feature selection methods , 2018, Energy.

[14]  Tanveer Ahmad,et al.  Deep learning-based fault diagnosis of variable refrigerant flow air-conditioning system for building energy saving , 2018, Applied Energy.

[15]  Madeleine Gibescu,et al.  Unsupervised energy prediction in a Smart Grid context using reinforcement cross-building transfer learning , 2016 .

[16]  Han Jun Kim,et al.  Development and application of the load responsive control of the evaporating temperature in a VRF system for cooling energy savings , 2016 .

[17]  Eric W. Lemmon,et al.  Equations of State for Mixtures of R-32, R-125, R-134a, R-143a, and R-152a , 2004 .

[18]  Santiago Velasco,et al.  Extended corresponding states expressions for the changes in enthalpy, compressibility factor and constant-volume heat capacity at vaporization , 2015 .

[19]  Xiaoyan Wang,et al.  An efficient VRF system fault diagnosis strategy for refrigerant charge amount based on PCA and dual neural network model , 2018 .

[20]  Savvas A. Tassou,et al.  Sensitivity of refrigeration system performance to charge levels and parameters for on-line leak detection , 2005 .

[21]  Woohyun Kim,et al.  Performance evaluation of a virtual refrigerant charge sensor , 2013 .

[22]  Jiong Li,et al.  Extending the virtual refrigerant charge sensor (VRC) for variable refrigerant flow (VRF) air conditioning system using data-based analysis methods , 2016 .

[23]  Elena Lucchi,et al.  Energy Efficiency, Heritage Conservation, and Landscape Integration: the Case Study of the San Martino Castle in Parella (Turin, Italy) , 2017 .

[24]  Jong Min Choi,et al.  Influence of the expansion device on the performance of a heat pump using R407C under a range of charging conditions , 2004 .

[25]  F. Varas,et al.  A hybrid formulation for fast explicit calculation of thermodynamic properties of refrigerants , 2012 .

[26]  Reinhard Radermacher,et al.  Integration of variable refrigerant flow and heat pump desiccant systems for the heating season , 2010 .

[27]  Jamal El Abbadi,et al.  Prediction of thermodynamic properties of refrigerant fluids with a new three-parameter cubic equation of state , 2016 .

[28]  Inhan Kim,et al.  Dynamic target high pressure control of a VRF system for heating energy savings , 2017 .

[29]  Honghyun Cho,et al.  Effects of refrigerant charge amount on the performance of a transcritical CO2 heat pump , 2005 .

[30]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[31]  Bryan P. Rasmussen,et al.  A comparison of static and dynamic fault detection techniques for transcritical refrigeration , 2017 .

[32]  A K Halm-Owoo,et al.  Applications of fault detection and diagnostic techniques for refrigeration and air conditioning: A review of basic principles , 2002 .

[33]  SchmidhuberJürgen.  Deep learning in neural networks , 2015 .

[34]  Shaobin Li,et al.  An efficient online wkNN diagnostic strategy for variable refrigerant flow system based on coupled feature selection method , 2019 .

[35]  Nikolaos Doulamis,et al.  Deep Learning for Computer Vision: A Brief Review , 2018, Comput. Intell. Neurosci..

[36]  Ying Li,et al.  Modeling evaluation of a direct-expansion solar-assisted heat pump water heater using R410A. , 2017 .

[37]  Woohyun Kim,et al.  Evaluation of the impacts of refrigerant charge on air conditioner and heat pump performance , 2010 .

[38]  Honghyun Cho,et al.  Refrigerant amount detection algorithm for a ground source heat pump unit , 2012 .

[39]  Eric W. Lemmon,et al.  Equation of State for the Thermodynamic Properties of trans-1,3,3,3-Tetrafluoropropene [R-1234ze(E)] , 2016 .

[40]  Eric F. May,et al.  Thermodynamic properties of hydrofluoroolefin (R1234yf and R1234ze(E)) refrigerant mixtures: Density, vapour-liquid equilibrium, and heat capacity data and modelling , 2019, International Journal of Refrigeration.

[41]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[42]  Ivor W. Tsang,et al.  Domain Adaptation via Transfer Component Analysis , 2009, IEEE Transactions on Neural Networks.

[43]  Man-Hoe Kim,et al.  Evaporating heat transfer of R22 and R410A in horizontal smooth and microfin tubes , 2005 .

[44]  Reinhard Radermacher,et al.  A review of recent development in variable refrigerant flow systems , 2015 .

[45]  Xin Zhou,et al.  Deep neural networks: A promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data , 2016 .

[46]  R JesúsE.Mora,et al.  Thermodynamic properties of refrigerants using artificial neural networks , 2014 .

[47]  Yacine Rezgui,et al.  Trees vs Neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption , 2017 .

[48]  Sukho Lee,et al.  Using deep learning approaches with variable selection process to predict the energy performance of a heating and cooling system , 2020 .

[49]  Masaki Onishi,et al.  Effective hyperparameter optimization using Nelder-Mead method in deep learning , 2017, IPSJ Transactions on Computer Vision and Applications.

[50]  Huanxin Chen,et al.  Refrigerant charge fault diagnosis in the VRF system using Bayesian artificial neural network combined with ReliefF filter , 2017 .

[51]  Savvas A. Tassou,et al.  Fault diagnosis and refrigerant leak detection in vapour compression refrigeration systems. , 2005 .

[52]  Eric W. Lemmon,et al.  A New Functional Form and New Fitting Techniques for Equations of State with Application to Pentafluoroethane (HFC-125) , 2005 .

[53]  Gianpiero Evola,et al.  Dynamic thermal and hygrometric simulation of historical buildings: Critical factors and possible solutions , 2020, Renewable and Sustainable Energy Reviews.