applied

sciences
Article
Forecasting of Power Demands Using Deep Learning
Taehyung Kang 1,† , Dae Yeong Lim 2,† , Hilal Tayara 3, * and Kil To Chong 2, *
1 IT Application Systems Engineering, Jeonbuk National University, Jeonju 54896, Korea; [email protected]
2 Advanced Electronics and Information Research Center, Jeonbuk National University, Jeonju 54896, Korea;
[email protected]
3 School of International Engineering and Science, Jeonbuk National University, Jeonju 54896, Korea
* Correspondence: [email protected] (H.T.); [email protected] (K.T.C.)
† These authors contributed equally to this work.




Received: 19 September 2020; Accepted: 13 October 2020; Published: 16 October 2020


Abstract: The forecasting of electricity demands is important for planning for power generator
sector improvement and preparing for periodical operations. The prediction of future electricity
demand is a challenging task due to the complexity of the available demand patterns. In this paper,
we studied the performance of the basic deep learning models for electrical power forecasting such as
the facility capacity, supply capacity, and power consumption. We designed different deep learning
models such as convolution neural network (CNN), recurrent neural network (RNN), and a hybrid
model that combines both CNN and RNN. We applied these models to the data provided by the
Korea Power Exchange. This data contains the daily recordings of facility capacity, supply capacity,
and power consumption. The experimental results showed that the CNN model outperforms the
other two models significantly for the three features forecasting (facility capacity, supply capacity,
and power consumption).

Keywords: power forecasting; deep learning; CNN; RNN; hybrid model

1. Introduction
The commercial electric power companies struggle to provide end-users with stable and safe
electricity. Therefore, designing efficient forecasting models is a vital step for the planning of the
operation of electronic power systems. The demand patterns of electricity could be affected by various
factors such as time, economy, social, and environmental factors [1,2].
The power forecasting models can be classified based on their predictive ability into short-term,
medium-term, and long-term. Short-term power forecasting models predict up to 1 day/week ahead
and used for scheduling the generation and transmission of electricity, medium-term power forecasting
models predict for 1 day/week to 1 year ahead and used for fuel preparation, long-term power
forecasting models predict more than 1 year and used for developing power supply and delivery
system [3–5].
Power demands prediction frameworks can be divided into statistical models, grey models,
and artificial intelligence models [6]. In statistical models, the correlation between the inputs and
the outputs is statistically figured out using empirical models such as log–linear regression models,
co-integration analysis and autoregressive integrated moving average (ARIMA), combined bootstrap
aggregation (bagging), and exponential smoothing. In the grey models, the researchers integrated a
partial theoretical structure with empirical data to build the structure and as result, a limited amount
of data are required to infer the behavior of the electrical systems. In artificial intelligence models,
they learn to model complex relationships between the outputs and inputs based on the available
training data.

Appl. Sci. 2020, 10, 7241; doi:10.3390/app10207241 www.mdpi.com/journal/applsci

Appl. Sci. 2020, 10, 7241 2 of 11

Different techniques have been applied in electricity demand forecasting including time series
models [7], holt-winters and seasonal regression [8], multiple linear regression [9,10], first-order
fuzzy time series [11], autoregressive integrated moving average (ARIMA) [12], seasonal ARIMA
(SARIMA) [2], support vector machine (SVM) [13], support vector regression [14], Least square SVM
(LSSVM) [1], and artificial neural network (ANN) [10,14,15].
For example, the work conducted by [11] used the monthly demand data from 1970 to 2009 as a
training dataset and 2010 data as a testing dataset. The electricity demand in Thailand was predicted by
including gross domestic product, maximum ambient temperature, and the population. These features
were used as input to a neural network [10]. The authors of [14] used Bayesian regularization in the
autoregressive neural network for electricity demand forecasting. Authors in [16] compared different
forecasting methods such as neural networks, fuzzy logic, and autoregressive process, and as a result,
they found that neural networks and fuzzy logic are more accurate that autoregressive processes.
Different techniques were combined by the authors of [17] in which they combined the support vector
machine, neural networks, autoregressive integrated moving average, and generalized regression
neural network. The authors of [18] proposed a traditional feed-forward neural network with one
output node that can predict peak load in one hour or one day in the future. Also, the researchers
explored radial basis function networks [19], recurrent neural networks [20], and self-organizing
maps [21].
All of these traditional methods have a limited performance as they require hand crafted features.
Therefore, with the advancement of deep learning, researchers have utilized different deep learning
architectures for several electricity demand forecasting problems [22–27]. Different resolutions have
been studied from multiple times a day such as monthly [22], weekly [27], daily [24–27], every two
hours [25], hourly [27], half-hourly [23–25], minute-by-minute [27]. These models studied different
targets such as business consumer [23,25], household consumer [24,27], other than business or
household consumer [22], and region/country such as UT Chandigarh, India, in [26].
In this paper, we studied the performance of deep learning models such as convolution
neural network (CNN) and recurrent neural network (RNN) for electricity demand forecasting.
Unlike previous works [28] that include year and month index to electricity consumption, we use only
the daily demand as an input to our predictive models. We designed different deep learning models
such as convolution neural network (CNN), recurrent neural network (RNN), and a hybrid model that
combines both CNN and RNN. We applied these models to the data provided by the Korea Power
Exchange. We built an independent model for each feature namely facility capacity, supply capacity,
and power consumption. The experimental results showed that the CNN model outperforms the other
two models significantly for the three features forecasting.

2. Materials and Methods

In this section, we introduce the dataset used for this study and the design of the proposed models.

2.1. Materials
In this paper, we used the data from the Korea Power Exchange. It contains recordings from
2003.01.01 to 2020.05.22. The available measurements were taken daily in this dataset—we have
6352 records. The available information is facility capacity, supply capacity, maximum power
consumption, supply reserve, and supply reserve ratio. Here, we are interested in predicting the
future demands of facility capacity, supply capacity, and maximum power consumption. The statistical
overview of these features is given in Table 1. We split that dataset into three parts. The training data
contains the records within the first 11 years (from 2003 to 2013), The validation data contains the
records within the following three years (from 2014 to 2016). Testing data contains the consumption
within the remaining years (from 2017 to 2020). Figure 1 shows the patterns of the available features in
the dataset within the study period.
Appl. Sci. 2020, 10, 7241 3 of 11

Table 1. The statistical overview of facility capacity, supply capacity, and power consumption.

Feature Count Mean STD Minimum Value (MW) Maximum Value (MW)
Facility Capacity (MW) 6352.0 83,447.03 20,918.82 53,800.0 126,887.0
Supply Capacity (MW) 6352.0 72,243.58 13,245.45 44,937.0 101,923.0
Power Consumption (MW) 6352.0 58,550.06 11,468.32 26,916.0 92,478.0

Figure 1. The characteristics of the features in the dataset within the period of the study.

2.2. Data Preprocessing and Preparation

The dataset was normalized to unit norm. To prepare the training, validation, and test sets,
we studied different scenarios by varying the history to be used for predicting future demands.
We tested different values of the history for predicting different future demands. The history values
were set to 7-days,15-days, 30-days, 45-days, and 60-days in the past. On the other hand, we tried to
predict future demands after 1-day, 7-days, 15-days, 30-days, 45-days, and 60-days. As a result, we had
to study 5 × 6 = 30 possible cases to see the best performing model that we can use for future demands
prediction. Figure 2 shows an example of preparing the dataset for 15-days in the past to predict the
demands 7-days ahead.

Figure 2. Example of data preparation.

2.3. The Proposed Models

We designed different deep learning models, namely convolution neural network (CNN) [29],
recurrent neural network (RNN) [30], and a hybrid model that combines CNN with RNN. We aimed
to find the best performing model according to the past and future data of our dataset.
CNN model is a type of deep neural network that is utilized in various domains. It is also considered
as a shift-invariant model as they have translation invariance characteristics and shared-weights
architecture. CNN models were widely and successfully used in different areas such as image and
Appl. Sci. 2020, 10, 7241 4 of 11

video classification, medical image analysis, natural language processing, bioinformatics, and time-series.
In this work, we designed a simple two-layer CNN model as shown in Figure 3. Each layer consists of
a 1-dimensional convolution layer with 32 filters and a filter size of 3, a non-linear activation function
which is the rectified linear unit (ReLU), and a max-pooling layer with window size and stride of 2.
The learned features from these two layers are then fed into a fully connected layer with a one-node
for prediction. The convolution layer is a 1-d convolution expressed in Equation (1) where I is the
input, k, and o are the indices of kernels and the output position, respectively, W f is the weight matrix
of S × N shape with S filters and N channels.
!
S −1 N −1
∑ ∑
f
Conv( I )ok = ReLU Wsn Io+s,n (1)
s =0 n =0

Figure 3. The architecture of the convultional neural network (CNN) model.

RNN is another typical deep learning model that is mainly used in natural language processing
and speech recognition [31,32]. It is used to understand the data’s sequential behavior and predict the
next likely outcomes [33]. In this paper, we have used a one bidirectional long short-term memory
(LSTM) layer with 16 nodes followed by a dropout layer [34] with a dropout probability of 0.2, and a
fully-connected layer with one node for prediction. Figure 4 shows the architecture of the RNN model.
Bi-LSTM has been used in different areas such as phoneme classification [35], speech recognition [36],
human action recognition [37], and machine translation [38]. Different gates are available in the LSTM
cell. The input gate is used to decide which information should be stored for the next layer and
update the current state. The forget gate is used to decide the information that should be removed
according to the previous inputs. The output gate decides which part of the state value should be
output. Thus, considering an input sequence {x}tT=1 , the LSTM has cell states {C}tT=1 , hidden states
{h}tT=1 and outputs a sequence {o}tT=1 . This can be expressed mathematically by Equation (2) where Wi ,
Wo , W f , Ui , Uo , U f are the weight matrices and bo , bc , bi , b f are the biases. Sigmoid and Tanh are the
activation functions. The  is the element-wise multiplication.
Appl. Sci. 2020, 10, 7241 5 of 11

ft = σ (b f + U f ht −1 + W f xt )
i t = σ ( bi + Ui h t − 1 + Wi x t )
ct = it  tanh(bc + Uc ht−1 + Wc xt ) + ft  ct−1 (2)
ot = σ (bo + Uo ot −1 + Wo xt )
ht = ot  tanh(ct )

Figure 4. The architecture of the recurrent neural network (RNN) model.

In addition, we have designed a hybrid model for future demand prediction. This model consists
of a 1-d convolution layer of 32 filters with a filter size of 3, ReLU, and a max-pooling layer. Then a
bidirectional LSTM layer with 16 nodes. The learned features from the hybrid model are then fed into
the fully connected layer with 1 node for prediction. This model is shown in Figure 5.

Figure 5. The architecture of the hybrid model.

For all of these models, we used the grid search algorithm for hyper-parameters tuning. We used
Keras framework for building and training the proposed models (https://keras.io/). The number of
Appl. Sci. 2020, 10, 7241 6 of 11

the epochs was set to 40 with early stopping based on validation loss. The RMSprop optimizer was
used for optimization with learning rate of 0.001 [39].

3. Results and Discussions

In this paper, we used mean absolute error (MAE) and R2 in order to evaluate the performance of
the proposed models. These parameters were calculated using the Scikit-learn tool (https://scikit-
learn.org/stable/).
We predicted future demands of facility capacity, supply capacity, and power consumption.
We tested the three developed tools namely CNN, RNN, and the hybrid model using different
values of histories and futures. Table 2 shows the best performing model with its past and future
configurations for the three features of the study. The ’-’ sign in Table 2 means that the model did not
fit the data at all. CNN model significantly outperforms RNN and hybrid models. We have extensively
searched for the best hyperparameters and architectures of RNN models but these models did not
converge. The main reason is that the size of the training dataset is not large enough to train the RNN
model. Thus, our future work will concentrate on collecting more data from Korean Power Exchange
to train more accurate forecasting models. Therefore, we will consider the CNN model for power
demands forecasting.
Figure 6 shows the heat maps of the R2 results for all past against future configurations in the
CNN model. For facility capacity forecasting, we can see that the CNN model performs outstandingly
for all combinations. The best R2 value is 0.992 for the past of 7 days and the future of 1 day. On the
other hand the minimum R2 is 0.820 for the past of 7 days and the future of 15 days. Thus, we can use
the developed CNN model for facility capacity forecasting using different past values for predicting
different future values.

Table 2. The best performance of the proposed models with different past and future configurations.

Feature Model Past (days) Future (days) R2 MAE

Facility Capacity (MW) CNN 7 1 0.992 0.025
Supply Capacity (MW) 7 1 0.851 0.214
Power Consumption (MW) 60 1 0.772 0.314
Facility Capacity (MW) RNN - - - -
Supply Capacity (MW) - - - -
Power Consumption (MW) - - - -
Facility Capacity (MW) Hybrid (CNN+RNN) - - - -
Supply Capacity (MW) 60 1 0.336 0.479
Power Consumption (MW) 7 1 0.627 0.409

For supply capacity forecasting, we can see that the CNN model performs well for some
combinations only, forecasting up to 15 days. The best R2 value is 0.851 for the past of 7 days
and the future of 1 day. Thus, we can use the developed model for short-term forecasting. The model
can forecast with 0.69 of R2 for the past of 7 days and the future of 7 days. Furthermore, the R2 is 0.43
for the past of 7 days and the future of 15 days.
For power consumption forecasting, we can see that the CNN model performs well for some
combinations only. The best R2 value is 0.772 for the past of 60 days and the future of 1 day.
Thus, we can use the developed model for short-term forecasting. The model can forecast with 0.68 of
R2 for the past of 45 days and the future of 7 days. Furthermore, the R2 is 0.59 for the past of 45 or
60 days and the future of 15 days.
Furthermore, the MAE of the CNN model outperforms the MAE results of the other models.
For instance, the MAE of facility capacity feature is 0.025. On other hand, the MAE of the supply
capacity of the CNN model is better by 0.265 than the hybrid model. Similarly, the MAE of the CNN
model in power consumption features is better by 0.095 of the hybrid model.
Appl. Sci. 2020, 10, 7241 7 of 11

0.99

0.99 0.98 0.82 0.98 0.97 0.94 0.85 0.69 0.43 -0.63 -1.8 -3.2 0

7
7
0.96

0.99 0.98 0.98 0.97 0.96 0.94 0.83 0.67 0.37 -1.5 -4.2 -4.5

15
1
15

0.93

0.99 0.98 0.98 0.97 0.94 0.91 0.84 0.67 0.28 -1.1 -3 -4

Past
Past

30
30

0.90 2

0.99 0.98 0.98 0.97 0.91 0.9 0.83 0.6 0.16 -2.2 -0.92 -2.2

45
45

0.87 3

0.99 0.98 0.97 0.95 0.91 0.89 0.82 0.62 -0.064 -1.2 -1.6 -3.6

60
60

0.84 4
1 7 15 30 45 60 1 7 15 30 45 60
Future Future

(a) Facility Capacity (b) Supply Capacity

0.75

0.68 0.6 0.49 0.2 -0.3 -0.67

7

0.50

0.72 0.6 0.53 0.23 -0.32 -0.67

15

0.25

0.74 0.66 0.57 0.26 -0.1 -0.67

Past
30

0.00

0.75 0.68 0.59 0.27 -0.2 -0.59

45

0.25

0.77 0.67 0.59 0.15 0.014 -0.31

60

0.50
1 7 15 30 45 60
Future

(c) Power Consumption

Figure 6. The heat map of R2 results for different past and future configurations in the CNN model.

In addition, we visualize the performance of CNN model for the three features in a sample future
interval as shown in Figures 7–9, respectively. It can be seen that the predicted results of facility
capacity and power consumption follow the observed values. On the other hand, the predicted results
of supply capacity are not always following the observed trend.

5.70

5.65

5.60
Facility Capacity (MW)

5.55

5.50

5.45

5.40

5.35

5.30
7/19/2018 9/7/2018 10/27/2018 12/16/2018 2/4/2019 3/26/2019 5/15/2019 7/4/2019 8/23/2019 10/12/2019
Date

OBSERVED PREDICTION

Figure 7. An example of the facility capacity prediction performance of the CNN model.
Appl. Sci. 2020, 10, 7241 8 of 11

Chart Title
4.00

3.50

3.00

Supply Capacity (MW)

2.50

2.00

1.50

1.00

0.50

0.00
6/4/2019 6/14/2019 6/24/2019 7/4/2019 7/14/2019 7/24/2019 8/3/2019 8/13/2019 8/23/2019
Date

OBSERVED PREDICTION

Figure 8. An example of the supply capacity prediction performance of the CNN model.
4.50

4.00

3.50
Power Consumption (MW)

3.00

2.50

2.00

1.50

1.00

0.50

0.00
6/4/2019 6/24/2019 7/14/2019 8/3/2019 8/23/2019 9/12/2019 10/2/2019
Date

OBSERVED PREDICTION

Figure 9. An example of the power consumption prediction performance of the CNN model.

In order to see the real performance of the proposed model, we compared it with a support vector
machine (SVM) and artificial neural networks (ANN). SVM was chosen as a benchmark model because
previous researchers have proven that SVM can produce satisfactory performance across various
power demands forecasting [40–42]. For a fair comparison, we also performed a hyperparameter
search for the penalty factor and gamma using grid search and found that the best performing penalty
factor and gamma are 1 and 0.001 for the facility capacity feature, 0.001 and 1 for the supply capacity
feature, and 100 and 0.001 for the power consumption feature.
ANN models have been used by many researchers and showed good performance such
as [10,14,15]. We have also performed a grid search for hyperparameter optimization. We designed a
two-layer ANN model where the first layer has 32 nodes followed by ReLU as a non-linear activation
function. Then a dropout layer with a drop rate of 0.5 was added. The second layer has one node for
prediction. Table 3 shows the comparison results between the proposed model and SVM and ANN
models. It can be seen that the CNN model outperforms ANN and SVM in facility capacity and supply
capacity features. However, in power consumption, SVM performs slightly better. It is known that
deep learning models require big datasets for training them therefore, our future work will include
collecting large datasets for more accurate forecasting models.

Table 3. Performance comparison between the proposed model and support vector machine (SVM)
and artificial neural networks (ANN) in terms of mean absolute error (MAE).

Feature CNN ANN SVM

Facility Capacity (MW) 0.025 0.502 1.285
Supply Capacity (MW) 0.214 0.262 3.218
Power Consumption (MW) 0.314 0.398 0.308
Appl. Sci. 2020, 10, 7241 9 of 11

4. Conclusions
Power demands forecasting is a challenging topic and important for future planning. In this paper,
we introduced the utilization of different deep learning models for future demands forecasting for
facility capacity, supply capacity, and power consumption. Different deep learning architectures were
studied, namely CNN, RNN, and the hybrid model that combines CNN with RNN. The experimental
results show that the CNN model outperformed RNN and hybrid model significantly. Furthermore,
we compared the performance of the CNN model with SVM and ANN models. The comparison
results showed that CNN performs generally better. The developed CNN model is a short-term power
demand forecasting model as it cannot forecast more than one day. The future plan is collecting more
training data from the Korea Power Exchange in order to train a more robust forecasting model that
can perform mid-term to long-term power demand forecasting.

Author Contributions: T.K. and D.Y.L. and H.T. prepared the dataset, conceived the algorithm, and carried out
the experiment and analysis. T.K., D.Y.L., and H.T. wrote the manuscript with support from K.T.C. All authors
discussed the results and contributed to the final manuscript. All authors have read and agreed to the published
version of the manuscript.
Funding: National Research Foundation of Korea: 2020R1A2C2005612; National Research Foundation of Korea:
NRF-2017M3C7A1044816
Acknowledgments: This work was supported in part by the National Research Foundation of Korea (NRF) grant
funded by the Korea government (MSIT) (No. 2020R1A2C2005612), in part by the Brain Research Program of the
National Research Foundation (NRF) funded by the Korean government (MSIT) (No. NRF-2017M3C7A1044816),
and in part by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded
by the Ministry of Education (No. 2019R1A6A3A01094685).
Conflicts of Interest: The authors declare that they have no competing interests.

References
1. Al-Shakarchi, R.G.; MM Ghulaim, M. Short-term load forecasting for baghdad electricity region. Electr. Mach.
Power Syst. 2000, 28, 355–371.
2. Sadeghi, K.; Ghaderi, S.; Azadeh, A.; Razmi, J. Forecasting electricity consumption by clustering data in
order to decrease the periodic variable’s effects and by simplifying the pattern. Energy Convers. Manag. 2009,
50, 829–836. [CrossRef]
3. Alfares, H.K.; Nazeeruddin, M. Electric load forecasting: Literature survey and classification of methods.
Int. J. Syst. Sci. 2002, 33, 23–34. [CrossRef]
4. Pedregal, D.J.; Trapero, J.R. Mid-term hourly electricity forecasting based on a multi-rate approach.
Energy Convers. Manag. 2010, 51, 105–111. [CrossRef]
5. Weron, R. Modeling and Forecasting Electricity Loads and Prices: A Statistical Approach; John Wiley & Sons Ltd.:
West Sussex, UK, 2007; Volume 403.
6. Hu, H.; Wang, L.; Peng, L.; Zeng, Y.R. Effective energy consumption forecasting using enhanced bagged
echo state network. Energy 2020, 193, 116778. [CrossRef]
7. Chen, G.; Li, K.; Chung, T.; Sun, H.; Tang, G. Application of an innovative combined forecasting method in
power system load forecasting. Electr. Power Syst. Res. 2001, 59, 131–137.[CrossRef]
8. Álvarez, C.; Añó, S. Stochastic load modelling for electric energy distribution applications. Top 1994,
2, 151–166. [CrossRef]
9. De Gooijer, J.G.; Hyndman, R.J. 25 years of time series forecasting. Int. J. Forecast. 2006, 22, 443–473.
[CrossRef]
10. Ediger, V.Ş.; Akar, S. ARIMA forecasting of primary energy demand by fuel in Turkey. Energy Policy 2007,
35, 1701–1708. [CrossRef]
11. Chen, H.; Canizares, C.A.; Singh, A. ANN-based short-term load forecasting in electricity markets.
In Proceedings of the 2001 IEEE Power Engineering Society Winter Meeting (Cat. No. 01CH37194),
Columbus, OH, USA, 28 January–1 February 2001; Volume 2, pp. 411–415.
12. Hahn, H.; Meyer-Nieberg, S.; Pickl, S. Electric load forecasting methods: Tools for decision making. Eur. J.
Oper. Res. 2009, 199, 902–907. [CrossRef]
Appl. Sci. 2020, 10, 7241 10 of 11

13. Niu, D.X.; Wang, Q.; Li, J.C. Short term load forecasting model based on support vector machine. In Advances
in Machine Learning and Cybernetics; Springer: Guangzhou, China, 2006; pp. 880–888.
14. Martín-Merino, M.; Román, J. A new SOM algorithm for electricity load forecasting. In International
Conference on Neural Information Processing; Springer: Hong Kong, China, 2006; pp. 995–1003.
15. Metaxiotis, K.; Kagiannas, A.; Askounis, D.; Psarras, J. Artificial intelligence in short term electric load
forecasting: A state-of-the-art survey for the researcher. Energy Convers. Manag. 2003, 44, 1525–1534.
[CrossRef]
16. Liu, K.; Subbarayan, S.; Shoults, R.; Manry, M.; Kwan, C.; Lewis, F.; Naccarino, J. Comparison of very
short-term load forecasting techniques. IEEE Trans. Power Syst. 1996, 11, 877–882. [CrossRef]
17. Bo, H.; Nie, Y.; Wang, J. Electric load forecasting use a novelty hybrid model on the basic of data preprocessing
technique and multi-objective optimization algorithm. IEEE Access 2020, 8, 13858–13874. [CrossRef]
18. Hippert, H.S.; Pedreira, C.E.; Souza, R.C. Neural networks for short-term load forecasting: A review and
evaluation. IEEE Trans. Power Syst. 2001, 16, 44–55. [CrossRef]
19. Gonzalez-Romera, E.; Jaramillo-Moran, M.A.; Carmona-Fernandez, D. Monthly electric energy demand
forecasting based on trend extraction. IEEE Trans. Power Syst. 2006, 21, 1946–1953. [CrossRef]
20. Srinivasan, D.; Lee, M. Survey of hybrid fuzzy neural approaches to electric load forecasting. In Proceedings
of the 1995 IEEE International Conference on Systems, Man and Cybernetics. Intelligent Systems for the 21st
Century, Yokohama, Japan, 20–24 March 1995; Volume 5; pp. 4004–4008.
21. Beccali, M.; Cellura, M.; Brano, V.L.; Marvuglia, A. Forecasting daily urban electric load profiles using
artificial neural networks. Energy Convers. Manag. 2004, 45, 2879–2900. [CrossRef]
22. Berriel, R.F.; Lopes, A.T.; Rodrigues, A.; Varejao, F.M.; Oliveira-Santos, T. Monthly energy consumption
forecast: A deep learning approach. In Proceedings of the 2017 International Joint Conference on Neural
Networks (IJCNN), Anchorage, AK, USA, 14–19 May 2017; pp. 4283–4290.
23. Chandramitasari, W.; Kurniawan, B.; Fujimura, S. Building deep neural network model for short term
electricity consumption forecasting. In Proceedings of the 2018 International Symposium on Advanced
Intelligent Informatics (SAIN), Yogyakarta, Indonesia, 29 August 2018, pp. 43–48.
24. Nair, A.S.; Hossen, T.; Campion, M.; Ranganathan, P. Optimal operation of residential EVs using DNN and
clustering based energy forecast. In Proceedings of the 2018 North American Power Symposium (NAPS),
Fargo, ND, USA, 9–11 September 2018; pp. 1–6.
25. Balaji, A.J.; Harish Ram, D.; Nair, B.B. A deep learning approach to electric energy consumption modeling.
J. Intell. Fuzzy Syst. 2019, 36, 4049–4055. [CrossRef]
26. Bedi, J.; Toshniwal, D. Deep learning framework to forecast electricity demand. Appl. Energy 2019,
238, 1312–1326. [CrossRef]
27. Kim, T.Y.; Cho, S.B. Predicting residential energy consumption using CNN-LSTM neural networks. Energy
2019, 182, 72–81. [CrossRef]
28. Oğcu, G.; Demirel, O.F.; Zaim, S. Forecasting electricity consumption with neural networks and support
vector regression. Procedia-Soc. Behav. Sci. 2012, 58, 1576–1585. [CrossRef]
29. LeCun, Y.; Bottou, L.; Bengio, Y.; Haffner, P. Gradient-based learning applied to document recognition.
Proc. IEEE 1998, 86, 2278–2324. [CrossRef]
30. Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997, 9, 1735–1780. [CrossRef]
[PubMed]
31. Sutskever, I.; Vinyals, O.; Le, Q.V. Sequence to sequence learning with neural networks. In Advances in
Neural Information Processing Systems; Curran Associates, Inc.: Montreal, QC, Canada, 2014; pp. 3104–3112.
32. Yao, K.; Zweig, G. Sequence-to-sequence neural net models for grapheme-to-phoneme conversion. arXiv
2015, arXiv:1506.00196.
33. Wang, Y.; Long, M.; Wang, J.; Gao, Z.; Philip, S.Y. Predrnn: Recurrent neural networks for predictive learning
using spatiotemporal lstms. In Advances in Neural Information Processing Systems; Curran Associates, Inc.:
Long Beach, CA, USA, 2017; pp. 879–888.
34. Srivastava, N.; Hinton, G.; Krizhevsky, A.; Sutskever, I.; Salakhutdinov, R. Dropout: A simple way to prevent
neural networks from overfitting. J. Mach. Learn. Res. 2014, 15, 1929–1958.
35. Graves, A.; Schmidhuber, J. Framewise phoneme classification with bidirectional LSTM and other neural
network architectures. Neural Netw. 2005, 18, 602–610. [CrossRef] [PubMed]
Appl. Sci. 2020, 10, 7241 11 of 11

36. Graves, A.; Jaitly, N.; Mohamed, A.R. Hybrid speech recognition with deep bidirectional LSTM.
In Proceedings of the 2013 IEEE workshop on automatic speech recognition and understanding, Olomouc,
Czech Republic, 8–12 December 2013; pp. 273–278.
37. Zhu, W.; Lan, C.; Xing, J.; Zeng, W.; Li, Y.; Shen, L.; Xie, X. Co-occurrence feature learning for skeleton based
action recognition using regularized deep LSTM networks. arXiv 2016, arXiv:1603.07772.
38. Sundermeyer, M.; Alkhouli, T.; Wuebker, J.; Ney, H. Translation modeling with bidirectional recurrent neural
networks. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing
(EMNLP), Doha, Qatar, 25–29 October 2014; pp. 14–25.
39. Tieleman, T.; Hinton, G. Lecture 6.5—RMSProp. In COURSERA: Neural Networks for Machine Learning;
Report; University of Toronto: Toronto, ON, Canada, 2012.
40. Son, H.; Kim, C. Short-term forecasting of electricity demand for the residential sector using weather and
social variables. Resour. Conserv. Recycl. 2017, 123, 200–207. [CrossRef]
41. Chen, Y.; Xu, P.; Chu, Y.; Li, W.; Wu, Y.; Ni, L.; Bao, Y.; Wang, K. Short-term electrical load forecasting using
the Support Vector Regression (SVR) model to calculate the demand response baseline for office buildings.
Appl. Energy 2017, 195, 659–670. [CrossRef]
42. Khosravi, A.; Koury, R.; Machado, L.; Pabon, J. Prediction of hourly solar radiation in Abu Musa Island
using machine learning algorithms. J. Clean. Prod. 2018, 176, 63–75. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional
affiliations.

c 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).