Improvement of Power Quality of Grid-Connected EV Charging Station Using Grid-Component Based Harmonic Mitigation Technique ^†

Abstract

Conventional approaches for designing distribution grids are often time-consuming and computationally expensive. To minimize power harmonics in a low-voltage network, there is a dire need of in-depth mathematical and technical calculations for each electrical equipment involved in the modeling of a distribution grid. In this study, a time- and resource-efficient distribution grid model is proposed, which is capable of improving power-quality impact of electric vehicle charging infrastructure. The proposed method uses mathematical equations, field measurement, data from equipment manufacturers, and distribution network operators to develop precise distribution grid model for the integration of bidirectional electric vehicle charging infrastructure. To prove the effectiveness of the proposed model, power-quality analysis of electric vehicle charging stations is conducted in the MATLAB/Simulink environment. As a result, the grid voltage THD has improved to 0.05% while the grid-connected current THD obtained is 0.88%. This signifies that by varying selection of technical parameters of electrical components of a distribution grid, power losses resulting in the form of harmonics can be improved.

1. Introduction

1.1. Background

Greenhouse gas (GHG) emissions are the unwanted by-product associated with fuels for energy needs. With the rapid increase in carbon prints and climate change severity, the world is ready to embrace transport electrification because of decreased gasoline usage and greenhouse emissions due to the replacement of combustion engine vehicles [1]. It is predicted that the electric vehicle usage on roads will reach 64% in the US and 45% in Australia by the end of 2030 [2]. The European Union resolution to achieve an 80–95% greenhouse gas emission-reduction goal is set to be reached by 2050 [3], while China is taking active measures towards climate change goals [4]. The introduction of electric vehicles into the grid brings a lot of challenges in terms of voltage instability and harmonics. Therefore, before moving towards low-carbon technology in the grid, various countries around the world have shown their realization and commitment towards smart grid implementation [4]. On one hand, various researchers have shed light on the challenges hindering the road to smart grid realization [5], while others have highlighted and addressed the modelling issues, challenges, and impact of integrating low-carbon technologies, i.e., electric vehicles in the grid [6,7,8,9]. In addition, some others have worked on different topologies, infrastructure, and operation modes of electric vehicles for the integration of electric vehicles into the grid [10,11,12,13].

According to data from China Charging Alliance, by the end of 2017, a total of 300,000 public charging stations, comprising 190,000 AC charging stations and 110,000 DC charging stations, had been built nationwide. As of March 2019, Chongqing has built 1900 charging and battery swapping stations and 14,000 charging stations, including 2800 DC charging stations and 300 AC charging stations. This municipality has 246 urban public charging stations and 3145 charging stations. A total of 52 charging stations have been built in 13 expressway service areas, and the charging service system has been gradually improved. The construction of charging facilities generally progresses in an orderly manner, but the phenomenon of vehicles without stations and stations without vehicles exists at the same time and is more prominent. At present, the vehicle-to-station ratio of electric vehicles in China is only 3.8:1, and the development scale of charging stations lags. At the same time, the large-scale use of electric vehicles has brought a series of impacts on the power system. Although charging during the low-power-consumption period can improve the utilization rate of the power grid, uncoordinated charging during the peak power-consumption period will have a great impact on the power system, and the charging process will generate large harmonics, which will affect the power system, so the power quality will be greater.

To summarize, with the continuous expansion of the new energy vehicle market, the problem of insufficient structural supply of charging infrastructure has become increasingly prominent, and the overall scale of charging stations still significantly lags behind market demand. On one hand, the scale of charging stations does not match the development of new energy vehicles in the same period, and a large number of charging stations will need to be added in the next few years. On the other hand, when a large number of charging stations are connected to the power grid, it will affect the balance of power consumption and affect power quality, forming harmonic influence and causing dynamic pollution of the power grid.

1.2. Literature Review

The problem that arises with the introduction of EVs to the grid needs to be addressed. The new technologies play an important role in saving the cost of building power systems, reaching the target of a level of quality and reliability of supply and solutions for the integration of electric vehicles resources without violating the network limits. Efficient decision-making tools are required to deal with the technical and physical limits of power grids. Therefore, the idea of the smart grid is that it has the ability to detect and deal with various perturbations by integrating smart technologies [5,6,7,8]. Moreover, various other aspects of grid implementation need to be considered [4,5], while EV modeling factors, limitations, and benefits should be fully realized [6,7,8,9]. Besides EV modelling, control strategies of EV integration and operation modes are crucial. In [14], a control algorithm for a bidirectional off-board electric vehicle charger based upon an adaptive notch filter has been proposed. For reference current generation, this scheme uses residential load current and PCC voltage. The authors have validated the control scheme using MATLAB/Simulink and a hardware prototype approach, which has successfully eliminated the need of PLL. However, it involves a complex control scheme [14]. Lenka et al. [15] has addressed EV charger control function with bidirectional functionality and AFM active filtering modes operating modes. The presented self-tuning filter-based algorithm works well as a grid current harmonics compensator and in the estimation of synchronizing voltage templates for reference current generation, unlike the conventional phase-locked loop. Another work aimed at the modification of the existing closed-loop PLL configuration by using the two cascaded PLL controller configuration is presented in [16]. In [17], a hall effect-based onboard EV charger with perfect harmonic cancellation control is proposed. The current for charging/discharging is set at 5/−5, and THD has been maintained at 3.1% and 3.2%, respectively. In [18], based upon adaptive direct power control theory, an electric vehicle charger is proposed, having a grid-side converter and a charger-side converter. A battery electric vehicle charger is designed with power factor improvement at the front end by replacing a conventional diode rectifier with a modified Landsman power factor converter in [19]. These are all just one aspect of looking into the problem of the grid’s power quality. Another perspective can be very handy; that is, the era of transport electrification can not only solve the problem of greenhouse emissions, but its advantages are far beyond, e.g., harmonic compensation [20]. Farahani et al. [20] have proposed a framework for the harmonic power market in which plug-in electric vehicles can take part as a player with enough economic incentives, and THD will also be relieved to a larger extent. Meanwhile, Hu et al. [21] have provided a comprehensive overview of harmonic sources modelling, which includes power electronics devices like EV charger controllers. Electric vehicles as harmonic compensators can be extremely useful for the distributors and consumers in terms of economic benefits.

The power-quality impact of electric vehicle charging may be positive or negative, depending on the different penetration levels. One such approach is addressed in [22], where modelling electric vehicle charger power quality impacts is defined in terms of the number of chargers considered. In this work, the authors have modelled an EV charger on actual measurements and data from equipment manufacturers and concluded that electric vehicle charging stations do have some impact on the grid either way, positive or negative. In [23], a 13-node low-voltage network has been modelled, which was 80% loaded with a total number of 65 charging points for electric vehicles of 3.7 kW each connected to this network. The charging point load was modeled on the basis of the results of other tests from which the distribution of higher harmonics generated by EV chargers was taken. Three scenarios of 5%, 15%, and 23% penetration scenarios were developed based on the number of electric vehicles charging. It came out to be the 23% penetration worst-case scenario with a 90% load on the grid. The total harmonic distortion voltage value did not exceed 1.2%, while the total harmonic distortion of current value was just 3%. To summarize, there were no negative impacts recorded on the quality of electricity, even when the worst-case scenario of 23% was made up of converter systems.

There are two modes available for charging: unidirectional and bidirectional. A unidirectional charging mode is also known as G2V, in which power is taken from the power grid to the vehicle. In bidirectional or V2G mode, the power is returned to the grid. Due to EVs’ capability of storing energy, they can act as energy storage, and power flow can be opposite (V2G). In this regard, various literatures cover the bidirectional control of EVs. The ac–dc converter with an additional loop to control the reactive component of power and a bidirectional converter implemented a PI controller with an additional loop to control the active component of power for investigating the provision of V2G in a residential setting is presented in [24]. Considering various operation modes of EVs like EVs as energy storage devices or an energy management system (V2G) have potential benefits for customers and system operators [25,26]. The interactivity between G2V and V2G operation modes with the grid must be controlled for mutual benefit. In [27], the work addresses the issue of the grid’s power quality by proposing a self-tuning filter and sliding mode control for an offboard EV battery charger. A mathematical, simulation, and prototype approach is introduced in [28] for the onboard charger controller, whereas a slightly different idea is presented in [29], which is a simple program tool developed to study the impacts of electric vehicle chargers on transformer life. Furthermore, it is found that there is a quadratic relationship between the transformer life consumption and the battery charger current THD [29]. The positive steady and transient results under CC and CV operation show the effectiveness of this idea by reducing switching stress, while in [30], a single-stage Cuk converter-based power factor correction is introduced instead of a two-stage conversion, which has poor efficiency and more components. A multiphase AC–DC converter without using PF corrector and active/passive filter circuits is presented in [31], while in [32], a multifunctional G2V and V2G charging system has been proposed where batteries are charged from the grid, and due to the ability to store energy, it is used as a bidirectional battery charger. The injection of harmonics into the grid has adverse effects, i.e., reducing the lifespan of motors, cables, transformers, and neutrals [33]. Measurement-based harmonic studies have been conducted in [34] using three mainstream EV models, Peugeot e- 208, Nissan LEAF e+, and Renault Zoe R90, for the investigation of the relationship between the charging current and the THD. A field study done in [35] employed two distinct smart charging rates, corresponding to maximum and minimum charging rates, while Ref. [36] has conducted research on the relationship between the charging algorithms, the state of charge, and the THD. The Global Real-Time Superlab (G-RTSLAB) at the Energy Center of Politecnico di Torino, in collaboration with the company Edison SpA, jointly conducted a study using RTS with PHIL methodology [37]. The comparison of this work with others has been summarized in

Table 1. Comparative analysis of the proposed method.

References	GCO	G2V	V2G	Reduction in Harmonics of Grid Current	Improved THD with EV Charging	Grid Electrical Components Modelling Based Approach	Efficiency	THD
[27]	✓	✓	✓	✓	✕	✕	-	I = 3.53%, 2.98%, 3.12%
[38]	✓	✕	✕	✕	✕	✕	-	-
[39]	✓	✓	✓	✓	✕	✕	-	-
[40]	✓	✓	✓	✕	✕	✕	-	V = 1.9%, I = 4.1%
[18]	✓	✓	✓	✓	✕	✕	-	V = 3.917%, I = 2.008%
[41]	✕	✕	✕	✕	✕	✕	-	-
[42,43]	✓	✓	✓	✕	✕	✕	-	- V = 2.02%, 2.05%, 2.06%, I = 3.17%, 2.98%, 3.32%
[44]	✓	✓	✓	✕	✕	✕	87%	I = 3.37%
[45,46]	✕	✕	✕	✕	✕	✕	80–90% 91.5%	I = 3.9%, 3.2%, 4.8%, I = 1.9%, 2.9% and 1.6%
[47]	✓	✕	✕	✕	✕	✕	-	-
[24]	✓	✓	✓	✕	✕	✕	-	-
[48]	✓	✓	✓	✕	✕	✕	-	-
[49]	✓	✓	✓	✕	✕	✕	96.1%	-
[50]	✓	✕	✓	✕	✕	✕	-	-
proposed method	✓	✓	✓	✓	✓	✓	97.4%	V = 0.05%, I = 0.88%

✓ means present, ✕ means not present.

.

1.3. Research Gap and Motivation

From the available literature review, it is evident that filter-based and control strategies dominate the current literature on improving EV charging power quality. Thus, there is a need to do rigorous and complex control and filtering designs of electric vehicle chargers. There are rarely any studies addressing harmonic mitigation through grid infrastructure. Most of the works focused on site solutions like optimizing active/passive filters and advanced converter controls employed in chargers to satisfy IEEE 519 [51] and IEC 61000 [52] harmonic limitations. However, grid component-based harmonic mitigation remains an underexplored area and is rarely implemented, thus highlighting a notable gap in the literature.

There are several factors explaining why grid component-based harmonic mitigation remains unexplored. Technical challenges, economic and policy disincentives, and practical deployment issues are some of the key reasons behind this lag. Passive filters or transformers are highly sensitive to grid impedance; failing to properly configure and design can introduce resonances that exacerbate power-quality problems. Also, in distribution systems, long cable runs can easily resonate with line inductances, amplifying harmonics, whereas active power-quality devices at the grid level can help in avoiding resonance, but they add considerable control complexity and cost to the infrastructure. On the other hand, utilities are not very willing to invest in costly grid-side mitigation, especially when international standards IEEE 519 and IEC 61000 compel EV chargers to limit harmonics at PCC. It is often considered more economical to require each EV charger or station to meet THD limits than to retrofit feeders. Additionally, practical deployment issues, which require network downtime, careful coordination, and ongoing maintenance, must be tuned to local grid conditions to be effective. All of these hurdles have kept researchers focused on the EV charger itself, with grid component-based solutions largely neglected.

None of the existing methods are focusing on how changing technical parameters of the distribution grid can improve the harmonics generated with the integration of electric vehicles. To the authors’ best knowledge, there is a clear gap in terms of the assessment of the power-quality impact of electric vehicle charging stations in connection with a grid’s electrical components. This study is targeted to show that by varying technical specifications of grid components, power harmonics generated with the integration of electric vehicles can be improved.

1.4. Contributions

Considering the above issues, the authors have come up with their own solution. Firstly, a low-voltage distribution system has been modeled using mathematical equations and the data acquired from distribution operators of Chongqing. Secondly, a reliable electric vehicle charging station has been proposed. Finally, the proposed system is designed in the MATLAB/Simulink platform and subjected to power-quality analysis.

The main contributions of this research paper are as follows:

• A comprehensive theoretical modeling of a reliable and accurate distribution grid model using mathematical equations and data from equipment manufacturers and distribution network operators has been designed. These data are not readily available from distribution operators and/or equipment manufacturers.
• Through deep research, potential harmonic distortions and resonance issues have been identified early in the design process. Based upon these insights, the technical selection of the grid’s electrical components has been done, i.e., transformers, filters, converters, and capacitor banks. These are specifically tuned/optimized to effectively mitigate harmonics and enhance overall system performance. It proves that the selection of grid’s electrical components plays an important role in improving power losses in the form of harmonics arising with EV charging.
• Comprehensive simulation results to validate the proposed method and clearly demonstrate its merits.

The paper is organized as follows: Section 2 contains the methodology adopted for this research work. The model validation done using the MATLAB/Simulink platform, along with power-quality analysis results, is presented in Section 3, followed by discussion in Section 4 and the conclusion in Section 5.

2. Methodology

The methodology adopted for this research work is presented in the form of flowchart in Figure 1.

2.1. Modelling Distribution Grid

For modeling a low-voltage AC distribution system, there is a dire need to consider technical specifications of all the electrical components involved. Technical guideline GB/T 32893-2016 deals with the specification of operation management for power users’ 10 kV substations and above [53]. The following section briefly deals with the grid’s electrical components’ technical parameters calculations.

2.1.1. Busbars Model

The voltage magnitude should abide by the allowed operational maximum and minimum voltage limits as provided by (1):

$$V_i^{min}\le V_i\le V_i^{max}$$

(1)

The current flowing through bus $i$ to bus j should remain within the rated current limit, as shown in (2):

$${0\le I}_{lij}\le {\mathrm{I}}_{lij}^{rated}$$

(2)

Finally, the apparent power can be expressed using (3), and its bounds are provided by (4):

$$S_{lij}=P_{lij}+{jQ}_{lij}$$

(3)

$${-\mathrm{V}}_i^{\mathrm{m}\mathrm{a}\mathrm{x}}{\left({\mathrm{I}}_{lij}^{rated}\right)}^T\le S_{lij}\le {\mathrm{V}}_i^{max}{\left({\mathrm{I}}_{lij}^{rated}\right)}^T$$

(4)

Using Equations (1)–(4), the four busbars 35 kV I, 35 kV II, 10 kV I, and 10 kV II calculations for V, I, and S have been done and summarized in

Table 2. Busbar parameters.

Interval Unit	Rated Voltage	Rated Current	Rated Frequency	Device Model	Steady Current	Stability Current	Stability Current Time	Bus Length	Cross-Section Specifications
	(kV)	(A)	(Hz)		(kA)	(kA)	(s)	(m)	(mm2)
35 kV I	40.5	2500	50	LGJ-300/25	80	31.5	4	50	333
35 kV II	40.5	2500	50	LGJ-300/25	80	31.5	4	50	333
10 kV I	12	1250	50	TMY-80 × 10	80	31.5	4	30	80 × 10
10 kV II	12	1250	50	TMY-80 × 10	80	31.5	4	30	80 × 10

.

It is worth mentioning that the calculations made for buses have kept in view the available bus material and models in China. Technical guidelines on selecting busbars followed GB/T 5585.1-2018 (copper or aluminium and its alloy bus bars for electrical purposes—Part 1: copper and copper alloy bus bars) [54] and GB/T 5585.2-2018 (copper or aluminium and its alloy bus bars for electrical purposes—Part 2: aluminium and aluminium alloy bus bars) [55].

2.1.2. AC Lines Model

To represent the series impedance matrix for line l between two buses, we have (5),

$${\mathrm{Z}}_l^{\mathrm{s}}=R_l^{\mathrm{s}}+{j\mathrm{X}}_l^{\mathrm{s}}$$

(5)

The capacitive and inductive reactance and resistance of low-voltage cable lines are considered. The calculation of AC lines model considered the acceptable voltage drops, current carrying capacity, power of individual loads, and the protection requirement against electric shock, which directly contributes to the construction of networks. The line models developed are labelled as l1, l2, l3, l4, l5, l6, l7, and l8, respectively. Technical guidelines on design of cables are GB 50217-2018 [56] (standard for design of cables of electric power engineering), GB 51302-2018 [57] (design standard for overhead insulated distribution lines). The corresponding cable model available complies with LGJ-120, LGJ-50, LGJ-35, JKLYJ-120, and SGJ-185, summarized in

Table 3. AC line parameters.

Interval	Voltage Level	Conductor Model	Line Length	Positive Sequence			Zero Sequence			Max Allowed Line Current
				R	X	B	R	X	B	40	25	10
	kV		km	Ω/km	Ω/km	S/km	Ω/km	Ω/km	S/km	A	A	A
l1	10	LGJ-120	6.8189	3.1746	0.0581	0.0002	0.1261	0.06	0.0001	295	380	395
l2	10	LGJ-50	35.691	0.58	0.0725	0.0001	0.1299	0.8	0.0001	170	210	-
l3	10	LGJ-120	5	3.1746	0.0581	0.0002	0.1261	0.06	0.0001	295	380	395
l4	10	-	-	-	-	-	-		-	-	-	-
l5	10	SGJ-185	2	2.0592	0.0533	0.0001	0.1251	0.15	0.0001	430	560	-
l6	10	JKLYJ-120	1.6608	3.1746	0.0581	0.0001	0.1261	0.06	0.0001	295	380	395
l7	10	LGJ-35	18.82	0.45	0.0612	0.0001	0.1268	0.06	0.0001	245	305	-
l8	10	LGJ-120	5	3.1746	0.0581	0.0002	0.1261	0.06	0.0001	295	380	395

.

2.1.3. Distribution Transformers Model

In distribution transformers, wye-delta connection is commonly used to step down a high voltage to a lower voltage. For distribution transformer modeling, the required parameters, $R_m$, $X_m$, $R_p$, $L_p$, $R_s$, and $L_s,$ have been calculated using transformer tests. Technical guidelines on selection of power transformers followed GB/T 17468-2019 (guide for choice power transformers), GB/T 10228-2023 (technical parameters and requirements for dry-type power transformers), GB/T 6451-2023 (technical parameters and requirements for oil-immersed power transformers), and GB/T 23755-2020 (three-phase site-combined power transformer) [58,59,60,61]. The detailed modelled medium voltage (MV) distribution transformers parameters are summarized in

Table 4. MV transformer parameters.

Label	Rated Voltage (HV) kV	Rated Voltage (LV) kV	Rated Current (LV) A	Rated Capacity MVA	Voltage Ratio	Connection Type	Losses (kW)		Short Circuit Impedance (%)	No Load Current Io (%)
							No Load Loss	Load Loss
T/F1	35 kV	10.5 kV	549.9	10	(35 ± 3 × 2.5%)/10.5	YNd11	8.87	46	7.39	0.19
T/F2	35 kV	10.5 kV	549.9	10	(35 ± 3 × 2.5%)/10.5	YNd11	8.87	0	7.39	-

.

2.1.4. Loads

The complex power to calculate load is provided by (6):

$$S_d=P_d+j{Q}_d$$

(6)

The load current I_d can then be calculated using the values of complex power and node voltage. From

Table 5. Load parameters.

Label	Voltage Level kV	Rated Frequency Hz	Active Power MW	Reactive Power MVAr	Phases	Connection	Model
L1	10 kV	50	0.72	0.3	three	wye	constant PQ
L2	10 kV	50	0.62	0.32	three	wye	constant PQ
L3	10 kV	50	0.25	−0.43	three	wye	constant PQ
L5	10 kV	50	0.16	0.08	three	wye	constant PQ
L6	10 kV	50	0.62	0.39	three	wye	constant PQ
L7	10 kV	50	1.09	0.37	three	wye	constant PQ
L8	10 kV	50	0.29	0.22	three	wye	constant PQ

, three-phase loads modeled for medium voltage level 35/10 kV are labelled as L1, L2, L3, L5, L6, L7, and L8, respectively.

2.1.5. Shunts

To maintain power factor close to unity, the shunt capacitor banks C1 and C2 are modelled at 10 kVI and 10 kVII buses, respectively. The shunt capacitor banks model parameters are summarized in

Table 6. Shunt parameters.

Label	Voltage Level kV	Rated Capacity MVAr	Rated Voltage kV	Rated CURRENT A	Rated Frequency Hz	Single Unit Capacity $\mathbf{k}\mathbf{V}\mathbf{a}\mathbf{r}$	No. of Units	Total Capacity $\mathbf{k}\mathbf{V}\mathbf{a}\mathbf{r}$	Capacitance Value µF
C1	10 kV	1.002	6.3509	105.00	50	334	3	1002	26.52
C2	10 kV	1.002	6.3509	105.00	50	334	3	1002	26.52

.

2.2. Bidirectional Electric Vehicle Charging Infrastructure

The topology of the designed electric vehicle charging station is shown in Figure 2. The initial concept of network topology is based upon the work in [62]. The configuration is completely revised, calculated, and tested as presented in Section 2.1 for the integration of electric vehicle charging station into the system. From Figure 2, it can be seen that a step-down distribution transformer is used, which converts 10 kV into 400 V to be fed to the charging station. A front-end AC–DC converter, which is basically a VSC transforming AC power from grid into DC power, further feeds power to a back-end bidirectional DC converter attached to EV battery with the grid. An LCL filter is used for damping higher harmonics in the power systems.

2.2.1. Parameter Selection for the Charging Station

The EV battery used here is rated at 400 V and 48 Ah. For bidirectional flow of power, $V_{dc}$ and $I_{batt}$ are controlled. The switching frequency considered for this operation is 10 kHz.

The grid-connected inverter converts the DC link bus voltage into a three-phase AC voltage, and vice versa. An LCL filter with damping resistor is connected at the terminals of the inverter for the purpose of harmonics reduction, smooth voltage, and current values. The charging station technical parameters used in this work are summarized in

Table 7. EVCS specifications.

Parameters	Specifications
Grid system	230 V, 50 Hz
Filter inductance	5 mH
Filter capacitor	20 µF
DC capacitance	5500 µF
DC link voltage	1000 V
Charging power rating	40 kVA
Battery storage	400 V, 48 Ah

.

Figure 2. The proposed grid-integrated electric vehicle charging station (GIEVCS).

Figure 2. The proposed grid-integrated electric vehicle charging station (GIEVCS).

LCL Filter

For connecting V2G charger with an AC grid, LCL filter block is used. Passive LCL filter reduces harmonics in grid-interfaced distributed power sources. For bidirectional power flow, the filters are designed as low-pass filters. The distorted signal generated by PWM modulation is filtered to a sinusoidal signal. For the design of inverter side with minimal inductance L, the parameters to be considered are provided as in Equation (7).

$$L_{inv}={\displaystyle \frac{V_{dc}}{8f_s{I}_{nom}{I}_{ripple}}}$$

(7)

The grid-side inductance is usually selected the same as the inverter-side inductance in order to improve the damping behaviour of harmonics. The maximal grid-side inductance is calculated using Equation (8).

$$L_{grid}=10\mathrm{\%}{\displaystyle \frac{V_{grid}^2}{2\pi f{P}_{nom}}}$$

(8)

The filter capacitance is calculated based on reactive power requirement. It is designed to keep power factor as close to unity as possible and is dependent on the AC grid frequency, power, and the grid voltage as in Equation (9).

$$C_{filter}={\displaystyle \frac{\mathrm{\%}{P}_{nom}}{3.2\pi f{V}_{grid}^2}}$$

(9)

Furthermore, to avoid resonances, the relationship between the grid frequency, the resonance frequency, and the switching frequency can be seen as in (10)

$$10f_{grid}\langle f_{res}\langle 0.5f_{sw}$$

(10)

The resonance frequency is calculated using Equation (11):

$$f_{res}={\displaystyle \frac{\sqrt{\frac{L_{inv}+L_{grid}}{L_{inv}{L}_{grid}{C}_{filter}}}}{2\pi }}$$

(11)

To avoid oscillations, a damping resistor is connected in series to the filter capacitor, provided in Equation (12):

$$R_d={\displaystyle \frac{1}{6\pi f_{res}{C}_{filter}}}$$

(12)

The designed filter is subjected to sensitivity analysis, considering ±20% variation in parameter values as depicted in

Table 8. Sensitivity analysis of LCL filter.

Parameter	Value	−20% Variation	+20% Variation
$L_{inv}$	5 mH	4 mH	6 mH
$L_{grid}$	5 mH	4 mH	6 mH
C	20 µF	16 µF	24 µF

.

From Figure 3, it is clear that the capacitor C is the most sensitive parameter affecting high-frequency response, while lowering C or L results in an increase in resonance frequency. For robust filtering, keeping the resonance frequency in 500–1500 Hz range is desirable.

AC–DC Converter

The LCL filter is connected with a three-phase rectifier circuit. This PWM-controlled converter maintains a constant DC link voltage and reactive power control. The control strategy of the AC–DC converter is shown in Figure 4.

DC–DC Converter

A bidirectional buck-boost converter is used here as interface between the dc link, EV battery, and the grid. It charges EV battery with constant current (CC) mode and discharges the battery with controlled current, as per requirements of the grid during V2G mode. The inductance L and the capacitance C of the buck-boost converter are calculated using Equations (13) and (14), respectively.

$$L_{dc-dc}={\displaystyle \frac{V_{out}\left(V_{in}-V_{out}\right)}{{∆I}_L{f}_{sw}{V}_{in}}}$$

(13)

$$C_{dc-dc}={\displaystyle \frac{{∆I}_L}{8f_{sw}{∆V}_{out}}}$$

(14)

The control strategy of DC–DC converter is shown in Figure 5.

2.2.2. Control System

The inverter control strategy adopted here is based on [63], depicted in Figure 4. A cascaded control in synchronous reference frame is utilized. This control structure involves a total of four control loops, i.e., two outer-voltage loops and two inner-current loops. The d-axis outer loop controls the DC bus voltage, and the inner loop controls active component of ac current. Likely, the voltage is regulated with the help of q-axis outer loop by adjusting the reactive component of current.

A constant current control strategy is adopted in this work as presented in Figure 5. The first important step of this control is to compare battery reference current with zero, which eventually decides the operation of DC converter between charging and discharging. Once the result is achieved, the reference current is compared with the measured battery current, and the error is passed through a PI controller, which generates the switching pulses for buck or boost converter operation. The battery control scheme is also adopted from [63], which can be seen in Figure 3. During charging, buck switch is turned on so that the converter steps down Vdc to charging voltage. In this state, the current flows through the inductor to the battery. This is called charging mode or G2V. When the switch is turned off, the current takes the opposite path and passes through the inductor and diode of the lower switch and completes the circuit. The battery voltage is described as the product of Vdc and duty ratio in buck mode.

When the reference current polarity is positive, the lower switch is turned on, and the converter acts as a boost converter. In this state, the current takes the path through inductor and completes the circuit via lower-diode switch and the capacitor. This mode is referred to as discharging mode or V2G.

3. Results

To rigorously check the influence of grid components on power quality, a total of five parameter sets have been generated using the methodology described in Section 2 in detail. The results are as follows:

3.1. Parameter Set 1

The proposed GIEVCS is designed in the MATLAB/Simulink platform, and simulation studies are conducted. Figure 6 shows the designed charging station under G2V and V2G operation modes. The transition of operation in two modes is analyzed via grid-connected V, I, DC link voltage, P_grid and EV battery parameters V_ev, I_ev, SOC, and P_ev. It can be seen from Figure 6a that the DC link voltage decreases until 2.5 s approximately under charging state initially, which means that the buck converter is switched on using the firing signal at the gate. After 2.5 s, the DC link voltage starts increasing while the SOC decreases, which shows that the system is in V2G working mode. As it can be seen in Figure 6b,c, the grid is working steadily and smoothly under G2V with Vg and Ig out of phase with each other, which shows that the battery is taking power from the grid until 2.5 s. Then, according to Figure 6d,e, it takes a transition and starts working in V2G mode after 2.5 s. The DC link voltage remains constant throughout this mode of operation.

3.2. Parameter Set 2

Figure 7 shows the designed charging station under G2V and V2G operation modes for Parameter Set 2. The transition of operation in two modes is analyzed via grid-connected V, I, DC link voltage, P_grid, and EV battery parameters V_ev, I_ev, SOC, and P_ev. It can be seen from Figure 7a that the DC link voltage decreases until 2.5 s approximately under charging state initially, which means that the buck converter is switched on using a firing signal at the gate. After 2.5 s, the DC link voltage starts increasing while SOC decreases, which shows that the system is in the V2G working mode. As can be seen in Figure 7b,c, the grid is working steadily and smoothly under G2V with Vg and Ig out of phase with each other, which shows that the battery is taking power from the grid until 2.5 s. Then, according to Figure 7d,e, it takes the transition and starts working in V2G mode after 2.5 s. The DC link voltage remains constant throughout this mode of operation.

3.3. Parameter Set 3

Figure 8 shows the designed charging station under G2V and V2G operation modes for Parameter Set 3. The transition of operation in two modes is analyzed via grid-connected V, I, DC link voltage, P_grid, and EV battery parameters V_ev, I_ev, SOC, and P_ev. It can be seen from Figure 8a that the DC link voltage decreases until 2.5 s approximately under charging state initially, which means that the buck converter is switched on using a firing signal at the gate. After 2.5 s, the DC link voltage starts increasing while the SOC decreases, which shows that the system is in the V2G working mode. As can be seen in Figure 8b,c, the grid works steadily and smoothly under G2V with Vg and Ig out of phase with each other, which shows that the battery is taking power from the grid until 2.5 s. Then, according to Figure 8d,e, it takes a transition and starts working in V2G mode after 2.5 s. The DC link voltage remains constant throughout this mode of operation.

3.4. Parameter Set 4

Figure 9 shows the designed charging station under G2V and V2G operation modes for Parameter set 4. The transition of operation in two modes is analyzed via grid-connected V, I, DC link voltage, P_grid, and EV battery parameters V_ev, I_ev, SOC, and P_ev. It can be seen from Figure 9a that the DC link voltage decreases until 2.5 s approximately under charging state initially, which means that the buck converter is switched on using a firing signal at the gate. After 2.5 s, the DC link voltage starts increasing while the SOC decreases, which shows that the system is in V2G working mode. As it can be seen in Figure 9b,c, the grid is working steadily and smoothly under G2V with Vg and Ig out of phase with each other, which shows that battery takes power from the grid until 2.5 s. Then, according to Figure 9d,e, it takes a transition and starts working in V2G mode after 2.5 s. The DC link voltage remains constant throughout this mode of operation.

3.5. Parameter Set 5

Figure 10 shows the designed charging station under G2V and V2G operation modes for Parameter Set 5. The transition of operation in two modes is analyzed via grid-connected V, I, DC link voltage, P_grid, and EV battery parameters V_ev, I_ev, SOC, and P_ev. It can be seen from Figure 10a that the DC link voltage decreases until 2.5 s approximately under charging state initially, which means that the buck converter is switched on using a firing signal at the gate. After 2.5 s, the DC link voltage starts increasing while the SOC decreases, which shows that the system is in V2G working mode. As can be seen in Figure 10b,c, the grid is working steadily and smoothly under G2V with Vg and Ig out of phase with each other, which shows that the battery is taking power from the grid until 2.5 s. Then, according to Figure 10d,e, it takes a transition and starts working in V2G mode after 2.5 s. The DC link voltage remains constant throughout this mode of operation.

EV battery Performance Evaluation for Parameter Sets 1 to 5:

Furthermore, Figure 11, Figure 12, Figure 13, Figure 14 and Figure 15 represent EV battery performance under these two modes of operation for Parameter Sets 1 to 5, respectively.

It is evident from Figure 11a–d that the EV battery charges initially until 2.5 s approximately so that $I_{ev}$ is negative while the SOC and $V_{ev}$ increase. Hence, the power is −12.5 kW approximately, as seen in Figure 11c.

From Figure 11a, the EV battery transitions to discharging after 2.5 s, so $I_{ev}$ is positive while the SOC and $V_{ev}$ decrease as in Figure 11d and Figure 11a, respectively, and power is positive 12.5 kW as in Figure 11c. It can be concluded that the charging station transitions smoothly between the G2V and V2G modes of operation.

Figure 12a–d shows EV battery performance under G2V and V2G transitions for Parameter Set 2.

Figure 13a–d shows EV battery performance under G2V and V2G transitions for Parameter Set 3.

Figure 14a–d shows EV battery performance under G2V and V2G transitions for Parameter Set 4.

Figure 15a–d shows EV battery performance under G2V and V2G transitions for Parameter Set 5.

3.6. Power Flow Analysis

Another way to look at the designed system performance response is the analysis of power being taken/fed to the grid. Figure 16 shows an affirmation of the proposed grid-connected electric vehicle charging station in bidirectional power transfer flow for all five parameter sets. It can be seen from Figure 16a,b that until 2.5 s, the grid power is positive, the EV power is negative, and the SOC increases, as in Figure 16c, which shows that the power is being taken from the grid. Hence, it works in G2V mode.

From 2.5–5 s, the grid power becomes negative in Figure 16a, EV power becomes positive, as can be seen in Figure 16b, and the SOC decreases in Figure 16c, which means that the power is now being provided to the grid and is in V2G mode. Also, it verifies the successful operation of grid under bidirectional electric vehicle charging. Hence, the best efficiency value calculated is 97.4%.

After deploying the proposed grid-tied electric vehicle charging system successfully for five datasets, power-quality analysis was performed. It is an established fact that the grid experiences voltage fluctuations and harmonics with the integration of electric vehicles due to their power electronics circuitry. For this process, the MATLAB/Simulink FFT analysis is performed. From Figure 17, the results of voltage and current signals and their THD for five datasets can be seen during the transition between the two modes of operation, respectively. According to IEEE 519 standard, the individual voltage THD should be under 5% for the low-voltage grid 1 kV. It is evident that neither of the employed datasets exceeds IEEE 519. The voltage THD ranges from 0.05–0.12%, while the current THD ranges from 0.88–1.35%. Therefore, it can be concluded that the voltage THD has improved significantly to 0.05% for the designed system (Dataset 5). The current THD is also under a reasonable value less than 1% for Datasets 2, 3, and 5, according to IEEE 519. The grid-connected current for a phase THD has improved significantly to 0.88% (Dataset 5). For Dataset 1 and Dataset 4, the current THD came out to be 1.06% and 1.35%, respectively. Therefore, it is clear that the proposed model improves the THD of the system to a desired value. This shows that by varying the selection of technical parameters of grid equipment, THD as a result of EV integration can be significantly improved.

3.7. Advantages of the Proposed Bidirectional Charger vs. Unidirectional Charger

In general, bidirectional chargers offer superior harmonic mitigation compared to unidirectional chargers due to their more advanced power electronics and control strategies based upon the literature and several research findings. The proposed bidirectional charger has certain advantages over unidirectional charger in terms of harmonic mitigation:

• Control Flexibility:

The proposed bidirectional charger incorporates sophisticated power converters that enable bidirectional power flow, i.e., charging and discharging modes. This bidirectional power-flow capability allows more flexible and precise control strategies, i.e., the filtering of harmonics during both modes of operation.

2.

Power-Quality Support:

Since bidirectional chargers have the capability to inject power back into the grid referred to as a V2G operation, they can be used as distributed energy resources to help mitigate harmonics on the power grid.

3.

Inverter Design:

The adopted inverter topology, VSI with the LCL filter, is more advanced than those in unidirectional chargers. It inherently produces lower harmonic distortion and can be tuned/optimized to adapt to varying grid conditions.

4.

Grid Interaction:

While unidirectional chargers mainly focus on delivering energy to EVs, the proposed bidirectional charging station can serve as a dual role, i.e., delivering and receiving power, which allows for dynamic harmonic mitigation. This virtue makes them valuable assets in smart grid environments where power quality and grid stability are crucial.

3.8. Comparative Study

A comparative analysis of the proposed method with state-of-the-art methods is provided in

Table 9. Comparison of proposed method with other existing methods.

Technique	Efficiency	Voltage THD	Current THD
STF and SMC-based control strategy [27]	-	-	3.53%, 2.98%, 3.12%
IGBT switch-based converters [40]	-	1.9%,	4.1%
Adaptive direct-power control theory [18]	-	3.917%,	2.008%
Positive-sequence components [43]	-	2.02%, 2.05%, 2.06%	3.17%, 2.98%, 3.32%
H-bridge [44]	87%	-	3.37%
Fully bridgeless canonical switching cell [45]	80–90%	-	3.9%, 3.2%, 4.8%
Interleaved Landsman converter [46]	91.5%	-	1.9%, 2.9%, 1.6%
SiC-based [49]	96.1%	-	-
SPV-based SAPF with UVT control [64]	-	-	2.5%
TOSSI-CTF [65]	-	-	2.3%
Proposed method	97.4%	0.05%	0.88%

. The proposed system has clearly outperformed other existing strategies in terms of efficiency and THD.

4. Discussion

The results discussion is summarized below:

i.

On the basis of mathematical calculations and real equipment data information gathered, a novel grid-integrated electric vehicle charging station has been modeled. The simulation tests were conducted to assess the impact of electric vehicle charging stations on the grid’s power quality. The development of reliable and accurate simulation models is possible, but at the same time, because of the solutions adopted, it is highly technical and complicated. Therefore, one-size-fits-all is ruled out. In short, it can be inferred that different grid topologies can have a different impact, either positive or negative, when new energy vehicles are integrated.

ii.

This work sheds light on the fact that in order to assess the impact of electric vehicle charging stations on the grid, not only do charging station components need to be modeled in detail, but also the grid’s topology and each of its components need to be calculated, designed, and verified via simulation analysis. These data are not readily available from distribution operators and/or equipment manufacturers.

iii.

Since there are no standards defined for the simultaneity factor, it is assumed that all the chargers are working simultaneously. In that case, the results showed that the voltage and current THD did not exceed 1%, which is acceptable, and they indicate that the charging station has no negative impact in the proposed scenario.

iv.

The V2G mode can be used as one of the elements of the smart grid, and it can also act as energy storage systems by reducing the need for power electronics circuitry, which results in power losses and harmonics not desired.

v.

To further test and analyse the introduction of the impact of electric vehicles on the grid, this model provides a good testing platform without the need for a costly hardware experimental setup. Since climate change has reached its peak point, there is a dire need to move towards low-carbon technologies in the transportation sector. The positive results provide a good base for using this model for the investigation of various other operations, i.e., new-energy vehicles and low-carbon technology integration into the grid.

5. Conclusions

In this paper, a reliable three-phase grid-integrated electric vehicle charging station has been proposed using mathematical equations and real equipment data information from the equipment manufacturing company. The model has been developed for five parameter sets using the MATLAB/Simulink platform. The employed parameter sets showed promising results, with voltage THD ranging from 0.05–0.12%, and current THD ranging from 0.88–1.35%. The voltage THD has improved significantly to 0.05% for the designed system, while the current THD is 0.88%, which is significantly low with 97.4% efficiency. Hence, the system is working well under different transition modes of electric vehicle charging station and varying parameter sets. Therefore, it can be inferred that grid component-based harmonic mitigation is highly recommended for implementation in real-world electric vehicle charging stations.

In future work, the authors will focus on the development of a renewable-energy-powered grid-integrated electric vehicle charging station. The authors see many advantages and opportunities offered by distributed generation-based charging stations with more benefits to the grid, consumers, and the environment. It has potential in reducing greenhouse gas emissions by using clean energy for charging, as well as decreasing reliance on fossil fuels for transportation. The economic benefits of this approach will reduce electricity costs for station operators (especially with solar), potential income by selling excess energy back to the grid, government incentives, and tax benefits for renewable installations. Grid benefits include a reduced peak load on the grid by supplying local renewable energy, which helps stabilize the grid through smart charging and energy management. Moreover, when paired with energy storage devices like batteries, solar-powered electric vehicle charging stations can store solar energy generated during the day and use it at night time, when people often charge their vehicles, making the whole system even more efficient. Another area would be exploring this technique on high-power fast-charging stations.