energies

Article
PV Power Prediction, Using CNN-LSTM Hybrid
Neural Network Model. Case of Study:
Temixco-Morelos, México
Mario Tovar ∗ , Miguel Robles and Felipe Rashid
Energy Systems Department, Instituto de Energías Renovables, Universidad Nacional Autónoma de México,
Priv. Xochicalco S/N Temixco, Morelos 62580, Mexico; mrp@ier.unam.mx (M.R.); frzs@ier.unam.mx (F.R.)
* Correspondence: matr@ier.unam.mx




Received: 21 October 2020; Accepted: 5 December 2020; Published: 10 December 2020


Abstract: Due to the intermittent nature of solar energy, accurate photovoltaic power predictions are
very important for energy integration into existing energy systems. The evolution of deep learning
has also opened the possibility to apply neural network models to predict time series, achieving
excellent results. In this paper, a five layer CNN-LSTM model is proposed for photovoltaic power
predictions using real data from a location in Temixco, Morelos in Mexico. In the proposed hybrid
model, the convolutional layer acts like a filter, extracting local features of the data; then the temporal
features are extracted by the long short-term memory network. Finally, the performance of the
hybrid model with five layers is compared with a single model (a single LSTM), a CNN-LSTM hybrid
model with two layers and two well known popular benchmarks. The results also shows that the
hybrid neural network model has better prediction effect than the two layer hybrid model, the single
prediction model, the Lasso regression or the Ridge regression.

Keywords: Neural Networks; CNN; LSTM; PV power predictions; Microgrids

1. Introduction
Nowadays renewable energy (RE) aims to be a real solution to solve the fossil fuel problem.
A very well known renewable energy is solar photo-voltaic (PV), which is also in fast growth. Global
PV capacity is estimated to increase significantly from 593.9 GW in 2019 to 1582.9 GW in 2030 following
significant capacity additions by China, India, Germany, the US and Japan [1]. Solar PV generation
plays an important role in the future of energy structure. According to some estimations of the Joint
Research Centre (JRC), solar PV power generation capacity may reach 1.4 TW by 2024 just in Europe [2].
Nevertheless, the intermittent nature of the solar resource poses big challenges for energy integration
into existing energy systems. Precise PV prediction is a good way to solve this problem [3].
On the other hand, over the past eight years, off-grid systems in the form of stand-alone solar
home systems (SHS’s) have proved the most popular and immediate solution for increasing energy
access in rural areas across the Global South [4]. In Mexico, there are aprox. 1.2 million people living
without access to the electrical grid [5]. SHS’s can be a real option to empower rural communities with
autonomous energy production [6]. In rural places with solar resource and with no access to the grid,
this is a viable solution.
The interconnection of existing stand-alone solar home systems can form micro grids [4]. This tiny
grid is formed by prosumers (households capable of producing and consuming electrical power,
Figure 1) and consumers (households only consuming power). In this new architecture of distributed
PV generation, many authors propose that consumers and prosumers can even trade self-produced
energy on a micro grid energy market. This involves an information layer that feeds data to an energy

Energies 2020, 13, 6512; doi:10.3390/en13246512 www.mdpi.com/journal/energies

Energies 2020, 13, 6512 2 of 15

management system (EMS) as it is shown in Figure 1. The EMS acts as the middleman between the
physical and the information layer. The EMS needs to have accurate prediction photo voltaic power
(PVP) models, consumption prediction models, an energy market model and other tools which are
essential in a smart energy management system for increasing reliability, sustainability, efficiency and
flexibility [7–10].

Figure 1. Micro-grid scenario with distributed generation (GD).

In general, there are three main prediction methods: statistical models, physical models and
machine learning models. The physical model relies on dynamics between solar radiation and laws of
physics [11]. The statistical model mainly depends on historical data, statistics and probability theory
to forecast future time series [12,13]. The machine learning models map directly from inputs to outputs,
they extract complex nonlinear features in a very efficient way [14,15]. In the machine learning models
we can highlight artificial neural networks (ANN) especially one type of ANN. Recurrent neural
networks (RNN), which is one of the most commonly used methods for forecasting time series [16,17].
The RNN has been studied in various applications like wind speed prediction [18], energy power
consumption prediction [19,20] or even traffic prediction [21], achieving excellent results.
However, one common problem in RNN (with gradient-based learning methods and
back-propagation), is the vanishing of gradient. Gradient vanishing occurs while training long
data sequences. This means that the gradient of the loss function approaches zero, making the network
hard to train [22]. Long Short Term Memory networks (LSTM) solve this problem [23,24].
PV prediction using LSTM models has been studied by many authors [14,17,22] achieving to
reduce the prediction error compared with other traditional methods.
In recent years, many researchers have combined CNN and LSTM models to extract temporal
and spacial features. In the medical field Gill et al. proposed a CNN-LSTM model to accurately
detect arrhythmias in the ECG [16]. Zhang C. Y. used deep belief networks (DBN) for wind speed
prediction [21] achieving better results than the traditional methods (such as SVR). Kim [25] proposed
a hybrid CNN-LSTM model for electric energy consumption achieving superior results than other
deep learning based methods. They found that extracting first the local features and then temporal
ones worked better than a LSTM-CNN model, performing with a medium square error (MSE) of
3.738. They also found that time series decomposition with deep learning models provides useful
visualizations to better understand the problem of predicting and analyzing energy consumption.
Energies 2020, 13, 6512 3 of 15

Wang et al. proposed an hybrid LSTM-CNN model for PV power prediction [26]. Accomplishing
four main contributions:

• They proposed a hybrid photo voltaic power prediction deep learning model which considers the
temporal and local features.
• The temporal features of the data were first extracted (using a LSTM model) and then the local
features using a CNN model.
• They reduced the complexity of the model by selecting the 1D-CNN model to extract the local
features, and then the data conversion link was eliminated. Therefore PV power prediction is
greatly facilitated.
• They compared with other models (CNN, LSTM, CNN-LSTM), to prove the validity of the model.

Wang et al. [26] proposed a hybrid model of one dimension for PV prediction. In this work
we choose a hybrid model with a stronger multi-layer architecture, this includes a 5D-CNN model
with max pooling and a 5D-LSTM model. Indeed, the five dimensional CNN and LSTM model will
consume more computational resources for training than a uni-dimentional model, but high accuracy
will be achieved [27,28]. Therefore, the computational time and some metrics such as the mean square
error (MSE) of the proposed model will be compared with other deep learning models.

2. The Dataset
We use data of a 1.7 kW photo voltaic solar system (PVSS) and from a meteorological station
(the ESOLMET station (Figure 2)) 120 m from the PVSS (Figure 3), in the IER-UNAM (Renewable
Energy Institute) in Temixco, Morelos, México. Latitude: 18◦ 500 25.6200 N, longitude: 99◦ 140 10.4900 W at
1253 m above sea level.
The electric variables measured from the PVSS after the inverter are the following: active power
generated [W], average phase current generated [A] and the AC Voltage generated [V]. The variables
available by the ESOLMET are: direct radiation [W/m2 ], global radiation [W/m2 ], diffuse radiation
[W/m2 ], temperature [◦ C], humidity [%], wind speed [m/s] and barometric pressure [mBar]. All these
ten variables will conform our dataset, and are shown in Figure 4.

Figure 2. ESOLMET weather station, IER, Temixco, Morelos [29].

The dataset consists of 52,428 values per variable. From the date: 01/01/2019 00:00:00:00 to the
date: 01/01/2020 00:00:00:00, with a resolution of 10 minutes. We used 80% of the data to train the
model (41,942 values) and 20% to validate our predictions(10,486 values).
Photo voltaic power generated (PVPG), is the variable we want to predict, in Figure 5. PVPG is
shown through the complete year. As it will be explained in detail in the next section, our model takes
an input of 10 variables and the proposed neural network throws a specific output(PVPG) in time.
Energies 2020, 13, 6512 4 of 15

Figure 3. Distance between ESOLMET and the PVSS in Temixco, Morelos.

Figure 4. ESOLMET Dataset, IER, Temixco, 2019 [29].

Figure 5. PV Generation in Temixco during 2019.

Energies 2020, 13, 6512 5 of 15

3. Proposed Hybrid Model

3.1. Local Feature Learning with CNN

CNN are very popular for extracting local features in images, for example. Convolution is the
main concept of a CNN. Our proposed CNN includes two parts: the convolutional layers (Equations (1)
and (2)) and the pooling layer (the main purpose of this layer is to reduce the number of parameters of
the tensor by reducing its size Figure 6) which helps to reduce computation time.

Figure 6. Convolution and Max-pooling process, image adapted from [26].

In the convolution layer, the previous layer features graph interacts with the convolutional kernel;
this interaction forms the output feature graph j of the convolutional layer. Each one of this output
feature graph j might contain a convolution with multiple input feature graphs.
The equations that define the convolution layer are:

y j = ( ∑ til −1 ⊗ wij ) + b j
(l ) (l ) (l )
(1)
i ∈c j

(l )
where c j is a set of input feature graphs. blj is the bias, ylj the output of the convolution and wij the
convolution kernel.
(l )
tlj = f (y j ) (2)

tlj the feature graph of the convolution layer l.

f ( x ) = max (0, x ) (3)

f is known as the activation function. In this work we use a rectified linear unit (Relu) defined in
Equation (3).

3.2. Temporal Feature Learning with LSTM

The lower layers of the proposed model are the LSTM. This layers store time information about
important characteristics of PV power generation extracted by the CNN. LSTM preserves long-term
Energies 2020, 13, 6512 6 of 15

memory by using memory units that can update the previous hidden state. This functionality makes it
possible to understand temporal relationships on a long-term sequence. This time recurrent neural
network was proposed by Hochreiter & Schmidhuber in 1997 [30]. Its internal memory unit and gate
mechanism overcome the gradient disappearance problem that occurs in training traditional recurrent
neural networks (RNN). The memory channel and the gate mechanism (which includes: forget gate,
input gate, update gate and output gate) are shown on Figure 7. The equations of the LSTM model are
the following [24]:
f t = σ (W f [ h t − 1 , X t ] + b f ) (4)

f t is the output value of the forget gate. And σ refers to the sigmoid activation function.

it = σ (Wi [ht−1 , Xt ] + bi ) (5)

it is the output value of the input gate.

gt = tanh(Wg [ht−1 , Xt ] + bg ) (6)

gt is the output value of the update gate.

c t = f t ∗ c t −1 + i t ∗ g t (7)

ct refers to the memory cell.

ot = σ(Wo [ht−1 , Xt ] + bo ) (8)

ot is the output value of the output gate.

ht = ot ∗ tanh(ct ) (9)

where ht is the output vector result of the memory cell at time t (see Figure 2). W f ,i,g,o are the weights
matrices and b f ,i,g,o the bias vectors.

Figure 7. LSTM cell.

3.3. CNN-LSTM Hybrid Model

The proposed CNN-LSTM structure as it is shown in Figure 8. consists of a five CNN and
LSTM layers and a full connected layer. The input of the neural network has ten variables, such as
temperature, irradiance, global radiation and others (see the section Dataset for more information
about the variables).
Energies 2020, 13, 6512 7 of 15

First, the upper layer of CNN-LSTM consists of CNN. The CNN layer can receive various variables
that affect PV power generated, such as voltage, intensity, and global radiation. On the other side,
the dataset is separated in two parts: 80% for training the model and 20% to validate the results. In the
validation part we used PVPG in Temixco during 2019 as shown on Figure 5.
The CNN consists of an input layer that accepts sensor variables as inputs, an output layer that
extracts features to LSTMs, and several hidden layers. The hidden layers used consists of: a convolution
layer, a ReLU layer,an activation function, or a pooling layer.
From the presented data in the input a unique output pattern is generated. The CNN extracts the
local features and the LSTM temporal part. With this structure, the neural network “learns” for every
input a weight that determines a specific output.

Figure 8. The CNN-LSTM general structure.

The CNN configuration in the hybrid model mainly consists of five convolutional layers and five
pooling layers. The number of convolution kernels is 64 ,128, 256, 128 and 64 respectively with a ReLU
activation function.
The LSTM model in the hybrid model contains five hidden layers, having respectively 64, 128,
256, 128 and 64 neurons. Finally, there is a dropout of 0.1 to prevent overfitting and finally two
fully-connected layers with 2048 and 1024 neurons respectively. For this experiment 1000 epochs were
selected. The specific parameter settings of the model are shown in the following Table 1.
All the hyper-parameters were designed by the trial and error method. This process means that
certain hyper-parameters are specified, then trained with 80% of the data and validated with other
20%, for the four seasons throughout the year. Hyper-parameters are then changed and the process is
repeated until they are finally optimized.
In order to demonstrate the good performance of the proposed model, the results obtained
by our model are compared with other models (LSTM and 2D CNN-LSTM) and two well known
benchmarks [31]. We will forecast PVPG for three different scenarios: summer, fall and winter.
These results will be further discussed in the last sections.
Energies 2020, 13, 6512 8 of 15

Table 1. Layer configuration of the proposed model.

Layer Configuration Activation Function

Convolution 1 filters = 64; kernel size = 3 ReLU loss function = mse
Max-pooling 1 kernel size = 2; stride = 2
Convolution 2 filters = 128; kernel size = 3 ReLU optimizer = adam
Max-pooling 2 kernel size = 2; stride = 2
Convolution 3 filters = 256; kernel size = 3 ReLU batch size = 64
Max-pooling 3 kernel size = 2; stride = 2
Convolution 4 filters = 128; kernel size = 3 ReLU
Max-pooling 4 kernel size = 2; stride = 2
Convolution 5 filters = 64; kernel size = 3 ReLU
Max-pooling 5 kernel size = 2; stride = 2
LSTM 1 Units = 64 Tanh, Sigmoid
LSTM 2 Units = 128 Tanh, Sigmoid
LSTM 3 Units = 256 Tanh, Sigmoid
LSTM 4 Units = 128 Tanh, Sigmoid
LSTM 5 Units = 64 Tanh, Sigmoid
Dropout dropout = 0.1
Fully connected Neurons = 2048
Fully connected Neurons = 1024

4. Performance Evaluation Metrics

For this work, we selected some metrics to evaluate the model. They are: MSE (Mean Square
Error), RMSE (Root Mean Square Error) and MAE (Mean Absolute Error). These metrics are defined as
follows [25]:
n
1
MSE =
n ∑ (Yi − Ybi )2 (10)
i =1
s
n
1
RMSE =
n ∑ (Yi − Ybi )2 (11)
i =1

∑in=1 Yi − Y
bi
MAE = (12)
n
where: Yi is the real PVPG value, Y
bi the predicted value and n the number of Yi .
MSE, RMSE and MAE have been used as a standard statistical metric to measure models
performance. This is an easily computable quantity because is sample-dependent [32,33].

5. Results

5.1. Performance Comparison with Other Deep Learning Models

In this work we propose a five CNN layer with max-pooling, a five layer LSTM structure and
a fully connected layer, also known as dense layer. We used the framework Keras and Tensorflow
in Python 3.6 to program the models. The computational process was accomplished in a personal
computer, with a 64 bit operating system, 8 GB of RAM with an Intel Core processor i7-4700MQ
(2.4 GHz, 6 MB cache and 4 cores).
A part of the dataset was used to train the model (80 %) and another part was used to verify the
predictions (20 %). Figures 9–11 show the graphs of the PV power prediction during summer, fall
and winter respectively. On the X axis, we have time with a 10-min time resolution and on the Y axis,
power in [W]. The blue curve are the actual measurements and the yellow curve is the prediction of
the five-layer hybrid model, the orange curve corresponds to the two-layer model and the green curve
is a five-layer LSTM model.
Energies 2020, 13, 6512 9 of 15

The metrics that evaluate the performance of the different models are shown in Table 2. In Table 3,
the computation time processing for each model is displayed (Time window prediction is of 10 min for
this experiment).

Figure 9. Prediction of PV generation during summer.

Figure 10. Prediction of PV generation during fall.

Figure 11. Prediction of PV generation during winter.

As it is shown in Figures 10–12, if we compare the single LSTM and the 2D CNN-LSTM models
with the five layer CNN-LSTM model we can observe that the five layer model can predict more
efficiently irregularities in Figure 13 is accomplished with a strong 5D CNN-LSTM network.
Energies 2020, 13, 6512 10 of 15

Figure 12. Prediction of PV generation during rainy days.

Table 2. Performance comparison of the proposed model with other deep learning models.

Model MSE RMSE MAE

5D-LSTM 0.0854788623 0.2923676835 0.1666727318
2D CNN-LSTM 0.059041463 0.242984493 0.1410550836
5D CNN-LSTM 0.006897012 0.083048253 0.0519295306

It is observed that the lowest MSE (0.0068) corresponds to the most robust five-layer model (5D
CNN-LSTM) This model also has the biggest time processing of almost 70 s.
On the other hand, the 5D-LSTM model and the 2D CNN-LSTM model obtained a very similar
RMSE (0.2923 and 0.2429 respectively). The same with the MAE, 0.1666 for the LSTM model, 0.1410 for
the 2D CNN-LSTM model and 0.0519 for the 5D CNN-LSTM model.

Table 3. Time processing.

Model Time [s]

5D-LSTM 9.1394
2D CNN-LSTM 8.0362
5D CNN-LSTM 69.1148

5.2. Performance Comparison with Competitive Benchmarks

We compare the performance of the proposed method with competitive benchmarks using
Temixco ESOLMET dataset [34]. Table 4 summarizes the performance of competitive benchmarks
(Time window prediction is of 10 min for this experiment.).
Yang et al. refer to benchmarking, a essential practice in solar prediction for comparing published
results [35]; some authors like Pedro et al., used some well known benchmarks based on linear
regression: Lasso and Ridge regression [31]. We select these two models for benchmarking due to their
high interpretability and availability in statistical software and in Python.
Furthermore, we compare competitive benchmarks with the proposed model, the results are
shown in Table 4. The forecasting horizons range from 10-min ahead to 180-min ahead. As it is
expected, when the horizon prediction increases the RMSE also does it in a linear way (Figure 13).
It can be seen that the proposed model can provide good prediction performance. Moreover, at all
prediction horizons in the study cases, the proposed method has the smallest RMSE metrics, which
shows the best prediction performance compared to other competitive benchmarks.
Energies 2020, 13, 6512 11 of 15

Table 4. Performance comparison of the proposed model with competitive benchmarks.

Model MSE RMSE MAE

Ridge 0.689931245 0.830620999 5.270156623
Lasso 0.626782312 0.791695845 4.928345520
5D CNN-LSTM 0.006897012 0.083048251 0.0519295306
Prediction horizon = 10 min.
Ridge 1.397775392 1.182275514 10.24442325
Lasso 1.288235656 1.135004694 9.10778134
5D CNN-LSTM 0.050930126 0.225677041 0.589563061
Prediction horizon = 30 min.
Ridge 3.155511992 1.776376084 23.83114626
Lasso 2.984087426 1.727451135 21.17831734
5D CNN-LSTM 0.21093012 0.459271299 1.7129381706
Prediction horizon = 60 min.
Ridge 4.597775392 2.144242381 39.22693581
Lasso 4.388235656 2.094811603 34.78419132
5D CNN-LSTM 0.531245673 0.728866018 4.691195306
Prediction horizon = 90 min.
Ridge 6.209805443 2.491948122 41.22371890
Lasso 6.008834922 2.451292500 40.10123278
5D CNN-LSTM 1.121145975 1.058841808 10.014495238
Prediction horizon = 120 min.
Ridge 8.158007231 2.856222545 50.532978262
Lasso 7.989883492 2.826638196 50.1027657134
5D CNN-LSTM 2.084572014 1.443804702 20.012299575
Prediction horizon = 150 min.
Ridge 11.01253333 3.318513723 61.82278104
Lasso 10.87665156 3.297976888 60.16298833
5D CNN-LSTM 4.231144792 2.056974669 29.059911306
Prediction horizon = 180 min.

Figure 13. RMSE statistics for various forecasting horizons.

Most of the related work [26,36–38], uses a single layer approach for PV power prediction.
Nevertheless, the results of this work show metrics like MSE, to be ten times lower for a multi-layer
CNN-LSTM deep neural architecture. This obviously comes with more processing time for training
the model, as it is shown in Table 3. However, this time is completely useful for all of the
Energies 2020, 13, 6512 12 of 15

prediction horizons presented, and can be dropped down if the model is trained with a GPU or
TPU processing unit.
Also results for different prediction horizons (from 10-min horizon to 180-min horizon) show the
RMSE of the proposed model to be lower than competitive benchmarks in all cases (see Table 4 and
Figure 13).

6. Conclusions
We use real data from the location 18◦ 500 24.100 N 99◦ 140 09.000 W (Temixco, Mexico) to train
and validate a 5D CNN-LSTM hydrid model. The dataset was composed by ten variables such as
temperature, solar radiation and others, for predicting PV power generation (PVPG). A five layer
CNN-LSTM hybrid model was compared with a single LSTM model and another CNN-LSTM hybrid
model of two layers. The error of a robust CNN-LSTM net (in this experiment with five layers) was:
MSE = 0.00689, as shown in Table 2. Therefore, we can say that there is a considerable correlation
between a robust neural network model with a highly accurate prediction effect.
We propose a CNN-LSTM model for precise prediction of PV power energy generation. The 5D
CNN-LSTM hybrid model accurately predicts PV power generation by extracting features from
variables that affect PVPG. The proposed model is compared to other machine learning methods to
demonstrate its usefulness. We used CNN-LSTM to learn trends in the PVPG and temporal information
from multivariate time-series.
The CNN-LSTM model proposed in this paper predicts irregularities in PV power generation
(PVPG) that could not be well learned by other existing machine learning models. In Figure 13 we see
how on a rainy day the 5D CNN-LSTM model (in green) is exact by predicting the irregular PVPG
trend in cloudy conditions.
The results of this paper represent that the 5D CNN-LSTM model predicts PVPG with high
accuracy, and shows the highest performance compared with the single LSTM model and the 2D
CNN-LSTM hybrid model. The results also shows the highest performance compared with two
existing competitive benchmarks.
Nevertheless, the computational time for a five layer hybrid model (69.1148 s) is bigger than
the two layer hybrid model (8.0362 s) or the single LSTM model (5.1394 s) as it is shown in Table 3.
A neural network prediction model should take in consideration that a high accuracy performance is
compromised by computational time processing.
On the other hand, the introduction of TPU’s (Tensor Processing Units) had include hardware
advantages with faster computational processing time [39], this experiments were conducted in a CPU
Intel Core i7, but robust hybrid neural networks models like the one presented in this work, should be
studied in new generation hardware like the Coral TPU accelerator introduced by Google [40], in order
to reduce time processing and increase performance.
Also, further studies should be done to implement dynamic prediction systems (generation and
consumption in households) with data available from an IoT network of devices in low cost and low
power in open-source platforms. Thus, aiming to create tools for a smart grid and efficient trading
energy management systems.

7. Future Work
The proposed model will be integrated in a Raspberry Pi 3 with a Coral TPU accelerator to have
a real time data prediction system. The general scheme is shown in Figure 14. The ESP32 sends
measurements to an IoT server (Thingsboard). This real time data is also available for the Raspberry Pi
to calculate new weights and update the hybrid model. The Coral TPU will provide the Raspberry Pi
the capability to run machine learning models. This kind of low cost system will be possible to use in
multiple prosumers in a micro-grid environment.
Energies 2020, 13, 6512 13 of 15

Figure 14. Prediction system with Raspberry Pi and Coral TPU accelerator.

Accurate PV power and electric energy consumption predictions are essential for energy
management systems in energy transaction architectures [8,41–43], where energy becomes a medium
of exchange. A low power prediction processing unit such as a Raspberry can increase the performance
of SHS’s in rural areas by introducing a trading energy prediction mechanism but more studies should
be done in this area.

Author Contributions: Conceptualization, M.R. and M.T.; methodology, M.R.; software, M.T. and F.R.; validation,
M.T.; formal analysis, M.T.; investigation, M.T.; resources, M.R.; data curation, M.T.; writing—original draft
preparation, M.T.; writing—review and editing, M.T.; visualization, M.T.; supervision, M.R.; project administration,
M.R.; funding acquisition, M.R. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Acknowledgments: This work was supported by the National Council of Science and Technology of México
(CONACYT) and the equipment of the Renewable Energy Institute of the National University of Mexico
(IER—UNAM).
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

RE Renewable energy
CNN Convolutional neural network
1D-CNN One dimension convolutional neural network
5D-CNN Five dimension convolutional neural network
LSTM Long short term memory
ReLU Rectified Liniar Unit
PV Photo voltaic
PVPG Photo voltaic power generated
PVSS Photo voltaic solar system
SHS Solar home system
JRC Joint Research Centre
ANN Artificial neural network
RNN Recurrent neural network
DRN Deep belief network
EMS Energy management system
SVR Support Vector Regression
ECG Electrocardiogram
UNAM Universidad Nacional Autónoma de México
IER Instituto de Energías Renovables
TPU Tensor Processing Unit
Energies 2020, 13, 6512 14 of 15

References
1. Data, G. Solar Photovoltaic (PV) Market, Update 2019—Global Market Size, Market Share, Average Price,
Regulations, and Key Country Analysis to 2030; GlobalData: London, UK, 2019.
2. Jaeger-Waldau, A. PV Status Report 2019, EUR 29938 EN; Publications Office of the European Union:
Luxembourg, 2019.
3. Sobri, S.; Koohi-kamali, S.; Rahim, N.A. Solar photovoltaic generation forecasting methods: A review
Number of Day. Energy Convers. Manag. 2020, 156, 459–497. [CrossRef]
4. Soltowski, B.; Campos-gaona, D.; Strachan, S.; Anaya-lara, O. Bottom-Up Electrification Introducing New
Smart Grids Architecture—Concept Based on Feasibility Studies Conducted in Rwanda. Energies 2019, 12, 2439.
[CrossRef]
5. Inegi; Sener, C. Encuesta Nacional Sobre Consumo de Energéticos en Viviendas Particulares. ENCEVI,
2018 Consulted on June 2020. Available online: https://www.inegi.org.mx/contenidos/programas/encevi/
2018/doc/encevi2018_presentacion_resultados.pdf (accessed on 10 June 2020).
6. Bahaj, A.; Blunden, L.; Kanani, C.; James, P.; Kiva, I.; Matthews, Z.; Price, H.; Essendi, H.; Falkingham, J.;
George, G. The Impact of an Electrical Mini-grid on the Development of a Rural Community in Kenya.
Energies 2019, 12, 778. [CrossRef]
7. Khorasany, M.; Donald Azuatalam, R.G.A.L.; Razzaghi, R. Transactive Energy Market for Energy
Management in Microgrids: The Monash Microgrid Case Study. Energies 2020, 13, 2010. [CrossRef]
8. Mengelkamp, E.; Kessler, S.; Gärttner, J.; Rock, K.; Orsini, L.; Weinhardt, C. Designing microgrid energy
markets. Appl. Energy 2017, 210, 870–880. [CrossRef]
9. El-baz, W.; Tzscheutschler, P.; Wagner, U. Integration of energy markets in microgrids: A double-sided
auction with device-oriented bidding strategies. Appl. Energy 2019, 241, 625–639. [CrossRef]
10. Bahrami, S.; Amini, M.H. A Decentralized Framework for Real-Time Energy Trading in Distribution
Networks with Load and Generation Uncertainty. arXiv 2017, arXiv:1705.02575v1.
11. Blaga, R.; Sabadus, A.; Stefu, N.; Dughir, C.; Paulescu, M. A current perspective on the accuracy of incoming
solar energy forecasting. Prog. Energy Combust. Sci. 2019, 70, 119–144. [CrossRef]
12. Haji, R.; Zargar, M.; Yaghmaee, M.H.; Member, S. Development of a Markov-Chain-Based Solar Generation
Model for Smart Microgrid Energy Management System IEEE Trans. Sustain. Energy 2020, 11, 736–745.
13. Gao, M.; Wang, K.; Chain, A.C.T.M. Probabilistic Model Checking for Green Energy Router System in
Energy Internet. In Proceedings of the GLOBECOM 2017—2017 IEEE Global Communications Conference,
Singapore, 4–8 December 2017.
14. Voyant, C.; Notton, G.; Kalogirou, S.; Nivet, M.L.; Paoli, C.; Motte, F.; Fouilloy, A. Machine learning methods
for solar radiation forecasting: A review. Renew. Energy 2017, 105, 569–582. [CrossRef]
15. Chandler, S.A.; Hughes, J.G. Smart Grid Distribution Prediction and Control Using Computational Intelligence.
In Proceedings of the 2013 1st IEEE Conference on Technologies for Sustainability (SusTech), Portland, OR, USA,
1–2 August 2013; pp. 86–89.
16. Gómez-Gil, P.; Ramírez-Cortes, J.M.; Pomares Hernández, S.E.; Alarcón-Aquino, V. A neural network scheme
for long-term forecasting of chaotic time series. Neural Process. Lett. 2011, 33, 215–233. [CrossRef]
17. Abdel-Nasser, M.; Mahmoud, K. Accurate photovoltaic power forecasting models using deep LSTM-RNN.
Neural Comput. Appl. 2019, 31, 2727–2740. [CrossRef]
18. Ba, Ü.; Filik, T. Wind Speed Prediction Using Artificial Neural Networks Based on Multiple Local Measurements
in Eskisehir. Energy Procedia 2017, 107, 264–269.
19. Le, T.; Vo, M.T.; Vo, B.; Hwang, E.; Rho, S. Improving Electric Energy Consumption Prediction. Appl. Sci.
2019, 9, 4237. [CrossRef]
20. Muralitharan, K.; Sakthivel, R.; Vishnuvarthan, R. Neural Network based Optimization Approach for Energy
Demand Prediction in Smart Grid AC PT US CR. Neurocomputing 2017, doi:10.1016/j.neucom.2017.08.017.
[CrossRef]
21. Li, Y.; Yu, R.; Shahabi, C.; Liu, Y. Diffusion convolutional recurrent neural network: Data-driven traffic
forecasting. In Proceedings of the 6th International Conference on Learning Representations, ICLR
2018—Conference Track Proceedings, Vancouver, BC, Canada, 30 April–3 May 2018; pp. 1–16.
22. Al-Dahidi, S.; Ayadi, O.; Adeeb, J.; Alrbai, M.; Qawasmeh, B.R. Extreme Learning Machines for Solar
Photovoltaic Power Predictions. Energies 2018, 11, 2725. [CrossRef]
Energies 2020, 13, 6512 15 of 15

23. Tang, Q.; Yang, M. ST-LSTM: A Deep Learning Approach Combined Spatio-Temporal Features for Short-Term
Forecast in Rail Transit. J. Adv. Transp. 2019, 2019, 8392592. [CrossRef]
24. Madondo, M.; Gibbons, T. Learning and Modeling Chaos Using LSTM Recurrent Neural Networks.
In Proceedings of the Midwest Instruction and Computing Symposium, Duluth, Minnesota, 6–7 April 2018.
25. Kim, T.Y.; Cho, S.B. Predicting residential energy consumption using CNN-LSTM neural networks. Energy
2019, 182, 72–81. [CrossRef]
26. Wang, K.; Qi, X.; Liu, H. Photovoltaic power forecasting based LSTM-Convolutional Network. Energy
2019, 189, 116225. [CrossRef]
27. Alex, G.; Schmidhuber, J. Offline Handwriting Recognition with Multidimensional Recurrent Neural
Networks. In Proceedings of the NIPS Proceedings 2008, Vancouver, BC, Canada, 8–11 December 2008.
28. Sainath, T.N.; Vinyals, O.; Senior, A.; Sak, H. Convolutional, long short-term memory, fully connected deep
neural networks. In Proceedings of the 2015 IEEE International Conference on Acoustics, Speech and Signal
Processing (ICASSP), Brisbane, Australia, 19–24 April 2015.
29. Solarimetric and Weather Station (Esolmet). National Autonomous University of Mexico Institute of Renewable
Energies. 2020. Available online: http://esolmet.ier.unam.mx/ESOLMET-IER.pdf (accessed on 10 June 2020).
30. Hochreiter, S.; Schmidhuber, J. Long Short-Term Memory. Neural Comput. 1997, 9, 1735–1780. [CrossRef]
31. Pedro, H.T.C.; Larson, D.P.; Coimbra, C.F. A comprehensive dataset for the accelerated development and
benchmarking of solar forecasting methods. J. Renew. Sustain. Energy 2019, 11, 036102. [CrossRef]
32. Donahue, J.; Hendricks, L.A.; Guadarrama, S.; Rohrbach, M.; Venugopalan, S.; Saenko, K.; Darrell, T.;
Austin, U.T.; Lowell, U.; Berkeley, U.C. Long-term Recurrent Convolutional Networks for Visual Recognition
and Description. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 39, 677–691. [CrossRef] [PubMed]
33. Santos Navarrete, M. Analysis and Prediction of Electricity Consumption in the Balearic Islands.
Bachelor’s Thesis, Polytechnic University of Madrid, Madrid, Spain, 2016.
34. UNAM, M.A.T.R. ESOLMET2019. Github. 2020. Available online: https://github.com/mariotovarrosas/
ESOLMET2019 (accessed on 10 June 2020).
35. Yang, D. A guideline to solar forecasting research practice: Reproducible, operational, probabilistic or
physically-based, ensemble, and skill (ROPES). J. Renew. Sustain. Energy 2019, 11, 022701. [CrossRef]
36. Al-Messabi, N.; Yun Li.; El-Amin, I.; Goh, C. Forecasting of photovoltaic power yield using dynamic neural
networks. In Proceedings of the 2012 International Joint Conference on Neural Networks (IJCNN), Brisbane,
Australia, 10–15 June 2012.
37. Li, G.; Xie, S.; Wang, B.; Xin, J.; Li, Y.; Du, S. Photovoltaic Power Forecasting With a Hybrid Deep Learning
Approach. IEEE Access 2020, 8, 175871–175880. [CrossRef]
38. Mordjaoui, M.; Haddad, S.; Medoued, A.; Laouafi, A. Electric load forecasting by using dynamic neural
network. Int. J. Hydrog. Energy 2017, 42, 17655–17663. [CrossRef]
39. Widergren, S. A Society of Devices. IEEE 2016, 14, 34–45.
40. Google. Coral USB Accelerator Consulted on August 2020. Available online: https://coral.ai/products/
accelerator/ (accessed on 20 August 2020).
41. Haghifam, S.; Zare, K.; Abapour, M.; Muñoz-delgado, G. A Stackelberg Game-Based Approach for
Transactive Energy Management in Smart Distribution Networks. Energies 2020, 13, 3621. [CrossRef]
42. Hammerstrom, D.J.; Ngo, H. A Transactive Network Template for Decentralized Coordination of Electricity.
In Proceedings of the 2019 IEEE PES Transactive Energy Systems Conference (TESC), Minneapolis, MN,
USA, 8–11 July 2019; pp. 1–5.
43. Ozmen, O.; Nutaro, J.; Starke, M.; Roberts, L.; Kou, X.; Im, P.; Dong, J.; Li, F.; Kuruganti, T.; Zandi, H.
Power Grid Simulation Testbed for Transactive Energy Management Systems. Sustainability 2020, 12, 4402.
[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/).