Modeling Method of Heat Pump System Based on Recurrent Neural Network
- 214 Downloads
Heat pump system is a complex interaction system of multiple mechanical, electrical, and control systems. Traditional modeling methods based on physical laws have large computational complexity and poor precision, which make it not suitable for control strategy optimization. To solve these problems, this paper presents a modeling method based on recurrent neural network (RNN). The network structure and training algorithm were determined according to actual needs. The RNN model was tested and verified on a ground source heat pump system in an office building of a university in Northeast China. The heat pump operation data were continuously monitored and collected, and input into the neural network with three layers. Part of the data set is used for training and the rest is used for testing. The results show that the model has high precision, indicating that this modeling method is effective. This method is considered to be repeatable and can be applied to other heat pump systems.
KeywordsHeat pump system Recurrent neural network Modeling RMSprop Adam
The authors wish to acknowledge the support of the Program of National Science and Technology of China during the Thirteenth Five-year Plan (2017YFB0604004-03).
- 1.Robert, A., Ristinen/Jack, J.: Kranshaar energy and the environment, 2nd edn, Wiley, Inc (2006)Google Scholar
- 2.Chen, W., Zeng, N., et al.: Conductivity prediction and control method of electrode boiler based on artificial neural network. Electron. Technol. Softw. Eng. 01, 71–72 (2019)Google Scholar
- 3.Samek, D.: Elman neural networks in model predictive control. In: European Conference on Modelling and Simulation, 577–581 2009Google Scholar
- 4.Cho, K., et al.: Learning phrase representations using RNN encoder–decoder for statistical machine translation, Empirical methods in natural language processing, pp. 1724–1734 (2014)Google Scholar
- 5.Yang, Y., Hua, C., et al.: Modeling method of vapor compression refrigeration system based on artificial neural network. J. Armored Force Eng. Inst. 30(05), 69–72 (2016)Google Scholar
- 6.Zhang, H.: Research and improvement of optimization algorithm in deep learning. Beijing University of Posts and Telecommunications, China (2018)Google Scholar