Energy Communities and Electric Mobility as a Win–Win Solution in Built Environment

Abstract

Recently, there has been an increasing effort to promote energy efficiency, using renewable energies and electric mobility to achieve a more sustainable future and even carbon neutrality by 2050. This paper aims to understand if combining these technologies leads to a win–win solution. For that, the system’s characteristics that will be used for the simulation were defined as a residential community consumption scenario with and without electric vehicles charging overnight. The simulation was completed in software, and eight scenarios were tested: high population density/low population density with/without electric mobility and hourly tariff/simple tariff. After these scenarios had been tested, the conclusion was that the low population density and hourly tariff without and with electric mobility were the best two cases economically (in terms of levelized cost of energy, net present costs, and savings) and environmentally, and the worst was high population density with hourly tariff and electric mobility. Other scenarios were then tested, including changes in the load curve, namely a commercial load curve, and changes in the load curve of electric vehicle chargers, mainly daytime charging. The conclusion was that even though the initial hypothesis did not lead to a win–win solution, with changes in the hypothesis, the integration of electric mobility in energy communities might lead to that.

1. Introduction

Nowadays, there exists a significant focus on and attempt to move away from fossil fuels and towards a more sustainable and cleaner form of energy to achieve a reduction in greenhouse gas emissions proposed by the Paris Agreement. In May 2019, the European Union finished the legislative files for the Clean Energy for All Europeans Legislative Package or Clean Energy Package (CEP). CEP consists of eight new laws and is essential for implementing the energy union strategy and meeting the 2030 climate and energy objectives. The objectives mentioned for 2030 in CEP are a 40% reduction in greenhouse gas (GHG) emissions, 32% of renewables in energy consumption, and 32.5% in energy efficiency [1]. This could also help achieve carbon neutrality in the long term in 2050.

Additionally, the rearrangement of Directive 2018/2001 (Renewable Energy Directive II or REDII) and Directive 2019/944 (Internal Electricity Market Directive or IEMD) defined two essential concepts: renewable energy communities (RECs) and citizen energy communities (CECs). With Regulation 2019/943 (the Internal Electricity Market Regulation or IEMR), the three directives establish an EU legal framework that supports community ownership [2]. According to the REDII, ECs are included in CEP because community energy gives a better acceptance of renewable energy, gives access to private capital, and gives more choices to the citizens besides giving them a role in the energy transition. In the IEMD, it is added that community energy helps to promote inclusion, gives affordable renewable energy to its members, fights energy poverty, allows new consumption patterns, and uses the latest technologies [2].

Even though some energy communities might need non-renewable energy resources, this research focuses on those that only use renewables since they contribute the most to carbon neutrality. Although they present great benefits, ECs still need to be improved due to their innovative character, which results in a lack of economic, technical, and social expertise. More information is also required regarding integrating electric mobility in energy communities.

Electric mobility is also an increasingly important topic, which, like energy communities, contributes to reducing GHGs. Electric mobility in Portugal is growing fast, which is expected to continue in the next five to ten years. Electric mobility shares some advantages with energy communities, such as decreased GHGs and increased energy efficiency.

In Portugal, the responsibility and authority given to each electric mobility network agent were defined in Decree-Law 90/2014 [3] to create competition within the electricity and charging station sectors for electric mobility. This regulation also coordinates the activities related to offering, managing, and using public charging stations while encouraging their incorporation into private spaces within the electric mobility network [4].

One example of a project that involves these two concepts in Portugal is COMSOLVE, by Coopérnico, which started on 15 July 2021 and ended on 30 June 2023. This project aims to manage energy communities with integrated electric mobility while having energy storage systems based on second-life batteries. It also created a platform where the estimated vs. actual production, the production vs. the consumption, and the charging schedule, including the duration and location, could be seen [5].

The primary purpose of this research is to understand if energy communities and electric mobility can be allies and present themselves as a win–win solution. Considering that they have similar advantages and goals, such as reducing GHG emissions and energy bills and gasoline bills, it makes sense to see if it is advantageous to have them together.

To do so, it will be used as the hypothesis that energy communities and electric mobility can be a win–win solution at an economic and environmental level. For that, some consideration must be made regarding the energy community and the electric charger’s model. It will be considered a residential community-type load curve, where the peak occurs between 6 p.m. and 9 p.m., and electric vehicles charge between 10 p.m. and 7 a.m. The energy community will only include photovoltaic energy, and the electric chargers will be assumed to be wallboxes.

The structure of this article is as follows: Section 2 is a literature review of energy communities and electric mobility; Section 3 presents the methodology for the work, including the selection of different equipment and the software methodology; the results of the simulation are presented in Section 4, and then, they are discussed in Section 5, where the work’s limitations can also be found. Lastly, the conclusions are summarized in Section 6.

2. Energy Communities and Electric Mobility Review

2.1. Energy Communities

An energy community (EC) is a group of citizens who produce and share energy. There are two types of energy communities included in the EU legislation (and in Portugal), which are defined in the revised Internal Electricity Market Directive (EU) 2019/944 (IEMD) [6] and the revised Renewable Energy Directive (EU) 2018/2001 (RED II) [7]: citizen energy communities and renewable energy communities, respectively. While CEC focuses mainly on returning the benefits to the local community or the groups of people involved in the energy project, connecting the social aspect of energy, RECs are geographically constrained, which forges a close bond between the community and the energy source [8].

According to [9], energy communities rely on actors with a specific role or more than one role in the community. The roles identified by the authors are initiator, energy service provider, and consumer. The consumer is the actor who receives and beneficiates from the energy commodity or service from another actor. The energy service provider is responsible for the generation, storage, and supply of energy, as well as the installation and maintenance of the equipment. Lastly, the initiators organize and coordinate the project so they may benefit from something other than the community. There is also an actor called prosumer, a mixture of consumers and energy service providers since they generate energy for their self-consumption and share it with other community members.

2.1.1. Economic Impacts of ECs

In an energy community, like in any other project, it is vital to understand the economic benefits that arise from its creation. By participating in a community, members can benefit from financial gains, for instance, paying less for energy when compared to the price of energy from the grid and being able to inject energy into the grid through feed-in tariffs [9].

Additionally, profits obtained from one project can be reinvested in community projects. Lastly, ECs can benefit the local community by promoting the creation of new jobs, developing new skills, and new regional value chains and industries [10].

However, benefits for the investors are only obtained because an economic feasibility analysis was previously performed to determine whether the project was profitable. The financial feasibility of an energy community depends on various factors, such as the payback time, the grid electricity prices, and the costs of distributed energy resources [11].

Moreover, Carreras and Steinmaurer [12] have identified three critical factors to determine whether the energy community is economically viable. These indicators are whether the overall energy costs for the community participants are lower than if the community did not exist, participant savings and coordination of the energy exchange, and whether the REC surplus covers the operators’ expenses.

2.1.2. Energy Communities in Portugal

In Portugal, the framework for collective self-consumption of renewable energy and energy communities (Comunidades de Energia Renovável) was established in the decree-law 162/2019 [13] on 25 October 2019. Regarding membership, potential activities, and the requirement to develop a legal person, the 2019 decree law adopts the main principles of the EU REDII.

Since 1 January 2020, individual and collective self-consumption projects and RECs have been possible if an intelligent counting system is installed at the same voltage level [14].

To start an energy community or a self-consumption project in Portugal, you must register and apply for DGEG (Direção-Geral de Energia e Geologia). This step is essential for managing and licensing the energy community and verifying whether self-consumption units are already in that area.

Regarding the energy supply, the supplier does not need a license to share electricity. However, the entity responsible for managing the community will need to coordinate activities with the DSO (distribution system operators) and be responsible for implementing and distributing electricity among the different participants [14].

Regarding grid access, the energy communities can be connected to the public grid, where the tariff to be covered is calculated by considering the voltage level used. On the other hand, self-consumers may establish an internal network with no access to the public network.

2.2. Electric Mobility

As it is known, the transport sector is one of the most significant contributors to greenhouse gas emissions, which adds to almost a quarter of the emissions. Therefore, there has been a transition from internal combustion engine vehicles to electric vehicles to reduce these emissions. This trend is also related to the European Commission’s goal of achieving a 40% reduction in GHG emissions by 2030, compared to the values of 1990 and a 60% reduction by 2050 in the transport sector [15].

Several European countries have tried creating strategies to transition from internal combustion engine vehicles to electric vehicles. Ruggieri et al. have extensively reviewed some measures that cities such as London, Hamburg, Oslo, Milan, Florence, and Bologna have implemented to promote this transition [16].

2.2.1. Market Trends

Electric vehicles are trendy in China, Europe, and the United States of America. Figure 1 shows that the growth of EVs is exponential, especially in China, where BEVs (battery electric vehicles) have grown from 0.2 million in 2017 to 10.7 million in 2022. Europe and the USA have also seen significant growth from 2017 to 2022, from 0.2 million to 4.4 million in the first case and from 0.2 million to 2.1 million in the second case [17].

Europe is the second largest market responsible for one-quarter of the EVs in circulation. Even though the growth from previous years was higher, in 2022, the sales went from 3 million (in 2021) to 4.4 million in BEVs and from 2.5 million to 3.4 million in PHEVs. Regarding the European countries where EVs are the most popular, Germany is the biggest market, followed by the UK and France. This growth is expected to continue since more incentives for the purchase of EVs were implemented [17].

2.2.2. V2X Technology

V2X (vehicle to everything) is a communication technology between the car and another entity, such as another vehicle, the grid, a home, or a building. This concept allows the alliance between EVs and smart grids, where the first one, through charging and discharging processes, can interact with the entities mentioned before to achieve economic or reliability purposes [18].

Regarding electric mobility, there are a variety of chargers that could be considered: dumb chargers, V1G, and V2G (which are included in V2X technology) are examples. V2G is a concept that represents the bidirectional transfer of energy between the vehicle and the grid. Therefore, the grid can give power to the electric car and receive energy stored in the vehicle’s battery.

In the case of V1G, which is unidirectional smart charging, the electricity only goes from the charging device to the vehicle. Therefore, the car cannot work as a battery. Additionally, since it is “smart”, it can only start charging when the prices are lower or when it is not a peak period, for example. Dumb charging, on the other hand, does not include any intelligent charging features.

2.3. Integration of Electric Mobility in Energy Communities

In the literature, a case study investigates the impact of introducing electric mobility in a local energy community. In the Savona Campus of the University of Genova, an intelligent polygeneration microgrid was implemented to provide thermal and electrical energy. This initiative was financed by the Italian Ministry of Education, University and Research (MIUR) [19].

This microgrid studied the impact of electric mobility on the design of energy communities. An optimization model was used to determine the size of the PV plant, energy storage, and charging stations for electric vehicles [20].

Regarding the PV plant, two types of PV modules are considered for decision variables: monocrystalline and polycrystalline. The optimized variables are the number of each type of PV module and the power generated by the PV plant.

Regarding the storage systems, only one type of battery is considered. In this case, the optimization variables are the number of batteries, the energy stored in the system, the charging and discharging power, and when it is charging or discharging. Lastly, three options were considered in the EV charging stations: traditional chargers, intelligent charging stations (V1G), and V2G chargers. The number and type of electric vehicles were not the objects of optimization, but their charging and discharging strategies were.

The conclusion was that the polycrystalline PV modules tend to be the most competitive ones due to their lower investment cost; battery storage is not considered if the goal is to have any economic optimization; when electric mobility is added, the overall optimization of the system is not only influenced by the number of vehicles but also by the type, since some vehicles could be or not be compatible with V2G chargers. Additionally, when more vehicles are added, the PV peak power will likely increase, as will the charger’s power.

The same authors have also written other papers about the Savona Campus LEC [21,22,23]. According to G. Piazza et al. [21], energy storage systems, PV plants, and a V2G charging station are already installed, and some EVs are already part of the carsharing system. In this case, the paper aims to study whether installing more PV, batteries, and chargers is profitable. S. Bracco et al.’s [22] paper seeks to examine the V2G and V2B projects in the Savona Campus. The work of G. Piazza et al. [23] is similar to G. Piazza et al. [21] as it also considers the existing energy system with PV plants, batteries, charging stations, and electric carsharing services and aims to minimize the maintenance and installation costs.

The conclusion was that bike-sharing is more profitable than carsharing, but it also covers 93% of the transportation demand, while carsharing only covers 39% [21]. Additionally, it was concluded that introducing V2G technology would lead to considerable savings. It also reinforces “the importance of modelling the e-mobility infrastructure within optimisation tools in a way as far as possible close to the real operation of vehicles and charging points” [23].

The estimation of EV demand was also in the literature, considering that the vehicles are only charged in their residential district from the grid. In this paper, it was assumed that the cars have a similar mobility profile as an ICEV, leaving the house around 7:45 a.m. and returning at 5:00 p.m. Contrary to other articles, this considers that the vehicle starts charging immediately when it arrives home, so no smart charging is considered. The results showed that the households’ power peak at night coincides with the EV charging power peak, leading to an increase of up to 77.9% of the total power demand [24].

Lastly, different charging options in energy communities were compared [25]. The dumb charging strategy is the most basic one, and the vehicles are charged for their next trip when they return to the community; therefore, it does not require any kind of intelligence. In the centralized strategy, the charging of the vehicles is scheduled by an intelligent scheduling unit, and the EVs can charge at variable rates and with interruptions. Lastly, in the decentralized strategy, EV charging is scheduled by its owners, and a coordination system is needed to maximize the use of power generated locally. Besides these three strategies, four options were considered: with V2G and without V2G, and fast or slow chargers.

In the dumb charging approach, every case resulted in prominent peaks since all vehicles charged simultaneously. In comparison, the centralized strategy has around a 14% reduction in the peak, and the decentralized strategy has a 65% reduction. Although the centralized approach, at first glance, looks like the one that results in the highest decreases, since the common goal is to decrease the peak, the decentralized strategy wins because there is a dynamic pricing model used to incentivize the coordination. The reduction in peak values increases the transformer’s lifetime and reduces the necessity of installing a transformer with higher capacity [25].

There is also a study carried out regarding an energy community in Copenhagen that includes both heat pumps and electric mobility [26]. It concludes that integrating EVs into the power system results in a higher penetration of renewable energies. However, this would only be possible with the use of smart charging.

Other authors also suggest the creation of a “Solar EV City”, in which integrated systems of electric vehicles (EVs) and rooftop photovoltaics provide citizens with inexpensive, dispatchable, and CO₂-free electricity [27]. This concept seems promising in terms of cost reduction and CO₂ emissions reduction. However, policies will be needed to support this.

Regarding price and business models, EVs’ charging and parking prices in residential areas were studied using multi-objective optimization [28]. The objective function is established to maximize all parties’ interests. The conclusion suggests that parking lot operators in residential areas are interested in sharing energy with community members.

Regarding cost reduction, a case study considered a local community with households, industries, business stores, and electric vehicles. The conclusion was that introducing P2V (physical to virtual) electricity transactions reduced the community’s costs. The P2V market was proposed so that electric cars and prosumers could negotiate electricity between them [29].

Another paper considers a community with stationary and mobile energy storage as flexible resources, whereas mobile energy storage is electric vehicles [30]. The paper finds a community with public and commercial buildings, battery storage, and several electric vehicles. Chargers in the buildings with charging and discharging power were considered for electric cars. The results showed that this model could better match the energy demand with the local PV generation and lower the costs.

Regarding charging EVs in residential areas, a study concludes that using load management is fundamental to avoid grid reinforcement [31]. Additionally, the standardization of the electric vehicle and knowing the battery’s state of charge are essential for stability.

Lastly, there is a systematic review of the main topics and research trends. The authors have analyzed multiple papers, defining that the main research topics groups were multi-stakeholder with cost reduction, grid perspective, price and business models, energy management and grid integration, multi-energy and holistic approaches, emission-reduction, scheduling, and game theory [32].

2.4. Research Gaps

Some limitations in the literature on the role of aggregators in energy communities were identified, mainly in researching incentives and measures to encourage the creation of aggregations [33]. The authors have also identified the need for more research regarding the sizing of energy communities. There were also gaps in the EC’s frameworks in different countries [34]. Regarding electric mobility, it was mentioned that there is a knowledge gap due to the lack of socio-economic assessment, macro-level assessment, and the lack of integrated modeling approaches [35]. The authors also mention the lack of studies on the end-of-life phase.

It was found that regarding energy communities, there is a need for research on the social impacts and how they can be quantified. One topic that still needs to be added to the literature is exploring how to balance cost reduction and carbon emissions reduction. These two realities seem contradictory. To reduce emissions, more solar PVs, for example, need to be installed along with an energy storage system, which would increase the already high upfront costs. Regarding electric mobility, it is believed that in the future, the lifetime of vehicle batteries and the complexity of V2G should be studied carefully since it could make a big difference in integrating electric mobility in energy communities. Lastly, only some case studies still illustrate the integration of electric mobility in energy communities, especially those that show the impact on GHG emissions or specific economic advantages.

3. Methodology

3.1. Energy and Electricity in Portugal

Firstly, the distribution of energy consumption in Portugal was studied using different energy types (electricity, natural gas, piped LPG, bottled LPG, biomass, heating gas oil, and solar thermal). Regarding the total energy consumption in the residential sector, electricity represents 46.4%, followed by biomass (18.4%) and natural gas (12.4%) [36]. Additionally, electricity is the type of energy that has the most expenses in one year. Therefore, using this value as a starting point to verify if energy communities lighten the electricity bill makes sense. The energy consumption by square meter was assumed to be 103.50 kWh/m², and the electricity consumption was 48.02 kWh/m² [37].

Secondly, the electricity price in Portugal was studied. There are different types of tariffs: the simple tariff, the bi-hour tariff, and the three-way tariff. It will only be considered the simple tariff, which does not depend on the time of the day, and the bi-hour tariff (or hourly tariff), which includes peak and off-peak periods.

For the simple tariff case, the average price found and the one that will be used is 0.14 €/kWh, and for the bi-hour case, the prices that will be used are 0.17 €/kWh for peak periods and 0.11 €/kWh for peak periods [38,39,40]. It will be considered that the off-peak periods start at 10 p.m. and end at 7 a.m.

3.2. Equipment

After essential inputs such as electricity consumption and price have been defined, it was necessary to determine the inputs of the different equipment. The equipment selected were PV modules, batteries, inverters, electric vehicle chargers, and communications and management systems. It will include the main characteristics, sizing approaches, and other topics to help extract conclusions and choices for the simulation chapter (Section 4).

3.2.1. PV

There are three main types of PV modules: monocrystalline silicon, polycrystalline silicon, and thin film. The ones chosen for the simulation are polycrystalline silicon since they have the best price/efficiency ratio. Of the three, they have the best price and can reach an efficiency of around 15%. Not only that but also from what can be learned from the campus of the University of Genova, polycrystalline silicon panels led to better economic and overall outcomes.

After doing market research, the conclusion was that the average cost of a polycrystalline panel was 734 €/kW, and the average OPEX was 50 €/year [41]. To calculate the purchase price, 12 polycrystalline panels from the market were selected with different output powers (from 30 W to 455 W). Each panel’s price per kW was calculated, obtaining results ranging from 400 €/kW to 1400 €/kW. After that, the average of all values was calculated.

One important expression regarding PV panels is the energy yield ($E_a$, which can be obtained in the following way [42]:

$$E_a={\sum }_{i=1}^n{\eta }_{global\left(i\right)}{P}_{DC\left(i\right)}(G,T){\mathsf{\Delta }t}_i,$$

(1)

$$\mathrm{Where}{P}_{DC}(G,T)=\frac{G}{G^r}{P}_p[1+{{\mu }_P}_p\left(T-T^r\right)]$$

(2)

G^r and T^r are the irradiance and temperature in standard test conditions (STCs), which are 1000 W/m² and 25 °C, respectively. η_global is the overall efficiency, considering the derate factors and the inverter efficiency. µ_Pp is the percentual variation in the peak power per degree Celsius. P_p is the peak power of the solar panel. For the preliminary study, it will be assumed that n = 12 and i represents the number of months of the year. ${\mathsf{\Delta }t}_i$ represents the number of hours for the month i, and the irradiance and temperature are calculated for each month.

Other important aspects to consider when modeling and sizing a PV system for an energy community are as follows:

• The estimation of the total energy consumption, the energy consumption per household and user;
• The estimation of power demands and domestic load profiles;
• The comparison of the domestic load profile with the PV production profile;
• The estimation of the percentage of self-consumption to achieve;
• The estimation of the W_p (watt-peak) necessary from the PVs;
• The number of required PVs.

3.2.2. Inverter

Three inverters are considered: off-grid battery-based, grid tie inverters, and hybrid. The hybrid inverters are a mixture of the previous two. They function as a grid tie inverter with a battery to store energy during the day and are used when the sun is down [43].

It will be considered hybrid inverters since they can exist with and without batteries. Market research was performed to obtain the average price of hybrid inverters. The conclusion was that it was 289.13 €/kW, and the OPEX was around 2% of the yearly purchase price.

In terms of sizing the inverter, it is essential to consider the DC/AC ratio. The goal is to choose an inverter with a P_AC such that the DC/AC ratio is between 1.1 and 1.3 [42]. For values higher than 1.3, there is the risk of clip losses.

3.2.3. Battery

There are four main types of batteries: lead–acid batteries, lithium-ion batteries, nickel–cadmium batteries, and flow. Even though it is more expensive, the battery chosen for the simulation will be the lithium-ion one since it has a better discharge rate, energy density, and higher lifespan. LiFePO₄ (lithium iron phosphate) solar batteries have been argued to be the best types of batteries used in solar power systems due to their safety, longevity, stability, and wide operating temperature range. Because of that, they will be the chosen ones. After doing market research, the conclusion was that the average price for lithium-ion batteries is 729.3 €/kWh.

Here are some practical terms to consider when choosing the battery size or the number of batteries. Firstly, what is believed to be the critical factor is the budget and the cost comparison of the system with and without the battery. It is no surprise that batteries represent a significant percentage of the energy community’s cost, so there is a possibility that the best solution, economically, is not to have batteries at all.

The depth of discharge (DoD) is also an important topic to address. The depth of discharge is the percentage of the discharged battery capacity compared to the total energy capacity. DoD is also related to the energy that can be used without compromising the battery’s lifespan. The information found the most is around 80% for lithium-ion batteries.

If batteries are beneficial, it is vital to know the daily energy consumption, as in the PV system modeling. Then, it is important to understand the autonomy period. In the case of energy communities, it will be the amount of time necessary to use the energy from the battery since the aim is not to create an off-grid community.

3.2.4. EV Charger

There are three levels of chargers (level 1, level 2, and level 3), level 1 being the most basic and level 3 the most advanced. Usually, the higher the level, the higher the power output, and the faster it is to charge the electric vehicle.

The battery inside an electric vehicle works with DC current; however, the chargers can supply AC or DC current. The first case is what happens in level 1 and level 2. In this case, the conversion of AC energy from the station to DC energy is performed inside the vehicle. The onboard charger is the device responsible for this conversion.

In DC charging, which happens in level 3, the onboard charger is not used since the charging stations supply DC energy that goes directly to the battery.

A wallbox is an EV charger that is compact and lightweight and can be either a level 1 or 2 charger. When this device is connected to an electric vehicle, the system verifies how much power can be supplied by the grid, how much power the wallbox can give, which is between 7.4 kW and 22 kW, and how much power the onboard charger of the vehicle can admit. The lowest value will determine the charging power and how fast it charges. There are also intelligent wallboxes that allow innovative features, such as defining how many kilometers the user wants to charge.

Searching the wallbox market and some informative commercial websites, information was found regarding the price of purchasing and installing wallboxes. It was found that the cost of buying the device varied from 500 € to 1500 € and the installation from 700 € to 1500 €, depending on whether the house has the right conditions to have a wallbox or not and what construction work is necessary to install it. It was also not possible to find a €/kW value since it was found that wallboxes with 22 kW output were significantly cheaper than 7.4 kW wallboxes and others that were substantially more expensive. Therefore, it will be assumed that each wallbox will cost 1000 €.

Regarding the electric vehicle, it will be chosen an electric car with a 40 kWh battery, compatible with an AC charger and, more specifically, with a wallbox. Considering a DoD of 80%, it is only recommended that the battery charges 32 kWh to promote a long battery life span.

Table 1. Summary of the characteristics of the different equipment.

Equipment	Chosen Type	Price
PV	Polycrystalline silicon	734 €/kW
Battery	Lithium-ion	729.3 €/kWh
Inverter	Hybrid	289.13 €/kW
EV charger	Wallbox	1000 €/unit

shows a summary of the characteristics selected for each equipment.

3.3. Proposed Approach

The business model that is going to be used is the following:

• Every member of the community needs to pay an initial investment that will cover the different components of the energy community (PV modules, inverters, batteries, EV chargers, management, and communication systems) as well as their installation. This cost is divided equally between the members.
• Every year, the members of the community contribute similarly to the operation and maintenance costs of the community.
• The members of the community do not have to pay to use the energy produced by the photovoltaic system.
• Although the energy produced is not equally divided between every member, meaning that the energy is distributed according to necessity, it will be assumed that every member has a similar energy consumption.
• It will be considered that the vehicles will only start charging after 10 p.m.
• It is assumed that the vehicles are not all charging simultaneously since it will cause a very high energy demand. That way, the cars will charge between 10 p.m. and 7 a.m. This is possible because wallboxes have smart charging features.
• It is possible to have batteries only for the EV chargers, considering that the vehicles start charging when solar production is lower or even nonexistent.
• The remaining energy is not sold back to the grid.

The scenarios that will be considered are the following:

• Bi-hour or simple time cycle;
• High population density area or low population density area;
• With and without electric mobility.

This approach has some restrictions that will be discussed in Section 5.

3.4. Methods in the Simulation

A preliminary study was carried out before the simulation. Expressions (1) and (2) were used to calculate the energy yield by 1 kWp. It was then used as a “solver” in Excel to determine the peak power value leading to different percentages of electricity produced/consumed.

The ratio between the electricity produced and consumed (Eprod/Econs) represents the system’s maximum potential. If the energy demand matched the energy production in time, Eprod/Econs = 1, meaning it produced the same amount of electricity consumed. Econs is a fixed value representing the electricity consumption of the specific tested area.

Some calculations were also made regarding the cost of PV panels according to the percentage of energy produced compared to the energy consumed. The preliminary study also considered the inverts’ costs, quantities, batteries, and EV chargers. However, it was only considered that the batteries supply the energy consumed by the EV chargers.

Lastly, Helioscope software (https://helioscope.aurorasolar.com/) was used to verify the approach taken previously. This step also determined the maximum number of PV panels and the maximum electricity that can be produced with them.

The software used for the simulation part was Homer Grid by Homer Energy. Although it has the limitations explained in Section 5.5, this software was chosen because it has the capacity to add electric mobility to energy communities. Additionally, there is a diversity of economic tools to evaluate the different results, such as net present cost (NPC), levelized cost of energy (LCOE), internal rate of return (IRR), payback time, and annual savings. Furthermore, it was considered intuitive to use, contributing to choosing it.

In the Homer Grid 1.11.1. software, the first step was selecting the type of load curve. The community-type curve was chosen since the goal was to create a community. After that, it was necessary to insert the daily energy consumption of the community. Figure 2 presents the curve for the low population density case, but the one for the high population density case has the same shape but different values. As can be seen, the vehicles are charging between 10 p.m. and 7 a.m., and the peak occurs between 6 p.m. and 9 p.m. for the household’s daily consumption.

It was necessary then to choose the maximum (“upper”) and minimum (“lower”) values of PV power, inverter power, and battery capacity, which correspond to the interval where the simulation is made. Therefore, the systems do not suggest solutions with values higher than the ones set for “upper” or lower than those set for “lower”. In the case of the PV, it was considered that the power would result in 10% of the energy consumption for the “lower” value. The “upper” value was considered the value obtained in a helioscope.

For the interval defined for the inverter in the Homer Grid software, the minimum (“lower”) was 0 kW (even though this value was only possible if the PV did not exist). The maximum (“upper”) was obtained by dividing the upper value of PV power output by 1.1 since it is the minimum value of the DC/AC ratio accepted. Regarding the batteries, the “lower” value was set to 0 kWh, meaning no batteries. The “upper” value was obtained through iterations in the software. In all cases, a scenario with batteries is always worse than without. Therefore, the “upper” value for batteries was progressively increased until it was worse than having only the grid.

4. High- and Low-Density Cases

4.1. High-Density Case

A condominium was used for the first scenario, which is a high population density case. This condominium comprises four buildings, each with 17 floors and 2 apartments per floor, which results in 136 apartments. The area of the roof was measured using Google Earth, and it was obtained that the useful area was 1605.87 m².

It was considered an electricity consumption of 553,002.6 kWh per year, considering 3360 kWh/year per household [36].

Firstly, a preliminary study was conducted to understand a reasonable number of PV panels and inverters by determining the amount of PV panels that would lead to different percentages of energy consumption. Based on expressions (1) and (2), the first step was calculating the PV’s peak power, leading to different Eprod/Econs values. The results are presented in

Table 2. Peak power needed to achieve different percentages of energy consumption in high population density cases.

Eprod/Econs	Pp (kW)	Ea (kWh/Year)
0.1	37.628	55,300.26
0.2	75.257	110,600.52
0.3	112.885	165,900.78
0.4	150.513	221,201.04
0.5	188.142	276,501.30
0.6	225.770	331,801.56
0.7	263.399	387,101.82
0.8	301.027	442,402.08
0.9	338.655	497,702.34
1	376.285	553,302.60

. This study also considered that one-fifth of the apartments owned one electric vehicle, and the number of EV chargers was the same as the vehicles. Batteries were also considered to support the energy consumed by the EV. It was concluded that having batteries was too expensive. It also used the Helioscope software to do a preliminary layout of the PV panels. The conclusion was that it fitted 166.5 kW of PV and 140 kW of inverter, which produces between 40% and 50% of the energy consumed (the energy yield was 249,200 kWh) [44].

Homer Energy used software called Homer Grid to do the simulations. In the cases where electric mobility was considered, chargers with 22 kW were assumed, and it was deemed that vehicles start charging after 10 p.m. and charge during the night. For the hourly tariff case, the following was considered: between 7 a.m. and 10 p.m., the energy tariff was 0.17 €/kWh, and between 10 p.m. and 7 a.m., the tariff was 0.11 €/kWh. A 0.14 €/kWh tariff was considered in the simple tariff cases.

Table 3. Summary of the results obtained for the high-density case in Homer Grid (adapted from [45]).

	Simple Tariff + Mobility	Hourly Tariff + Mobility	Simple Tariff − Mobility	Hourly Tariff − Mobility
PV (kW)	153	123	148	125
Inverter (kW)	84.8	75.5	84.9	77.6
NPC (€)	855,309	1,010,328	850,170	802,795
LCOE (€/kWh)	0.124	0.128	0.114	0.110
IRR (%)	21%	16%	22%	16%
Payback time (year)	4.5	5.8	4.4	5.3
Savings (€)	26,473	18,410	26,040	18,590

shows the main results obtained for the four scenarios in the high population density case. It is important to note that the simulation also considered batteries. However, optimization concluded that the best option was not to have batteries since those cases presented worse results economically.

4.2. Low-Density Case

For the low population density case, 31 houses in a residential area in Alto de São João, Lisbon, were chosen. One particularity that was looked for was a similar functional area for installing solar PV panels. The useful area was 1964.87 m².

It was considered an electricity consumption of 156,262.9 kWh/year, 3360 kWh/year per household [36], and a factor of 1.5 to account for the fact that the houses are significantly bigger than the average.

Firstly, a preliminary study was conducted, and two conclusions were made. The first one is the same as previously, which was that batteries were too expensive. This is not a surprise since the same conclusion was obtained in the local energy community of the Savona Campus of the University of Genova. The second one is that it seemed to be more advantageous to have panels in low population density cases. Both these questions will be further explored in the simulation. It also used the Helioscope software to prepare a preliminary layout of the PV panels. The conclusion was that it fitted 133.8 kW of PV and 110 kW of inverter, which led to an energy production that was more than 100% of the energy consumed (the energy yield was 163,400 kWh) [44]. The PV’s peak power was calculated using expressions (1) and (2), leading to different Eprod/Econs values. The results are presented in

Table 4. Peak power needed to achieve different percentages of energy consumption in low population density cases.

Eprod/Econs	Pp (kW)	Ea (kWh/Year)
0.1	10.63	15,626.29
0.2	21.26	31,252.58
0.3	31.90	46,878.86
0.4	42.53	62,505.15
0.5	53.16	78,131.44
0.6	63.80	93,757.73
0.7	74.43	109,384.02
0.8	85.06	125,010.30
0.9	95.69	140,636.59
1	106.33	156,262.88

.

In the simulation part, it was considered the same assumptions.

Table 5. Summary of the results obtained for the low-density case in Homer Grid (adapted from [45]).

	Simple Tariff + Mobility	Hourly Tariff + Mobility	Simple Tariff − Mobility	Hourly Tariff − Mobility
PV (kW)	42.2	79.6	42.2	79.6
Inverter (kW)	24.1	55.3	24.1	55.3
NPC (€)	241,818	180,172	241,818	180,204
LCOE (€/kWh)	0.122	0.0773	0.115	0.0665
IRR (%)	21%	30%	21%	30%
Payback time (year)	4.4	3.2	4.5	3.2
Savings (€)	7402	18,935	7402	18,931

shows the main results obtained for the four different scenarios in the low population density case. It is important to note that the simulation also considered batteries. However, optimization concluded that the best option was not to have batteries since those cases presented worse results economically.

4.3. Comparing Estimation Results: Homer Grid and Preliminary Study

Simple tariff scenarios were used when comparing the preliminary research with the software results. In this comparison, only the energy produced, the DC/AC ratio, and the percentage of renewable energy produced compared to the energy needed are discussed. The values related to the energy produced were obtained in the software.

The PV quantity presented in Table 2 and Table 4, which was closest to the optimized values obtained in Homer Grid, was selected. The inverter and energy produced associated with that PV quantity were selected, which are presented in the same tables. Lastly, the ratio of energy produced by PV panels divided by the energy consumption was calculated.

As seen in

Table 6. Comparison of the preliminary study with the simulations.

	Low-Density Case			High-Density Case
	Homer Grid + Mobility	Homer Grid − Mobility	Preliminary Study	Homer Grid + Mobility	Homer Grid − Mobility	Preliminary Study
PV (kW)	42.2	42.2	42.53	153	148	150.513
Energy produced (kWh)	68,786	68,786	62,856	248,773	240,456	221,335
Inverter (kW)	24.1	24.1	32.9	84.8	84.9	115.85
DC/AC ratio	1.75	1.75	1.3	1.80	1.74	1.3
Renewable prod/cons	0.30	0.42	0.4	0.28	0.42	0.4

, overall, the energy produced by the PV panels is very similar to the one predicted in the preliminary study, showing that even though it was used a fast estimation to obtain that, it was carried out with a reasonable approximation. The same can be said regarding the renewable production/consumption ratio, especially in scenarios without electric mobility. This does not come as a surprise since it is expected that energy consumption increases when the charging of the vehicles is added; therefore, this ratio is lower.

On the other hand, the DC/AC ratio is higher than the interval expected of 1.1 and 1.3. In all cases, this ratio is higher than 1.7. It leads to the belief that there might be some moments when clipping losses occur. However, it is understood that having more inverter capacity would most likely increase the overall cost of the system.

5. Potential and Limitations

5.1. Economic Analysis

What was defined to be the best and the worst case for each scenario was the following:

• The lower the NPC, the better.
• The lower the LCOE, the better.
• The higher the IRR, the better.
• The lower the payback time, the better.
• The higher the utility bill savings, the better.

Based on this and Table 3 and Table 5, the high population density with an hourly tariff (HT) and electric mobility case seems to be the worst in all categories, except for savings. On the other hand, the low population density with HT and with and without electric mobility seems to be the best case, with very similar values. However, it is important to remember that economic data do not consider the cost of purchasing, maintaining, and installing the wallboxes. Therefore, it is believed that considering the addition of this factor, the case with electric mobility would become slightly worse.

One interesting thing noticed is that homologous cases with a simple tariff (ST), regardless of being in a high or low population scenario, have similar LCOE, IRR, and payback time values. For example, both cases with ST and mobility have very close values on these parameters, and the same thing happens for both cases with ST without mobility. Regarding IRR and payback, all four cases have the same results.

5.2. Emissions Analysis

Besides an economic analysis, it is vital to perform an environmental analysis (

Table 7. Emission analysis of all cases tested.

		Carbon Dioxide
		Initial (kg/Year)	Best Option (kg/Year)	Reduction (%)	Energy Produced by PV (%)
High Density	ST + Mobility	546,574	412,277	24.57	26.9
	HT + Mobility	555,739	442,832	20.32	22.0
	ST − Mobility	349,480	217,650	37.72	39.6
	HT − Mobility	349,480	234,648	32.86	34.6
Low Density	ST + Mobility	142,664	105,110	26.32	28.5
	HT + Mobility	142,664	66,820	53.16	44.9
	ST − Mobility	98,759	61,205	38.03	39.9
	HT − Mobility	98,759	22,914	76.80	59.1

) since this is one dimension of energy communities.

The first thing that can be noticed is that the high population density scenario cases have much higher initial and “best option” (being the best option, which results in Homer Grid) emissions. In terms of CO₂ emissions, high population density with HT and electric mobility is the one that has the highest level per year and additionally is the one that has the lowest reduction in emissions when the system changes from grid (initial) to grid + PV (best option).

Contrarily, the low population density with hourly tariff and without mobility case is the best scenario regarding CO₂ emissions in the PV + grid architecture, has the highest percentual decrease when this option is used, and the highest percentage of energy produced by solar panels. The second-best option is the low population density with HT and electric mobility. Once again, the scenarios in low-population density areas show better results.

Something interesting to notice is that the cases with electric mobility tend to have higher emissions and lower reductions compared with their homologous without electric mobility. It is believed that this happens because charging vehicles increases energy consumption. Additionally, it increases consumption when there is no sunlight, increasing the need to purchase energy from the grid. Since the power from the grid is not 100% renewable, using fossil fuel sources increases the overall emissions of CO₂.

5.3. Is high- and Low Density a Win–Win?

After analyzing the results obtained for the high population density and the low population density regarding economic factors, such as NPC, LCOE, IRR and savings, and environmental factors, such as the CO₂ emissions and the load profiles, it seems that low density is more probable to present a win–win solution than the high-density population scenario.

One of the reasons for this is the limited roof area available for energy generation compared to the amount of energy consumed. This is because high population density areas are usually built in height, while low population density areas are built at the ground level. Since the first one consumes considerably more energy than the second one, it makes sense that having an energy community in a low population density scenario is more profitable if both have similar areas. On the other hand, places with higher population density tend to have more EVs, which increases consumption even more.

Additionally, since consumption is higher in areas with high population density, the initial emissions are also higher. Additionally, even when PV panels are introduced, a large percentage still comes from the grid. Because of that, the emissions are still higher than in the case of low population density.

However, one solution for this was implementing PV panels in the facade of the buildings. Although it is more seen in office buildings, in the future, it will most likely be also seen in residential buildings. This solution will increase energy production, especially in cases with high population density, which presents a vast facade.

5.4. Alternative Consumption Patterns

Initially, it was considered a community load curve with lower values between midnight and 4 a.m., intermedium values between 8 a.m. and 4 p.m., and higher values between 6 p.m. and 9 p.m. This curve is valid for low- and high-density cases, and the software suggested it for energy communities. That curve is presented in Figure 2a and is related to the low-density one but could be adapted to a high-population scenario. It was also assumed that the EV chargers have a consumption curve where they consume energy between 10 p.m. and 7 a.m. The curve in Figure 2b is related to the low population density but could be adapted to the high-density case.

On the other hand, it could be interesting to understand what changes when the consumption pattern changes. A predefined load curve for energy communities was used to perform the different alternative simulations the Homer Grid software provided. That curve is represented in Figure 2a. However, Homer Grid had other types of predefined load curves, such as residential (Figure 3a) and commercial (Figure 3b). Additionally, it was also defined two other load profiles, one like the residential one, but where the primary time zones where the energy is consumed are early morning and late afternoon/night (Figure 3c), and the other one that represents the consumption when the day is spent at home. It has an additional peak during lunchtime (Figure 3d). All these curves were adapted from the low population density consumption values.

Taking the five different load curves into consideration (the community load curve presented in Figure 2a and the four alternatives given in Figure 3), the data from the PV production in the low population density with HT cases were used to calculate the output for an average day. The excess of renewable energy and grid purchases were then studied.

First, the average production curve for one day was calculated to calculate the excess PV energy (denominated as “Ex”), considering the inverter output power. Then, this information was subtracted from the PV production to determine the energy demanded for each hourly timestamp. Of course, when the result was negative, it was assumed to be zero. The average day’s values for each hour were summed to obtain the total excess of PV energy. It is essential to understand that this value is not a reality since many more losses must be considered when calculating.

A similar process was used to calculate the grid purchases (denominated as “GP”). Contrary to what was described before, it was subtracted from the energy demand of PV production. It was also considered as zero in the negative results. To obtain the total grid purchase for each case, the average values for each hour were summed. While the results were the same for the cases with and without EVs in excess energy, the same did not happen for the grid purchases. The results are presented in

Table 8. Excess energy and grid purchase for each alternative case (kWh).

With/Without Mobility
Ex (Community)	Ex (Residential)	Ex (Commercial)	Ex (Residential 2)	Ex (Day at Home)
117.40	133.71	33.53	180.68	144.66
Without Mobility
GP (Community)	GP (Residential)	GP (Commercial)	GP (Residential 2)	GP (Day at home)
238.62	254.93	154.74	301.89	265.87
With Mobility
GP (Community)	GP (Residential)	GP (Commercial)	GP (Residential 2)	GP (Day at home)
442.58	458.88	358.70	505.85	469.83

.

By analyzing the results, the commercial case had the lowest grid purchases, both with and without electric mobility, and presented the lowest excess of energy. This happens because the highest demand happens between 8 a.m. and 4 p.m. when the PV production is the highest. However, it is essential to consider one thing. Considering a commercial scenario with vehicles charging at night does not make sense since EV owners do not tend to leave their cars at work during that time.

Because of that, it was necessary to introduce an alternative charging pattern: daytime charging. This could represent what happens in the commercial environment and when the vehicle is left at home during the day. The results regarding the combination of the different scenarios with daytime charging are presented in

Table 9. Excess energy and grid purchase for a daytime charging routine (kWh).

Ex (Community)	Ex (Residential)	Ex (Commercial)	Ex (Residential 2)	Ex (Day at Home)
15.50	23.12	0	22.04	21.04
GP (Community)	GP (Residential)	GP (Commercial)	GP (Residential 2)	GP (Day at Home)
340.67	348.30	325.17	347.21	346.21

.

Combining daytime charging with a commercial load curve can achieve a more significant match between consumption and production. In this scenario, there is no excess production, and the grid purchases are also lower since no vehicles charge during the night. Considering the case “day at home”, it was also noticeable that the excess of energy and the grid purchases dropped. Additionally, it can also be seen in Table 9 that the excess of energy in the “day at home” scenario decreases significantly. The grid purchases decrease as well. The option of charging the vehicle during the day when the owners spend the day at home also seems to be a good option since the energy excess generated during the day can be used to charge the vehicles.

Considering the default EV charging model in Homer Grid, which was charging between 6 a.m. and 5 p.m., the low population density with HT and electric mobility case in the community’s default curve was tested. The results are the following (

Table 10. Results obtained for daytime charging in Homer Grid.

Architecture			Cost				Project Economics
PV (kW)	Battery (kWh)	Converter (kW)	NPC (€)	LCOE (€/kWh)	Operating Cost (€/Year)	CAPEX (€)	IRR (%)	Payback Time (Years)
123	N/A	86.7	153,727	0.0659	5694	80,118	30	3.2

):

These results seem even better than the ones obtained for the low population density with HT and no mobility, at least economically. Regarding CO₂ emissions, it counts 24,789 kg/year, slightly higher than the LPD HT without mobility.

This leads to the belief that energy communities and electric mobility might be a win–win solution related to the charging schedule.

Looking at all the different load curves for households and electric vehicles, this study cannot give an undebatable answer on integrating electric mobility in energy communities since it would have been necessary to have a robust technical analysis to complement the economic and environmental analysis. For instance, it was not considered that it may be required to increase the capacity of the supply distribution network due to having multiple vehicles charging during the night, which creates a very high energy demand. This is a clear example of how economic analysis should be complemented by a technical one since increasing the capacity of the supply distribution network results in a significant investment.

However, some practical implications can be applied to modeling energy communities. Firstly, considering the residential factors, energy communities should be approached more from a neighborhood perspective. This includes neighborhoods with low population density and the possibility of distributing the excess energy during the day to local shops, schools, churches, etc. This will enhance the sense of community and promote local valorization; the remaining energy will not be thrown away.

Some considerations need to be made regarding modeling electric mobility in energy communities. The first one is the possibility of integrating batteries or not. This would be a deal breaker, and it is believed that when batteries become more affordable, this integration will become more accessible in the residential sector. Another recommendation that would be given before starting an EC with electric mobility is questioning; firstly, how many neighborhoods have an electric vehicle and their charging patterns. Some people might typically charge their car during the day at work. Others prefer to charge as soon as they get home, and others who work at night will instead charge their vehicle during the day at home. Managing the different charging patterns will help determine how beneficial it is to have chargers in the community.

5.5. Other Types of Integration

This research is focused on integrating electric mobility in energy communities because householders have electric vehicles incorporated into the community’s energy consumption. However, this approach is not the only one that can be considered.

The first approach that could have been taken, which is closer to the chosen one, is car sharing, which is, in fact, one of the activities of electro-mobility. For this, two options are seen: having a similar system as the one presented earlier, where the energy consumption includes not only the EV chargers but also the household appliances, or having a system where the energy production is sized only according to the energy consumed by the chargers, which can be seen as more of “carsharing community”. However, one of the downsides of car sharing is the people’s mindset. Most people are not yet convinced that they might not need to own a vehicle to do their everyday tasks, and this is one of the main challenges nowadays for integrating carsharing into electric mobilities.

The second approach would be like Coopernico’s use of energy storage systems based on second-life batteries. In this case, the reused batteries would be from electric vehicles. Considering that old EV batteries only have 20% of their initial capacity, and if batteries have 40 kWh capacity, it will be 8 kWh that is not thrown away. Mainly because, as seen previously, batteries could resolve the mismatch between the demand and the production, mainly in EV charging. However, this type of solution is still in the early stages. Another option that is believed to be essential to consider, not only for energy communities but for the environment itself, could be recycling the EV batteries, which is, of course, no easy task, but could be worth it to save valuable components, such as nickel and cobalt [46].

Besides reusing the old EV batteries in the EC system, there are other forms to encourage people to own EVs. It was known that consecutive charging/discharging cycles lead to the degradation of the battery. However, this should not discourage members from acquiring electric vehicles for their energy community.

One solution is battery leasing, where the owner buys the vehicle but not the battery. Instead, the owners lease the battery to a third party and need to pay a monthly fee. Usually, the leasing company is responsible for the maintenance or replacement of the battery if it is not working correctly. There are also options, such as battery swapping stations. This is not common yet, but there are stations where the discharged battery of the car is substituted by a charged one. Lastly, mentioning the warranties for EV batteries is essential, as it seems to be an assurance for EV owners.

5.6. Limitations in the Research

Some limitations were detected while performing the simulations in the software:

• Firstly, a weakness that was noticed is that it is impossible to include the costs associated with EV chargers, in this case, of the wallboxes in the simulations.
• Another thing that could be very helpful is seeing how CO₂ changes during the day and, therefore, relating those emissions to the presence of vehicles charging.
• For the hourly tariff, it was considered that between 10 p.m. and 7 a.m., the electricity price was 0.11 €/kWh, and between 7 a.m. and 10 p.m. the price was 0.17 €/kWh. It was not considered having vehicles charging, for example, between 7 p.m. and 6 a.m. This scenario could have been helpful in understanding when the best hours to charge the car are from the moment it arrives in the afternoon to the moment it leaves in the morning.
• Lastly, unlike what was planned in Section 3, the software cannot include management and communication systems; therefore, its costs were not considered.

It is also essential to address the limitations of the initial hypothesis. As seen before, the hypothesis was studying the financial benefits of adding electric mobility in energy communities through wallboxes that charge the vehicles during the night in a residential EC, which has some limitations. Firstly, it was assumed that all users have similar electricity consumption, which is false. All households have different consumption patterns, which depend on their day-to-day life. The same can be applied to the charging patterns of vehicles.

Additionally, the possibility of selling the remaining electricity to the grid was not considered, but this could be a factor that would benefit the community economically. It is believed that selling energy back to the grid leads to financial gains for the members, which, although positive, is not one of the main goals of an energy community. This would be better suited to a collective self-consumption activity rather than an EC. Additionally, the fact that the energy is not sold back to the grid allows the inclusion of small shops or schools nearby to enter the community and potentially use the remaining power. This enhances the social dimensions of energy communities.

Regarding the chargers, they were assumed to be wallboxes with smart charging features. V2G was not considered due to its complexity. However, it should not be discarded as a new hypothesis. Lastly, it was assumed that all vehicles charge during the night, which is a limited hypothesis since it does not explore when it is more beneficial to charge the EVs. Lastly, the cost of installing the different equipment was not explicitly addressed. These values were hard to find and estimate. Additionally, it was considered that all EVs have the same battery, which is not a reality since different vehicles have different batteries.

To further enhance the understanding of the impact of electric vehicle charging on the power supply system, it is essential to simulate other scenarios. For instance, if some electric vehicles are charged during the day from installed PV systems and not at night, the load at night could be significantly reduced. This is mainly due to demand, which could create load failure during daytime hours. Therefore, equalizing the power system’s load schedule with electric vehicle charging could be a highly effective strategy.

To ensure the best use and management of the powering network management, rechargeable batteries are installed inside the power supply system of multi-story residential buildings if increasing the capacity of the supply distribution network will require significant capital investment.

Another critical thing to note is that this case is applied explicitly to Portugal. It considers Portuguese legislation regarding electric mobility and energy communities and the electricity prices applied to this country. Therefore, the business model used might need to be adapted to suit each country’s legislation. The electricity tariff would also need to be adapted.

Lastly, one of the significant limitations of this work is that the capacity of the grid to supply a large number of vehicles charging during the night was not considered, and there were restrictions on the supply of cable lines and power transformers. The conclusion was that, economically, it was more profitable not to have batteries, and because of that, these restrictions should be accounted for in the following research.

5.7. Other Limitations

Besides the software and hypothesis limitations, it is crucial to consider the limitations in integrating electric mobility in energy communities.

Firstly, the social limitations can be considered. It is believed that the fact that some members benefit from the EV chargers and others do not will create disparities within the community. This is because some members consume less energy than others, which could create disagreements when contributing to the equipment’s maintenance and installation. On the other hand, considering the application of energy communities in the residential sector, all neighbors must accept the creation of the community. However, there is already some resistance in the population. As it is known, ECs already have a high upfront cost, and because of that, if more people participate, the price is lower for each other, and the creation of an EC becomes more appealing. Lastly, it seems that it is not yet straightforward and advantageous enough incentives and policies to make energy communities more popular in the residential sector.

Regarding economic barriers, one of the most important to mention is the high upfront costs. Even though some companies manage ECs and have members not paying for equipment installation as part of their business model, this is not a standard practice. For some communities with lower incomes, the high costs could be a determinant factor in not creating a community. Adding electric mobility into the EC increases this cost even more. Additionally, those who own an electric vehicle might need to improve their contracted power to account for the extra consumption of the car. This will most likely increase the electricity price and, because of that, the cost of charging the vehicle.

Lastly, it is essential to mention the concerns regarding privacy concerns. Since EV owners could potentially be the ones who present the higher energy consumption, depending on the business model, it could be necessary to have access to the charging history for billing purposes. On the other hand, depending on the charger being used, it can use the vehicle’s data, such as energy consumption and charging patterns, to optimize the process, which could compromise the owner’s privacy.

6. Conclusions and Further Developments

6.1. Conclusions

This research aims to answer whether integrating electric mobility, in the form of car charging stations, in energy communities is a win–win solution. The simplified answer is that it might be. The truth is that it depends on many factors. So, it is now important to summarize some of these aspects.

The first one that is important to mention is the population density factor. It is understood that this factor influences the integration of electric mobility in energy communities and the existence of energy communities. As seen before, areas with high population density tend to be expressed by buildings with many floors; therefore, the construction is being carried out at a height. Because of that, the rooftop area tends to be lower than it would have been if the construction was carried out at the ground level. Therefore, the energy generated by the PV panels could not cover the demand. Regarding electric mobility itself, naturally, areas with higher population density will have a higher number of electric vehicles and chargers. This leads to the first conclusion that energy communities with electric mobility could be a win–win solution in low-density areas but not in high-density ones.

Because of that, the second important factor is the presence of batteries. This is a two-end sword because while batteries could help resolve the mismatch between the production and consumption problem, they introduced additional costs that are considerably high. During the day, at least in the low population density case, there is a surplus of electricity that is very much needed during the night, even more in the cases with electric mobility, and a way of doing that is through batteries. Therefore, the second conclusion is that if economic factors are not a constraint, batteries make energy communities and electric mobility a win–win solution.

The third factor that is important to mention is the load curve itself. Regardless of whether they have electric mobility, grid purchases are the lowest in a commercial-type load curve scenario. Even though the object of study of this research was the residential sector, this leads to the conclusion that energy communities might work better in a more industrial/commercial scenario. Regarding electric mobility, it is not believed that it makes a significant difference in a commercial or residential scenario, in case the vehicles are charging during the night, since during this time in both scenarios, the demand could be higher.

This leads to the fourth factor, the demand curve of the EV chargers. As a hypothesis, however, it was considered that the vehicles charge during the night. This hypothesis might be different from what leads to energy communities and electric mobility combined being a win–win solution. When compared to vehicles charging at night, daytime charging presents better results.

To conclude, the combination that is believed would make energy communities and electric mobility a win–win solution could be low population density with daytime vehicles charging or low population density with night-time charging of the cars but with batteries. Other scenarios, such as commercial ones with daytime charging, could also be beneficial. Still, it would need to be studied carefully since, in this case, the energy consumption of the building plus the chargers could overcome the PV production during the day. Having this said, the hypothesis studied, night-time charging in a residential community scenario, is not a win–win but an acceptable solution. This means that it is a good scenario. It is almost as good as not having electric mobility and should not be discarded, but being nearly as good does not make it a win–win solution.

Some practical proposals that can be implemented are considering a low population density area, the number of EVs and their routines and charging patterns should be explored. As seen previously, one of the critical factors for the successful integration of electric mobility in energy communities is the load curve of the charger. Therefore, studying this matter carefully before starting the project makes sense. A few pieces of information could be relevant to collect, such as the preferred charging hours, the model of the EV, the type and capacity of the battery, the kind of charger that can be connected to it, and the overall usage of the vehicle.

Additionally, when starting a community in a residential neighborhood, the surrounding area should be investigated to find a small business or school to receive or buy the excess electricity produced during the day. This type of place tends to have a load curve like the production curve.

6.2. Further Developments

Regarding further developments, it is suggested that factors not considered during this research, such as the installation costs of the different devices, the management and communication systems, and selling the excess energy back to the grid, be introduced in the subsequent investigation. A different initial hypothesis is suggested to be tested, which would be settled in a low population density environment since it is the most economically beneficial scenario. Here, the main goal would be to find which time of the day is the best for charging vehicles. It would also be suggested to study whether the introduction of V2G technology would be beneficial.

Using electric vehicles could create challenges in managing energy systems. This depends strongly on the energy intensity of batteries in electric cars. Since high-energy intensity batteries could make the management of energy systems complex, one limitation of the study is considering a defined battery. This limitation is because of high battery variation, so further developments and research are relevant. In several countries, dynamic electricity tariffs are applied at fast charging stations, significantly reducing the maximum electricity consumption of electric vehicles from the distribution network. Electric cars and battery charging are a challenge to energy networks that must be considered in future research.

The commercial and office zone scenario should also be studied more carefully. Even though this article focused on the residential sector, it was possible to conclude that a commercial scenario would be advantageous in creating an energy community with electric mobility. Therefore, parking lots where vehicles can be charged during the day should be considered to have a more significant match between demand and consumption.

Economic or fiscal conditions must be developed to promote PV with a synergistic link with electric mobility to ensure an effective win–win for the economy and environment. This includes introducing or enlarging PV systems in multi-story residential buildings, private households, and other building typologies integrating parking and chargers with energy management systems.