Wevj 15 00048

Article
Neural Sliding Mode Control of a Buck-Boost Converter Applied

to a Regenerative Braking System for Electric Vehicles
Jose A. Ruz-Hernandez 1, * , Ramon Garcia-Hernandez 2 , Mario Antonio Ruz Canul 3 , Juan F. Guerra 2 ,
Jose-Luis Rullan-Lara 1 and Jaime R. Vior-Franco 1
1 Facultad de Ingenieria, Universidad Autonoma del Carmen, C.56 No.4 Esq. Avenida Concordia Col. Benito
Juarez, Ciudad del Carmen 24180, Campeche, Mexico; jrullan@pampano.unacar.mx (J.-L.R.-L.);
180643@mail.unacar.mx (J.R.V.-F.)
2 Tecnologico Nacional de Mexico, Instituto Tecnologico de La Laguna, Blvd. Revolución y Av. Instituto
Tecnologico de La Laguna s/n, Torreon 27000, Coahuila, Mexico; rgarciah@lalaguna.tecnm.mx (R.G.-H.);
m.jfguerrac@correo.itlalaguna.edu.mx (J.F.G.)
3 Centro Universitario de Ciencias Exactas e Ingenierias, Universidad de Guadalajara, Blvd. Marcelino Garcia
Barragan, Guadalajara 44430, Jalisco, Mexico; mario.ruz2798@alumnos.udg.mx
* Correspondence: jruz@pampano.unacar.mx
Abstract: This paper presents the design and simulation of a neural sliding mode controller (NSMC)
for a regenerative braking system in an electric vehicle (EV). The NSMC regulates the required
current and voltage of the bidirectional DC-DC buck–boost converter, an element of the auxiliary
energy system (AES), to improve the state of charge (SOC) of the battery of the EV. The controller
is based on a recurrent high-order neural network (RHONN) trained using the extended Kalman
filter (EKF) and the unscented Kalman filter (UKF) as the tools to train the neural networks to obtain
a higher SOC in the battery. The performance of the controller with the two training algorithms is
compared with a proportional integral (PI) controller illustrating the differences and improvements
obtained with the EKF and the UKF. Furthermore, robustness tests considering Gaussian noise
and varying of parameters have demonstrated the outcome of the NSMC over a PI controller. The
Citation: Ruz-Hernandez, J.A.; proposed controller is a new strategy with better results than the PI controller applied to the same
Garcia-Hernandez, R.; Ruz Canul, buck–boost converter circuit, which can be used for the main energy system (MES) efficiency in an
M.A.; Guerra, J.F.; Rullan-Lara, J.-L.;
EV architecture.
Vior-Franco, J.R. Neural Sliding Mode
Control of a Buck-Boost Converter
Keywords: electric vehicles; regenerative braking; neural sliding mode control; Kalman filter;
Applied to a Regenerative Braking
bidirectional DC-DC buck–boost converter; recurrent high-order neural networks
System for Electric Vehicles. World
Electr. Veh. J. 2024, 15, 48. https://
doi.org/10.3390/wevj15020048
Academic Editors: Jianfei Chen, Liyan 1. Introduction

Zhu, Chen Duan and Haibo Huang
The energy storage in electric vehicles (EVs) fully operated with clean fuels is an
Received: 27 September 2023 extremely relevant issue in the automotive industry despite their huge contribution to the
Revised: 22 December 2023 environmental impact. In [1], an environmental study of many EVs has demonstrated the
Accepted: 25 December 2023 advantages and positive impact that result in a lower number of CO2 emissions in a certain
Published: 2 February 2024 period of analysis compared to conventional fossil fuel vehicles. The commitment with
the environmental impact has developed new strategies to recycle battery packs such as
lithium ion batteries [2], for which demand has increased in recent years. The lead acid
battery recycling process has demonstrated great performance, as well as the necessary
Copyright: © 2024 by the authors.
chemical procedure.
Licensee MDPI, Basel, Switzerland.
In recent years, many approaches to improve energy consumption have been devel-
This article is an open access article
oped and new alternatives to fuel EVs have emerged. The primary goal of implementing
distributed under the terms and
conditions of the Creative Commons
regenerative braking systems in EVs is to recover the kinetic energy produced during
Attribution (CC BY) license (https://
motor deceleration, resulting in increased energy storage capacity and increased driving
creativecommons.org/licenses/by/ distance [3]. In [4], a regenerative braking system with one single braking pedal decreased
4.0/). the energy consumption and improved the commercial cost of EVs.
World Electr. Veh. J. 2024, 15, 48. https://doi.org/10.3390/wevj15020048 https://www.mdpi.com/journal/wevj

World Electr. Veh. J. 2024, 15, 48 2 of 39
The combination of a supercapacitor and battery bank as an energy supply for an EV

has demonstrated an improvement in regenerative braking performance, saving energy
and increasing efficiency [5].
In [6], a proposed design of a boost converter significantly reduces the number of
elements compared to existing models, improving energy performance and reducing cost
and size. Furthermore, it allows for the adjustment of converter parameters such as pulse
width modulation (PWM) to achieve the power required for the system. DC converters
have been implemented with good results in many applications. In [7] a sliding mode
control (SMC) is implemented in DC-DC converters such as the buck and boost circuit.
The control law is proposed, considering the continuous-time buck and boost converter
mathematical model. To reduce chattering, a multiphase controller is developed. However,
the analysis of these converters is made separately, and only current control can guarantee
stability, as it is the main dynamic to ensure the tracking of a desired voltage reference.
In [8], a bidirectional DC-DC converter is operated in a photovoltaic system to take its
energy and charge a battery through the converter.
Extensive work is presented in [9], where possible failures that can affect a photovoltaic
system, such as the shading fault that leads to the output voltage being lower than expected,
and the failure of the DC converter to provide the desired voltage, are studied. Thus,
an SMC with the IncCond technique is designed to track the desired voltage under both
failures, providing the maximum available solar power. Additionally, a global SMC based
on Lyapunov stability analysis is presented in [10], where the main goal of applying this
controller is to track the desired output voltage value of a buck converter and reduce the
chattering generated with the conventional SMC.
Another study is presented in [11], where a strategy based on inverse optimal control
is employed to adjust the output voltage of a boost converter. In addition, the design of
two observers is developed to reduce the quantity of sensors. Furthermore, the inverse
optimal controller is compared to an SMC. An extremely interesting work is presented
in [12], where the two training filters are implemented to obtain the optimal SOC to avoid
overcharge of a Li-ion battery used in commercial electric vehicles. Many approaches
have been developed to enhance the performance of the EV and the energy use in the
DC converters. One of the more important properties of electric vehicles is the ability to
regeneratively brake as a bridge that allows for the recovery of energy while braking and
the improvement of the efficiency of the storage system [13].
A PI was used to adjust the dynamics of a bidirectional buck-boost in a regenerative
braking system with a supercapacitor as an auxiliary energy device in [14]. In [15], the
mileage of an EV has been increased by implementing a nonlinear predictive control
strategy with two stages to regulate the torque needed during braking. In addition, the
optimization of the torque improves the energy conservation during braking.
To minimize chattering produced by the control to regulate the optimal slip value
in an antilock braking system, a sliding mode fuzzy controller is implemented using an
exponential reaching law. Furthermore, the results are compared to a bang–bang controller,
with the SM controller outperforming the bang–bang controller in [16].
Another interesting case is the one depicted in [17], where artificial intelligence is
the main tool to develop a “comfort regenerative braking” using neural networks with
the backward propagation algorithm, taking data from the acceleration and deceleration
limits and using them to map the optimal behavior of the EV. Regenerative braking has
also been improved in [18], where state estimation is achieved using neural networks
and the least squares algorithm, and is later used in the proposed control. This study
demonstrated an improvement in the recovery of 9.62% energy under specified conditions.
Moreover, an NSMC with an EKF identifier is being developed to improve the EV system
with regenerative braking without considering the DC motor behavior and solely managing
the current and voltage of the buck–boost converter within the EV system in [19]. Finally, in
an EV framework, a neural inverse optimal control (NIOC) is accomplished by employing
a reference generator with the speed of a DC motor that only decelerates once during the
entire operation in [20].
The simulation of an EV with a regenerative braking system is presented in this
paper, where the above-mentioned system is controlled employing a neural sliding mode
controller (NSMC) based on system identification implementing RHONNS trained with
EKF and UKF training algorithms to reduce the state estimation error and achieve good
performance of the EV system. In this work, the NSMC is used to track the full dynamics of
the regenerative braking system using a reference generator that is related to a DC motor
that operates at different speeds, resulting in a signal that changes between acceleration
and deceleration, and allows us to appreciate the operating modes of the bidirectional
DC-DC buck–boost converter more accurately. Considering these changes in the speed of
the motor is important, as these approximate the simulation to a real case of an EV driving
performance. Furthermore, this approach will enable us to analyze the regenerative braking
system of electric vehicles utilizing multiple robust intelligent control techniques based on
Kalman filters, leading to novel findings.
The proposed NSMC regulates the required current and voltage of the DC-DC buck–
boost converter, an element of the AES, in order to improve the MES efficiency, particularly
in the SOC of the EV battery. Successful energy recovery during braking system operation
is ensured when correct NSMC performance is achieved.
This article’s main novelty is the use of intelligent control algorithms, such as the
NSMC, which is developed using RHONN identification and the EKF and UKF as train-
ing tools for neural networks, resulting in a higher SOC in the EV battery. The con-
troller has been validated in simulation for various tasks, including tracking of a varying
time–signal trajectory and tracking of a trajectory generated from a DC motor with mul-
tiple acceleration and deceleration changes. The performance of the RHONN with the
two training algorithms, EKF and UKF, is validated using a chirp signal. The initial weights
of the RHONN are chosen randomly. Afterwards, the performance of the proposed NSMC
with EKF and UKF identifiers is compared to a PI controller demonstrating the differences
and improvements obtained, including Gaussian noise and varying parameters.
The structure of this paper is outlined in Section 2, which explains the steps taken to
obtain the results of the simulation and the procedure followed in this study. Section 3
presents the electrical circuit of the DC converter, including a brief explanation of its
operation. Section 4 outlines the mathematical preliminaries, including the equations
used to design the neural controller. Section 5 describes the modeling of the buck–boost
converter system and the integration of RHONN identification into the NSMC design.
Section 6 shows the simulation results for each step of the article, including the neural
controller test and a comparison of neural identification with EKF and UKF. Additionally,
a robustness test of the NSMC controller is done under two scenarios. Section 7 presents
the mean square error (MSE) and energy of the error signal (EES), comparing their results
obtained during the tracking with the UKF, EKF, and PI considering the arrangement used
in [14]. Finally, Section 8 discusses the obtained results and proposes future work. The
conclusions of this work based on the results are presented in Section 9.
2. Materials and Methods

The simulation has followed particular steps to obtain the results presented in this article.
• The main objective of this paper is to improve the energy storage and power con-
sumption of the regenerative brake system. To achieve this, the current and voltage
dynamics can be controlled using a bidirectional converter. The regenerative braking
system and the neural controller have been simulated and verified using the Simscape
Electrical from Matlab (Matlab R2020a, Simulink, 1994–2024 ©The MathWorks, Inc.,
Natick, MA, USA).
• The simulation implements the mathematical representation of the buck–boost con-
verter from [21].
• The system identification is developed taking the RHONN philosophy in [22,23] to

identify the dynamics to regulate the buck–boost converter. The RHONN is validated
using a chirp signal with initial weights chosen randomly in the range of (0, 1).
• The RHONN philosophy is employed to design the identification process, which is
then trained with the EKF and UKF algorithms.
• The NSMC is developed to track the trajectories in the AES chosen for the case study.
In the first scenario, a signal is selected to validate the identification and regulation of
the buck–boost converter dynamics. The signal is variable, and designed to operate
within the range of the AES (supercapacitor) and MES (battery bank) parameters. In
the second scenario, the complete control framework implemented in the simulation is
verified. The DC motor parameters are considered for building a reference generator
that tracks the EV system requirements.
• The results of the NSMC with EKF and UKF are presented, and the enhancement of
the battery banks with the implemented AES is demonstrated.
• The NSMC with EKF and UKF is validated to demonstrated the robustness of the
system under different conditions. Additionally, the comparison of the NSMC’s
robustness with a PI controller based on the control strategy [14] is presented.
• The MSEs and EESs obtained with UKF and EKF are compared with a PI control
strategy [14] to identify the best performance in regenerative braking simulation and
robustness of the control strategies.
3. DC-DC Bidirectional Buck–Boost Converter

A fundamental device in the regenerative braking system is the power converter, such as
the boost converter, buck converter, step-down converter, step-up converter, etc. However,
the buck–boost converter is chosen in this case. Additionally, this converter is a fundamental
element of the AES, which helps reduce the energy consumption during braking.
Buck–boost converters are the result of the combination of two configurations of DC
power converters, which generally work separately in different applications. Furthermore,
the configuration selected for the converter can be a bidirectional DC buck–boost converter
related to the requirements of the power system [24], and this configuration is perfectly
coupled with energy storage and the control of battery charge and discharge applications.
The regenerative braking system contains several elements that allow it to operate
correctly. First of all, there is a battery bank, which is responsible for powering the DC
motor that moves the vehicle. Next, we have the responsible elements of growing or
reducing the voltage supplied to the EV. These elements are the supercapacitor and the DC
buck–boost converter. The connections of the system are illustrated in Figure 1.
+
T2 D2
L ic +
Vc C
Vsc
T1 D1
− −
Battery Bank Supercapacitor
Figure 1. DC buck-boost converter circuit.
Here, L is an inductor, C is a capacitor, T1 and T2 represent the insulated gate bipolar

transistors (IGBT), and the D1 and D2 represent the diodes. A regulated signal such as the
PWM is the input of the transistors that transfer the current from or to the supercapacitor.
Buck mode: When this mode is activated, there is a reduction in the output voltage
related to the input voltage. T1 is deactivated and T2 is activated, resulting in the transfer of
the energy coming from the capacitor voltage Vc to the supercapacitor voltage Vsc . Once T2
is deactivated, the energy stored in the capacitor flows C to the supercapacitor employing
the current capacitor ic . Furthermore, the inductor L is charged with a fragment of the same
current capacitor ic energy. Likewise, when T2 is deactivated, the L current is discharged
into Vc via D1 , guiding this current energy in the flow of C [14].
Boost operation: During activation of this mode, the output voltage increases concern-
ing the input voltage. T1 is ON and T2 is OFF, and the energy coming from supercapacitor
Vsc is transferred to battery bank Vc. As soon as T1 is activated, the capacitor’s energy is
taken and stored in L. On the other hand, when T1 is activated, the energy in L is driving
into the capacitor C across D2 into the battery bank.
While the braking operation is occurring, the electricity generated in the DC motor
is managed with the brake into the battery bank or supercapacitor. The boost function
is turned ON when the EV is accelerating, while the buck function is activated when the
EV is decelerating. The DC-DC converter operating in both modes makes it easier for the
supercapacitor to increase and decrease energy. Consequently, the SOC in the battery bank
is improved.
4. Mathematical Preliminaries
4.1. Discrete-Time Sliding Mode Control
The SMC allows it to work with robustness over some perturbations, being one of
their more attractive properties over the last few decades. Consider the following nonlinear
system [25]:
ẋ = f ( x, u, t) (1)
with a f ( x ), | f ( x )| < f o function, which is a bounded constant and a switching function u
as the control law, which is used to reduce the error e = r − x to zero, where the reference
input is r , and u is established as

u0 i f s ( x ) ≥ 0
u= (2)
− u0 i f s ( x ) < 0
where s( x ) and u0 are defined as the sliding surface and the upper control bound, respec-
tively. Figure 2 describes a continuous-time system with scalar SMC, where the x (t) state
starts from an initial x (t = 0) point achieving the trajectory of the sliding surface s( x ) = 0
within finite time tsm and remaining on the surface afterwards.
s(x)=0
x(t=0)
x(tsm)
Figure 2. Continuous-time system trajectory when SMC is applied.
When for each sampling point t j = k∆t, k = 1, 2 . . . the function is derived, and a
representation in discrete-time of the continuous time system (1) is
x k +1 = F ( x k ) (3)
with the starting condition (tsm ), the trajectory equated to the sliding manifold when
s( x (t)), or k m ≥ tsm /∆t [25].
s( xk ) = 0 ∀k ≥ k sm (4)
This operation can be defined as discrete-time sliding mode. Afterwards, from (1), and
considering a scenario in which for any constant control u and whatever initial condition
x (0), its closed solution can be written as [26]:
x ( t ) = F ( x (0), u ) (5)
For the constant state control signal to attain s( xk+1 ) ,it is required to identify uk at
each sample point k. The corresponding discrete-time representation of the system is the
following [26]:
x k +1 = F ( x k , u k ) (6)
uk = u( xk ) (7)
At each sample point, the sliding manifold is achieved, that is, s( xk+1 ),
∀k = 0, 1, · · · ∗ ∗∗ is completed. This can be corroborated since F ( x (0), u) tends to x (0) as
∆t → 0; the function u( x (0), ∆t) may need more control resources than are available for u0 .
Using the discrete-time SMC, the sliding manifold is described as
s( xk ) = xk − xre f ,k (8)
Evaluating at sample time (k + 1), the sliding manifold is determined:
sk+1 = F ( xk , uk ) − xre f ,k+1 (9)
Now is calculated the equivalent control ueq ( xk , k ), as in [25]:

h i
ueq ( xk , k) = − F ( xk , k) − xre f ,k+1 (10)
A stabilizing term, determined as un ( xk , k), should be added to asymptotically achieve

the sliding manifold [26].
un ( xk , k) = −(Ssc sk ) (11)
where Ssc is the Schur matrix.
The following control law is determined considering the boundedness of the control
signal ∥ uc ( xk , k ) ∥< u0 , u0 > 0 [25]:
i f ∥uc ( xk , k)∥ < u0



 uc ( xk , k )
u( xk , k ) = (12)
ueq ( xk ,k )
 u0 i f ∥uc ( xk , k)∥ ≥ u0

∥ueq (xk ,k)∥
where ∥•∥ represents the Euclidean norm of uc ( xk , k) and is defined as:
uc ( xk , k) = ueq ( xk , k) + un ( xk , k ) (13)
4.2. Discrete-Time Recurrent High-Order Neural Networks

Recurrent neural networks (RNNs) have been utilized in recent times to approximate
and identify the mathematical models of complex systems, according to [22]. RHONNs
have proven that they are effective at identifying nonlinear systems, where identification is
achieved through modifying the parameters of a suitable system representation attached to
an adaptive law. The approximated state variable of a nonlinear system using a RHONN
identifier with a series-parallel configuration is obtained by [23]:
χi,k+1 = ωiT ϕi ( xk ) + ω iT φi ( xk , uk ) (14)

where χi,k+1 is defined as the state of the ith neuron that identifies the ith component of the
state vector xk , where xk = [ x1,k , . . . , xn,k ], ωi,k ∈ ℜ Li are the neural networks’ adjustable
synaptic weights and ω i,k are fixed weights, ϕi and φi are linear functions of the state vector,
and uk is the vector input of the RHONN model. Moreover, u ∈ ℜm and ϕi ( xk , uk , k ) ∈ ℜ Li
is defined as [22]  dij(1) 
∏ ζ
 j∈ I1 ijd
 
ϕi1 
ij(2) 
 ϕik    ∏ j ∈ I2 ij
ζ 
ϕi ( xk , uk , k) =  ..  =  (15)
 
.

 .    .. 

 
ϕi L ,k dij( L )
i ∏ j∈ IL ζ ij i
i
with dij,k defined as non-negative integers; Li as the number of the connection; I1 , I2 , . . . , ILi
as a collection of non-ordered subsets 1, 2 . . . , n + m; n determined as the state dimension;
m as the input dimension; and ζ i defined as [22]:
S ( x1 )
 
 
ζ i1
 . 
 ..   .. 
 .   
   S xn 
ζ i =  ζ in  = 
    (16)
 .   u1,k 

 ..   . 
 .. 
ζ in+m
um,k
T ] u is the network’s input vector. Additionally, the hyperbolic

where u = [u1.k , u2,k , . . . , um,k
function S(.) is defined as
S( xk ) = αi tanh( β i xk ) (17)
where xk is set as the state variable; while α and β are positive constants.
A ith RHONN identifier scheme is presented in Figure 3.
Figure 3. Scheme of a RHONN identifier.
To reduce the error between the estimation of states, the RHONN identifier requires a
training algorithm. The explanation of the two training methods proposed is presented below.
4.3. Extended Kalman Filter

The EKF has turned into a typical method applied in many nonlinear systems’ iden-
tification. Only if the state equations of the system have determined the linearization,
considering a Gaussian random, can variable operation be possible [27,28].
The algorithm model is defined as follows:
wi,k+1 = wi,k + ηi Ki,k ei,k (18)

Ki,k = Pi,k Hi,k Mi,k (19)
T
Pi,k+1 = Pi,k − Ki,k Hi,k Pi,k + Qi,k (20)
ei,k = xi,k − χi,k i = 1, 2, · · · , n (21)
h i −1
T
Mi,k = Ri,k + Hi,k Pi,k Hi,k (22)
where ei ∈ R is defined as the identification error to be minimized, ηi is a parameter

of the training algorithm design, Ki,k ∈ R Li ×m is the Kalman matrix, Qi,k ∈ R Li × Li and
Ri,k ∈ Rm×m are set as positive definite constant matrices, and Pi ∈ R Li × Li is an adjustable
diagonal matrix. Finally, Hi ∈ R Li ×m is an adjustable matrix, which is defined as the
resulting state derivative with respect to the adjustable weights of the neural identifiers.
More extensive details and a stability proof explanation of the RHONN and the EKF
training algorithm are presented in [22,29].
4.4. Unscented Kalman Filter

The EKF leverages Jacobian matrices for linear approximations. Its effectiveness hinges
on two key factors: firstly, accurate “first-order” approximations give rise to suboptimal
terms directly affected by the nonlinearity of the system. Secondly, the computation of
Jacobian matrices in practical applications can become complex. Such disparities can cause
low performance or divergence. The UKF addresses the limitations generated by the
EKF using a deterministic sampling approach. The limitations generated by the EKF are
addressed with the UKF employing a deterministic sampling approach.
A method used to calculate the statistics of a random variable that undergoes a
nonlinear transformation is defined by Juliar Uhlmann in [30] and defined as the unscented
transformation (UT). This is used as an extension of the recursive estimation of states in the
UKF, considering the Kalman filter framework.
The UKF identification algorithm equations can be described as follows:
jyy −1

j jxy
Ki,k = Pi,k Pi,k ,
j j j j
wi,k+1 = wi,k + Ki,k ei,k , (23)
j j j jyy j ⊤
Pi,k+1 = Pi,k − Ki,k Pi,k Ki,k ,
with
2L ⊤
∑ ηic
jxy j j- j j-
Pi,k = Xi,k- − x̂i,k Yi,k- − ŷi,k ,
i =0
2L ⊤
Pi,k = Ri,k + ∑ ηi
jyy j jc j j- j j-
Yi,k- − ŷi,k Yi,k- − ŷi,k ,
i =0 (24)
2L ⊤
Pi,k = Qi,k + ∑ ηi
j j jc j j- j j-
Xi,k- − x̂i,k Xi,k- − x̂i,k ,
i =0
j j j
ei,k = Xi,k − xi,k
j j jyy
where ei,k is the identification error, Pi,k is the matrix of prediction error covariance, Pi,k is
jxy
the covariance of predicted output matrix, Pi,k is the output matrix and the cross-covariance
j jc j
of state, wi,k is the j − th weight of the i − th subsystem, ηi is a parameter of design, Xi,k
j
is the j-th state of the plant, and xi,k is the j-th state of the neural network. L is defined as
j j
the number of states while Ki,k is the Kalman gain matrix, Qi,k is the measurement noise
j j−
covariance matrix, Ri,k is the state noise covariance matrix, x̂i,k is the predicted state mean,
j− j j
ŷi,k is the predicted output mean, and Xi,k|k−1 and Yi,k|k−1 are sigma-points propagated
through prediction and observation, respectively.
Table 1 explains the implementation step-by-step and the comparison of the EKF and
UKF. More extensive details and the stability analysis of the UKF training algorithm for
neural identification are explained in [31].
Table 1. Differences between EKF and UKF training algorithms.
Extended Kalman Filter (EKF) Unscented Kalman Filter (UKF)

Design parameters
Analytical Jacobian matrices Scaling parameters
Fn , Hn α, β, κ, λ, ηim , ηic
Initialization
Qn , Rn , x̂0 , P0
Prediction
=

Xi,k- f Xi,k , uk
x̂k- = f ( x̂k , uk ) 2L
x̂k- = ∑ ηim Xi,k−1
i=0
Pk- = Fk Pk Fk⊤ + Qk 2L
= ∑ ηim

Pk- Qk + Xi,k- − x̂k-
i=0
⊤
Xi,k- − x̂k-
Observation
=

Yi,k- h Xi,k , uk
ŷk- = h( x̂k- ) 2L
ŷk−1 = ∑ ηim Yi,k-
i=0
yy
Pn = Hn Pk- Hn⊤ + Rn xy
2L
= ∑ ηim

Pn Xi,k- − x̂k-
i=0
xy ⊤
Pn = Pk- Hn⊤ Yi,k- − ŷk-
2L
yy
= ∑ ηim

Pn Rn + Yi,k- − ŷk-
i=0
⊤
Yi,k- − ŷk-
Update
xy yy −1

Kn = Pn Pn
x̂n = x̂k- + Kn (yn − ŷk- )
yy
Pn = Pk- − Kn Pn Kn⊤
5. System Modeling and Sliding Mode Neural Control

5.1. Buck–Boost Converter Model
In this application, boost and buck converters were combined for developing the
DC-to-DC converter. The first is used in charge-only situations, whereas the second is
utilized in discharge-only situations. The work in [21] is used for the definition of the boost
converter model.
ts ts
x1,k+1 = (1 − ) x − x2,k (25)
RC 1,k c
ts
x2,k+1 = x2,k + Ubtt uc (26)
L
The buck converter model is also defined by [21]:
ts ts
x1,k+1 = (1 − ) x1,k + x2,k (27)
RC c
ts
x2,k+1 = x2,k + Ubtt uc (28)
L
where the converter output voltage is defined as x1,k , the output current is x2,k , Ubtt is the
voltage in the battery, uc is the set as the input vector, L is the inductance measured in
Henry ( H ), R is the load resistance measured in ohms (Ω), C is the capacitor in measured
in farads ( F ), and finally, ts is the sample time.
5.2. Neural Controller Design

Following the procedures established in Section 4 and using Equation (14), a RHONN
has been employed to close in on the buck–boost converter behaviors in order to manage
the current flow and assure the charging and discharging operating modes. Given the
RHONN’s adaptive nature and the similarity of the buck and boost converter models,
a single identifier is suggested for both situations as
χ̂1,k = ω11,K S( x1,K ) + ω12,K S( x2,K )

+ w1,3 S( x1,K )S( x2,K ) + ϖ1 x2,K + u( x1,k ) (29)
χ̂2,k = ω21,K S( x2,K ) + ω22,K S( x1,K )
+ ω23,K S( x1,K )S( x2,K ) + ϖ2 u( x2,k ) (30)
where the estimated dynamics of the vector [ x1,k , x2,k ] T are [χ̂1,k , χ̂2,k ] T . As defined above,
x1,k is the measured voltage in the converter’s output, and x2,k is set as the converter output
current. the control signal chosen for each state of the system is u, ϖ1 and ϖ2 are constant
fixed weights. A representation of the identification design configuration is shown in
Figure 4.
UNKNOWN
SYSTEM xk
e -
u! KALMAN
FILTER
wK +
NEURAL
IDENFIER
Figure 4. Neural identification representation strategy.
5.3. Discrete Time Sliding Mode Control

The sliding mode controller is synthesized once the buck–boost converter’s system
identification is carried out. Due to the presented neural model’s triangular construction,
only the final dynamics require control [25].
For the converter to function effectively in transferring energy to the battery bank,
both voltage and current dynamics have to be controlled.
Following the steps in (6)–(13), the sliding surface at the sample point k + 1 of the x̂1,k
dynamics control is obtained as
Voltage NSMC
s x1,k+1 = ω2,1 (k )S( x2 ) + ω2,2 (k )S( x1 )

+ w1,3 S( x1,K )S( x2,k ) + ϖ1 x2,k + u( x1,k ) − x1re f ,k+1 (31)
The equivalent x̂1,k control is calculated as follows:
1
ueq ( xk , k) = − [ω (k)S( x2 ) + ω2,2 (k)S( x1 )
ϖ1 2,1
+ w1,3 S( x1,k )S( x2,k ) − x1re f ,k+1 ] (32)
i f ∥uc ( x1 , k)∥ < u0



 u c ( x1 , k )
u( x1,k ) = (33)
ueq ( x1 ,k )
 u0 i f ∥uc ( x1 , k)∥ ≥ u0

∥ueq (x1 ,k)∥
The uc ( x1 , k) is set as
uc ( x1 , k) = ueq ( x1 , k) + un ( x1 , k) (34)

with un ( x1 , k) = − Ssc s x1,k+1 and Ssc is defined as a square matrix with real entries and
eigenvalues conditioned with ∥Ssc ∥ < 1, and u0 is the control upper bound with the
condition u0 > 0.
Current NSMC
The sliding surface an control development for x̂2,k is obtained as
s x2,k+1 = ω2,1 (k)S( x2 ) + ω2,2 (k)S( x1 )

+ w2,3 S( x1 )S( x2 )ϖ2 uk − xre f ,k+1 (35)
Then, the equivalent current control is obtained as follows:
1
ueq ( x2 , k ) = − [ω (k)S( x2 ) + ω2,2 (k)S( x1 )
ϖ2 2,1
+ w2,3 S( x1 )S( x2 ) − xre f ,k+1 ] (36)
and the NSMC is implemented as follows
i f ∥uc ( x2 , k)∥ < u0



 u c ( x2 , k )
u ( x2 , k ) = (37)
ueq ( x2 ,k )
 u0 i f ∥uc ( x2 , k)∥ ≥ u0

∥ueq (x2 ,k)∥
The uc ( x2 , k) is set as
uc ( x2 , k) = ueq ( x2 , k) + un ( x2 , k) (38)

with un ( x2 , k ) = − Ssc s x2,k+1 , where Ssc is again considered as a square matrix with real
entries and eigenvalues conditioned with ∥Ssc ∥ < 1, and u0 is the control upper bound
with the established condition of u0 > 0.
5.4. Reference Generator Development

The regenerative braking system results in energy consumption through DC motor
behavior and requires the required voltage and current in the AES. A reference generator
has been implemented in the simulation to put the whole regenerative braking system
in motion.
A reference generator was created to provide the intended value for the buck−boost
current and voltage. The energy stored in the supercapacitor as a function of the energy
produced by the DC motor is described as the charge reference. A definition considering
the work and energy theorem can be consulted in [32]. The mathematical formula for the
statement “the work done between point A and point B on a particle increases the kinetic
energy” is described as:
Z B
WA → B = Pdt (39)
A
Applying this theorem on the DC motor, for a time t ∈ [kδ, (k + 1)δ)] the energy can
calculated as [14]: Z t
E(t) = Pk dξ + Ek (40)
kδ
P = τω is the DC motor power, with the torque and speed defined as τ and ω,
respectively. The supercapacitor can operate in two modes, charge and discharge. The
equation below determines the supercapacitor energy during charging mode as
Z t
re f −c
EC (t) = −k p sat1 ( P)dξ + Eck (41)
kδ
The following equation defines the supercapacitor energy during discharging:

Z t
re f −d
EC (t) = kp sat2 ( P)dξ + Eck (42)
kδ
re f −c re f −d
with EC (t), EC (t) and 0 < k p < 1 defined as the reference energy charge, the refer-
ence energy discharge, and the lost energy, respectively. sat1 represents a function restricted
in (−∞, 0), and sat2 represents the same function with the restriction in (0, ∞). The sum
of both energies in the capacitor references (41) and (42), results in the supercapacitor
energy reference.
re f −c re f −d
EC = EC + EC (43)
Finally, the obtained reference for the buck−boost converter voltage is written as
r
2EC
Vcr = (44)
C
where Vcr is defined as the reference signal computed for the voltage in the supercapacitor,
Ec is the stored energy in the supercapacitor, and C is the capacitance value of the superca-
pacitor. After the Vcr is developed, this can be used in the buck–boost converter control,
where the resulting u( x2 , k) is used to estimate the duty cycle in the transistor T1 and T2 .
The duty cycle is determined as follows:


 (∥u( x2 , k)∥ 0) if u ( x2 , k ) < 0



Duty Cycle ( Pwm1 Pwm2 ) = (0 ∥u( x2 , k)∥) if u ( x2 , k ) > 0 (45)




(0 0) if u ( x2 , k ) = 0

Figure 5 shows a representation of the regenerative braking system, which includes the
developed reference generator block and the NSMC. This representation is subsequently
introduced into the simulation. The performance of the DC motor determines how effec-
tively the regenerative braking systems work; therefore, a PI controller whose gains are
calculated based on the Ziegler–Nichols method and the experience is used to help the DC
motor follows a specific speed reference.
Z t
uω = k ω p (ωre f − ω ) + k ωi (ωre f − ω ) (46)
0
where ωre f is the speed motor reference and kω p and kωi are the controller gains, respectively.
x!
+
T2 D2 x"
L ic +
Vc C
Vsc
T1 D1
- -
Super Capacitor Buck-Boost Converter x!

Bank
c1,k
w RHONN
x" c2,k
Identifier
t
REGENERATIVE Battery
DC
BRAKING Bank Motor NEURAL
SYSTEM IDENTIFICATION
c1,k c2,k
w t
− Vcr
− Sx1,k+1
−
u(x!,k))
Sx2,k+1
wref DC Motor speed uw
Control Reference Generator NSMC NSMC
+ + VOLTAGE + ireference CURRENT
INTELLIGENT CONTROL
Pwm!
Pwm" PWM Generator
Figure 5. Regenerative braking system control representation.
6. Simulation Results
The results of different analyses made in this work are described in this section.
First, the validation of the RHONN with both UKF and EKF is illustrated using a chirp
signal to guarantee the correct operation and estimation of the identifier over a rapidly
varying signal.
Subsequently, two scenarios are presented, in which the developed NSMC is applied.
In the first simulation test, the control and identification of the system is validated using
a time-varying signal operating in a voltage range that the AES and MES can allow. This
part illustrates the trajectory tracking of the NSMC without a voltage and current reference
generated through the “reference generator”, explained in Section 5.
The final section of the simulation results demonstrates the complete operation of the
regenerative braking system, including the simultaneous operation of the identification,
NSMC, reference voltage and current generators, and the DC motor. Additionally, a robust-
ness test is implemented in order to validate the correct operation of the NSMC controllers
with some changes in the signal dynamics and the parameters of the buck–boost converter.
The proposed control schemes, including the corresponding energy systems used,
are implemented and validated with the help of the Matlab/Simulink toolbox “Simscape
Electrical” of Matlab. In addition, the simulation parameters used for the build-up of the
energy systems AES and MES are enumerated in Table 2.
Table 2. Parameters of the AES and MES.
Description Unit
Buck–boost converter resistance R 50 Ω
Buck–boost converter inductance L 13 mH
Buck–boost converter capacitance C1 2 mF
Buck–boost converter capacitance C2 1 µF
Supercapacitor voltage Vsc 350 V
Battery bank voltage Vc 500 V
Initial SOC 80%
Sampling time (ts ) 0.1 µ seg
6.1. RHONN Validation

The RHONN developed in Section 4 is validated using a chirp signal with the param-
eters shown in Table 3 for each estimated dynamics. The initial weights of RHONN are
randomly chosen in a range of (0, 1).
Table 3. Chirp signal parameters.
Chirp Signals Parameters

Controller Initial Frequency Target Time (s) Frequency at Target Time (Hz)
Chirp signal for voltage 0.5 Hz 25 0.6 Hz
Chirp signal for current 0.5 Hz 10 0.5 Hz
The aim of identifying chirp signals is to guarantee the correct functioning of the
RHONN with both training algorithms. Figures 6 and 7 show the identification of the
proposed signal for the UKF, where x1,real and x2,real are the chirp signals used as the
voltage and current input, respectively, and χ̂1,k and χ̂2,k are the estimated dynamics with
the RHONN. On the other hand, the validation of the identification with EKF is illustrated
in Figures 8 and 9 using the same chirp signals as described in Table 3.
1.5
1
Voltage(Volt)
0.5
0
−0.5
−1
−1.5
0 10 20 30 40 50
t [Sec]
Figure 6. Chirp signal identification for voltage x1 with UKF.
1.5
1
Current(Amp)
0.5
0
−0.5
−1
−1.5
0 10 20 30 40 50
t [Sec]
Figure 7. Chirp signal identification for current x2 with UKF.
6.2. NSMC Trajectory Tracking Results Using UKF

This test allows for the corroboration of the neural controller with the RHONN trained
with UKF. The objective here is that the controller could track a desired trajectory operated
in a permitted range of the AES. The chosen signal to track works in a range of 350 V to
380 V. First, the signal starts at 355 V for 10 s; after that, the voltage decreases to 350 V and
lasts to t = 20 s. Then, the signal increases the value to 365 V. Finally, at t = 30 s, the signal
operates as a sine wave between 355 V and 375 V and stops at t = 50 s.
1.5
Voltage(Volt) 1
0.5
0
−0.5
−1
−1.5
0 10 20 30 40 50
t [Sec]
Figure 8. Chirp signal identification for voltage x1 with EKF.
1.5
1
Current(Amp)
0.5
0
−0.5
−1
−1.5
0 10 20 30 40 50
t [Sec]
Figure 9. Chirp signal identification for current x2 with EKF.
The outcome of the trajectory tracking is shown in Figure 10, where the signal control
is tracked with a minimum error at t = 20 s, where the controller adjustment works fast to
reach the desired value. The current tracking trajectory is shown in Figure 11, where the
current operates related to the controlled voltage, resulting in a signal that operates between
−200 and 200 Amps. Furthermore, the controller did not show significant behavior during
fast changes in the signal, as seen around t = 20 s.
380
Voltage(Volt)
370
360
350
340
0 10 20 30 40 50
t [Sec]
Figure 10. Voltage dynamics trajectory tracking with UKF.
200
100
Current(Amp)
0
−100
−200
0 10 20 30 40 50
t [Sec]
Figure 11. Current dynamics trajectory tracking with UKF.
Additionally, during the control of the dynamics, the weights of the identifier are
adjusted over the tracking of the desired trajectory. The data obtained from the weight
adjustment are illustrated in Figure 12, and represent the computation made during the
tracking of the trajectory. Furthermore, the identification of the voltage and current dy-
namics is illustrated in Figure 13. It is important to note that the states identified are the
dynamics that take place over the AES, and are tracked subsequently with the NSMC. The
proper development of the identification enables the controller’s proper operation, which
leads to the improvement of the braking system’s energy recovery capacity.
Weights UKF
500
−500
11 12 13 21 22
−1000
0 10 20 t [sec] 30 40 50
Figure 12. Weight adjustment with UKF.
380
Current (Amp) Voltage (Volt)
360
340
0 10 20 30 40 50
200
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 13. System dynamics identification with UKF.

The calculation and explanation of the MSE and EES results can be found in Section 7,
comparing the UKF and EKF with a PI identification differences during the control trajectory
tracking with the NSMC.
6.3. NSMC Trajectory Tracking Results Using EKF

The implementation of the NSMC using the EKF is presented. Figure 14 illustrates the
tracking trajectory of the proposed signal where the adjustments made by the controller
when the signal changes their parameters are more significant that the one with the UKF.
Figure 15 illustrates the trajectory of current x2 dynamics tracking using the NSMC
with the EKF. The controller results demonstrated a great performance to achieve the values
of the tracked signal.
380
370
Voltage(Volt)
360
350
340
0 10 20 30 40 50
t [Sec]
Figure 14. Voltage dynamics trajectory tracking with EKF.
200
100
Current(Amp)
−100
−200
0 10 20 30 40 50
t [Sec]
Figure 15. Current dynamics trajectory tracking with EKF.
The weights of the RHONN identifier during the control of the desired signal with
varying time and the adjustments made by the neural network to acquire the required
weight values are presented in Figure 16. Additionally, the identified dynamics with the
EKF are shown in Figure 17.
6.4. Regenerative Braking System Control Using a Reference Generator with UKF
The complete functioning of the regenerative braking system will be addressed in this
section of the simulation results. This test is significant because it relates the entire regen-
erative braking system by using the DC motor’s speed variations to provide a signal that
can be employed to determine and then implement the NSMC. The DC motor parameters
are described in Table 4. First, the DC motor operates by varying the speed, which works
in the range of 0 to 80 rad/s, with the objective of replicating an EV driving operation
where the speed increases and decreases in different instants. In order to ensure the correct
calculation of the reference generator is necessary to control the DC motor speed. A PI
controller is applied to control the motor speed, as depicted in Figure 18.
4 Weights EKF
10
1.5
1
−1
−1.5 11 12 13 21 22
0 10 20
t [sec] 30 40 50
Figure 16. Weight adjustment with EKF.
380
360
340
0 10 20 30 40 50
400
200
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 17. System dynamics identification with EKF.
Table 4. Parameters of the DC motor.
Description Unit
Armature resistance R a 2.581 Ω
Armature inductance L a 0.028 H
Field resistance R f 281.3 Ω
Field inductance L f 156 H
Field-armature mutual inductance L a f 0.9483 H
Power Pw 5 Hp
DC voltage 240 v
Rated speed rpm 1750 rpm
Field voltage 300 v
DC motor speed control

100
80
Speed (rad/s)
60
40
20
0
0 10 20 t (Sec) 30 40 50
Figure 18. Motor speed control with PI.
After the desired speed of the motor is controlled, the reference generator block
computes the voltage required to operate the AES system and tracks the signal of voltage
provided through the reference obtained. The neural controller in this case uses the UKF to
identify the system as the first step, and then with the estimated dynamics, the tracking of
the system dynamics related to the DC motor speed action is accomplished. The tracking
voltage Vc r is the reference obtained with the reference generator block, and Figure 19
shows the tracking of accomplished voltage related to the motor speed trajectory. Moreover,
we can highlight that when the speed in the motor increases, the voltage x1,k decreases. On
the other hand, when the speed of the motor decreases, the voltage increases.
In addition, Figure 20 illustrates the performance of the current dynamics during the
regulation with NSMC. The current reference (ire f erence ) is obtained through the reference
generator, as described in Figure 5.
450
Voltage(Volt)
400
350
300
0 10 20 30 40 50
t [Sec]
Figure 19. AES voltage during regenerative braking with UKF.
After that, the trajectory tracking is acquired over the 50 s of simulation time. The SOC
of the battery has been compared with and without AES. The importance of the AES is that
it allows for slow discharge on the battery. The results obtained are illustrated in Figure 21,
where the achieved signal in the AES and the regenerative braking system are operating,
which results in a better SOC in the battery. At the beginning of the simulation, the system
achieves the desired reference voltage in approximately 2 s, as seen in Figure 10. However,
during the rest of the signal tracking, the results overtake the SOC obtained without the
AES connected to the EV bank of batteries.
The battery voltage is compared in Figure 22 with and without the AES. From the
results, we can observe that the battery decreases its voltage when the AES is not connected.
Moreover, the voltage in the battery increases and maintains its voltage value above the
initial 500v when the AES is connected.
1000
Current(Amp)
0
−1000
0 10 20 30 40 50
t [Sec]
Figure 20. AES current during regenerative braking with UKF.
80.1
State of Charge(%)
80
79.9
With AES Without AES

79.8
0 10 20 30 40 50
t [Sec]
Figure 21. SOC comparison of the battery with and without AES using UKF.
550
540
Voltage (Volts)
530
520
510
500
490
0 10 20 30 40 50
t [Sec]
Figure 22. Comparison of the battery bank voltage with and without AES using UKF.
6.5. Regenerative Braking System Control Using a Reference Generator with EKF
To compare the results later, we used the identical signals in this section of the project’s
tests as we did with the UKF. Figure 23 illustrates that the first trajectory tracking of the
Vcr has a difference in the adjustment during the first seconds due to the initial conditions
of x1,k and has another behavior compared with the one in the Figure 19. Similarly, the
ire f erence current tracking trajectory displays a similar behavior in the same section of the
simulation.
The SOC and voltage in the battery are described in Figures 24 and 25, respectively,
and display a similar behavior as the NSMC with the UKF. However, the differences in
NSMC with UKF and NSMC with EKF are quantitatively compared in Section 7.
450
Voltage(Volt)
400
350
300
0 10 20 30 40 50
Current(Amp)
2000
0
−2000
0 10 20 t [Sec] 30 40 50
Figure 23. AES voltage and current during regenerative braking using EKF.
80.1
State of Charge(%)
80
79.9

79.8
0 10 20 30 40 50
t [Sec]
Figure 24. MES battery bank SOC comparison with and without AES using EKF.
550
540
Voltage (Volts)
530
520
510
500
490
0 10 20 30 40 50
t [Sec]
Figure 25. MES battery bank voltage comparison with and without AES using EKF.
6.6. Robustness Test

The reliability of a control system is very important to assure the correct performance
of the proposed scheme during variations on the system and disturbances that are not
considered initially in the system’s proposed model. To validate the robustness of the
NSMC, two scenarios have been applied. The first one is the system operated by adding
noise to the buck–boost converter using a Gaussian random signal, while the second is a
variation of the parameters of the buck–boost converter circuit.
6.6.1. Gaussian Noise

The Gaussian random signal is used to generate random values that will help to
validate the robustness of the NSMC, and for this test, the chosen mean value is equal to
0 and variance is 1. The noise is added to the measured voltage and current output, and
Equations (25)–(28) of the buck–boost converter model, considering the Gaussian noise
signal, can be defined as
ts ts
x1,k+1 = (1 − ) x1,k − x2,k + Gn (47)
RC c
ts
x2,k+1 = x2,k + Ubtt uc + Gn (48)
L
The buck converter model is also defined as
ts ts
x1,k+1 = (1 − ) x + x2,k + Gn (49)
RC 1,k c
ts
x2,k+1 = x2,k + Ubtt uc + Gn (50)
L
where Gn is the Gaussian noise added to the dynamics of the buck–boost converter.
Robustness Test with NSMC and UKF
The voltage dynamics affected by the Gaussian noise signal are depicted in Figure 26 a,b,
which are a closer view of the real signal compared with the signal with noise. Additionally,
Figure 27 describes the dynamics of the current, where the noise signal in (a) and (b) is a
closer look at the signal affected by the Gaussian noise.
The control signal of the voltage dynamics is illustrated in Figure 28 and the current
dynamics in Figure 29. The results demonstrated the robustness of the NSMC using UKF
where the desired trajectory is followed still without issues. Additionally, in Section 7, the
MSE and EES analysis supports the illustrated results.
(a)
450
(b)
410
Voltage(Volt)
400
Voltage(Volt)
350
400
0.3 t[Sec] 1
300
0 10 20 30 40 50
t [Sec]
Figure 26. (a) Voltage dynamics with Gaussian noise signal. (b) is a zoom of (a).
(a)
1000
(b)
500 550
Current(Amp)
Current(Amp)
−500
−1000
520
0.5 t [Sec] 2
−1500
0 10 20 30 40 50
t [Sec]
Figure 27. (a) Current dynamics with Gaussian noise signal. (b) is a zoom of (a).
450
Voltage(Volt)
400
350
300
0 10 20 30 40 50
t [Sec]
Figure 28. Tracking trajectory with noise signal for voltage x1 with UKF.
1000
500
Current(Amp)
−500
−1000
−1500
0 10 20 30 40 50
t [Sec]
Figure 29. Tracking trajectory with noise signal for current x2 with UKF.
Robustness Test with NSMC and EKF

The robustness of the NSMC with the EKF has been tested for the current and voltage
of the buck–boost converter. Figures 30 and 31 are the signals of the dynamics with the
Gaussian noise added to the output signal. Additionally, Figures 32 and 33 illustrate good
results in the control of the voltage and current with the noise added to the converter.
(a)
450
(b)
430
Voltage(Volt)
Voltage(Volt)
400
350
410
0.3 t [Sec] 1
300
0 20 40
t [Sec]
Figure 30. (a) Voltage dynamics with Gaussian noise signal. (b) is a zoom of (a).
(a)
1000
(b)
730
Current(Amp)
Current(Amp)
0
−1000
710
0.5 t [Sec] 1.5
0 20 40
t [Sec]
Figure 31. (a) Current dynamics with Gaussian noise signal. (b) is a zoom of (a).
450
Voltage(Volt)
400
350
300
0 10 20 30 40 50
t [Sec]
Figure 32. Tracking trajectory with noise signal for voltage x1 with EKF.
1000
500
Current(Amp)
−500
−1000
−1500
0 10 20 30 40 50
t [Sec]
Figure 33. Tracking trajectory with noise signal for current x2 with EKF.
Robustness Test with PI

Using the same signal illustrated in Figures 30 and 31, a PI controller based on the
control strategy presented in [14], which is not an intelligent controller because the neural
identifier is not performed to estimate the dynamics of the buck–boost converter, is used
for the robustness test. The voltage and current are controlled each, with a PI controller
resulting in some issues for the dynamics control, as illustrated in Figures 34 and 35.
The results demonstrated that the PI is not completely reliable for the tracking trajectory
considering Gaussian noise. In addition, Section 7 presents the MSE and EES for the three
configurations of the NSMC.
450
Voltage(Volt) 400
350
300
250
0 10 20 30 40 50
t [Sec]
Figure 34. Tracking trajectory with noise signal for voltage x1 with PI.
2000
Currrent(Amp)
−2000
−4000
0 10 20 30 40 50
t [Sec]
Figure 35. Tracking trajectory with noise signal for current x2 with PI.
6.6.2. Changes in the Parameters of the Buck–Boost Converter

The second scenario consists of changing the buck–boost converter parameters, which
have been used during the different stages of the paper without changes. Now, a variation
of its capacitance values is applied to verify the robustness of the NSMC. The new values
are illustrated in Table 5. The same NSMC with UKF and EKF is used for the robustness
test. The PI controller arrangement of the first robustness test is performed again in this
second scenario.
Table 5. Parameters of the AES and MES with changes for the robustness test.
Description Unit
Buck–boost converter resistance R 50 Ω
Buck–boost converter inductance L 13 mH
Buck–boost converter capacitance C1 14 mF
Buck–boost converter capacitance C2 6µF
Supercapacitor voltage Vsc 350 V
Battery bank voltage Vc 500 V
Initial SOC 80%
Sampling time (ts ) 0.1 µ seg
NSMC with UKF for the tracking of dynamics with changes in buck-boost converter
The results of the validation of the NSMC are illustrated in Figures 36 and 37 for
voltage and current, respectively. The NSMC still regulates the dynamics of the buck–boost
converter, ensuring the energy consumption.
450
Voltage(Volt)
400
350
300
0 10 20 30 40 50
t [Sec]
Figure 36. Tracking trajectory with changes in the buck–boost converter for voltage x1 with UKF.
1000
500
Current(Amp)
−500
−1000
−1500
0 10 20 30 40 50
t [Sec]
Figure 37. Tracking trajectory with changes in the buck–boost converter for current x2 with UKF.
NSMC with EKF for the tracking of dynamics with changes in buck–boost converter
The same parameters described in Table 5 are used to validate the robustness of the
NSMC with EKF illustrated in Figures 38 and 39. As expected, the controller still regulates
the dynamics of the buck–boost converter despite the changes in its parameters.
450
Voltage(Volt)
400
350
300
250
0 10 20 30 40 50
t [Sec]
Figure 38. Tracking trajectory with changes in the buck–boost converter for voltage x1 with EKF.
(a)
1000
Current(Amp)
0
−1000
0 10 20 30 40 50
t [Sec]
Figure 39. Tracking trajectory with changes in the buck–boost converter for current x2 with EKF.
The results obtained from the test to validate the robustness of the NSMC demonstrate
its correct operation under non-ideal scenarios. This includes scenarios where noise and
other signals not initially considered are added to the signal for regulation. However, it
is important to note that the NSMC, when used with UKF and EKF, operates differently,
and provides results that require further analysis, as does the PI controller. In Section 7, we
conduct a detailed analysis where we calculate the MSE and EES.
7. Comparative Analysis between Proposed NSMC Variants and PI Controller

To validate the simulation results and draw a comprehensive conclusion of the work
presented in this paper, it is essential to compute the MSE and EES values. The primary
objective of this comparison is to determine which of the two filtering techniques, UKF
or EKF, provides better results for the electric vehicle regenerative braking system by
calculating a numerical value. Furthermore, the results obtained are compared with those
of a PI controller with the same structure in [14].
The MSE is obtained as follows:
n
1
MSE =
n ∑ (Yi − Ŷi )2 (51)
i =1
and the EES is calculated as follows:

n
EES = ∑ (Yi − Ŷi )2 (52)
i =1
where MSE is the calculated mean square error, EES is the calculated energy of the error signal,
and in both cases, n is the number of data points to evaluate the computation, Yi represents
the values obtained evaluated at the data points, and Ŷi represents the reference values at the
data points.
In this case, we use the values of the proposed reference signal and the real values of
the states x1 and x2 to obtain the following results.
7.1. MSE and EES of a Time-Varying Signal Controlled with NSMC

First, the MSE and EES results correspond to the control results obtained with the time-
varying signal of Sections 6.2 and 6.3. The tracking trajectory with the NSMC using EKF
performs better than the NSMC using UKF and PI. On the other hand, the PI controller has
a better outcome than the NSMC with UKF in the tracking trajectory for the x1,k dynamic.
Figure 40 presents the three methods compared for the x1,k dynamic, while Figure 41 shows
the three methods applied in the track trajectory for x2,k .
Table 6 displays the MSE results for the x1 tracking trajectory, where the NSMC with
EKF has better results over the NSMC with UKF and PI. However, the PI controller offers
better performance than the NSMC with UKF.
380
Voltage(Volt) 370
360
350
340
0 10 20 t [Sec] 30 40 50
Figure 40. Tracking trajectory with the three controllers for voltage X1 .
200
100
Current(Amp)
−100
−200
0 10 20 t [Sec] 30 40 50
Figure 41. Tracking trajectory with the three controllers for current X2 .
Moreover, MSE and EES results for the x2 tracking trajectory are illustrated in Table 7,
where EKF again presents better performance compared to the NSMC with UKF and PI.
However, PI again illustrates better results over the NSMC with UKF.
Table 6. MSE and EES in x1 without regenerative braking system.
MSE and EES of Tracking Trajectories in x1

Controller MSE Value EES Value
NSMC with UKF 72.9529 3.647 × 108
NSMC with EKF 1.0585 5.2923 × 106
PI 1.7403 8.7016 × 106
The error signals in both dynamics of the system are demonstrated in Figures 42–44
taking the results obtained during the tracking trajectory with UKF, EKF, and PI, respec-
tively. The x1 error signal shows that the values obtained remain around zero throughout
the entire simulation. Although the error signal in x2 has some instants where the value
moves away from zero, the error values remain around zero.
Table 7. MSE and EES in x2 without regenerative braking system.

NSMC with UKF 1.0427 × 104 5.2134 × 1010
NSMC with EKF 0.1362 6.8084 × 105
PI 270.1132 1.3506 × 109
error
Current(Amp) Voltage(Volt)
20
0
−20
0 10 20 30 40 50
20
0
−20
0 10 20 t [Sec] 30 40 50
Figure 42. Error in both dynamics with UKF.
error
20
−20
0 10 20 30 40 50
20
−20
0 10 20
t [Sec] 30 40 50
Figure 43. Error in both dynamics with EKF.
error
20
0
−20
0 10 20 30 40 50
20
0
−20
0 10 20
t [Sec] 30 40 50
Figure 44. Error in both dynamics with PI.

7.2. MSE and EES of the Complete Regenerative Braking System Controlled with NSMC
In this part, the compared results have been taken when the whole connected re-
generative braking system is operating. As explained in Section 6, the results Vcr and
ire f erence are from the reference generator block, using the parameters provided during
the acceleration and deceleration of a DC motor. Two figures of the tracking trajectory,
considering the parameters of the described scenario, are illustrated in Figures 45 and 46
for x1 and x2 , respectively.
450
400
Voltage (Volt)
350
300
250
0 10 20 t[Sec] 30 40 50
Figure 45. Tracking trajectory with the three controllers for voltage X1 with regenerative braking system.
1000
Current (Amp)
−1500
0 10 20 t [Sec] 30 40 50
Figure 46. Tracking trajectory with the three controllers for current X2 with regenerative braking system.
In Table 8, the tracking of the x1 voltage trajectory with the regenerative braking
system is presented. In addition, the NSMC with the EKF presents significantly better
performance than the NSMC with the UKF and the PI controller. These results are different
concerning those presented in Tables 6 and 7; the NSMC with UKF showed better results
compared to the PI.
Table 8. MSE and EES in x1 with regenerative braking system.

NSMC with UKF 45.2714 2.2636 × 108
NSMC with EKF 1.0321 5.1604 × 106
PI 132.4538 6.6227 × 108
In Table 9, the tracking of the x2 current trajectory is presented. Additionally, NSMC

with UKF performs much better than NSMC with the EKF and PI controller.
Table 9. MSE and EES in x2 with regenerative braking system.

NSMC with UKF 1.9679 9.8395 × 106
NSMC with EKF 8.9223 4.4611 × 107
PI 1.8035 × 105 9.0176 × 1011
With the results obtained during the UKF, EKF, and PI tracking trajectory, the error
signals of the EV system dynamics are demonstrated in Figures 47–49. The x1 error signal
shows that the values obtained remain around zero during most of the simulation and
present only a small adjustment at the beginning. On the other hand, the x2 error values
also present an adjustment during the first seconds of the simulation, but eventually, the
value of the error is maintained around zero.
error
100
0
−100
0 10 20 30 40 50
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 47. Error in both dynamics during regenerative braking with UKF.
error
200
0
−200
0 10 20 30 40
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 48. Error in both dynamics during regenerative braking with EKF.
7.3. MSE and EES of Robustness Test

The results of the MSE and EES of the robustness test presented in Section 6 are
presented. Two new tables were added to compare the obtained results. Table 10 presents
the results of the voltage dynamics with Gaussian noise for UKF, EKF, and PI. Furthermore,
Table 11 shows the MSE and EES of the tracking trajectory with the same additional noise
to the signal. In addition, similar to the case in Table 8, the NSMC with EKF presents
significantly better performance than the NSMC with UKF and the PI controller for x1 .
error

400
200
0
−200
0 10 20 30 40 50
4000
2000
0
−2000
−4000
0 10 20 t [Sec] 30 40 50
Figure 49. Error in both dynamics during regenerative braking with PI.
Table 10. MSE and EES in x1 with Gaussian noise.
MSE and EES of Tracking Trajectories with Gaussian noise in x1

NSMC with UKF 73.4682 3.6734 × 108
NSMC with EKF 45.7588 2.2879 × 108
PI 132.9545 6.65 × 108
Table 11. MSE and EES in x2 with Gaussian noise.
MSE and EES of Tracking Trajectories with Gaussian noise in x2

NSMC with UKF 1.0435 × 104 5.2174 × 1010
NSMC with EKF 2.780 × 104 1.390 × 1010
PI 1.80 × 105 9.02 × 1011
Additionally, in the obtained results the NSMC with the UKF performs much better
than NSMC with the EKF and PI controller for x2 . The error signals of the system dynamics
with Gaussian noise and UKF, EKF and PI are demonstrated in Figures 50–52.
error
100
Current (Amp) Voltage(Volt)
−100
0 10 20 30 40 50
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 50. Error in both dynamics with Gaussian noise with UKF.
error
200
−200
0 10 20 30 40 50
200
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 51. Error in both dynamics with Gaussian noise with EKF.
Error
200
−200
0 10 20 30 40 50
4000
2000
0
−2000
−4000
0 10 20 t [Sec] 30 40 50
Figure 52. Error in both dynamics with Gaussian noise with PI.
7.4. MSE and EES of Robustness Test with Changes in the Buck–Boost Converter Parameters
The MSE and EES obtained for the test made with the parameters of the buck–boost
converter changed in Section 6 are presented in Table 12 for the voltage dynamics and in
Table 13 for the current dynamics. As a result, regarding the voltage x1 , similar to the case
in Table 8 the NSMC with EKF presents significantly better performance than the NSMC
with UKF and the PI controller for x1 .
Table 12. MSE and EES in x1 with changes in buck–boost converter parameters.
MSE and EES with Changes in Buck–Boost Converter Parameters in x1

NSMC with UKF 73.7266 3.6863 × 108
NSMC with EKF 46.0728 2.3036 × 108
PI 134.9163 6.7458 × 108
Furthermore, the MSE and EES for the x2 current dynamics are depicted in Table 13,
where the results conclude that the NSMC with UKF performs much better than NSMC
with the EKF and PI controller for x2 . The error signals for the system dynamics with
the UKF, EKF and PI with the buck–boost converter parameters changes are illustrated in
Figures 53–55.
Table 13. MSE and EES in x2 with changes in buck–boost converter parameters.
MSE and EES with Changes in Buck–Boost Converter Parameters in x2

NSMC with UKF 1.0488 × 104 5.2439 × 1010
NSMC with EKF 2.8186 × 103 1.4093 × 1010
PI 1.8475 × 105 9.2374 × 1011
error
100
−100
0 10 20 30 40 50
0
−200
−400
0 10 20 30 40 50
Figure 53. Error in both dynamics with changes in buck–boost converter parameters with UKF.
error
100
−100
0 10 20 30 40 50
0
−200
−400
0 10 20 t [Sec] 30 40 50
Figure 54. Error in both dynamics with changes in buck–boost converter parameters with EKF.
Error
Current(Amp) Voltage (Volt)
200
−200
0 10 20 30 40 50
4000
2000
0
−2000
−4000
0 10 20 30 40 50
t [Sec]
Figure 55. Error in both dynamics with changes in buck–boost converter parameters with PI.
8. Discussion
This paper presented the comparison of the performance of three controllers: NSMC
with EKF, NSMC with UKF, and the PI controller. The novelty of the paper is the design
and application via simulation of a hybrid intelligent control scheme using sliding modes
and RHONNs trained with EKF and UKF. Our stated control strategy manages the AES,
defined as a combination of a supercapacitor and buck–boost converter, to recoup the
energy lost during braking and power the MES.
The proposed intelligent controller is a new strategy with better results than a PI
controller applied to the same bidirectional buck–boost converter circuit, which can be
used for a regenerative braking system in an EV. The performance of three controllers is
compared using two scenarios: the first one without considering a reference generator for
trajectory tracking and the second one with a reference generator for trajectory tracking for
regenerative braking. For both scenarios, NSMC trained with EKF has better performance,
improving the SOC of the battery of the EV.
The states of the AES have been approximated successfully using an RHONN identifier
trained by an EKF and UKF for different operation modes. The decision to use an RHONN
for the identifier was based on the nonlinear systems dynamics estimation facilities. The
RHONN has been validated with a chirp signal for both training algorithms.
To make an extensive study of the NSMC and the EKF and UKF training algorithms,
different tests have been achieved. The identifier using EKF and UKF is implemented for
tracking trajectories without the DC motor considered in the EV system during different
AES operation modes. The system identification and synaptic weight adjustment figures
help to corroborate the correct performance of the RHONN identifier.
The NSMC takes the estimated states from the RHONN identifier to define the sliding
manifold where the controller will operate. The objective of this controller is to manage the
DC buck–boost converter’s dynamics during charging and discharging operation modes.
In the first scenario, time-varying signals are employed to validate the NSMC without still
taking into consideration the DC motor in the EV system but considering the system’s
operation range.
In a second scenario, the NSMC is implemented to operate with a completely con-
nected regenerative braking system where the speed of the DC motor is the main parameter
to define a reference with the desired trajectory to provide the necessary energy to supply
the EV during the acceleration and deceleration. Although this research focuses mainly on
simulating the proposed control scheme, from this perspective, this particular case enables
us to confirm the effectiveness of the suggested NSMC in a scenario that is closer to the
actual operation of regenerative braking.
Additionally, in both cases, the NSMC was shown to enhance the EV regenerative
braking system when the car is accelerating and decelerating. Furthermore, the neural
controller with UKF and EKF present different and motivating results with and without the
reference generator block but the one with better results has been the NSMC with EKF. The
most important advantage obtained is in the SOC, which presents a better performance
where the percent of charge decreases slowly with the regenerative braking system in
comparison without the regenerative braking system.
9. Conclusions
The MSE results obtained in Section 7 allow us to determine the framework in which
EKF and UKF offer better control synthesis results than a conventional control technique
such as PI. This can be a result of the favorable differences between the two scenarios
presented in the simulation. The first case employed a time-varying signal with bounded
values to track, while the second case used a signal that is the result of other dynamics
that can induce fast adjustments to achieve the EV requirements during the acceleration
and deceleration.
To validate the robustness of the NSMC, two scenarios have been applied. The first
one is the system operated adding a Gaussian random signal, where the mean value is
equal to 0 and the variance is 1 for x1 and x2 .
The noise is added to the measured voltage and current output. The second scenario
involves varying parameters on the converter buck–boost circuit. Results are consistent
with the tracking trajectory without the Gaussian noise and the variation of the parameters.
We are encouraged to extend the results of this study to obtain an NIOC based on the
UKF algorithm to train a RHONN identifier and to control the AES dynamics enhancing
the regenerative braking and energy consumption behavior in more scenarios, as well as to
carry out a comparative study between NSMC and NIOC based on Kalman filters.
As a future work, the authors and coworkers will continue this project by considering
the parameters and dimensions of an EV, as well as the typical driving cycles and new
challenges and problems to solve.
Author Contributions: Conceptualization, J.A.R.-H.; methodology, J.A.R.-H.; software, M.A.R.C.,

R.G.-H. and J.F.G.; investigation, J.R.V.-F., M.A.R.C. and J.A.R.-H.; formal analysis, M.A.R.C., R.G.-H.,
J.F.G. and J.A.R.-H.; validation, J.A.R.-H., J.-L.R.-L. and R.G.-H.; writing—original draft prepa-
ration, M.A.R.C.; writing—review and editing, M.A.R.C., J.F.G., J.A.R.-H., R.G.-H., J.R.V.-F. and
J.-L.R.-L.; funding acquisition, J.A.R.-H. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was funded by Universidad Autónoma del Carmen. Project number FING/
1ERP2023/03.
Data Availability Statement: Data are contained within the article.
Acknowledgments: The first author thanks to Universidad Autonoma del Carmen for the support
and providing the facilities to develop this project.
Conflicts of Interest: The authors declare no conflicts of interest.
Abbreviations
The following abbreviations are used in this manuscript:
NSMC Neural Sliding Mode Controller

SMC Sliding Mode Controller
NIOC Neural Inverse Optimal Controller
IOC Inverse Optimal Control
PWM Pulse Width Modulation
UKF Unscented Kalman Filter
EKF Extended Kalman Filter
EV Electric Vehicle
RHONN Recurrent High-Order Neural Network
MES Main Energy System
AES Auxiliary Energy System
DC Direct Current
PI Proportional–Integral
SOC State of Charge
MSE Mean Squared Error
EES Energy of the Error Signal
Variable Description
i = 1, 2, 3, . . . , n index for name states
k = 1, 2, 3, . . . , n index for name RHONN states
xi,k neural network adjustable synaptic weights
ωi,k fixed weights
T
xk = x1,k , x2,k , . . . , xn,k state vector
state of the ith neuron that identifies the ith component of
Xi,k+1
the state vector xk
ϕi , φi linear functions of the state vector

uk vector input of the RHONN model
dij,k non-negative integers
Li number of the connection
I Li collection of non-ordered subsets 1, 2, . . . , n + m
n the state dimension
m the input dimension
ζi defined as [23]
Extended Kalman Filter
ei ∈ R the identification error to be minimized
ηi a parameter of the training algorithm design
Ki,k ∈ R Li × Li the Kalman gain matrix
Qi,k ∈ R Li × Li the measurement noise covariance matrix
Ri,k ∈ Rm×m the state noise covariance matrix
Pi ∈ R Li × Li the prediction error matrix
the measurement matrix, which is defined as the resulting
Hi ∈ R Li ×m state derivative with respect to the adjustable weights of
the neural identifiers
Unscented Kalman Filter
j
ei,k the identification error
j
Pi,k the prediction error covariance matrix
jyy
Pi,k the covariance of predicted output matrix
jxy
Pi,k the output matrix
j
wi,k the j-th weight of the i-th subsystem
jc
ηi a parameter of design
j
Xi,k the j-th state of the plant
j
xi,k the j-th state of the neural network
L the number of states
j
Ki,k the Kalman gain matrix
j
Qi,k the measurement noise covariance matrix
j
Ri,k the state noise covariance matrix
j−
xi,k the predicted state mean
j−
yi,k the predicted output mean
j
Xi,k|k−1 sigma-points propagated through prediction
j
Yi,k|k−1 sigma-points propagated through observation
Buck–Boost Converter Model
x1,k the buck–boost converter output voltage
x2,k the buck–boost converter output current
Ubtt the voltage in the battery
L the inductance in henry (H)
R the load resistance in ohms (Ω)
C the capacitor in farads (F)
ts the sample time
Gn Gaussian Noise
References
1. Wang, N.; Tang, G. A Review on Environmental Efficiency Evaluation of New Energy Vehicles Using Life Cycle Analysis.
Sustainability 2022, 14, 3371. [CrossRef]
2. Hantanasirisakul, K.; Sawangphruk, M. Sustainable Reuse and Recycling of Spent Li-Ion batteries from Electric Vehicles:
Chemical, Environmental, and Economical Perspectives. Glob. Chall. 2023, 7, 2200212. [CrossRef] [PubMed]
3. Yoong, M.K.; Gan, Y.; Gan, G.; Leong, C.; Phuan, Z.; Cheah, B.; Chew, K. Studies of Regenerative Braking in Electric Vehicle; IEEE:
Kuala Lumpur, Malaysia, 2010; pp. 40–45. [CrossRef]
4. Li, Z.; Shi, Z.; Gao, J.; Xi, J. Research on Regenerative Braking Control Strategy for Single-Pedal Pure Electric Commercial Vehicles.
World Electr. Veh. J. 2023, 14, 229. [CrossRef]
5. Indragandhi, V.; Selvamathi, R.; Gunapriya, D.; Balagurunathan, B.; Suresh, G.; Chitra, A. An efficient regenerative braking
system based on battery-ultracapacitor for electric vehicles. In Proceedings of the 2021 Innovations in Power and Advanced
Computing Technologies (i-PACT), Kuala Lumpur, Malaysia, 27–29 November 2021; pp. 1–5. [CrossRef]
6. Khalid, H.; Mekhilef, S.; Mubin, M.B.; Seyedmahmoudian, M.; Stojcevski, A.; Rawa, M.; Horan, B. Analysis and design of
series-LC-switch capacitor multistage high gain DC-DC boost converter for electric vehicle applications. Sustainability 2022,
14, 4495. [CrossRef]
7. Utkin, V. Sliding mode control of DC/DC converters. J. Frankl. Inst. 2013, 350, 2146–2165. [CrossRef]
8. Pandey, K.K.; Kumar, M.; Kumari, A.; Kumar, J. Bidirectional DC-DC buck-boost converter for battery energy storage system and
PV panel. In Proceedings of the Modeling, Simulation and Optimization: Proceedings of CoMSO 2020, Silchar, India, 3–5 August
2020; Springer: Singapore, 2021; pp. 681–693. [CrossRef]
9. Hosseinzadeh, M.; Salmasi, F.R. Determination of maximum solar power under shading and converter faults—A prerequisite for
failure-tolerant power management systems. Simul. Model. Pract. Theory 2016, 62, 14–30. [CrossRef]
10. Mobayen, S.; Bayat, F.; Lai, C.C.; Taheri, A.; Fekih, A. Adaptive global sliding mode controller design for perturbed DC-DC buck
converters. Energies 2021, 14, 1249. [CrossRef]
11. Montoya, O.D.; Gil-González, W.; Riffo, S.; Restrepo, C.; González-Castaño, C. A Sensorless Inverse Optimal Control plus Integral
Action to Regulate the Output Voltage in a Boost Converter Supplying an Unknown DC Load. IEEE Access 2023, 11, 49833–49845.
[CrossRef]
12. Priya, R.P.; Sakile, R. State of charge estimation of lithium-ion battery based on extended Kalman filter and unscented Kalman
filter techniques. Energy Storage 2023, 5, e408. [CrossRef]
13. Zhang, L.; Cai, X. Control strategy of regenerative braking system in electric vehicles. Energy Procedia 2018, 152, 496–501. [CrossRef]
14. Manríquez, E.Q.; Sanchez, E.N.; Toledo, M.E.A.; Muñoz, F. Neural control of an induction motor with regenerative braking as
electric vehicle architecture. Eng. Appl. Artif. Intell. 2021, 104, 104275. [CrossRef]
15. Salari, A.H.; Mirzaeinejad, H.; Mahani, M.F. A new control algorithm of regenerative braking management for energy efficiency
and safety enhancement of electric vehicles. Energy Convers. Manag. 2023, 276, 116564. [CrossRef]
16. Guo, J.; Jian, X.; Lin, G. Performance evaluation of an anti-lock braking system for electric vehicles with a fuzzy sliding mode
controller. Energies 2014, 7, 6459–6476. [CrossRef]
17. Hwang, M.H.; Lee, G.S.; Kim, E.; Kim, H.W.; Yoon, S.; Talluri, T.; Cha, H.R. Regenerative braking control strategy based on AI
algorithm to improve driving comfort of autonomous vehicles. Appl. Sci. 2023, 13, 946. [CrossRef]
18. Chen, Z.; Xiong, R.; Cai, X.; Wang, Z.; Yang, R. Regenerative Braking Control Strategy for Distributed Drive Electric Vehicles
Based on Slope and Mass Co-Estimation. IEEE Trans. Intell. Transp. Syst. 2023, 24, 14610–14619. [CrossRef]
19. Ruz-Canul, M.A.; Djilali, L.; Ruz-Hernandez, J.A.; Sanchez-Camperos, E.N. Neural Sliding mode control of a regenerative braking
system for electric vehicles. J. Innov. Des. 2022, 6, 6–15. [CrossRef]
20. Ruz-Hernandez, J.A.; Djilali, L.; Ruz Canul, M.A.; Boukhnifer, M.; Sanchez, E.N. Neural Inverse Optimal Control of a Regenerative
Braking System for Electric Vehicles. Energies 2022, 15, 8975. [CrossRef]
21. Djilali, L.; Sanchez, E.N.; Ornelas-Tellez, F.; Avalos, A.; Belkheiri, M. Improving Microgrid Low-Voltage Ride-Through Capacity
Using Neural Control. IEEE Syst. J. 2020, 14, 2825–2836. [CrossRef]
22. Sanchez, E.N.; Alanis, A.Y.; Loukianov, A.G. Discrete-Time High Order Neural Control: Trained with Kalman Filtering; Springer
Science & Business Media: Cham, Switzerland, 2008. [CrossRef]
23. Alanis, A.Y.; Sanchez, E.N.; Loukianov, A.G. Discrete-time adaptive backstepping nonlinear control via high-order neural
networks. IEEE Trans. Neural Netw. 2007, 18, 1185–1195. [CrossRef]
24. Majumder, R.; Ghosh, A.; Ledwich, G.; Zare, F. Power management and power flow control with back-to-back converters in a
utility connected microgrid. IEEE Trans. Power Syst. 2009, 25, 821–834. [CrossRef]
25. Utkin, V.; Guldner, J.; Shi, J. Sliding Mode Control in Electro-Mechanical Systems; CRC Press: Boca Raton, FL, USA, 2017. [CrossRef]
26. Sanchez, E.; Djilali, L. Neural Control of Renewable Electrical Power Systems; Springer Nature: Cham, Switzerland, 2020. [CrossRef]
27. Chui, C.K.; Chen, G.; Chui, C.K.; Chen, G. Extended Kalman filter and system identification. In Kalman Filtering: With Real-Time
Applications; Springer: Cham, Switzerland, 2017; pp. 115–137. [CrossRef]
28. Wan, E.A.; Van Der Merwe, R. The unscented Kalman filter for nonlinear estimation. In Proceedings of the Proceedings of the
IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No. 00EX373), Lake Louise,
AB, Canada, 4 October 2000; pp. 153–158. [CrossRef]
29. Rovithakis, G.A.; Chistodoulou, M.A. Adaptive Control with Recurrent High-Order Neural Networks: Theory and Industrial Applications;
Springer Science & Business Media: Cham, Switzerland, 2012. [CrossRef]
30. Julier, S.J.; Uhlmann, J.K. New extension of the Kalman filter to nonlinear systems. In Proceedings of the Signal Processing,
Sensor Fusion, and Target Recognition VI, Orlando, FL, USA, 21–25 April 1997; SPIE: Bellingham, WA, USA, 1997; Volume 3068,
pp. 182–193. [CrossRef]
31. Guerra, J.F.; Garcia-Hernandez, R.; Llama, M.A.; Santibañez, V. UKF-Based Neural Training for Nonlinear Systems Identification
and Control Improvement. IEEE Access 2022, 10, 114501–114513. [CrossRef]
32. Zohuri, B. Scalar Wave Driven Energy Applications; Springer: Cham, Switzerland, 2019. [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.

Wevj 15 00048

Uploaded by

Copyright:

Available Formats

Wevj 15 00048

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wevj 15 00048

Uploaded by

Copyright:

Available Formats

Article

Neural Sliding Mode Control of a Buck-Boost Converter Applied

Academic Editors: Jianfei Chen, Liyan 1. Introduction

World Electr. Veh. J. 2024, 15, 48. https://doi.org/10.3390/wevj15020048 https://www.mdpi.com/journal/wevj

The combination of a supercapacitor and battery bank as an energy supply for an EV

2. Materials and Methods

• The system identification is developed taking the RHONN philosophy in [22,23] to

3. DC-DC Bidirectional Buck–Boost Converter

Here, L is an inductor, C is a capacitor, T1 and T2 represent the insulated gate bipolar

Figure 2. Continuous-time system trajectory when SMC is applied.

Evaluating at sample time (k + 1), the sliding manifold is determined:

sk+1 = F ( xk , uk ) − xre f ,k+1 (9)

Now is calculated the equivalent control ueq ( xk , k ), as in [25]:

A stabilizing term, determined as un ( xk , k), should be added to asymptotically achieve

i f ∥uc ( xk , k)∥ < u0

where ∥•∥ represents the Euclidean norm of uc ( xk , k) and is defined as:

4.2. Discrete-Time Recurrent High-Order Neural Networks

χi,k+1 = ωiT ϕi ( xk ) + ω iT φi ( xk , uk ) (14)

T ] u is the network’s input vector. Additionally, the hyperbolic

Figure 3. Scheme of a RHONN identifier.

4.3. Extended Kalman Filter

wi,k+1 = wi,k + ηi Ki,k ei,k (18)

where ei ∈ R is defined as the identification error to be minimized, ηi is a parameter

4.4. Unscented Kalman Filter

Table 1. Differences between EKF and UKF training algorithms.

Extended Kalman Filter (EKF) Unscented Kalman Filter (UKF)

5. System Modeling and Sliding Mode Neural Control

5.2. Neural Controller Design

χ̂1,k = ω11,K S( x1,K ) + ω12,K S( x2,K )

Figure 4. Neural identification representation strategy.

5.3. Discrete Time Sliding Mode Control

s x1,k+1 = ω2,1 (k )S( x2 ) + ω2,2 (k )S( x1 )

The equivalent x̂1,k control is calculated as follows:

i f ∥uc ( x1 , k)∥ < u0

s x2,k+1 = ω2,1 (k)S( x2 ) + ω2,2 (k)S( x1 )

Then, the equivalent current control is obtained as follows:

and the NSMC is implemented as follows

i f ∥uc ( x2 , k)∥ < u0

5.4. Reference Generator Development

The following equation defines the supercapacitor energy during discharging:

Super Capacitor Buck-Boost Converter x!

Figure 5. Regenerative braking system control representation.

Table 2. Parameters of the AES and MES.

6.1. RHONN Validation

Table 3. Chirp signal parameters.

Chirp Signals Parameters

6.2. NSMC Trajectory Tracking Results Using UKF

Figure 12. Weight adjustment with UKF.

Figure 13. System dynamics identification with UKF.

6.3. NSMC Trajectory Tracking Results Using EKF

Figure 16. Weight adjustment with EKF.

Table 4. Parameters of the DC motor.

DC motor speed control

With AES Without AES

With AES Without AES

6.6. Robustness Test

6.6.1. Gaussian Noise

Robustness Test with NSMC and EKF

Robustness Test with PI