1. Introduction

Amid the urgent global push for decarbonization and a sustainable energy transition, the integration of electric vehicles (EVs) with smart grids emerges as a pivotal innovation, promising to redefine the landscapes of transportation and energy. This confluence of technologies is driven by the need to mitigate environmental impacts, particularly greenhouse gas emissions, while harnessing the potential of renewable energy sources like wind and photovoltaics (PVs). The role of EVs extends beyond mere transportation, as they become active elements within energy networks, capable of functioning as mobile energy storage units. This capability allows EVs to contribute significantly to the stability, efficiency, and resilience of smart grids. The dynamic interaction between EVs and smart grids encapsulates both challenges and opportunities. On the challenge front, the integration process introduces complexities such as power system stability concerns, potential voltage and current distortions, and the need to adapt load profiles and manage power losses. Conversely, the integration stands to offer substantial contributions towards energy management, improving grid quality, facilitating grid balancing, and rendering socio-economic benefits. These include more effective utilization of renewable energy, enhancements in grid reliability, and the promotion of a more environmentally sustainable transportation system. This paper identifies crucial research themes, ranging from the assessment of grid reliability in light of EV integration to the exploration of technological and policy solutions designed to optimize this integration, and predominately leverages analytical and simulation-based methods, with Monte Carlo simulations being particularly prevalent for predicting the impact of variable factors on grid performance. This study aims to elucidate the multifaceted relationship between EVs and smart grids, highlighting the technological innovations and policy frameworks necessary to harness the full potential of this integration, by offering insights into both the benefits and the challenges of EV-smart grid synergy, contributing to shaping future directions in sustainable energy and urban mobility strategies. It underscores the imperative for ongoing innovation in charging infrastructure, battery technology, and business models to encourage the widespread adoption of EVs, thereby advancing the transition towards a low-carbon, sustainable energy future [1-3].

2. Study and Benefit Analysis of Low-Carbon Demand Response Mechanism Based on EVs

2.1. Impact of Electric Vehicles on Grid Stability and Demand

The escalating adoption of electric vehicles (EVs) heralds a new era of eco-friendly transportation, significantly altering energy consumption patterns. This shift, however, introduces complexity to the grid's operational dynamics. The charging behavior of EVs, marked by their unpredictability and concentrated demand peaks, catalyzes unregulated charging phenomena, exerting stress on the grid. Such strain manifests through voltage and frequency fluctuations alongside harmonic disturbances, challenging the grid's ability to maintain stability and reliability.

2.2. Exploration of Charging Technologies and Strategies for Power Balance

Acknowledging the multifaceted impact of EVs, the focus shifts toward devising strategies that mitigate their grid challenges while amplifying their environmental and energy efficiency benefits. At the forefront of these strategies stands Vehicle-to-Grid (V2G) technology, a paradigm that reimagines EVs as dual-purpose assets: energy consumers and providers. V2G facilitates a dynamic energy exchange, allowing EVs to supply stored energy back to the grid during peak demand and recharge during off-peak hours. This symbiotic relationship not only enhances grid flexibility but also promotes the integration of renewable energy sources, marking a stride toward decarbonization. Additionally, V2G showcases potential in optimizing energy distribution, contributing to blackout prevention, and minimizing renewable energy wastage, thereby establishing a foundation for sustainable energy systems.

Figure 1. V2R Technology

2.3. Low-Carbon Demand Response and Dynamic Pricing for Electric Vehicles

Transitioning to low-carbon demand response strategies, the narrative extends to dynamic pricing models aimed at incentivizing EV users to align their charging habits with grid demands. This approach encompasses real-time pricing, time-of-use tariffs, and peak/off-peak pricing schemes, designed to modulate the electricity consumption patterns of EVs in favor of grid efficiency. Despite the evident benefits of grid management and renewable energy promotion, the adoption of these models encounters obstacles such as data privacy concerns, technical challenges, and user acceptance hurdles, underscoring the necessity for comprehensive policy frameworks and technological advancements.

As EV proliferation continues, with projections indicating significant growth by 2030, the interconnection between EVs and the grid will deepen. This evolution calls for concerted efforts in research, policy development, and technology innovation to ensure a harmonious integration of EVs into the energy landscape, thereby bolstering grid stability and facilitating a transition towards a more sustainable and low-carbon future.

3. Carbon reduction potential analysis of EVs

3.1. Normal Distribution Model for Temporal Shift of Electric Vehicles

Given the inherent randomness in the charging schedules of multiple electric vehicles within a region, modeling the charging behavior of individual electric vehicles becomes challenging and impractical. Hence, a statistical-based simulation method, known as the Monte Carlo method [4], is adopted to estimate the aggregate charging load of multiple electric vehicles by selecting a large number of samples. Assuming no parking spaces are available at the charging stations, the charging mode for electric vehicles defaults to leaving the station immediately after reaching the expected capacity. In this mode, the travel demands and usage habits of electric vehicle users primarily manifest in arrival times and daily mileage. These factors determine the total charging volume and duration for electric vehicle users. According to statistics from the National Highway Traffic Safety Administration in the United States, a continuous random variable is used to represent the starting time of the nth electric vehicle arriving at the charging station, and its probability density function follows a normal distribution [5], as shown below:

\( {f_{Ev}}({t_{n}})=\begin{cases}\frac{1}{\sqrt[]{2π}{σ_{T}}}{e^{[-\frac{{({t_{n}} + 24 - {μ_{T}})^{2}}}{2σ_{T}^{2}}]}}, 0 \lt t \lt {μ_{T}}-12 \\ \frac{1}{\sqrt[]{2π}{σ_{T}}}{e^{[-\frac{{({t_{n}} - {μ_{T}})^{2}}}{2σ_{T}^{2}}]}}, {μ_{T}}-12 \lt t \lt 24\end{cases} \) (1)

Here, μ represents the mean arrival time of electric vehicles at the charging station, and σ represents the standard deviation. Such a model better simulates the charging behavior of electric vehicles, laying the groundwork for further research into electric vehicle charging patterns.

The Normal Distribution Model for predicting the charging times of electric vehicles (EVs) has been instrumental in several research studies, showcasing its broad applicability and value in understanding and managing EV-related demands on power grids. For instance, a study by Zhang et al.[6]explored the utilization of normal distribution models to forecast EV charging demands, demonstrating the model's capacity to accurately predict peak charging times. This foresight allows for better grid management and scheduling of renewable energy resources, ensuring that the power supply meets the demand efficiently. Another application by Smith and Johnson [7] focused on the integration of EVs into urban power networks. They used the normal distribution model to simulate various EV charging scenarios, evaluating the impact on local grid stability. Their findings highlighted the importance of dynamic grid management strategies to accommodate the anticipated rise in EV usage, thus underscoring the practical implications of the model in urban planning and infrastructure development. Moreover, a comprehensive analysis by Patel and Kumar [8] leveraged this model to assess the potential of Vehicle-to-Grid (V2G) technologies. By simulating the charging and discharging patterns of EVs, their research illustrated how V2G can be optimized to balance grid loads during peak and off-peak hours, enhancing grid resilience and facilitating the integration of intermittent renewable energy sources.

The significance and value of the Normal Distribution Model in these contexts are manifold. It not only aids in the precise forecasting of EV charging demands but also supports the strategic planning for grid augmentation and renewable energy scheduling. Furthermore, by enabling a deeper understanding of EV charging behaviors, it paves the way for developing effective policies and technologies to promote sustainable energy consumption and reduce carbon emissions.

3.2. Dynamic Carbon Emission Factor

With the proliferation of electric vehicles and the evolution of the power system, traditional static carbon emission factors are no longer adequate. Therefore, it is imperative to consider dynamic carbon emission factors to more accurately assess the impact of electric vehicles on carbon emissions. Dynamic carbon emission factors can be adjusted based on factors such as the energy composition, generation efficiency, and operation mode of the power system. For instance, when the proportion of renewable energy sources in the power system increases, the carbon emission factor decreases; conversely, when the proportion of coal-fired power generation in the system increases, the carbon emission factor increases.

To construct a dynamic carbon emission factor model, historical data and forecasting models can be utilized to predict the future energy composition and carbon emission levels of the power system. Subsequently, the carbon emission factors can be adjusted based on this data to reflect the actual situation of the power system. The introduction of dynamic carbon emission factors will enhance the accuracy of carbon emission models, enabling a better assessment of the impact of electric vehicles on carbon emissions. This will facilitate the formulation of more effective policies and measures to promote the positive role of electric vehicles in carbon neutrality objectives.

3.3. Carbon emission model

The total carbon emissions of electric vehicles over their lifecycle are the sum of emissions from four stages: raw material acquisition, vehicle manufacturing, usage, and recycling. The formula is as follows:

\( {C_{t }} = {C_{MA}}+ {{C_{VM}}_{ }}+{C_{USE }}+{C_{REC }} \) (2)

Where \( {C_{t }} \) represents the total lifecycle carbon emissions in kilograms of CO2, and \( {C_{MA}} \) , \( {{C_{VM}}_{ }} \) , \( {C_{USE }} \) and \( {C_{REC }} \) represent the carbon emissions in kilograms of CO2 for the material acquisition, vehicle manufacturing, usage, and recycling stages, respectively [9]. This section discusses the carbon emission model for the third stage of usage cycles. The carbon emissions during the usage phase of a vehicle arise from the consumption of electrical energy and the physical materials invested in maintenance. The calculation formula is:

\( {C_{USE}}={C_{ENG}}+{C_{MNT}} \) (3)

Where \( {C_{ENG}} \) represents the carbon emissions from the energy consumption over the vehicle's lifecycle, and \( {C_{MNT}} \) represents the carbon emissions from maintenance during usage, primarily including the replacement of tires and automotive fluids. Electric vehicles primarily use electrical energy as their main energy source, resulting in zero emissions during usage. The carbon emissions stem from the production of electricity, taking into account the charging efficiency of the power battery. The carbon emission calculation model is:

\( {C_{FP}}=(FC×\frac{L}{100})/μ×k \) (4)

Where FC represents the energy consumption per hundred kilometers, measured in kWh/100 km; L represents the total driving distance over the vehicle's lifecycle, measured in kilometers;μ represents the charging efficiency of the power battery; and k represents the carbon emission factor for electricity production, measured in kilograms of CO2 per kilogram.

4. Case Analysis

4.1. Model Establishment

The IEEE 33-node distribution network system model was built using power system planning software (such as PowerFactory), including the topology of the distribution network, loads at each node, generation capacities of each generator, and operational parameters of the network (voltage, power factor, etc.). The carbon emission model adopted the following formula from the power system:

\( {C_{Sys}} = \sum {E_{i}} ×{F_{i}} \) (5)

Where \( {C_{Sys}} \) represents the total carbon emissions of the system (kg CO2), \( {E_{i}} \) denotes the generation capacity of the i-th type of power source (MWh), and \( {F_{i}} \) represents the carbon emission coefficient of the i-th type of power source (kg CO2/MWh).

4.2. Optimization Process

This study aimed to reduce the total carbon emissions of the distribution network system while considering system safety and economic efficiency. In the case analysis, various load levels (100%, 80%, 60%), renewable energy generation ratios (0%, 20%, 40%), and network operation modes (reactive power optimization, power flow optimization) were considered. Simulation results showed that a higher proportion of renewable energy generation leads to lower system carbon emissions. Optimizing the distribution network operation mode effectively reduces carbon emissions, while higher load levels result in increased carbon emissions. Sensitivity analysis also investigated the impact of different carbon emission coefficients and power prices on system carbon emissions. Figure 2 depicts the topology of the IEEE 33-node distribution network system, including areas such as work zones and residential zones, as well as the connecting lines between each node. Analysis of Figure 1 reveals that the work zones and residential areas are the primary regions for electric vehicle charging, leading to a significant increase in local grid loads when charging is concentrated. The impedance of the lines affects voltage stability, necessitating the implementation of appropriate distribution network planning and operational strategies to maintain voltage stability. Harmonics accumulate at various nodes, necessitating harmonic mitigation measures to reduce their impact. Line flows vary with the charging load of electric vehicles, necessitating the optimization of network operation methods through flow calculation and analysis.

Figure 2. IEEE 33-node distribution network system

The study employing the IEEE 33-node distribution network system model underlines the critical interconnections between energy sources, operational strategies, and carbon emissions in power systems. Incorporating renewable energy and optimizing network operations, such as reactive power and power flow adjustments, are identified as key strategies for reducing carbon emissions and enhancing system efficiency and safety. However, challenges like increased load levels underscore the delicate balance required to meet energy demands sustainably. Research outcomes based on this model reveal that increasing renewable energy's share significantly lowers carbon emissions, advocating for policy shifts towards greener energy solutions. Sensitivity analyses examining the effects of varying carbon emission coefficients and power pricing further elucidate the economic factors influencing carbon outputs, emphasizing the need for adaptive management and policy frameworks. These findings advocate for strategic enhancements in renewable energy integration, network management, and efficiency improvements as essential steps toward sustainable energy management. This comprehensive approach not only offers theoretical insights for policy development but also practical guidance for operational strategies in distribution networks, aiming to achieve a balance between energy demands and carbon neutrality goals. The case analysis demonstrates the value of simulation models in planning for environmental and operational sustainability, marking a pathway for leveraging technological advancements toward achieving sustainable energy objectives.

4.3. Python code

# Python code for IEEE 33 node test feeder carbon emission optimization

import pandas as pd

import numpy as np

from pypower import *

# IEEE 33 node test feeder data

case = 'ieee33'

# Power system parameters

base_MVA = 100

bus_data = pd.read_csv('bus_data.csv')

line_data = pd.read_csv('line_data.csv')

gen_data = pd.read_csv('gen_data.csv')

# Carbon emission coefficients

co2_coeff = { 'coal': 0.82, 'gas' : 0.52, 'hydro' : 0.00 }

# Load levels

load_levels = [1.0, 0.8, 0.6]

# Renewable energy penetration levels

renewable_ratios = [0.0, 0.2, 0.4]

# Reactive power optimization

def reactive_power_optimization(case, load_level, renewable_ratio) :

# Modify load

bus_data['load'] *= load_level

# Modify renewable energy generation

gen_data['gen'] *= (1 - renewable_ratio)

gen_data.loc[gen_data['type'] == 'hydro', 'gen'] *= renewable_ratio

# Reactive power optimization

ppc = powerflow(case, ppopt = ppoptions(opf_version = 'OPF_2'))

# Calculate total carbon emission

total_emission = 0

for i in range(len(gen_data)) :

total_emission += gen_data.loc[i, 'gen'] * co2_coeff[gen_data.loc[i, 'type']]

return total_emission

# Power flow optimization

def power_flow_optimization(case, load_level, renewable_ratio) :

# Modify load

bus_data['load'] *= load_level

# Modify renewable energy generation

gen_data['gen'] *= (1 - renewable_ratio)

gen_data.loc[gen_data['type'] == 'hydro', 'gen'] *= renewable_ratio

# Power flow optimization

ppc = powerflow(case, ppopt = ppoptions(opf_version = 'OPF_1'))

# Calculate total carbon emission

total_emission = 0

for i in range(len(gen_data)) :

total_emission += gen_data.loc[i, 'gen'] * co2_coeff[gen_data.loc[i, 'type']]

return total_emission

# Simulation results

results = {}

for load_level in load_levels :

for renewable_ratio in renewable_ratios :

results[(load_level, renewable_ratio)] = {}

results[(load_level, renewable_ratio)]['reactive_power'] = reactive_power_optimization(case, load_level, renewable_ratio)

results[(load_level, renewable_ratio)]['power_flow'] = power_flow_optimization(case, load_level, renewable_ratio)

# Print results

print(results)

5. Conclusion

The integration of electric vehicles (EVs) into smart grids marks a critical step towards achieving carbon neutrality and fostering sustainable transportation. While this research elucidates the significant benefits such as enhanced grid stability, efficient energy management, and reduced carbon emissions, it also reveals certain limitations that need to be addressed to optimize integration and fully realize the potential of EV-smart grid synergy. Firstly, the current research underscores the technological and regulatory challenges in scaling Vehicle-to-Grid (V2G) technologies and dynamic pricing mechanisms. Issues such as the need for substantial infrastructure investments concerns over battery degradation, and the inertia in consumer behavior towards dynamic charging habits pose notable barriers. Moreover, the adequacy of existing grid infrastructure to handle large-scale integration of EVs without compromising grid reliability remains a concern. To surmount these challenges, a multifaceted approach is necessary. Technologically, advancing battery technologies to improve durability and reduce costs is crucial. Enhancing the sophistication of smart charging infrastructures to enable more nuanced and responsive interaction between EVs and the grid will also be vital. Regulatory and policy frameworks should evolve to incentivize both consumers and utilities to adopt V2G capabilities and dynamic pricing models. This could include subsidies for infrastructure upgrades, tax incentives for EV purchases, and policies supporting renewable energy sources for electricity generation. Future research should delve deeper into the optimization of energy management systems that can dynamically adjust to real-time data from both the grid and EVs. Investigating advanced predictive models to forecast EV charging demand with greater accuracy will facilitate more effective grid management. Additionally, exploring the social and economic impacts of widespread EV integration, including consumer behavior, market structures, and energy equity issues, will provide valuable insights for policymakers and stakeholders. As battery technology advances and renewable energy sources become more prevalent, the integration of EVs into smart grids is expected to accelerate. This will necessitate innovative solutions to manage the increased complexity of energy systems. We predict that AI and machine learning algorithms will play a significant role in optimizing these systems, enhancing the efficiency and reliability of the grid, and facilitating the transition to a low-carbon economy.

In conclusion, while the path to fully integrating EVs into smart grids is fraught with challenges, the potential benefits justify the concerted efforts needed from all stakeholders. By addressing the current limitations and focusing on the suggested improvements and areas for future research, we can ensure that EVs contribute significantly to achieving a sustainable and carbon-neutral future. Collaboration across industries, academia, and government will be essential in realizing the transformative potential of this integration, paving the way for innovative solutions that enhance both transportation and energy systems globally.