Short-Term Firm-Level Energy-Consumption Forecasting for Energy-Intensive Manufacturing: A Comparison of Machine Learning and Deep Learning Models

To minimise environmental impact, to avoid regulatory penalties, and to improve competitiveness, energy-intensive manufacturing firms require accurate forecasts of their energy consumption so that precautionary and mitigation measures can be taken. Deep learning is widely touted as a superior analytical technique to traditional artificial neural networks, machine learning, and other classical time-series models due to its high dimensionality and problem-solving capabilities. Despite this, research on its application in demand-side energy forecasting is limited. We compare two benchmarks (Autoregressive Integrated Moving Average (ARIMA) and an existing manual technique used at the case site) against three deep-learning models (simple Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU)) and two machine-learning models (Support Vector Regression (SVR) and Random Forest) for short-term load forecasting (STLF) using data from a Brazilian thermoplastic resin manufacturing plant. We use the grid search method to identify the best configurations for each model and then use Diebold–Mariano testing to confirm the results. The results suggests that the legacy approach used at the case site is the worst performing and that the GRU model outperformed all other models tested.

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

[2]  Kadir Kavaklioglu,et al.  Modeling and prediction of Turkey’s electricity consumption using Support Vector Regression , 2011 .

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

[4]  Ran Li,et al.  Deep Learning for Household Load Forecasting—A Novel Pooling Deep RNN , 2018, IEEE Transactions on Smart Grid.

[5]  Haydar Demirhan,et al.  A bagging algorithm for the imputation of missing values in time series , 2019, Expert Syst. Appl..

[6]  Jürgen Schmidhuber,et al.  Learning to forget: continual prediction with LSTM , 1999 .

[7]  J. Ben Hadj Slama,et al.  Day-ahead load forecast using random forest and expert input selection , 2015 .

[8]  Haydar Demirhan,et al.  Missing value imputation for short to mid-term horizontal solar irradiance data , 2018, Applied Energy.

[9]  Ben Richard Hughes,et al.  Deep learning techniques for energy forecasting and condition monitoring in the manufacturing sector , 2020, Energy and Buildings.

[10]  Monjur Mourshed,et al.  Forecasting methods in energy planning models , 2018 .

[11]  Abbas S. Milani,et al.  Support vector regression modelling and optimization of energy consumption in carbon fiber production line , 2018, Comput. Chem. Eng..

[12]  C. Willmott,et al.  Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance , 2005 .

[13]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[14]  Ponnuthurai Nagaratnam Suganthan,et al.  Empirical Mode Decomposition based ensemble deep learning for load demand time series forecasting , 2017, Appl. Soft Comput..

[15]  Hongseok Kim,et al.  Deep Neural Network Based Demand Side Short Term Load Forecasting , 2016 .

[16]  Yacine Rezgui,et al.  A modular optimisation model for reducing energy consumption in large scale building facilities , 2014 .

[17]  A. Shabri,et al.  A comparison of time series forecasting using support vector machine and artificial neural network model , 2010 .

[18]  Iiro Harjunkoski,et al.  Integration of Energy-cost Optimization and Production Scheduling Using Multiparametric Programming , 2016 .

[19]  Heesung Yoon,et al.  Comparative Evaluation of ANN- and SVM-Time Series Models for Predicting Freshwater-Saltwater Interface Fluctuations , 2017 .

[20]  Ying Liu,et al.  Energy consumption modelling using deep learning technique — a case study of EAF , 2018 .

[21]  Bart De Schutter,et al.  Forecasting spot electricity prices Deep learning approaches and empirical comparison of traditional algorithms , 2018 .

[22]  Luh-Maan Chang,et al.  Prediction of Air-Conditioning Energy Consumption in R&D Building Using Multiple Machine Learning Techniques , 2020, Energies.

[23]  Stephan M. Wagner,et al.  Modeling carbon footprints across the supply chain , 2010 .

[24]  Ali Azadeh,et al.  Annual electricity consumption forecasting by neural network in high energy consuming industrial sectors , 2008 .

[25]  D. R. Cutler,et al.  Utah State University From the SelectedWorks of , 2017 .

[26]  Ravi Sankar,et al.  Time Series Prediction Using Support Vector Machines: A Survey , 2009, IEEE Computational Intelligence Magazine.

[27]  Ping-Huan Kuo,et al.  A High Precision Artificial Neural Networks Model for Short-Term Energy Load Forecasting , 2018 .

[28]  Chuan Li,et al.  Improving forecasting accuracy of daily enterprise electricity consumption using a random forest based on ensemble empirical mode decomposition , 2018, Energy.

[29]  Wei-Chang Yeh,et al.  Forecasting stock markets using wavelet transforms and recurrent neural networks: An integrated system based on artificial bee colony algorithm , 2011, Appl. Soft Comput..

[30]  Fabrice Rossi,et al.  Mean Absolute Percentage Error for regression models , 2016, Neurocomputing.

[31]  Zhou Lu,et al.  Energy consumption and economic growth: New evidence from the OECD countries , 2018, Energy.

[32]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[33]  Reyhan Aydoğan,et al.  Algorithm selection and combining multiple learners for residential energy prediction , 2019, Future Gener. Comput. Syst..

[34]  Francis X. Diebold,et al.  Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold–Mariano Tests , 2012 .

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