Assessment of long short-term memory and its modifications for enhanced short-term building energy predictions

Abstract Given the need for timely and reliable management of power distribution systems and smart grids , it is of great significance to develop a quick and accurate short-term building energy prediction model. Currently, the deep learning method, i.e., long short-term memory network (LSTM), is widely used for short-term building energy prediction. To further enhance the prediction accuracy and reduce the computational cost, previous studies have investigated improved LSTM models with modified structures such as LSTM-Attention, and LSTM-CNN. However, there is a lack of systematic assessment of these LSTM-based building energy forecast models considering the influencing factors such as model parameters tuning, modelling data volume, building type, climate features. Further, there is a lack of research on the combination of LSTM together with Attention and convolutional neural network (CNN) modifications. To address these research gaps, comparative evaluations of pure LSTM and five improved LSTM models (i.e., LSTM-CNN, CNN-LSTM, LSTM-Attention, CNN-Attention-LSTM, and LSTM-Attention-CNN) were performed in this study. These models were validated using the open-source data sets from the Building Data Genome Project 2 . Comparative studies were conducted on 60 randomly selected buildings from four different climate zones consisting of six different building types; evaluations were performed using either one-year or two-year energy consumption data . Further, the prediction performance of these models after parameter tuning was assessed in terms of prediction accuracy and computational time. The results demonstrated that, after parameter optimisation , LSTM models exhibited reduced root mean square error (RMSE) by 6.2%–29.2%. When only one-year data were used for modeling, CNN-LSTM decreased the average RMSEs of LSTM by as much as 2.9%. When two-year data were used for modelling, LSTM-ATT exhibited more stable prediction performance than the other models and decreased the average RMSE of LSTM by 5.6% at most.

[1]  Wenjie Gang,et al.  Assessment of deep recurrent neural network-based strategies for short-term building energy predictions , 2019, Applied Energy.

[2]  Xiangfei Kong,et al.  Heating load prediction based on attention long short term memory: A case study of Xingtai , 2020 .

[3]  Sylvain Robert,et al.  State of the art in building modelling and energy performances prediction: A review , 2013 .

[4]  Yongjun Sun,et al.  Statistical investigations of transfer learning-based methodology for short-term building energy predictions , 2020 .

[5]  Jin Wen,et al.  Review of building energy modeling for control and operation , 2014 .

[6]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[7]  Leopoldo Eduardo Cárdenas-Barrón,et al.  How does an industry manage the optimum cash flow within a smart production system with the carbon footprint and carbon emission under logistics framework? , 2019, International Journal of Production Economics.

[8]  Farrokh Janabi-Sharifi,et al.  Review of modeling methods for HVAC systems , 2014 .

[9]  R. Sendra-Arranz,et al.  A long short-term memory artificial neural network to predict daily HVAC consumption in buildings , 2020 .

[10]  Elie Azar,et al.  Occupant-centric miscellaneous electric loads prediction in buildings using state-of-the-art deep learning methods , 2020 .

[11]  Mary Ann Piette,et al.  Building thermal load prediction through shallow machine learning and deep learning , 2020, Applied Energy.

[12]  Weilin Li,et al.  Short-term electrical load forecasting using the Support Vector Regression (SVR) model to calculate the demand response baseline for office buildings , 2017 .

[13]  Ljupco Kocarev,et al.  Deep belief network based electricity load forecasting: An analysis of Macedonian case , 2016 .

[14]  Kibeom Ku,et al.  Building electric energy prediction modeling for BEMS using easily obtainable weather factors with Kriging model and data mining , 2018 .

[15]  Kaile Zhou,et al.  Load demand forecasting of residential buildings using a deep learning model , 2020 .

[16]  Ao Li,et al.  Advanced data analytics for enhancing building performances: From data-driven to big data-driven approaches , 2020, Building Simulation.

[17]  Sung-Bae Cho,et al.  Predicting residential energy consumption using CNN-LSTM neural networks , 2019, Energy.

[18]  Dolaana Khovalyg,et al.  Short-term energy use prediction of solar-assisted water heating system: Application case of combined attention-based LSTM and time-series decomposition , 2020 .

[19]  Teresa Wu,et al.  Short-term building energy model recommendation system: A meta-learning approach , 2016 .

[20]  Nelson Fumo,et al.  Methodology to estimate building energy consumption using EnergyPlus Benchmark Models , 2010 .

[21]  Ning Zhang,et al.  Probabilistic individual load forecasting using pinball loss guided LSTM , 2019, Applied Energy.

[22]  Mary Ann Piette,et al.  Data fusion in predicting internal heat gains for office buildings through a deep learning approach , 2019, Applied Energy.

[23]  Vivek Srikumar,et al.  Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks , 2018 .

[24]  Tao Lu,et al.  Modeling and forecasting energy consumption for heterogeneous buildings using a physical -statistical approach , 2015 .

[25]  Xiaoning Xu,et al.  Using long short-term memory networks to predict energy consumption of air-conditioning systems , 2020 .

[26]  Yuan Zhang,et al.  Short-Term Residential Load Forecasting Based on LSTM Recurrent Neural Network , 2019, IEEE Transactions on Smart Grid.

[27]  Xiaowei Luo,et al.  Spatial granularity analysis on electricity consumption prediction using LSTM recurrent neural network , 2019, Energy Procedia.

[28]  Paul Raftery,et al.  The Building Data Genome Project 2, energy meter data from the ASHRAE Great Energy Predictor III competition , 2020, Scientific Data.

[29]  Fu Xiao,et al.  Research and Applications of Data Mining Techniques for Improving Building Operational Performance , 2018 .

[30]  Nora El-Gohary,et al.  A review of data-driven building energy consumption prediction studies , 2018 .

[31]  Zhong-kai Feng,et al.  A hybrid short-term load forecasting model based on variational mode decomposition and long short-term memory networks considering relevant factors with Bayesian optimization algorithm , 2019, Applied Energy.

[32]  Manisa Pipattanasomporn,et al.  Robust short-term electrical load forecasting framework for commercial buildings using deep recurrent neural networks , 2020 .

[33]  Yang Zhao,et al.  A hybrid deep learning-based method for short-term building energy load prediction combined with an interpretation process , 2020 .

[34]  Fu Xiao,et al.  An interactive building power demand management strategy for facilitating smart grid optimization , 2014 .

[35]  Rita Streblow,et al.  Development and validation of grey-box models for forecasting the thermal response of occupied buildings , 2016 .

[36]  Jacob H. Stang,et al.  Load prediction method for heat and electricity demand in buildings for the purpose of planning for mixed energy distribution systems , 2008 .

[37]  Zeyu Wang,et al.  Random Forest based hourly building energy prediction , 2018, Energy and Buildings.

[38]  Fu Xiao,et al.  A short-term building cooling load prediction method using deep learning algorithms , 2017 .

[39]  Eenjun Hwang,et al.  Recurrent inception convolution neural network for multi short-term load forecasting , 2019, Energy and Buildings.

[40]  Wan He,et al.  Load Forecasting via Deep Neural Networks , 2017, ITQM.

[41]  Jing Wu,et al.  Analysis on the Influence of Building Envelope to Public Buildings Energy Consumption Based on DeST Simulation , 2015 .