Sector Coupling and Flexibility Measures in Distributed Renewable Energy Systems: A Comprehensive Review

Abstract

Distributed energy systems (DESs) are crucial for renewable deployment, but decentralised generation substantially increases flexibility requirements. Flexibility is framed as a system property that emerges from the coordinated operation of demand, storage and dispatchable generation across multi-energy carriers. Demand response schemes and demand-side management can provide flexibility, but their effective potential is constrained by user participation. Sector-coupling strategies and energy storage systems enable temporal and cross-sector decoupling between renewable generation and demand. Electrochemical batteries are technically mature and well suited for short-term balancing, but costs and environmental impacts are significant. Power-to-Heat with heat pumps and thermal energy storage is a cost-effective solution, especially when combined with low-temperature district heating. Electric vehicles, when operated under smart-charging and vehicle-to-grid schemes, can shift large charging demands feeding energy into the grid, facing battery degradation and infrastructure costs. Power-to-Gas and Power-to-X use hydrogen and electrofuels as long-term storage but are penalised by low round-trip efficiencies and significant capital costs if power-to-power with fuel cells is applied. On the supply side, micro-CHP can provide dispatchable capacity when fuelled by renewable fuels and combined with seasonal storage. Costs and efficiencies are strongly scale-dependent, and markets, regulation, digital infrastructure and social acceptance are key enablers of flexibility.

1. Introduction

The potential risks associated with human-induced climate change can lead to irreversible damage to ecosystems and cause negative impacts on people’s economic and social conditions [1]. Global efforts are needed to limit global warming below 1.5 °C, and the rapid reduction of greenhouse gas (GHG) emissions in energy systems is the priority for the coming years [2].

Most countries have defined roadmaps for decarbonising their energy and economic systems. This will lead to a rapid deployment of renewable energy sources (RESs) to replace conventional fossil fuels [3].

However, conventional energy systems base system flexibility on the generation side. When most of the generation is not programmable, it is necessary to develop flexibility within the system itself and on the demand side [4].

Initially, this challenge has been addressed by seeking solutions exclusively within the electricity sector [5]. Nevertheless, the large-scale RES integration involves several issues related to the fluctuations and intermittence of renewable generation, and those are unlikely to be solved in an efficient and cost-effective manner with electric batteries alone [6].

In the last decade, research concerning the system flexibility has been interested in the electric carrier conversion in order to use more convenient energy storage systems, in accordance with the sector-coupling concept [7].

Different definitions, such as Smart Energy Systems [8], Multi-energy Systems [9] and Integrated Energy Systems [10], have been proposed in the literature in order to identify a holistic approach to energy systems that overcomes the single-sector view.

The general concept guiding these definitions is that by exploiting synergies between sectors and different energy carrier conversions, it is possible to identify the best configuration for the electricity sector as well as for the entire energy system.

Furthermore, the RES deployment will not only take place by means of large centralised systems but also by installing numerous small distributed systems [11].

The concept of “distributed” has been introduced in the literature to overcome the concept of “decentralised” [12]. Indeed, the latter is simply the opposite of “centralised” and indicates a geographical proximity between production and consumption [13].

On the contrary, the deployment of Distributed Energy Systems (DESs) involves a new techno-economic paradigm concerning changes in the energy system configuration, not only in the generation side but also in the management, storage and demand side, as well as changes in the institutional and economic framework [14].

The European Union has recently introduced Renewable Energy Communities (RECs) into its legislative framework in order to encourage citizen participation and enable their members to collectively produce, consume, store and sell renewable energy [15].

With the emergence of this new paradigm, issues related to the non-programmability of renewables will have to be addressed locally in order to minimise the electricity injection into the grid. Indeed, existing power networks are able to accommodate up to 40% of renewable electricity [16,17]. Distributed energy systems based on renewable generation have to be managed with the aim to maximise local self-consumption [18]. Therefore, the RECs’ establishment cannot be separated from the implementation of small-scale application of sector-coupling technologies and flexibility measures.

1.1. Scope of the Review

This review analyses and discusses the state of the art, perspectives and main challenges of applications of sector-coupling strategies and flexibility measures in DESs. While most existing reviews address flexibility solutions at system or national scale, this work explicitly concentrates on small-scale applications, where generation, storage and demand coexist and where grid constraints and local self-consumption become crucial criteria for DES design. The decentralisation process fostered by DESs is taken as a starting point to investigate how the need to maximise local renewable energy self-consumption reshapes the techno-economic role of flexibility options at the distributed level.

This review encompasses both demand-side solutions, flexibility measures based on storage systems, and flexible generation options in small-scale multi-energy systems. It covers a broad set of technologies and strategies for providing system flexibility, such as electrochemical batteries, electric vehicles and charging strategies, Power-to-Heat with heat pumps and thermal energy storage, Power-to-Gas and hydrogen-based options, small-scale combined heat and power and other supply-side flexible generators, as well as demand-side management schemes and digital market infrastructures. For each option, the analysis emphasises how performance, costs and integration constraints change when moving from utility-scale to DES-scale deployment. Particular attention is given to capital cost, efficiencies, operating constraints, and environmental impacts, and to the implications of modularity for distributed energy planning.

Beyond the purely technical and economic characterisation, the work also addresses the design of local markets and peer-to-peer trading schemes, and the potential role of digital technologies such as blockchain in enabling decentralised coordination. By combining technical, energy, environmental and economic perspectives, the review aims to provide a comprehensive and holistic assessment of flexibility solutions for small-scale distributed systems, identify critical gaps in current knowledge, and outline future research directions.

To support this critical review, the authors scrutinised the most recent peer-reviewed literature on distributed renewable energy systems, with specific attention to technology options, sizing approaches, flexibility concepts, and market/regulatory enablers. The Scopus database was used as the primary source to identify peer-reviewed journal and conference papers, considering a time window mainly spanning 2012–2025 (with earlier seminal works included when necessary to frame key concepts). Searches were performed through an iterative, topic-driven approach, using keyword families aligned with the main sections of the manuscript.

Specifically, a first block of context-related terms (non-exhaustively: “distributed energy system”, “district energy system”, “multi-energy system”, “integrated energy system”) was combined with a second block addressing flexibility, markets and policy enablers (non-exhaustively: “sector coupling”, “energy flexibility”, “power system flexibility”, “demand response”, “local flexibility market”, “aggregator”, “ancillary services”, “peer-to-peer energy trading”, “blockchain”, “carbon pricing”, “emissions trading”, “subsidies”), as well as technology-specific terms (e.g., “electrolyser”, “hydrogen”, “battery energy storage”, “thermal energy storage”, “heat pump”, “power-to-X”).

Title/abstract screening was followed by full-text assessment to ensure relevance to the scope (distributed/district applications and their techno-economic, operational, or market/regulatory implications). Backward and forward snowballing from key references was used to complement database searches and mitigate omissions across the heterogeneous topics covered by the manuscript.

1.2. Outline

This paper is structured as follows: Section 2 introduces the DES concept and discusses the emerging necessity of maximising local energy self-consumption in DESs. Section 3 defines the concept of system flexibility from a multi-energy perspective and positions sector coupling within this broader framework. Section 4 reviews demand-side flexibility options and demand response schemes in DESs. Section 5 analyses sector-coupling strategies and storage-based flexibility measures, including electrochemical batteries, electric vehicles and smart charging strategies, Power-to-Heat with heat pumps and thermal energy storage, and Power-to-Gas and hydrogen-based options, with a focus on their techno-economic and environmental performance at small scale. Section 6 examines supply-side flexibility and the role of small-scale dispatchable generation, in particular combined heat and power systems, in highly renewable DESs. Section 7 discusses emerging market architectures, regulatory frameworks and digital platforms required to coordinate distributed resources, with a focus on local markets, peer-to-peer trading and blockchain-based solutions. Finally, Section 8 summarises the main findings of the review and identifies key research gaps and policy implications for the deployment of sector-coupling strategies and flexibility measures in DESs.

2. Distributed Energy Systems

2.1. Concept and Definition of Distributed Energy Systems

DESs are local, modular energy infrastructures that combine distributed generation, conversion, storage and flexible demand at different scales [19]. The different technologies operate close to end users, interacting with higher-level networks, especially with the power grid. They extend earlier notions of decentralised or distributed generation, which focused mainly on small single-vector generators, towards integrated multi-vector systems where electricity, heating, cooling, gas and hydrogen are generated, managed and optimised with local loads and network constraints [20].

Recent reviews classify DESs along multiple dimensions, including primary resources and technologies (CHP-based, renewable-dominated or hybrid), energy vectors (single-electricity versus multi-energy hubs), grid connection (islanded versus grid-connected), temporal role (backup, peak-shaving or continuous supply) and spatial scale from individual buildings and microgrids to districts and regional clusters [19,20].

Within this framework, multi-energy DESs explicitly exploit cross-sector synergies by integrating renewable sources and multiple storage technologies to smooth variability and improve overall efficiency [21]. Specifically, sector-coupling strategies make it possible to expand the range of options beyond electrochemical batteries to include other forms of distributed energy storage systems such as thermal storage, hydrogen or electric vehicles [22].

DES applications in the literature span from building-scale systems, to industrial microgrids, as well as neighbourhood-level energy hubs and regional distributed energy systems [23,24].

Considering multiple energy vectors, energy storage systems and different scales, the design and operation of DESs pose significant modelling and planning challenges [25]. The complexity of defining DES optimal design requires representing, sizing and managing RES generation, multi-energy storage dynamics and interactions with energy networks at the same time [26].

Recent bibliometric analyses highlight a rapidly growing body of research around hybrid DESs, optimisation methods and control strategies [27].

Beyond techno-economic aspects, DES deployment is shaped by regulatory frameworks, market design and social acceptance. Especially in the growing context of RECs, neighbourhood-scale studies underscore the importance of perceived local benefits, fair allocation of costs and co-ownership models, while also identifying regulatory and structural barriers [28].

In this context flexibility emerges as a priority in order to integrate distributed RES into the energy system [29].

2.2. Energy Self-Consumption as a Necessity for Distributed Energy Systems

In the first stage of renewables’ deployment, local excess was not considered an issue as it could be easily accommodated on the distribution network. Several support schemes also incentivised the production of renewables through net metering systems, according to which, for certain quantities, the feeding of renewable energy into the grid was well remunerated and thus cost-effective. The network was seen as a storage system capable of integrating the small quantities electricity fed in. However, when many feed-in points coexist on the same distribution grid, they can lead to a large excess and an imbalance in grid parameters.

In a future characterised by the development of numerous energy communities, they cannot rely on the grid, but will have to manage the generation excess within the community itself in order to self-consume the energy produced. When renewable energy is produced in a residential community, it is not completely self-consumed on site. Therefore, without storage systems, there is high overgeneration or under-sizing of renewable plants. The issue of self-consumption is well known in recent literature and presents itself as a primary necessity in the design of distributed energy systems.

A formal definition of self-consumption is provided in Ref. [30]. Accordingly, the instantaneous energy self-consumption $E_{SC}\left(t\right)$ is the minimum of the instantaneous electricity demand $E_D\left(t\right)$ and the instantaneous RES generation $E_P\left(t\right)$.

$$E_{SC}\left(t\right)=\min \left\{E_P\left(t\right),E_D\left(t\right)\right\}$$

(1)

On an annual basis, the energy self-consumption ($E_{SC})$can be defined as follows:

$$E_{SC}={\int }_{t=t_1}^{t_2}{E}_{SC}\left(t\right)={\int }_{t=t_1}^{t_2}\min \left\{E_P\left(t\right),E_D\left(t\right)\right\}$$

(2)

Furthermore, the concepts of Self-Consumption Ratio ($SCR$) and Self-Sufficiency Ratio (SSR) have often been used in the literature [31,32]. The SCR can be defined as the ratio between the renewable energy self-consumption and the overall RES production over one year period.

$$SCR={\displaystyle \frac{E_{SC}}{E_P}}$$

(3)

The $SSR$ can be defined as the ratio between the $E_{SC}$ and the annual REC’s electricity demand ($E_D$).

$$SSR={\displaystyle \frac{E_{SC}}{E_D}}$$

(4)

Figure 1 shows the $SCR$ and $SSR$ as functions of the ratio between $E_P$ and $E_D$, in a case study of a residential renewable energy community in Rome and considering only PV as renewable generation [33]. The analysis is computed on an hourly basis, and no energy storage systems are included in order to highlight the temporal mismatch between PV production and electricity demand.

As the annual ratio between $E_P$ and $E_D$ increases, $SCR$ falls steeply, from nearly 1 at low PV penetration to about 0.4 when $E_P$ equals $E_D$. $SSR$ rises only gradually and levels off at roughly 0.4 in the absence of storage. Thus, even when annual PV production matches annual demand, less than half of the PV generation is actually consumed on site and only about 40% of the electricity demand is met instantaneously by PV. The gap is driven by the temporal mismatch between daytime PV peaks and evening-dominant residential loads.

Therefore, by simply oversizing PV quickly yields diminishing returns for SSR and depresses $SCR$, because incremental generation increasingly occurs outside demand windows. Robust techno-economic assessments of flexibility options must therefore be based at least on hourly profiles to capture actual technologies’ operation and marginal benefits.

3. System Flexibility

Several different definitions have been given to the flexibility concept, and their evolution reflects the progressive broadening of the problem addressed.

Early works rooted in power-system operation interpreted flexibility mainly through the lens of operational adequacy (e.g., ramping and reserve capability) needed to follow net-load variations over multiple time scales [34].

With the increasing penetration of variable renewable generation, flexibility has been further framed as the ability of the power system to respond to supply–demand variations within economic and technological boundaries, stressing the role of constraints and costs in enabling such response [35]. In parallel, the focus has expanded from generation-side measures to demand-side and prosumer-oriented actions, where flexibility is increasingly associated with the ability to react to external signals (e.g., market, grid or control signals) by modulating demand or generation, and by exploiting temporal shifting strategies [36]. This shift is consistent with recent contributions that explicitly connect flexibility to the possibility of acting on the temporal profile of energy demand (i.e., when energy is used), rather than only on its magnitude, highlighting temporal shifting as a distinct dimension of flexibility [37].

For instance, in Ref. [38], flexibility is intended as “the ability to respond to external signals (sub-hourly and upwards) by technologies capable of electricity demand or generation in the district energy–electricity system interface”.

Cochran et al. [39], focus on balancing the electricity grid, defining this concept as “the ability of a power system to respond to change in demand and supply”.

Also, in Ref. [40], flexibility is only addressed within the limits of the electricity sector and considered as “the capability of a power system to maintain continuous service in the face of rapid and large swings in supply or demand, whatever the cause”.

According to Heggarty et al. [41], flexibility can be defined as “the ability to cope with variability and uncertainty in generation and demand”.

Furthermore, as energy systems move toward integration and electrification, the concept has progressively extended beyond the electricity sector, embracing cross-sectoral interactions and sector-coupling pathways (e.g., electricity–heat–gas–mobility), where flexibility can be provided by multi-energy assets and coordinated operation across vectors [42].

A general definition should take into account the whole energy system, beyond the limits of the electricity sector, precisely to emphasise sector-coupling strategies as a measure of flexibility. Therefore, the general concept of flexibility can be expressed as the ability of the entire energy system to match energy supply and demand at any given time, across all energy sectors.

This definition emphasises that flexibility is not a property of a single technology, but an emergent attribute of the whole system, depending on how conversion, storage and control strategies are coordinated.

Time is an important dimension of flexibility. Very short-term flexibility (milliseconds to seconds) underpins frequency control; short-term flexibility (minutes to hours) supports ramping and intra-day balancing; medium-term flexibility (days to weeks) is needed to manage weather-driven variability; long-term flexibility (months to seasons) compensates for seasonal mismatches [43].

The need for flexibility becomes critical as the share of variable RES (VRES) increases. VRESs introduce additional variability and forecast uncertainty, leading to more frequent overgeneration.

A possible distinction between types of flexibility can be made between generation-side, grid-side and demand-side flexibility.

For the purposes of this review, a classification in three main categories has been adopted:

• Demand-side flexibility: captures the ability of end-use loads to adjust their consumption patterns when exposed to economic or control signals, while maintaining acceptable levels of service and comfort for final users.
• Storage- and sector-coupling-based flexibility: denotes the capability of storage devices and cross-vector conversion technologies to decouple in time and carrier the balance between supply and demand by temporarily storing energy (in electrical, thermal or chemical form) or shifting it between vectors (e.g., Power-to-Heat, Power-to-X).
• Supply-side flexibility: denotes the capability of controllable generation and conversion units to vary their net power output over time in response to system needs, within their technical operating limits (ramping, minimum load, start-up time, efficiency).

4. Demand-Side Flexibility

Demand-side flexibility can be broadly defined as the ability of end-users to adjust their electricity consumption in terms of time, location or intensity in response to external signals (prices, control commands, network constraints) or internal objectives (self-consumption, comfort, cost minimisation), while preserving the underlying energy services [44]. In DES, demand-side flexibility is valuable because it allows prosumers to better align their demand with local renewable generation, reduce congested feeders and reduce the need for costly flexibility on the supply side [45].

From a system perspective, different types of demand flexibility can be distinguished according to the shape changes they induce in the load profile. According to Ref. [45], different categories of demand-side flexibility can be defined as follows:

• Load shedding: the temporary curtailment of non-critical demand during scarcity events.
• Load shifting: involves moving flexible uses from high-price or congested periods to off-peak hours.
• Load modulation: a short-term up- and down-regulation around a reference profile for balancing and ancillary services.
• On-site generation-driven flexibility: the adaptation of the demand to coincide with local RES generation.

Classical demand-side management literature further refines these concepts into strategies such as peak clipping, valley filling, load shifting, strategic conservation, strategic load growth and flexible load shaping, which all correspond to specific ways of redistributing demand in time and amplitude [37].

Demand-side flexibility in DESs is mainly realised through demand-side management (DSM) actions and demand response (DR) schemes. A common and widely adopted distinction in the DR literature is between implicit and explicit DR. Implicit DR is associated with price-based instruments, such as Time-of-Use, Critical Peak Pricing, Real-Time Pricing and Variable Peak Pricing, that modify the retail tariff structure in order to encourage prosumers to reschedule their flexible loads in response to dynamic prices [46]. Explicit DR, by contrast, is implemented through incentive-based schemes, including direct load control programmes, interruptible or curtailable contracts, and capacity or ancillary-service products, in which aggregators or distribution system operators enter into contracts for predefined volumes of flexibility from portfolios of small consumers and activate them through direct control signals or standardised dispatch requests [47].

For each typology of energy end-use, a further distinction can be drawn between programmable and non-programmable demand. The first group comprises loads whose operating time and, in some cases, power level can be planned in advance or managed automatically, such as electric vehicle charging and domestic hot-water production [48]. The second group includes non-programmable uses, for example lighting or cooking. Even in this case, however, gradual changes in user behaviour and improvements in appliance efficiency can lower overall consumption and peak power requirements.

The effective demand-side flexibility implementation depends mainly on a set of digital and organisational infrastructures. For instance, smart metering and advanced metering infrastructures, home and building energy management systems, IoT-based monitoring and control of appliances and aggregators allow prosumers to participate in flexibility markets and provide system services [49]. However, many works highlight that flexibility potential is often constrained by social and behavioural factors, making citizen involvement and acceptance of such schemes a key factor in their implementation [46,47].

5. Sector-Coupling Strategies and Energy Storage Systems

In this section, different sector-coupling and storage options are examined in detail, highlighting their technical characteristics, costs and integration constraints in DESs. The following subsections focus on electrochemical batteries, Power-to-Vehicle strategies, Power-to-Heat solutions and Power-to-X technologies. Costs reported in this section are referred to the European market context, harmonised to real EUR 2024, and are intended as installed CAPEX, i.e., inclusive of installation and associated balance-of-system items when applicable.

5.1. Electric Batteries

Rechargeable Electrochemical Batteries (EBs), or secondary batteries, are the most-studied technology for storing renewable generation in distributed energy systems. Therefore, the scientific production, both as case studies and reviews, is very extensive [50].

The reasons for their role as a reference technology are manifold. Most of the available options have been commercial technologies for several decades, have very high round-trip efficiency values, enable bi-directional grid balancing and are modular and available in small sizes [51]. Furthermore, the issue of renewable intermittency has often been addressed in the past decades within the limits of the electricity sector [52]. A wide variety of technologies have been developed over the years.

Lead–acid batteries are a mature technology currently used mainly in automotive applications and characterised by low cost, low specific energy and relatively short lifetime [53].

Nickel–cadmium (Ni-Cd) batteries are also a mature technology, with high energy density; however, they are subject to “memory effect”. Furthermore, cadmium is highly toxic.

Nickel–metal hydride (Ni-MH) batteries are a variant of the previous ones, with higher energy density and no presence of toxic materials [54].

Sodium-based batteries operate at high temperature (300–350 °C) and are characterised by high energy density and long-life cycle [55].

Lithium-ion (Li-ion) batteries are the most used among electrochemical energy storage technologies, especially for small-scales stationary applications. They are characterised by high energy density, high round-trip efficiency and long lifetime [56]. Despite Li-ion batteries referring to a wide group of lithium-based technologies, different material combinations affect technical and economic parameters [57]. The main barrier for their deployment is the still too high costs. Notwithstanding, a significant cost reduction of Li-ion batteries is expected in the next decade [58].

Flow batteries replace electrodes with tanks of liquid electrolytes containing the ions that react in a cell stack. Thus, the power rating and storage capacity of these batteries are independent. Furthermore, there is no depth of discharge, and self-discharge issues are minimal. Nevertheless, the starting and stopping phases are very slow (about 7 min), because it is necessary to drain off the two electrolyte solutions from the cell stack when the flow is to be interrupted [59]. Different flow battery technologies, such as the Vanadium redox battery (VRB) and Zinc–bromine battery (Zn-Br), have been developed in the last years. Among these, VRB is the most advanced technology and in the commercial stage, but much research is still needed to improve some key parameters such as energy density and cost.

A synoptic comparison of the main parameters of EBs is presented in

Table 1. Comparison of the main parameters of electrochemical batteries for stationary applications [56,60].

	Lead–Acid	Ni-Cd	NaS	Li-Ion	VRB	Zn-Br
Maturity	Mature	Commercial	Commercial	Commercial	Early Commercial	Demonstration
Power Capacity (kW)	1–40,000	10–40,000	50–50,00	10–100,000	30–5000	50–2000
Storage Capacity (kWh)	100–100,000	0.01–1500	1000–100,000	0.01–100,000	10–10,000	50–4000
Efficiency (%)	70–90%	60–75%	70–90%	85–95%	60–85%	60–75%
Response time (ms)	5–10 ms	1–10 ms	1 ms	>20 ms	<1 ms	<1 ms
Self-discharge rate (%/day)	0.033–0.3	0.067–0.6	0.05–20	0.1–0.3	0.2	0.24
Suitable storage duration	min − days	min − days	s − h	min − days	h − months	h − months
Lifetime (years)	5–15	10–20	10–15	5–15	5–10	5–10
Lifetime (cycles at 80% DOD)	400–2000	2000–3500	2500–4500	2000–10,000	10,000–13,000	2000–10,000

.

For most of the electrochemical batteries, self-discharge losses are consistent; therefore, their application is viable for short-term storage [61]. Stationary applications can provide several services for balancing and managing distribution power grids, such as ancillary services, voltage support, frequency regulation and, more generally, power quality, in both on-grid and off-grid systems [62].

The costs of EBs will decrease in the near future due to increasing technological maturity and economies of scale. In Figure 2, current and future cost ranges for the main EB technologies have been represented on the basis of Refs. [58,60,63].

The costs of EBs for small-scale applications are dependent on the considered power capacity. A specific cost curve for lithium-ion batteries has been developed in Ref. [68] and is depicted in Figure 3.

Furthermore, the environmental impact of stationary EBs cannot be neglected. Rahman et al. [69] reviewed several life cycle assessment studies on energy storage systems for stationary applications. According to them, the main environmental issues of electrochemical batteries are correlated with their replacement due to their short lifetime and with the disposal of chemicals [69].

5.2. Power-to-Vehicle

Electrification of end uses is seen as the priority to speed up the decarbonisation process of energy systems, and transport will have to quickly convert vehicles and infrastructure in the coming decades. Advances in batteries and economies of scale are rapidly reducing the prices of electric vehicles (EVs) on the market. Moreover, several countries are strengthening policies to support their deployment.

In the development of 100% renewable energy systems, light transport will have to be mostly electrified, while decarbonising a share of heavy vehicles will require alternative fuel deployment [70]. According to Ref. [71], in order to reach the decarbonisation targets by 2050, EVs will have to account for more than 80 per cent of all road transport.

The EV share increase in future energy systems involves operational challenges representing both opportunities and issues for energy systems management.

Large-scale integration of EVs involves additional electric loads, implying the risk of critical variation in power grid parameters (such as grid congestion, power losses and voltage drop) [72]. Furthermore, EV charging load is very complex to predict, adding further variability on the demand side [73,74].

Nevertheless, EVs can represent a means for balancing intermittent renewable generation [75]. Therefore, EV charging management can both solve some demand-side uncertainty problems and foster the RES penetration in future energy systems [76]. This topic has been extensively analysed in the literature in recent years [77]. Most of those studies agree on the need for introducing an EV aggregator, a new entity with the responsibility of coordinating and managing the charging time scheduling and the eventual discharging one.

According to Ref. [78], three EV management strategies in distribution networks can be identified:

• Uncoordinated charging;
• Smart charging;
• Smart charging and discharging, i.e., bidirectional vehicle-to-grid (V2G).

Uncoordinated charging represents the lack of a common management strategy. EVs, once they are connected to the grid, start to charge until the batteries are completely full.

Smart charging, also called Unidirectional V2G, Vehicle-1-Grid or more generally Power-to-Vehicle, consists in the management of a single power flow direction between the grid and EVs. An EV aggregator controls the charging scheduling varying time and power. In such a way, EV fleets can provide ancillary services to the local electricity network, balancing the VRES generation and enhancing the distributed energy system flexibility [79].

The technical infrastructure is not expensive, since only simple controllers are required in addition to the classical charging devices [80]. A potential barrier to the implementation of that strategy is represented by the participation of EV owners [81]. Indeed, energy trading policies and incentive schemes are needed to encourage the citizens to be involved.

The last step is represented by the V2G system, i.e., bidirectional flow between EVs and the power grid. In that strategy, EVs can also operate as temporary electric storage systems and, when it is required, they can inject electricity into the power grid and solve some intermittency issues of VRES [82]. Thereby, EVs can be used for peak load shaving and load levelling services, improving the flexibility in the local power grid.

One important barrier for V2G implementation is represented by the rapid degradation of electric batteries due to the numerous charging and discharging cycles [83]. That issue can also discourage the EVs owners from participating in V2G schemes. Furthermore, the IT system for managing V2G strategy is more complex and expensive than the V1G one, requiring additional hardware [84]. Nevertheless, V2G applications turn out to be more interesting in off-grid energy systems, where the need for bidirectional electricity storage systems is higher [85].

Lanz et al. [86] analysed the “levelized cost of charging” for different technological options in 30 European countries. Furthermore, their work provides an overview of potential solutions for residential and commercial users and an extensive cost database. According to them, the cost of charging is strictly dependent on the power level and the utilization rate. In Figure 4, the cost ranges of single charging stations by power level for typical residential district options have been summarised [87]. Moreover, for those charging infrastructure options, the equipment needed per charging plug for smart charging and V2G solutions are in the range of EUR 500–1000 and EUR 4000–6000, respectively.

5.3. Power-to-Heat

Power-to-Heat represents the conversion of renewable electricity into thermal energy. Heat is generated efficiently by means of Heat Pumps (HPs). Their coefficient of performance (COP) ranges between 3 and 5 [67], depending on both hot and cold heat sink temperatures. By using a variable speed drive to the compressor side, COP is not substantially affected by partial loads up to 25% of the rated ones.

The advantage of that strategy is to exploit a commercially established and low-priced technology. Furthermore, the management of the thermal vector is much easier than the other energy carriers. For those reasons, the Power-to-Heat strategy can be considered as one of the best solutions for integrating the RES excess into distributed energy systems.

Power-to-Heat systems can provide system flexibility both through demand response strategies and thermal energy storage (TES) systems [88]. Heat can be easily stored in water tanks for thermal load shifting without violating indoor comfort [89]. Thereby, the RES excess can be converted in heat and stored in TES systems in an efficient way.

In Ref. [6], different energy storage solutions have been compared. Their findings show how, under certain conditions, TES is approximately one hundred times more cost-effective than traditional EBs.

Innovative TES systems have been studied in the recent years. Phase Change Materials (PCMs) can exploit the latent heat as a storage means in residential districts [90]. Such systems are characterised by more compact storage design; however, their costs are still higher than traditional solutions [61]. Seasonal TES systems are usually based on underground solutions, but these are large systems, not always suitable for residential communities [91].

In Ref. [92], the factors affecting flexibility potential of Power-to-Heat systems have been analysed. The flexible operation is mainly correlated to the HP size, the TES size and typology, the thermal demand and its profile and the dynamic system properties. Due to those endogenous factors, Power-to-Heat applications often have limitations that do not allow for the full integration of local RES excess.

The performance of HPs is affected by the different heat source and hot sink options. Due to thermodynamic efficiency, COP decreases as the temperature difference between the source and the sink increases.

The most commonly used heat source is air, due to the simplicity of the installation scheme and the source availability. Nevertheless, air leads to strong seasonal variations in COP. To overcome such issue and increase overall efficiency, it is sometimes possible to exploit soil or ground water as heat source [93].

In residential buildings, water is normally used as a heat sink. Yet, the required temperature depends on the heating distribution and storage system. Traditional air-to-water HPs have high performance when supplying heat at low temperatures. However, sometimes, as in the case of historic buildings, it is not possible to make changes to the distribution system, and it is necessary to supply heat at high temperatures.

Finally, the actual flexibility that such systems can provide is highly dependent on external characteristics, building physics and building-plant interactions [89].

An important distinction can be made between centralised and decentralised Power-to-Heat options [94]. The decentralised approach envisages individual HPs for each building and, conversely, the centralised approach includes electricity conversion by using centralised HPs and then thermal energy is distributed by means of district heating (DH) networks.

As demonstrated in several works [95], DH turns out to be the best solution in urban areas with high heat density, and individual HPs have to be limited to rural areas where the heat density is low. The frontier of research for DH is represented by the fourth generation of DH (4GDH), as defined by Lund et al. [96].

Among them, 4GDH is a smart thermal grid supplying low-temperature (50°–70°) heat for space heating and domestic hot water (DHW) in order to reduce thermal losses and link heat and electricity sectors [97]. Indeed, DHW can be supplied by low temperature DH respecting water comfort temperature as well as legionella-safe temperature in circulation lines [98]. As a system, 4GDH represents a viable and cost-effective solution for integrating renewable generation in local energy systems [99]. Furthermore, it is a crucial means for decarbonising the building sector and an essential element of smart energy systems planning [100].

Recently, the concept of fifth generation DH has been introduced in the literature [101]. Nonetheless, no unique definition has been provided for that concept, and most studies using it deal with combined heating and cooling networks. Furthermore, the use of the term “generation” turns out to be inappropriate, since that concept does not go beyond the 4GDH, and the two approaches present several similarities [102].

The costs of Power-to-Heat technologies for small-scale applications are dependent on the installed size. A specific cost curve for air-to-water HPs has been developed in Ref. [68], according to data made available by the Danish Energy Agency catalogue [56], and depicted in Figure 5.

The CAPEX of TES systems is also dependent on their size. For traditional hot water tanks, Martínez-Lera et al. [104] proposed a specific cost function depending on the storage volume (V):

$${CAPEX}_{TES}=4042·V^{0.506}$$

(5)

5.4. Power-to-X

Power-to-Hydrogen (PtH) systems convert renewable electricity into hydrogen by means of water electrolysis. Such a solution is an interesting option for balancing local grids and integrating RES excess. The role of hydrogen becomes even more important in energy systems with a high-RES share. Indeed, those conditions require long-term energy storage systems to cope with the seasonality of generation and high-RES excess.

Several works have analysed the role of hydrogen in distributed energy systems. Fonseca et al. [105] provided a systematic review in order to highlight the main trends in the application of PtH technologies in distributed energy systems.

Most studies in the literature concern the joint application of electrolysers and fuel cells in smart grids. Nevertheless, the re-conversion of hydrogen into electricity for grid balancing purposes is characterised by very low round-trip efficiency [106]. More interesting is the application of hydrogen to decarbonise other energy carriers’ demand and to couple sectors.

Hydrogen can be converted in other alternative fuels that are easier to manage and integrate with the existing energy infrastructure [107]. The general conversion of electricity in hydrogen and its possible further synthesis in other electrofuels is often referred to in the literature as Power-to-X (PtX). Several options have been taken in consideration, such as synthetic natural gas, dimethyl ether (DME), methanol and ammonia [107]. Nevertheless, the reactor’s size for the electro-fuel synthesis is very large, so that solution is not cost-effective for small-scale local applications [108]. To exploit it, hydrogen should be stored and transported from several centres in order to optimise the electro-fuels’ production process.

Hydrogen can be a key vector for decarbonising the local transport sector [109]. In Ref. [110], the potential of small-scale hydrogen refuelling stations (HRSs) with onsite production is investigated, demonstrating the techno-economic feasibility of integrating autonomous HRSs in urban environments. Furthermore, Alavi et al. [111] analysed the possibility of integrating fuel cell electric vehicles in a microgrid, providing both flexibility and hydrogen end-use at local-scale. However, the lack of dedicated infrastructure and specifically widespread HRSs can represent a barrier for its application in the urban mobility field [112].

A viable solution for small-scale application is hydrogen blending with natural gas (NG). Hydrogen can be blended into local NG distribution networks at low volumetric fractions with no significant changes in gas grid parameters [113]. In such a way, gas grids can be exploited as a medium for storing and transporting hydrogen, avoiding the need for dedicated infrastructure [114]. Furthermore, that option is suitable for the residential sector due to the reduction in explosion risks and other safety issues related to the hydrogen storage.

The combined implementation of Power-to-Gas and Power-to-Heat systems in residential neighbourhoods has also been analysed, showing how synergies between the two strategies can be exploited [68]. Indeed, the Power-to-Heat strategy reduces system costs, and all non-integrated excess can be converted into hydrogen to decarbonise the gas grid. In this way, it is possible to exploit the individual advantages of the strategies and overcome the limitations of both.

There are different electrolyser technologies, which can be classified according to the electrolyte and the operating temperature [115]. Most electrolysers operate at low temperatures in the range of 50 °C to 90 °C.

The most mature technology on the market is Alkaline electrolysis. The technology has been developed for different scales, from a few kW up to several MW. Such electrolysers can operate at partial load between 20% and 100% of the design capacity [116]. Nevertheless, alkaline electrolysers must be operated continuously, as this technology requires several minutes to restart after a shutdown [117]. Hydrogen can be produced with a gas quality of about 99.5–99.9% [118]. The running and maintenance costs are high, yet the unit investment costs are the lowest among alkaline technologies. Moreover, the lifetime is about 90,000 operating hours, the highest among the different options [119].

A more recent technology is Polymer Electrolyte Membrane (PEM) electrolysis. Those electrolysers base their operation on two half-cells separated by a proton exchange membrane, which allows high proton conductivity and low gas crossover [120]. The low membrane thickness results in a compact system design, high power density and high-pressure operation [121]. This technology seems to be the most suitable for integration into distributed energy systems due to a faster cold start, more flexible operations and better coupling with intermittent renewables [122]. Moreover, the hydrogen purity is higher than alkaline electrolysers, reaching values equal to 99.999% [123]. The main negative aspects are the high cost of components, which results in high CAPEX values, and the low lifetime [120].

A still non-commercial and developing technology is Anion-Exchange Membrane (AEM) electrolysis. Such electrolysers have the advantage of replacing traditional noble metal electrocatalysts with low-cost transition metal catalysts [124]. Nevertheless, considerable efforts still need to be made to improve power efficiency, membrane stability and the overall cost of the technology. They operate at low temperature (50–70 °C) and medium-high pressure (around 30° bar) [125].

Solid Oxide Electrolysis Cells (SOECs) are a technology based on water splitting reaction. Operating at high temperatures (600–900 °C), the water–steam phase change reduces electricity consumption by saving the energy required to provide the latent heat for vaporisation [126]. Therefore, SOECs have the highest conversion efficiency among electrolytic technologies [127]. Nevertheless, the high temperatures result in fast material degradation and limited long-term stability [116]. Furthermore, high temperature and steam handling may pose safety issues for SOEC application in residential districts. Different materials can be applied in SOEC structures. The most common electrolyte is composed by a dense yttria-stabilized zirconia (YSZ) [128]. The solid electrolyte could be an important cost-reducing factor in the future. However, nowadays, this technology is in a pre-commercial version, and the unit costs are much higher than other electrolysers [129]. A major advantage of SOECs is the possibility to operate in reverse mode as a fuel cell. Therefore, in the recent past, a few studies have analysed the potential role of SOECs for grid balancing in distributed energy systems [130].

In general, all technologies are available in small sizes, and their modular nature lends itself to small-scale residential applications. A synoptic comparison of the main parameters of electrolysis technologies is presented in

Table 2. Comparison of the main parameters of electrolysis technologies [116,119,124,131,132].

	Alkaline	PEM	AEM	SOEC
Electrolyte	Alkaline solution	Solid polymer membrane	Polymeric anion exchange membranes	ZrO2 ceramic doped with Y2O3
Electrical Efficiency (% LHV)	63–70%	56–60%	58–62%	74–81%
Temperature range (°C)	60–80	70–90	50–70	600–900
Pressure range (bar)	1–30	30–80	1–30	1–15
H2 purity (%)	99.9%	99.999%	99.99%	99.99%
Lifetime (h)	100,000	30,000–90,000	NA	10,000–30,000
Power Capacity	Up to several MW	Up to hundreds of kW	Up to tens of kW	Up to hundreds of kW
Load Range (%)	10–110%	0–160%	5–100%	20–100%
Technology maturity	Commercial	Commercial for small scale	R&D	Pre-commercial

.

Power-to-Hydrogen systems are still not commercially established, and their current cost is very high. Nevertheless, the ambitious international targets for electrolyser capacity and hydrogen deployment will speed up the learning process and reduce the costs of the technology. In Figure 6, the cost range for the main electrolyser technologies with forecasts to 2030 and 2050 has been represented on the basis of Refs. [119,129,133,134,135].

The size of electrolysers has a significant impact on their specific cost. In Ref. [68], a curve describing the electrolyser CAPEX as a function of the rated power has been proposed. Such a curve is depicted in Figure 7.

6. Supply-Side Flexibility

Conventional energy systems have always based their flexibility on the supply-side. In renewable systems, the role of dispatchable generation should be to balance intermittent generation and ensure grid reliability [136]. In detail, such systems play a key role in managing the rapid ramping due to RES generation.

Small-scale applications of combined heat and power (CHP) plants in distributed energy systems have been much studied in recent years.

Cogeneration systems already offer a set of established advantages in distributed energy systems, combining high overall efficiencies with fast dynamic response and local controllability. These characteristics make CHP particularly suited to provide supply-side flexibility, as electricity and heat can be produced close to demand while supporting renewable integration and alleviating stress on distribution networks [137].

The role of cogeneration is expected to evolve towards more flexible and intermittent operation, with reduced full-load hours and stronger interaction with storage and demand-side flexibility. At the same time, a progressive shift towards renewable and low-carbon fuels, such as sustainable biomass, hydrogen and synthetic fuels, is seen as essential to preserve CHP as a dispatchable option in highly renewable energy systems [138].

Such systems also allow further coupling of the electricity and gas grid [139]. Therefore, when exploited with TES systems, they provide an additional source of flexibility and improve the RES integration in the energy system. Numerous studies show that integrating CHP with thermal energy storage and advanced control can significantly expand the operating range of these units and increase their ability to deliver upward and downward regulation in DESs [140].

The state of the art of CHP systems is extensively addressed by different publications [141,142,143]. These studies underline that future CHP systems will operate with reduced full-load hours, more frequent cycling and a progressive shift towards renewable or low-carbon fuels, while maintaining an important role in providing local flexibility and capacity adequacy [144].

The fuel used to feed these plants must be taken into consideration. Indeed, the RES share increase on the electricity grid may eliminate the energy and emission savings provided by gas-driven CHP plants [145]. Therefore, the choice, even for balancing systems, should be directed towards technologies powered by renewable fuels [144,146].

Bio-fuels can be used in several small-scale CHP technologies, such as internal and external combustion engines and micro-gas turbines. However, it is necessary to take into account the availability of this resource on the territory and the sustainability of the biomass production chain. Therefore, it is not always possible to rely on such a solution, especially in an urban environment.

Hydrogen or other electro-fuels can be applied for the reconversion to electricity by means of fuel cells [147]. Such systems are characterised by higher electrical efficiency and power-to-heat ratios than other CHP technologies. This aspect can be useful in the context of a rapid electrification of end-uses in urban districts. Therefore, fuel-cell-based micro-CHP thus represents a dispatchable, low-carbon supply-side flexibility option that can be integrated with local hydrogen production and storage [148].

Recent studies have highlighted that the long-term role of fuel-cell-based micro-CHP in distributed energy systems is strongly influenced by durability and degradation phenomena. In particular, advanced approaches for state-of-health estimation and degradation tracking have been proposed to improve lifetime prediction and operational reliability of hydrogen and fuel cell systems under realistic operating conditions [149]. These aspects are crucial for assessing the actual flexibility contribution of fuel cells over their lifetime.

In parallel, accurate modelling of fuel cell behaviour requires robust parameter estimation techniques. Recent work has addressed this issue by developing advanced parameter identification methods for PEM fuel cell voltage models, enabling improved representation of performance and dynamic behaviour under varying operating conditions [150]. Such approaches contribute to reducing modelling uncertainty and enhancing the reliability of system-level simulations of fuel-cell-based flexibility options.

Despite the cogeneration set-up, the low round-trip efficiency of a Power-to-Gas-to-Power system should limit the use of fuel cells for grid balancing. Notwithstanding, especially in off-grid systems, the seasonality of RES cannot be managed exclusively with electric batteries, and fuel cells can therefore exploit long-term storage. Recent techno-economic studies of islanded and high-RES systems show that hybrid configurations, in which short-term balancing is provided by batteries while hydrogen or other electro-fuels provide long-term storage, minimise overall system costs and curtailment [151]. According to their results, the best configuration is to combine the two storage systems, using electric batteries for daily balancing and hydrogen for seasonal storage.

7. New Energy Markets, Regulation and Blockchain Technology

Beyond technological solutions, the effective deployment of flexibility is strongly shaped by market design and regulation, which determine who is allowed to provide flexibility, which products are traded, and how flexibility providers are remunerated. In the EU context, recent regulatory frameworks explicitly aim to enable demand response and flexibility participation, stressing the role of cost-reflective network tariffs, consumer empowerment and the evolving responsibilities of DSOs/TSOs in local procurement of services [152].

Furthermore, a growing body of work highlights that unlocking small-scale flexibility requires dedicated rules for new market actors such as independent aggregators (including the definition of contractual relationships and compensation mechanisms), since the absence of clear frameworks may become a practical barrier to participation [153].

Economic instruments also act as key drivers. Carbon pricing and emissions trading can strengthen the value of low-carbon flexibility, while the interaction between renewable subsidies and short-term price spreads may affect the business case of storage and flexibility assets [154]. Price-based and incentive-based schemes (e.g., dynamic tariffs, time-varying pricing, incentive DR) are frequently discussed as enabling tools for demand-side flexibility, but their effectiveness depends on metering infrastructure and the regulatory treatment of tariffs and data [155].

In this evolving context, the decentralisation process involves several actors with an active role, increasing the number of possible actions and the complexity of system management. In order to allow all stakeholders to move actively in the network, the secure exchange of data, to which both energy flows and economic transactions are linked, is necessary. Blockchains or distributed ledger technologies can be a means of managing decentralised transactions without the need for centralised management [156]. These new transactional layers operate on top of the physical DES infrastructure and are required to coordinate large numbers of small prosumers, aggregators and distribution system operators, and to manage the flexibility technologies discussed in previous sections [157].

Blockchain technologies are systems based on shared and distributed data structures or ledgers, which can be read and modified by multiple nodes in a network [158]. The advantage of such technology is that digital transactions can be stored securely without using a central point of authority [159]. Indeed, to make a change in the ledger in the absence of a central authority, the nodes (i.e., the parties involved) must reach a consensus. All transactions are recorded and linked to previous ones by cryptography. In the context of energy markets, permissioned blockchains and smart-contract platforms are increasingly investigated as a way to automate verification and settlement of local energy and flexibility trades, while ensuring traceability and reducing transaction costs [160].

As the number of participants grows, the possible interactions increase, and management becomes extremely complex. The so-called peer-to-peer (P2P) trade, based on blockchain technologies, allows transactions between market participants without the involvement of a third party [161]. Recent reviews distinguish between pure P2P, community-based and hybrid local market designs, which differ in how trades are organised, how network constraints are represented and how responsibilities are shared between prosumers, aggregators and DSOs [162].

Blockchain is a relatively new, trans-disciplinary and fast-developing area of research in recent years [163]. Several algorithms and structures are in development, and some real projects have already started. According to Ref. [157], the potential of this technology is very high, and the impact on the energy sector can be crucial to realise the decentralisation of systems. However, several barriers still have to be overcome to improve scalability and security levels. Furthermore, the legal and regulatory framework will have to be adapted in a consistent manner [164]. In addition, concerns about energy consumption of some blockchain implementations, data privacy and the allocation of responsibilities among market actors point to the need for careful techno-economic and institutional design [157].

It should also be taken into account that the different energy markets today are mostly separate. In smart energy systems, where the interconnections between energy sectors will be numerous due to the spread of sector-coupling technologies, markets will also have to be more interconnected and interrelated [165]. For sector-coupled DESs, a key research question is how to coordinate these local electricity and flexibility markets with emerging markets for heat, gas and hydrogen so that multi-vector flexibility can be efficiently valued across interlinked infrastructures [165].

8. Concluding Remarks and Research Gaps

This review has examined how sector coupling and flexibility measures can enable the large-scale integration of renewable energy in DESs, with a focus on small-scale applications such as buildings, districts and renewable energy communities. Moving from decentralised generation to truly distributed, multi-vector systems requires changes not only in generation assets, but also in how storage, demand, markets and institutions are organised. In this context, flexibility emerges as a system-level property that must be planned and operated at local scale to maximise renewable self-consumption and to avoid transferring integration problems to higher-voltage networks.

The contribution of this review is to frame system flexibility in a multi-energy perspective and to assess techno-economic parameters of main technologies and strategies. Flexibility is defined as the ability of the entire energy system to match energy supply and demand at any given time, across all energy sectors, and not simply as a property of a single technology. Within this framework, demand-side flexibility, storage- and sector-coupling-based flexibility, and supply-side flexibility have been analysed. The literature surveyed clearly shows that no single option can provide all the flexibility required in high-renewable DESs in a cost-effective way. Robust solutions rely on a complex interaction between different technologies and strategies in order to provide flexibility from fast balancing to seasonal storage.

Residential and tertiary loads can provide demand-side flexibility through several distinct services when properly characterised and controlled. A compact taxonomy has been adopted in the literature, distinguishing between load shedding, load shifting, load modulation and on-site generation-driven flexibility. These categories can be implemented through a variety price-based and explicit incentive-based demand response schemes. Demand-side measures often represent the lowest-cost flexibility resource, as they mainly rely on digital infrastructure rather than on large physical assets. However, their practical potential is constrained by user comfort, behavioural responses, social acceptance and the design of tariff and market arrangements.

Sector-coupling strategies and energy storage systems enable temporal and cross-sector decoupling between renewable generation and final energy use. Electrochemical batteries, especially lithium-ion systems, are technically mature and well suited for short-term balancing, but their cost, resource constraints and environmental impacts limit their role as the sole storage backbone. Power-to-Heat with heat pumps and thermal energy storage stands out as a particularly cost-effective and efficient option for integrating renewable surpluses at building and district scale, especially when combined with low-temperature district heating. Electric vehicles and controlled charging provide both substantial flexibility potential and an additional pathway for electrification, though issues such as battery degradation, user participation and infrastructure costs remain critical. Power-to-Hydrogen and more general Power-to-X options offer long-term and cross-sector flexibility but are penalised by low round-trip efficiencies and significant capital costs; their most promising role in DESs appears to be the decarbonisation of transport fuels and high-temperature heat, rather than frequent power-to-power cycling.

Supply-side flexibility in high-renewable DESs may still rely on small-scale dispatchable generation, such as micro-CHP and polygeneration, which can provide fast response and electricity–heat coupling. Looking ahead, renewable fuels and hybrid solutions combining batteries for short-term balancing with hydrogen or electro-fuels for seasonal storage appear promising for islanded or weakly connected DESs, despite current cost and efficiency penalties.

A cross-cutting result of the review is that cost and efficiency trade-offs are strongly scale- and context-dependent. Technologies that appear attractive in large, centralised systems may not be optimal in DESs, where modularity, part-load operation, space constraints and local network limits become critical.

Markets, regulation and digital infrastructure are key enablers of flexibility in DESs. Local flexibility markets, peer-to-peer trading schemes and community-based arrangements are being actively explored to value small-scale flexibility and to allocate costs and benefits among prosumers, aggregators and system operators. Blockchain and distributed ledger technologies offer new ways of implementing decentralised trading and settlement, but remain at an early stage with significant open questions regarding scalability, energy use, governance and regulation.

An existing research gap regards integrated multi-energy modelling of DESs. Indeed, too many works still focus solely on the electricity sector, neglecting the interconnections between electricity, heat, gas, hydrogen and e-mobility. Furthermore, in order to model such complex multi-energy and multi-vector systems and optimise their configuration, energy modelling tools are needed, suitable for the district level, which allow the optimisation of the size and operation of numerous technologies, integrating sector-coupling strategies and multiple energy storage systems and analysing the system at least at hourly temporal resolution. Such models should better capture network constraints, operational details of storage and conversion technologies, and the stochastic nature of demand and renewable supply.

Furthermore, despite numerous indicators proposed in the literature, a consistent and widely accepted set of metrics for quantifying flexibility in DESs is still lacking. Developing such metrics would facilitate the comparison of options and the design of flexibility products for local markets.

Many of the technologies analysed in this review have the potential to significantly reduce costs in the coming decades thanks to learning processes and economies of scale. Therefore, research into innovative materials, technologies and strategies is crucial to providing new and more efficient solutions for flexibility. Real-world pilots for the demonstration and monitoring of sector-coupled DESs play a key role in this process. Testing projects are essential to validate modelling assumptions, understand operational challenges and refine business models for combined flexibility options.

Further studies are also necessary to understand the social acceptability of such flexibility schemes and citizen participation. It will therefore be important to develop empirical evidence on actual user responses to dynamic tariffs, automation and aggregation in DESs. Future work should combine engineering models with social science methods to co-design demand response schemes that are both technically effective and socially acceptable.

Furthermore, market and regulatory innovation for multi-vector flexibility should be further analysed in order to develop solutions to coordinate local flexibility markets with wholesale markets, network tariffs and climate policies. Alongside, as DESs become more digital and data-intensive, questions of interoperability, cyber-security, privacy and data ownership will become central. Blockchain-based solutions should be evaluated not only for their technical feasibility but also for their real added value relative to simpler architectures.

In conclusion, this review confirms that sector coupling and flexibility measures are indispensable for enabling high shares of renewable energy in distributed renewable energy systems, but also shows that their deployment is far from a purely technical optimisation problem. Cost-effectiveness, environmental performance and social acceptance all depend on how technologies are combined, how institutions and markets are designed and how citizens are involved. Addressing these complex and interlinked dimensions will be crucial to unlock the full potential of DESs.