sustainability

Article
Research on Peak Load Prediction of Distribution Network
Lines Based on Prophet-LSTM Model
Zhoufan Chen 1,2 , Congmin Wang 3 , Longjin Lv 4 , Liangzhong Fan 2 , Shiting Wen 2, * and Zhengtao Xiang 1, *

1 School of Electrical and Information Engineering, Hubei University of Automotive Technology,

Shiyan 442000, China; [email protected]
2 Shool of Computer and Data Engineering, NingboTech University, Ningbo 315000, China
3 State Grid Zhejiang Electric Power Co., Ltd., Ningbo Power Supply Company, Ningbo 315000, China
4 School of Finance and Information, Ningbo University of Finance and Economics, Ningbo 315000, China;
[email protected]
* Correspondence:[email protected] (S.W.); [email protected] (Z.X.)

Abstract: The increasing demand for precise load forecasting for distribution networks has become
a crucial requirement due to the continual surge in power consumption. Accurate forecasting of
peak loads for distribution networks is paramount to ensure that power grids operate smoothly
and to optimize their configuration. Many load forecasting methods do not meet the requirements
for accurate data and trend fitting. To address these issues, this paper presents a novel forecasting
model called Prophet-LSTM, which combines the strengths of the Prophet model’s high trend fitting
and LSTM model’s high prediction accuracy, resulting in improved accuracy and effectiveness of
peak load forecasting. The proposed algorithm models the distribution network peak load using the
Prophet-LSTM algorithm. The researchers then analyzed the experimental data and model of the
algorithm to evaluate its effectiveness. We found that the Prophet-LSTM algorithm outperformed
the Prophet and LSTM models individually in peak load prediction. We evaluate the proposed
model against commonly used forecasting models using MAE (mean absolute error) and RMSE
(root mean square error) as evaluation metrics. The results indicate that the proposed model has
better forecasting accuracy and stability. As a result, it can predict the peak load of distribution
Citation: Chen, Z.; Wang, C.; Lv, L.;
networks more accurately. In conclusion, this study offers a valuable contribution to load forecasting
Fan, L.; Wen, S.; Xiang, Z. Research for distribution networks. The proposed Prophet-LSTM algorithm provides a more precise and
on Peak Load Prediction of stable prediction, making it a promising approach for future applications in distribution network
Distribution Network Lines Based on load forecasting.
Prophet-LSTM Model. Sustainability
2023, 15, 11667. https://doi.org/ Keywords: distribution grid; Prophet-LSTM; load forecasting; time series analysis
10.3390/su151511667

Academic Editors: Nicu Bizon and

Mohamed A. Mohamed
1. Introduction
Received: 5 May 2023
With the development of power systems, distribution network load forecasting has
Revised: 9 July 2023
received more and more extensive attention as one of the important research areas of
Accepted: 26 July 2023
power grid management and dispatching [1]. The rapidly growing demand for electricity
Published: 28 July 2023
in urban areas poses a significant challenge to their power grids. The resulting increase
in peak loads makes it increasingly difficult to maintain a stable grid configuration and
operation [2,3]. Due to the lack of peak load monitoring distribution networks, overloads
Copyright: © 2023 by the authors. are a frequent concern in relatively small capacity urban distribution lines, which can lead to
Licensee MDPI, Basel, Switzerland. power outages, line losses, and component burnouts [4]. Accurate power load forecasting
This article is an open access article can therefore provide decision makers with important information to maintain grid line
distributed under the terms and stability and achieve optimal power dispatch [5]. Currently, power load forecasting is
conditions of the Creative Commons usually performed with historical data as the main component to forecast future load
Attribution (CC BY) license (https:// data. Load forecasting is commonly divided into short-term (daily and weekly), medium-
creativecommons.org/licenses/by/ term (monthly), and long-term (annually) categories depending on the forecast horizon [6].
4.0/).

Sustainability 2023, 15, 11667. https://doi.org/10.3390/su151511667 https://www.mdpi.com/journal/sustainability

Sustainability 2023, 15, 11667 2 of 16

This research paper aims to forecast peak distribution network line currents over a two-
month period, which can be classified as a medium-term load prediction. However,
given the duration of the prediction, the short-term load forecasting method can also
be used. Medium-term power load forecasting plays a critical role in power system
planning, operation, and management. Medium-term load forecasting can help power
system planning departments determine power demand and resource allocation, guide
decision makers to make effective decisions in energy construction, power generation
planning, renewable energy utilization, and grid upgrading, and can provide a reference
basis for power generation dispatch and grid operation [7,8]. The research scope of power
load forecasting is defined by a tree diagram (Figure 1).
8VXDOO\IRUHFDVWWKHSRZHUORDGLQWKHQH[WIHZKRXUVRUGD\V

6KRUWWHUP)RUHFDVW
$VVLVWVLQUHDOWLPHRSHUDWLRQRIWKHSRZHUV\VWHP
DGMXVWVSRZHUJHQHUDWLRQDQGWUDQVPLVVLRQDQGGLVWULEXWLRQLQUHDOWLPH
DYRLGVORDGLPEDODQFHLQWKHSRZHUV\VWHPDQGHQVXUHVVDIHDQGVWDEOHRSHUDWLRQRIWKH
JULG

8VXDOO\IRUHFDVWVWKHGHPDQGIRUHOHFWULFLW\ORDGVLQWKHFRPLQJGD\VWRPRQWKV

3RZHUORDGIRUHFDVWLQJ 0LGWHUP)RUHFDVW
+HOSVHOHFWULFXWLOLWLHVZLWKSRZHUGLVSDWFKDQGSODQQLQJRYHUORQJHUWLPHVFDOHV
7KHIRUHFDVWUHVXOWVFDQEHXVHGIRUXWLOLWLHV ORQJWHUPHQHUJ\SXUFKDVHDQGLQYHVWPHQW
GHFLVLRQVDQGIRUWKHDSSURSULDWHFRQILJXUDWLRQDQGH[SDQVLRQRIHTXLSPHQWDQGLQIUDVWUXFWXUH˗
VXSSRUWVWKHDGYDQFHPHQWRIVXVWDLQDEOHGHYHORSPHQWJRDOV

7\SLFDOO\IRUHFDVWVHOHFWULFLW\ORDGVRYHUVHYHUDOPRQWKVRU\HDUVZLWKDZLGHWLPHFRYHUDJH

/RQJ7HUP)RUHFDVW
+HOSVZLWKFRPSUHKHQVLYHSRZHUSODQQLQJ˗
/RQJWHUPSRZHUORDGIRUHFDVWVFDQEHXVHGDVDQLPSRUWDQWEDVLVIRUPDUNHWDQDO\VLV˗
/RQJWHUPSRZHUORDGIRUHFDVWLQJVXSSRUWVVXVWDLQDEOHGHYHORSPHQW

Figure 1. Tree diagram classification field.

Load forecasting is usually performed using traditional forecasting models and models
with artificial intelligence. Traditional forecasting models include autoregressive (AR),
mean shift (MA), and ARIMA models, which are simple to use and fast to compute.
However, it can only analyze the time-dependent relationships between the data, and, in
addition, the power load data have the characteristics of nonlinearity and periodicity, which
cannot deal with these problems well, and the prediction accuracy and effect using the
above methods are not good. Artificial intelligence methods including SVM, evolutionary
algorithms, RNN, LSTM, etc., are also widely used for load prediction, but there are
convergence and complexity problems. This paper will be discussed specifically in the
related work section.
This paper explores the prediction of peak loads on distribution network lines by
combining the Prophet and LSTM algorithms. The model is designed to handle various
influencing factors, such as daily, weekly, annual, and seasonal cycles, and holidays, and can
correct prediction errors through time series analysis to eliminate periodic and trend errors,
thus improving accuracy and stability. Through training and testing using experimental
data, the Prophet-LSTM algorithm’s effectiveness and accuracy in peak load prediction on
distribution network lines are verified.
The main contributions of this paper are as follows:
• In this paper, a Prophet-LSTM-based load forecasting method is proposed, learning
data trends based on the Prophet model to improve the data trend fit, while using the
high prediction accuracy of LSTM model for prediction, and further improving the
prediction results through the BP network to improve the prediction accuracy and
effectiveness of the model.
• In Prophet model training, we optimize the parameters of the Prophet model based
on PSO (particle swarm optimization algorithm), which can better capture the data
mutation points.
Sustainability 2023, 15, 11667 3 of 16

• A complete set of experiments was designed to compare the forecasting method

proposed in this paper with common forecasting methods, and to demonstrate that
the method proposed in this paper can achieve better forecasting results compared
with other methods.

2. Related Work
This paper categorizes load forecasting methods into three main categories: traditional
forecasting models [9], machine learning [10], and deep learning [11]. Each category is
presented in detail in the following sub-summaries.

2.1. Traditional Forecasting Models

Traditional forecasting methods are primarily based on historical time series data,
utilizing simple models that have a relatively quick calculation speed. For example, the
method based on regression analysis takes load data and influencing factors as independent
and dependent variables, and load forecasting is performed by establishing regression
equations. Among the traditional methods, time series methods are commonly used in
load forecasting, including autoregressive (AR) [12,13], moving average (MA) [14], autore-
gressive integrated moving average (ARIMA) [15,16], seasonal autoregressive integrated
moving average (SARIMA) [17] model, and autoregressive integrated moving average
model with external inputs (ARIMAX) [18] model. These methods all rely on past data
to forecast, with simple structures, without considering the correlation between data, and
only treating historical and predicted data as simple mathematical relationships, with
low forecasting accuracy. They are also unable to better handle nonsmooth, nonlinear
data. Compared with the above methods, the Prophet model can perform nonlinear data
processing and capture data trends, especially in power load time series forecasting.

2.2. Machine Learning

With the rapid development of machine learning, machine learning methods have been
widely applied to image processing, and certain achievements have been made in power
load forecasting. Common machine learning methods include support vector machines
(SVMs), random forests (RFs), and evolutionary algorithms (EAs). There is a lot of work
on SVMs for load forecasting that provides ideas for the accuracy of the forecast [19–21].
Pang et al. [22] used gray relational analysis and support vector machines to study short-
term electrical load forecasting and proved the validity of the method. Liu et al. [23]
presented a method for electric load forecasting using Elman neural network and par-
ticle swarm optimization algorithm, significantly improving load forecasting accuracy.
Chen et al. [24] proposed a new SVR model based on demand response, and the experi-
mental results showed that the proposed model has some generality. Barman et al. [25]
considered climatic conditions in their experiments and performed load forecasting based
on an SVR model and locust optimization algorithm, and the results showed a major
breakthrough in forecasting accuracy. Random forests, which integrate multiple decision
trees internally and, thus, constitute a classifier, have the advantages of fewer parame-
ters, high generalization ability, and high forecasting accuracy [26], and are also used by
some researchers in load forecasting. For example, Wu et al. [27] improved the RF model
and proposed a segmentation algorithm to improve the forecasting accuracy and robust-
ness. Although the above machine learning methods are effective in processing nonlinear
data, they have weak feature extraction capability and low accuracy in forecasting highly
random data.

2.3. Deep Learning

In recent years, deep learning has been gaining more and more weight in load forecast-
ing. Deep learning neural networks have more complex structural models, stronger learn-
ing ability, generalization ability, etc. [28–30]. In the literature [31], a long short-term mem-
ory (LSTM) network with more advantages than recursive neural network (RNN) is used
Sustainability 2023, 15, 11667 4 of 16

for power load forecasting, which overcomes the problems such as the gradient explosion
of RNN and improves the accuracy of forecasting. In addition, Motepe et al. [32] proposed
a new hybrid artificial intelligence and deep learning system for forecasting the load on the
distribution network in South Africa. Li et al. [33] developed a combined forecasting model
that uses LSTM and XGBoost methods to improve the accuracy of electric load forecasting.
Pham et al. [34] developed a hybrid forecasting method integrating singular spectrum
analysis (SSA) and deep neural network techniques. Similarly, Ciechulski et al. [35] used
artificial neural networks and SSA techniques to investigate efficient and accurate methods
for 1 h and 24 h power load pattern forecasting. Deep-learning-based load forecasting
methods show good promise in improving grid intelligence [36].

2.4. Summary
We analyzed the current common load forecasting methods and related works, and
summarized the advantages and disadvantages of different forecasting methods through
Table 1. Since a single forecasting method cannot simultaneously meet the requirements of
high fitting of data trends and high accuracy of predicted data, we propose the Prophet-
LSTM model based on these advantages and disadvantages, which is able to fit the data
trend better and improve the accuracy at the same time. There are periodic trends in power
load data, and the Prophet algorithm is an effective method to fit these trends accurately
compared to other methods [37,38]. Parizad et al. proposed a comprehensive two-stage
STLF (short-term load forecasting) method based on the Prophet algorithm [39], which can
effectively forecast data trends. The paper [40] presents a Prophet-model-based electric
current load forecasting, which performs well in terms of forecasting accuracy and model
fit. Although not much work has been carried out using the Prophet model for power load
forecasting, we can infer from the above works that this approach is effective in dealing
with modeling and data fitting. Also, we add an LSTM model and a BP network to improve
the accuracy of load forecasting.

Table 1. Advantages and disadvantages of load forecasting methods.

Common Methods Advantages Disadvantages

Low forecasting accuracy; unable to
Traditional forecasting methods The model is simple and the calculation
handle nonsmooth and nonlinear data
(AR, MA, ARIMA) is fast.
better; poor fitting effect.
The feature extraction capability is weak
The models are effective in dealing with
Machine learning (SVM, RF, EA) and the accuracy of prediction for highly
nonlinear data.
random data is low.
The ability to cope with large-scale,
Deep learning algorithms have a large
high-dimensional, nonlinear load data
Deep learning (RNN, LSTM, GRU, TCN) computational resource footprint and
and to predict future load conditions
poor interpretation.
more accurately.

3. Research Theory and Methodology

3.1. Prophet Model
Facebook has released an open-source time series forecasting algorithm called the
Prophet algorithm. This algorithm is capable of managing multidimensional time series
data, supporting varying units of measurement while ensuring forecast accuracy without
sacrificing computational speed. It is widely used in business and scientific research for
its significant advantages in time series data forecasting and analysis. Prophet uses an
additive model to describe the nonlinear trend, observing periodicity and holiday effects,
which enables accurate forecast time series intuitive parameters. It is a decomposable
model consisting of a trend, seasonal, and holiday model. What sets Prophet apart from
other classical time series forecasting models is its periodic decomposition of the series
Sustainability 2023, 15, 11667 5 of 16

before forecasting. This approach highlights holiday effects and trends in the data, making
it particularly robust in the presence of missing data, sudden changes in trends, or outliers.
The Prophet model uses Equation (1) and a time series decomposition approach to
forecast time series data, dissecting the data into four essential components [41].

y(t) = g(t) + s(t) + h(t) + ε(t) (1)

Equation (1) requires g(t), which symbolizes the modification function employed to suit
the nonperiodic trend in the data. The change function can consist of various options, such
as the logistic regression function and segmented linear function. For instance, the trend
term built on logistic regression can be indicated by Equation (2):
C
g(t) = (2)
1 + e(−k(t−m))
Equation (2) includes C, which denotes the carrying capacity of the model, k, representing
the growth rate, and m, the data offset. As t varies, g(t) gradually approaches the upper
threshold C. Additionally, the segmented linear function is another alternative for the
change function, which can be defined by Equation (3):
 
g(t) = (k + a(t)δ) · t + m + a(t)T γ (3)

Equation (3) uses δ as a symbol for the change in growth rate over time, and m as the bias
term. The change point, represented by c j , refers to the time when the growth rate k shifts.
The algorithm specifies the location and number of change points. Furthermore, a(t) is
a time series produced by a binary function, and γ is another time series with the same
length as a(t).
In the Prophet model, s(t) represents the periodic term and is used to capture the
periodicity of the data, typically on a weekly, monthly, or yearly basis. The periodic term
s(t) can be fitted with a flexible Fourier series to model the periodic trend of the data. The
Fourier series is calculated as follows:
N     
2πnt 2πnt
s(t) = ∑ an cos + bn sin (4)
n =1
P P

In Equation (4) of Prophet, P represents the period of the time series, which is typically in
years. The parameter 2n represents the number of periods used in the algorithm.
In Prophet, h(t) represents the holiday component of the time series. In real-world
scenarios, major events or holidays can significantly impact load forecasting for distribution
networks, and their periodicity may not be straightforward to determine. To account for
this, the model includes a holiday variable h(t) that affects the time series in the period
before and after these events, and is expressed as follows:
L
h(t) = ∑ K i 1 ( t ∈ Di ) . (5)
i =1

Z (t) = [1(t ∈ D1 ) · · · 1(t ∈ DL )]. (6)

h(t) = Z (t)κ, κ ∼ Normal(0, ν). (7)
In Equations (5)–(7) of Prophet, Ki represents the impact of the time period before and after
a particular holiday, where i denotes the holiday and Di denotes the window period. The
value of the standard deviation ν determines the magnitude of the impact of the holiday
on the model. A larger value of ν indicates that the holiday has a greater impact, while a
smaller value indicates a smaller impact. The error term ε(t) follows a normal distribution.
Figure 2 illustrates the flow of the Prophet model loop.
Sustainability 2023, 15, 11667 6 of 16

0RGHOLQJ

$QDO\VWLQ
9LVXDOL]DWLRQ WKHORRS
)RUHFDVW
RISUHGLFWHG
(YDOXDWLRQ
UHVXOWV
$XWRPDWHG

6XUIDFH
3UREOHPV

Figure 2. Prophet model loop structure diagram.

The Prophet algorithm is primarily based on three time-varying factors: daily, weekly,
and seasonal variation. First, the time series data are segmented into multiple time periods,
and a separate linear regression equation is fitted within each period, capturing the trend
and cyclical patterns of different periods. Next, the algorithm incorporates seasonal and
holiday components into the model to account for the impact of events and important dates
on the time series. Finally, the predicted values produced by the model are used to forecast
future loads.

3.2. PSO Method

The particle swarm optimization (PSO) algorithm was developed by J. Kennedy and
R. C. Eberhart [42]. It is an evolutionary algorithm that takes inspiration from bird foraging.
PSO has been increasingly used in solving multidimensional nonlinear optimization prob-
lems owing to its benefits like easy implementation, fast convergence, and high accuracy.
The PSO algorithm considers solutions to an optimization problem as “particles”, and a
group of these particles is called a “particle swarm”. The PSO algorithm overcomes many
of the limitations of traditional optimization algorithms and is easy to implement. It also
avoids some of the problems present in traditional algorithms, such as the need to solve
for the derivative of the function and the risk of becoming stuck in a local optimum. The
algorithm is also widely used. During the iteration, the velocity v(i+1) j and position x(i+1) j
of each particle are updated for the next time by Equation (8).



v(i+1) j =wvij + c1 R1 Pj_ best − xij + c2 R2 Gbest − xij
(8)
x(i+1) j = xij + v(i+1) j

In Equation (8), j(1 6 j 6 n) represents the particle number; i (1 6 i 6 m) represents

the number of current iteration; n, m represent the total number of particles and the
total number of iterations, respectively; w is the inertia weight; c1 , c2 are the learning
factors; R1 , R2 are random numbers from 0 to 1. The particle population calculates the
fitness of each particle in this iteration according to the objective function Fij , from which
the individual extreme value Pj_ best and the population extreme value Gbest are obtained.
Finally, the operation is terminated according to the number of iterations, and the parameter
corresponding to the population extrema (Gbest ) is calculated as the optimal solution by the
objective function Fij .
Sustainability 2023, 15, 11667 7 of 16

3.3. LSTM Model

LSTM (long short-term memory) is a deep learning algorithm based on RNN designed
to capture and learn long-term dependent information. With excellent sequence modeling
capability, LSTM achieves promising results in prediction and speech recognition tasks. The
primary structure of LSTM includes a storage cell that stores and propagates information
explicitly at different time steps. The storage cell of LSTM leverages different cell states at
different instances to remember contextual information. In contrast to RNN, which faces
the problem of gradient disappearance with longer input sequences, LSTM introduces
techniques such as forgetting gates, input gates, and output gates to handle long-term
dependencies and resolve the issue of gradient vanishing/exploding. Its inherent ability to
store and process sequence information and update itself with continuous inputs makes it
an optimal choice for time series data prediction tasks. Figure 3 illustrates the addition of c,
a memory unit, to the recurrent layer of a traditional RNN, forming the core idea behind
the long- and short-term memory neural network. Figure 4 shows the LSTM cell structure

511RXWSXWOD\HU 2 2 /670RXWSXWOD\HU

&
511ORRSOD\HU /670ORRSOD\HU
K
K

511LQSXWOD\HU /670LQSXWOD\HU
; ;

Figure 3. LSTM structure diagram.

KW

F W  FW

WDQK

IW LW RW

6LJPRLG 6LJPRLG WDQK 6LJPRLG

KW
K W 

[W

Figure 4. LSTM cell structure.

The first step in LSTM is to decide which information should be forgotten from the cell
state by leveraging the Sigmoid function. The function determines which bits of information
should be discarded from the cell state, where values closer to 0 are forgotten, and those
nearer to 1 are retained.
 
f t = σ x t U f + h t −1 W f + b f (9)
Sustainability 2023, 15, 11667 8 of 16

The second step in LSTM is to determine which new pieces of information should be stored
in the cell state. The input gate decides whether it should be updated, and the data is
assigned a new vector with candidate values C̃t through the tanh function.
 
i t = σ x t U i + h t − 1 W i + bi . (10)

C̃t = tanh( xt U c + ht−1 W c + bc ). (11)

In the third step, the LSTM old cell state information is updated:
Ct = Ct−1 ⊗ f t ⊕ it ⊗ Ct (12)

Finally, in the output gate, the LSTM determines which information should be output. Here,
Ot decides which cell states should produce output, and the cell states are passed through
a tanh layer (with values ranging from −1 to 1) and multiplied by Ot at the output gate.

Ot = σ ( x t U o + h t −1 W o + bo ) . (13)

ht = Ot ⊗ tanh(Ct ). (14)

4. Model Construction and Prediction

Before training the model, it is important to prepare relevant data to test the model
and analyze the prediction results. In this paper, we gathered current data for a distribution
network line, specifically line current values collected once every 5 min, for one year. Then,
we constructed a dataset with current as the load and used the Prophet-LSTM algorithm
to build a prediction model. To develop the dataset, we extracted data for daily current
peaks over 368 days by preprocessing the historical load data, which included interpolating
and smoothing missing points. The dataset contained 368 sample points, with each point
comprising three indicators—maximum load value, minimum load value, and average
load value. This allowed us to explore the highest load value of the distribution network
lines—i.e., the peak load of the lines based on the actual demand. Before implementing the
Prophet-LSTM model, we tested the Prophet and LSTM models separately for comparison
using the dataset.

4.1. Prophet Model Construction

The Prophet model is well suited for fitting time series data, and it was used in this
study to model peak line load data and perform data prediction. The Prophet algorithm
requires two main parameters: “ds” column (date stamp), in YYYY-MM-DD format, and
the “y” column containing the variable value. In this study, the Prophet algorithm was
used to predict the peak load of distribution network lines. To construct the Prophet model
and handle the experimental data characteristics, the following steps were taken: data
preprocessing—raw data were screened and cleaned based on maximum load values to
remove invalid sample points and outliers. Prophet model construction—the data were
loaded into the Prophet model, including time and load values, and the model was built
and executed. Model training and parameter optimization—the Prophet model was trained
using historical data, and model parameters were adaptively optimized based on experi-
mental results. Prediction results were then output, and the Prophet algorithm generated a
graph of prediction results, showing the Prophet model’s fitting and trend curve.
We optimize the Prophet model parameters using PSO. The method for optimizing
Prophet parameters using PSO is as follows: (1) Partitioning the electric load dataset into a
training set and a test set: set the parameters of the Prophet model to be optimized; initialize
the hyperparameters such as the total number of particles n, the maximum number of
iterations m, the learning factors c1 and c2 , and the inertia weights w in the PSO method.
(2) At each iteration, each particle in the swarm is used as the parameter value to be
optimized in the Prophet model: Prophet models on the training set and predictions on the
test set; the root mean square error (RMSE) of the predicted values on the test set is used as
the target value for each particle iteration in the particle swarm; compute the current Pj_ best
Sustainability 2023, 15, 11667 9 of 16

and Gbest by RMSE. (3) After the last iteration is completed, the particle corresponding
to Gbest is used as the optimal solution of the Prophet model for the parameters to be
optimized.The flow chart for optimizing the parameters of Prophet model based on the
PSO method is shown in Figure 5.

(OHFWULFLW\ORDG
GDWDVHWGLYLVLRQ

3DUDPHWHUVRIWKH3URSKHWPRGHOWREH
RSWLPL]HG˗6HWWKHK\SHUSDUDPHWHUVRI
WKH362DOJRULWKP

,QLWLDOL]HSDUWLFOH
SRVLWLRQDQGYHORFLW\

%XLOG3URSKHWPRGHOEDVHGRQWUDLQLQJVHW

8SGDWHSDUWLFOHSRVLWLRQ
7HVWVHWSUHGLFWLRQ
DQGYHORFLW\

'HWHUPLQLQJLQGLYLGXDODQG
JURXSH[WUHPHV

-XGJPHQW
LWHUDWLRQ
FRQGLWLRQV

2XWSXWJURXS
H[WUHPHV

Figure 5. PSO optimized Prophet parameter flow.

According to the selection range of the number of populations n in the optimization

algorithm (100 6 n 6 300), n = 300 is set in this paper. It is found that the number of
iterations has a large impact on the experimental running time, so the number of iterations
for each particle is taken as 20 and the inertia weight w is taken as 0.8; the learning factors
c1 and c2 are both set as 0.5. The PSO method parameters are shown in Table 2. We adjust
the trend parameters and seasonal parameters of Prophet model to minimize RMSE as
the objective function, select key parameters, optimize the parameters by particle swarm
optimization algorithm (PSO), set changepoint_range = 0.37, and adjust the model fitting
flexibility changpoint_prior_scale = 0.67. The parameters after PSO optimization are given
in Table 3. The change of the objective function (RMSE) value optimized by the PSO method
is shown in Figure 6, where the RMSE value becomes smaller and smaller by iterative
optimization of the PSO method (the Prophet model optimized by the PSO method is still
essentially the Prophet model, hereafter referred to as Prophet model).
Sustainability 2023, 15, 11667 10 of 16

Table 2. PSO method parameters.

Parameters Description
n = 300 Number of populations
m = 20 Number of iterations
w = 0.8 Inertia weights
c1 = c2 = 0.5 Learning factors

Table 3. Prophet section parameter settings.

Parameters Description
growth = linear Set the growth model; this paper is set to linear model (linear).
changepoint_prior_scale = 0.67 Flexibility of the growth trend model.
holidays = holidays Set special dates and holidays.
weekly_seasonality = True Analyze weekly seasonality of data.
n_changepoints = 25 Set the number of potential variables.
The location of the change point needs to be set in a time series
changepoint_range = 0.37
as long as the first changepoint_range.

Figure 6. Fitness change curve.

Figure 7 illustrates the model’s predictive performance on the past data and the
following two months or so. The y-axis represents the current value in amperes (A), while
the x-axis indicates time. The dark blue line depicts the model predicted values (yhat),
whereas the shaded blue area represents the upper and lower limits of the predicted values,
and the black dots are the true data. As observed, the peak load during January to February,
coinciding with the Spring Festival holiday, is lower, while during July, it is higher, which
indicates the seasonal trend of the data. The model data mostly lie within the prediction
confidence interval, thus fitting the current trend and allowing accurate forecasting.

Figure 7. General trend of line peak forecast.

Sustainability 2023, 15, 11667 11 of 16

The graph in Figure 8 compares the model’s predicted values with the real values.
Here, “actual” represents the real value, and “predict” represents the predicted value. It
can be observed that the real value exhibits a cyclical trend, which is estimated to be about
two weeks by the curve. The model can better capture these apparent changes in the early
stage and make timely adjustments; however, it is not capable of making good changes in
the later stage due to the long prediction window.

Figure 8. Prophet comparison of actual data and forecast data.

This study utilized the Prophet algorithm to predict the maximum load of distribution
network lines, with a training-to-testing data ratio of 8:2. The effectiveness and accuracy
of this approach were further demonstrated by analyzing and predicting experimental
data. Although the Prophet model optimized using PSO performs well in fitting the
data, it suffers from the problem of low accuracy, and we add the LSTM model to solve
this problem.

4.2. LSTM Model Construction

The collected point-to-point load data were first analyzed and processed into time se-
ries data that conformed to the input format required by the LSTM model. Subsequently, the
Keras framework was employed to build the LSTM model, modifying the model structure
and adjusting the hyperparameters to enhance the prediction accuracy of the model.
The prediction of load data by the LSTM model is not only influenced by historical data
but also by neighboring data. The window size (w) affects the model input, whereas the
level factor (h) affects the model output (indicating the distance for predictions). Figure 9
displays the load dataset (dataset), training results (train), and test results (test). Consider-
ing the periodic pattern of the data and the length of the time series, the window size was 7
and the level factor was 1, and the update gradient was based on Adam optimizer. Finally,
a load prediction experiment was conducted using the trained model, with training and
test data at a ratio of 8:2. By calculation, we find that the accuracy of the load prediction
data is higher in the training and test sets. We combine the advantages of the Prophet
model and the LSTM model to build the Prophet-LSTM model.

Figure 9. LSTM comparison of actual data and forecast data.

Sustainability 2023, 15, 11667 12 of 16

4.3. Prophet-LSTM Combined Model

In this paper, we propose an optimal combined prediction model based on the Prophet
model and LSTM neural network for time series prediction. The predicted values of these
two models are used as input to a BP neural network to reduce errors and increase accuracy
gradually. The processing flow is shown in Figure 10. We divide the electric load dataset
into training sets, and determine the Prophet model parameters and LSTM model weights
through training set training. The prediction is performed by the trained Prophet model
and LSTM model, and the prediction results of the two models are fed into the BP neural
network as feature values to obtain more accurate prediction results.

'LVWULEXWLRQ
QHWZRUNORDGGDWD

([WUDFWLRQRI
SHDNGDWD

%XLOGLQJWKH
GDWDVHW

'LYLGLQJWKHWUDLQLQJVHWDQG
WHVWVHW

7UDLQLQJVHW

3URSKHWPRGHO /670PRGHO

2XWSXWSUHGLFWHG 2XWSXWSUHGLFWHG
YDOXH YDOXH

%311

)LQDO
SUHGLFWHG
YDOXH

Figure 10. Prophet-LSTM combined model flow chart.

As shown in Figures 11 and 12, we compared the prediction results of six models with
the true values. We observed that the predicted values of the Prophet-LSTM model were
closer to the true values compared to the other models. This indicates that the Prophet-
LSTM model had higher prediction accuracy and its predicted values were closer to the
true curve.
In order to test the accuracy of the model’s predictions, this paper uses root mean
square error (RMSE) and mean absolute error (MAE).
!
1 n
n i∑
∗ ∗
MAE(y, y ) = |y − y | (15)
=1
v
u !2
u1 n
∗
RMSE(y, y ) = t
n ∑ (y − y∗ ) (16)
i =1
Sustainability 2023, 15, 11667 13 of 16

Based on these two metrics, the metrics for Prophet, LSTM, GRU [43], TCN [44], Trans-
former [45], and Prophet-LSTM were computed and are presented in Table 4.

Figure 11. Comparison of prediction curves of different models.

Figure 12. Overall comparison.

Table 4. Comparison of metrics of different models.

Model MAE RMSE

Prophet 15.74 19.26
LSTM 12.32 15.55
GRU 11.67 15.45
TCN 12.63 17.57
Trans f ormer 12.73 18.44
Prophet-LSTM 8.569 11.68
Sustainability 2023, 15, 11667 14 of 16

The statistical indicators in Table 4 of this paper show that the proposed Prophet-LSTM
model outperforms the Prophet, LSTM, GRU, TCN, and Transformer alone. These metrics
show that the Prophet-LSTM model has the smallest mean absolute error (MAE) and root
mean square error (RMSE) values, which further supports the conclusion that the proposed
model is more accurate in predicting the target variable.

5. Conclusions
This paper proposes a novel hybrid prediction model, namely, Prophet-LSTM, for
forecasting the peak load of distribution network lines.The performance of the proposed
model is verified by power load data and the test results show that the Prophet-LSTM model
has a lower MAE value (8.569) and RMSE value (11.68) compared to the single Prophet,
LSTM, GRU, TCN, and Transformer models. This model has significant practical value for
load prediction, which can improve the operational efficiency and economic efficiency of the
power system, reduce the cost of electricity consumption and load volatility, and guarantee
the stability and reliability of power supply. Although there are many load forecasting
methods, the existing analysis mainly focuses on the time series data themselves with a
single predictive factor. Therefore, future research in this area should focus on adding more
influencing factors, such as calendar, weather information, etc., and optimizing the models
to obtain more accurate prediction results.

Author Contributions: Conceptualization, S.W. and Z.X.; methodology, Z.C.; data collection, L.L.;
software, Z.C.; investigation, Z.C.; resources, C.W.; writing—original draft preparation, Z.C. and L.L.;
writing—review and editing, Z.C.; project administration, L.F.; funding acquisition, Z.C., C.W., L.L.,
L.F., Z.X. and S.W. All authors have read and agreed to the published version of the manuscript.
Funding: This work is supported by Ningbo Science and Technology Special Innovation Projects
with Grant Nos. 2022Z095 and 2022Z235.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Grabner, M.; Wang, Y.; Wen, Q.; Blažič, B.; Štruc, V. A Global Modeling Framework for Load Forecasting in Distribution Networks.
IEEE Trans. Smart Grid 2023, 14, 1476–1484. [CrossRef]
2. Chen, H.; Zhang, B.; Geng, H.; Wang, M.-M.; Gao, H. Demand Response during the Peak Load Period in China: Potentials,
Benefits and Implementation Mechanism Designs. Comput. Ind. Eng. 2022, 168, 108117. [CrossRef]
3. Chen, H.; Yan, H.; Gong, K.; Yuan, X.-C. How will Climate Change Affect the Peak Electricity Load? Evidence from China.
J. Clean. Prod. 2021, 322, 129080. [CrossRef]
4. Zhang, M.; Sun, Q.; Yang, X. Research on the Assessment of the Capacity of Urban Distribution Networks to Accept Electric
Vehicles Based on the Improved TOPSIS Method. IET Gener. Transm. Distrib. 2021, 15, 2804–2818 . [CrossRef]
5. Lanka, V.V.S.; Roy, M.; Suman, S.; Prajapati, S. Renewable Energy and Demand Forecasting in an Integrated Smart Grid. In
Proceedings of the 2021 Innovations in Energy Management and Renewable Resources (IEMRE), Online Conference, 25–26 March
2021; pp. 1–6.
6. Panda, S.K.; Jagadev, A.K.; Mohanty, S.N. Forecasting Methods in Electric Power Sector. Int. J. Energy Optim. Eng. 2018, 7, 1–21.
[CrossRef]
7. Niu, D.; Ji, Z.; Li; W. Research and application of a hybrid model for mid-term power demand forecasting based on secondary
decomposition and interval optimization. Energy 2021, 234, 121145. [CrossRef]
8. Nti, I.K.; Teimeh, M.; Nyarko-Boateng, O.; Adekoya, A.F. Electricity load forecasting: A systematic review. J. Electr. Syst. Inf.
Technol. 2020, 7, 13. [CrossRef]
9. Liu, Z.; Zhu, Z.; Gao, J.; Xu, C. Forecast Methods for Time Series Data: A Survey. IEEE Access 2021, 9, 91896–91912. [CrossRef]
10. Guo, W.; Che, L.; Shahidehpour, M.; Wan, X. Machine-Learning Based Methods in Short-Term Load Forecasting. Electron. J. 2021,
34, 106884. [CrossRef]
11. Lim, B.; Zohren, S. Time-Series Forecasting with Deep Learning: A Survey. Philos. Trans. R. Soc. A 2021, 379, 2020A0209.
[CrossRef]
Sustainability 2023, 15, 11667 15 of 16

12. Haben, S.; Giasemidis, G.; Ziel, F. Short term load forecasting and the effect of temperature at the low voltage level. Int. J. Forecast.
2019, 35, 1469–1484. [CrossRef]
13. Weron, R. Modeling and Forecasting Electricity Loads and Prices: A Statistical Approach; Wiley: Hoboken, NJ, USA, 2007; Volume 403.
14. Box, G.E.P.; Jenkins, G.M. Time series analysis: Forecasting and control. J. Time Ser. Anal. 2010, 31, 249–250.
15. Zhu, X.; Shen, M. Based on the ARIMA Model with Grey Theory for Short Term Load Forecasting Model. In Proceedings of the
2012 International Conference on Systems and Informatics (ICSAI2012), Yantai, China, 19–20 May 2012; pp. 564–567. [CrossRef]
16. Gupta, A.; Kumar, A. Mid Term Daily Load Forecasting using ARIMA, Wavelet-ARIMA and Machine Learning. In Proceedings
of the 2020 IEEE International Conference on Environment and Electrical Engineering and 2020 IEEE Industrial and Commercial
Power Systems Europe (EEEIC/I&CPS Europe), Madrid, Spain, 9–12 June 2020; pp. 1–5. [CrossRef]
17. Yin, C.; Liu, K.; Hu, K.; Yang, Z.; Yang, L.; Zhao, N. SARIMA-Based Medium- and Long-Term Load Forecasting. Strateg. Plan.
Energy Environ. 2023, 42 , 283–306. [CrossRef]
18. Shilpa, G.N.; Sheshadri, G.S. Electrical Load Forecasting Using Time Series Analysis. In Proceedings of the 2020 IEEE Bangalore
Humanitarian Technology Conference (B-HTC), Vijiyapur, India, 8–10 October 2020; pp. 1–6. [CrossRef]
19. Jiang, H.; Zhang, Y.; Muljadi, E.; Zhang, J.J.; Gao, D.W. A Short-Term and High-Resolution Distribution System Load Forecasting
Approach Using Support Vector Regression with Hybrid Parameters Optimization. IEEE Trans. Smart Grid 2018, 9, 3341–3350.
[CrossRef]
20. Hou, K.; Shao, G.; Wang, H.; Zhang, Q.; Yan, W. Research on Practical Power System Stability Analysis Algorithm Based on
Modified SVM. Prot. Control Mod. Power Syst. 2018, 3, 11. [CrossRef]
21. Lu, H.; Azimi, M.; Iseley, T. Short-Term Load Forecasting of Urban Gas Using a Hybrid Model Based on Improved Fruit Fly
Optimization Algorithm and Support Vector Machine. Energy Rep. 2019, 5, 666–677. [CrossRef]
22. Sun, W.; Pang, X.; Liu, W.; Wang, Y.; Luan, C. Short-term Power Load Forecasting Based on Grey Relational Analysis and Support
Vector Machine. In Proceedings of the 2022 4th International Conference on Industrial Artificial Intelligence (IAI), Shenyang,
China, 24–27 August 2022; pp. 1–5.
23. Liu, S.; Zhang, L.; Xia, H. Research on Power Load Prediction Based on Elman Network with Improved Particle Swarm
Optimization. J. Phys. Conf. Ser. 2022, 2338, 012076. [CrossRef]
24. 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]
25. Barman, M.; Dev Choudhury, N.B.; Sutradhar, S. A Regional Hybrid GOA-SVM Model Based on Similar Day Approach for
Short-Term Load Forecasting in Assam, India. Energy 2018, 145, 710–720. [CrossRef]
26. Uriarte, R.B.; Tiezzi, F.; Tsaftaris, S.A. Supporting Autonomic Management of Clouds: Service Clustering with Random Forest.
IEEE Trans. Netw. Serv. Manag. 2016, 13, 595–607. [CrossRef]
27. Wu, X.; He, J.; Zhang, P.; Hu, J. Power System Short-Term Load Forecasting Based on Improved Random Forest with Grey
Relation Projection. Dianli Xitong Zidonghua/Autom. Electr. Power Syst. 2015, 39, 50–55.
28. Xu, H.; Fan, G.; Kuang, G.; Song, Y. Construction and Application of Short-Term and Mid-Term Power System Load Forecasting
Model Based on Hybrid Deep Learning. IEEE Access 2023, 11, 37494–37507. [CrossRef]
29. Cui, G.; Song, D.; Yu, S.; You, Y. Medium-Term Load Prediction Based on GALightGBM-LSTM. In Proceedings of the 2022 9th
International Forum on Electrical Engineering and Automation (IFEEA), Zhuhai, China, 4–6 November 2022; pp. 1–5. [CrossRef]
30. Kumar, S.; Hussain, L.; Banarjee, S.; Reza, M. Energy Load Forecasting using Deep Learning Approach-LSTM and GRU in Spark
Cluster. In Proceedings of the 2018 Fifth International Conference on Emerging Applications of Information Technology (EAIT),
Dehradun, India, 27–28 October 2018; pp. 1–4.
31. Cui, C.; He, M.; Di, F.; Lu, Y.; Dai, Y.; Lv, F. Research on Power Load Forecasting Method Based on LSTM Model. In Proceedings
of the 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC), Chengdu, China, 28–30 August
2020; pp. 1657–1660.
32. Motepe, S.; Hasan, A.N.; Stopforth, R. Improving Load Forecasting Process for a Power Distribution Network Using Hybrid AI
and Deep Learning Algorithms. IEEE Access 2019, 7, 82584–82598. [CrossRef]
33. Li, C.; Chen, Z.; Liu, J.; Li, D.; Gao, X.; Di, F.; Li, L.; Ji, X. Power Load Forecasting Based on the Combined Model of LSTM and
XGBoost. In Proceedings of the 2019 the International Conference on Pattern Recognition and Artificial Intelligence, Wenzhou,
China, 19–22 July 2019; Association for Computing Machinery: New York, NY, USA, 2019; pp. 46–51.
34. Pham, M.H.; Nguyen, M.N.; Wu, Y.K. A Novel Short-Term Load Forecasting Method by Combining the Deep Learning With
Singular Spectrum Analysis. IEEE Access 2021, 9, 73736–73746. [CrossRef]
35. Ciechulski, T.; Osowski, S. High Precision LSTM Model for Short-Time Load Forecasting in Power Systems. Energies 2021, 14, 2983.
[CrossRef]
36. Xu, Y.; Feng, Y. Short-Term Power Load Forecasting Based on Deep Learning. Master’s Thesis, School of Electrical Engineering
and Computer Science, Stockholm, Sweden, 2022.
37. Singh, R.P.; Gao, P.X.; Lizotte, D.J. On hourly home peak load prediction. In Proceedings of the 2012 IEEE Third International
Conference on Smart Grid Communications (SmartGridComm), Tainan, Taiwan, 5–8 November 2012; pp. 163–168.
38. Nag, A.; Ranjan, R.; Vinoth Kumar, C.N.S. An Approach on Cyber Crime Prediction Using Prophet Time Series. In Proceedings of
the 2022 IEEE 7th International Conference for Convergence in Technology (I2CT), Mumbai, India, 7–9 April 2022; pp. 1–5.
Sustainability 2023, 15, 11667 16 of 16

39. Parizad, A.; Hatziadoniu, C.J. Using Prophet Algorithm for Pattern Recognition and Short Term Forecasting of Load Demand
Based on Seasonality and Exogenous Features. In Proceedings of the 2020 52nd North American Power Symposium (NAPS),
Virtual, 11–14 April 2021; pp. 1–6.
40. Lv, L.; Luo, L.; Yang, Y. Distribution Line Load Predicting and Heavy Overload Warning Model Based on Prophet Method.
Sustainability 2022, 14, 13697. [CrossRef]
41. Korstanje, J. The Prophet Model. In Advanced Forecasting with Python: With State-of-the-Art-Models Including LSTMs, Facebook’s
Prophet, and Amazon’s DeepAR; Apress: Berkeley, CA, USA, 2021; pp. 253–272.
42. Kennedy, J.; Eberhart, R. Particle Swarm Optimization. In Proceedings of the ICNN’95—International Conference on Neural
Networks, Perth, WA, Australia, 27 November–1 December 1995; Volume 4, pp. 1942–1948. [CrossRef]
43. Xia, M.; Shao, H.; Ma, X. A Stacked GRU-RNN-Based Approach for Predicting Renewable Energy and Electricity Load for Smart
Grid Operation. IEEE Trans. Ind. Inform. 2021, 17, 7050–7059. [CrossRef]
44. Jiang, W. Deep Learning Based Short-Term Load Forecasting Incorporating Calendar and Weather Information. Internet Technol.
Lett. 2022, 5, 383. [CrossRef]
45. Wang, C.; Wang, Y.; Ding, Z. A Transformer-Based Method of Multienergy Load Forecasting in Integrated Energy System. IEEE
Trans. Smart Grid 2022, 13, 2703–2714. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.