Prediction of Building’s Thermal Performance Using LSTM and MLP Neural Networks
Source abstract
Abstract
Accurate prediction of building indoor temperatures and thermal demand is of great help to control and optimize the energy performance of a building. However, building thermal inertia and lag lead to complex nonlinear systems is difficult to model. In this context, the application of artificial neural networks (ANNs) in buildings has grown considerably in recent years. The aim of this work is to study the thermal inertia of a building by developing an innovative methodology using multi-layered perceptron (MLP) and long short-term memory (LSTM) neural networks. This approach was applied to a public library building located in the north of Spain. A comparison between the prediction errors according to the number of time lags introduced in the models has been carried out. Moreover, the accuracy of the models was measured using the CV(RMSE) as advised by AHSRAE. The main novelty of this work lies in the analysis of the building inertia, through machine learning algorithms, observing the information provided by the input of time lags in the models. The results of the study prove that the best models are those that consider the thermal lag. Errors below 15% for thermal demand and below 2% for indoor temperatures were achieved with the proposed methodology.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Building Energy and Comfort Optimization · Energy Load and Power Forecasting · Wind and Air Flow Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mechanical Engineering, Heat Engines and Fluids Mechanics, Industrial Engineering School, University of Vigo, Maxwell s/n, 36310 Vigo, Spain
Defense University Center, Spanish Naval Academy, Plaza de España, s/n, 36920 Marín, Spain
Department of Applied Mathematics I. Telecommunications Engineering School, University of Vigo, 36310 Vigo, Spain
Funding
Ministerio de Ciencia, Innovación y Universidades
References (62)
Show source-provided references
1. Zero energy buildings and sustainable development implications—A review10.1016/j.energy.2013.01.070 ↗
2. Official Journal of the European Union (2010). Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the Energy Performance of Buildings, EU.
3. European Commission (2012). Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on Energy Efficiency, Amending DIRECTIVES 2009/125/EC and 2010/30/EU and Repealing Directives 2004/8/EC and 2006/32/EC., EU.
4. Towards nearly zero-energy residential neighbourhoods in the European Union: A case study10.1016/j.rser.2020.110198 ↗
5. Modeling and forecasting building energy consumption: A review of data-driven techniques10.1016/j.scs.2019.101533 ↗
6. A review on time series forecasting techniques for building energy consumption10.1016/j.rser.2017.02.085 ↗
7. Methodology to estimate building energy consumption using EnergyPlus Benchmark Models10.1016/j.enbuild.2010.07.027 ↗
8. A review on modeling and simulation of building energy systems10.1016/j.rser.2015.12.040 ↗
9. Using building energy simulation and geospatial modeling techniques to determine high resolution building sector energy consumption profiles10.1016/j.enbuild.2008.01.005 ↗
10. Comparison between detailed model simulation and artificial neural network for forecasting building energy consumption10.1016/j.enbuild.2008.06.013 ↗
11. University of Wisconsin—Madison (1975). Solar Energy, L. TRNSYS, a Transient Simulation Program, The Laboratory.
12. EnergyPlus: Creating a new-generation building energy simulation program10.1016/S0378-7788(00)00114-6 ↗
13. James, J. (2020, September 01). Hirsch & Associates (JJH). DOE-2 (version 2.2-047d). Available online: http://www.doe2.com/.
14. ASHRAE (2014). Guideline 14-2014—Measurement of Energy, Demand, and Water Savings, ASHRAE.
15. A Survey of Methods for Explaining Black Box Models
16. Runge, J., and Zmeureanu, R. (2019). Forecasting Energy Use in Buildings Using Artificial Neural Networks: A Review. Energies, 12.10.3390/en12173254 ↗
17. Oh, S. (2020). Comparison of a Response Surface Method and Artificial Neural Network in Predicting the Aerodynamic Performance of a Wind Turbine Airfoil and Its Optimization. Appl. Sci., 10.10.3390/app10186277 ↗
18. An artificial neural network model for rainfall forecasting in Bangkok, Thailand10.5194/hess-13-1413-2009 ↗
19. Using artificial neural network models in stock market index prediction10.1016/j.eswa.2011.02.068 ↗
20. Modelling of solar energy potential in Nigeria using an artificial neural network model10.1016/j.apenergy.2008.12.005 ↗
21. Short-Term Residential Load Forecasting Based on LSTM Recurrent Neural Network10.1109/TSG.2017.2753802 ↗
22. Artificial neural network model as a potential alternative for groundwater salinity forecasting10.1016/j.jhydrol.2010.12.016 ↗
23. Turbidity Prediction in a River Basin by Using Artificial Neural Networks: A Case Study in Northern Spain10.1007/s11269-013-0487-9 ↗
24. Neural networks applied to discriminate botanical origin of honeys10.1016/j.foodchem.2014.11.121 ↗
25. Gil-Cordero, E., and Cabrera-Sánchez, J.-P. (2020). Private Label and Macroeconomic Indexes: An Artificial Neural Networks Application. Appl. Sci., 10.10.3390/app10176043 ↗
26. Predicting Tumor Category Using Artificial Neural Networks
27. Neural networks applied in chemistry. I. Determination of the optimal topology of multilayer perceptron neural networks10.1002/cem.1401 ↗
28. Prediction of tension properties of cork from its physical properties using neural networks10.1007/s00107-015-0885-1 ↗
29. Artificial neural network model for forecasting sub-hourly electricity usage in commercial buildings10.1016/j.enbuild.2015.11.045 ↗
30. Machine learning approaches for estimating commercial building energy consumption10.1016/j.apenergy.2017.09.060 ↗
31. Predicting heating demand and sizing a stratified thermal storage tank using deep learning algorithms10.1016/j.apenergy.2018.06.064 ↗
32. Lipton, Z., Kale, D., Elkan, C., and Wetzel, R. (2015). Learning to Diagnose with LSTM Recurrent Neural Networks. arXiv.
33. Mikolov, T., and Zweig, G. (2012, January 2–5). Context Dependent Recurrent Neural Network Language Model. Proceedings of the 2012 IEEE Workshop on Spoken Language Technology, SLT 2012, Miami, FL, USA.10.1109/SLT.2012.6424228 ↗
34. Poulose, A., and Han, D.S. (2020). UWB Indoor Localization Using Deep Learning LSTM Networks. Appl. Sci., 10.10.3390/app10186290 ↗
35. Sak, H., Senior, A., and Beaufays, F. (2014). Long Short-Term Memory Based Recurrent Neural Network Architectures for Large Vocabulary Speech Recognition. arXiv.10.21437/Interspeech.2014-80 ↗
36. From Feedforward to Recurrent LSTM Neural Networks for Language Modeling10.1109/TASLP.2015.2400218 ↗
37. Sundermeyer, M., Schlüter, R., and Ney, H. (2012). LSTM Neural Networks for Language Modeling, Science Department RWTH Aachen University.10.21437/Interspeech.2012-65 ↗
38. Prediction for Tourism Flow based on LSTM Neural Network10.1016/j.procs.2018.03.076 ↗
39. Duan, Y., and Yisheng, L. (2016, January 1–4). Travel Time Prediction with LSTM Neural Network. Proceedings of the 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC), Rio De Janeiro, Brazil.10.1109/ITSC.2016.7795686 ↗
40. Sequential Fault Diagnosis Based on LSTM Neural Network10.1109/ACCESS.2018.2794765 ↗
41. Predicting residential energy consumption using CNN-LSTM neural networks10.1016/j.energy.2019.05.230 ↗
42. Marino, D.L., Amarasinghe, K., and Manic, M. (2016, January 23–26). Building energy load forecasting using Deep Neural Networks. Proceedings of the IECON 2016—42nd Annual Conference of the IEEE Industrial Electronics Society, Florence, Italy.10.1109/IECON.2016.7793413 ↗
43. Calibrated simulation of a public library HVAC system with a ground-source heat pump and a radiant floor using TRNSYS and GenOpt10.1016/j.enbuild.2015.09.006 ↗
44. Sensitivity analysis of a vertical geothermal heat exchanger dynamic simulation: Calibration and error determination10.1016/j.geothermics.2017.06.012 ↗
45. Review on Methods to Fix Number of Hidden Neurons in Neural Networks10.1155/2013/425740 ↗
46. Chen, K., Yang, S., and Batur, C. (2012, January 29–31). Effect of multi-hidden-layer structure on performance of BP neural network: Probe. Proceedings of the 2012 8th International Conference on Natural Computation, Chongqing, China.10.1109/ICNC.2012.6234604 ↗
47. Shin-ike, K. (2010, January 18–21). A two phase method for determining the number of neurons in the hidden layer of a 3-layer neural network. Proceedings of the SICE Annual Conference 2010, Taipei, Taiwan.
48. Doukim, C., Dargham, J., and Chekima, A. (2010, January 10–13). Finding the Number of hidden Neurons for an MLP Neural Network Using Coarse to Fine Search Technique. Proceedings of the 10th International Conference on Information Science, Signal Processing and their Applications (ISSPA 2010), Kuala Lumpur, Malaysia.10.1109/ISSPA.2010.5605430 ↗
49. Vujicic, T., Matijević, T., Ljucovic, J., Balota, A., and Sevarac, Z. (2016). Comparative Analysis of Methods for Determining Number of Hidden Neurons in Artificial Neural Network. Artif. Intell. Rev., 48.
50. Landslide risk analysis using artificial neural network model focusing on different training sites
51. Theoretical analysis of batch and on-line training for gradient descent learning in neural networks10.1016/j.neucom.2009.05.017 ↗
52. Li, M., Soltanolkotabi, M., and Oymak, S. (2020, January 3–5). Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks. Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, Palermo, Italy.
53. Bilbao, I., and Bilbao, J. (2017, January 5–7). Overfitting problem and the over-training in the era of data: Particularly for Artificial Neural Networks. Proceedings of the 2017 Eighth International Conference on Intelligent Computing and Information Systems (ICICIS), Cairo, Egypt.10.1109/INTELCIS.2017.8260032 ↗
54. A comparison of deep networks with ReLU activation function and linear spline-type methods10.1016/j.neunet.2018.11.005 ↗
55. Bock, S., and Weiß, M. (2019, January 14–19). A Proof of Local Convergence for the Adam Optimizer. Proceedings of the 2019 International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary.10.1109/IJCNN.2019.8852239 ↗
56. Li, M., Zhang, T., Chen, Y., and Smola, A. (2014, January 24–27). Efficient mini-batch training for stochastic optimization. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, New York, NY, USA.10.1145/2623330.2623612 ↗
57. Zaremba, W., Sutskever, I., and Vinyals, O. (2014). Recurrent Neural Network Regularization. arXiv.
58. Daily Sediment Yield Modeling with Artificial Neural Network using 10-fold Cross Validation Method: A small agricultural watershed, Kapgari, India
59. Cacabelos, A., Eguía, P., Febrero Garrido, L., and Granada, E. (2017). Development of a new multi-stage building energy model calibration methodology and validation in a public library. Energy Build., 146.10.1016/j.enbuild.2017.04.071 ↗
60. Framework for Approaching the Minimum CV(RMSE) using Energy Simulation and Optimization Tool10.1016/j.egypro.2016.06.157 ↗
61. Kuo, P.-H., and Huang, C.-J. (2018). A High Precision Artificial Neural Networks Model for Short-Term Energy Load Forecasting. Energies, 11.10.3390/en11010213 ↗
62. Pilgrim, M., and Willison, S. (2009). Dive into Python 3, Springer.10.1007/978-1-4302-2416-7 ↗