An Overview of Power System Flexibility: High Renewable Energy Penetration Scenarios

Abstract

Power system flexibility is becoming increasingly critical in modern power systems due to the quick switch from fossil fuel-based power generation to renewables, old-fashioned infrastructures, and a sharp rise in demand. If a power system complies with financial restrictions and responds quickly to unforeseen shifts in supply and demand, it can be considered flexible. It can ramp up production during periods of high demand or increase it during unanticipated or scheduled events. The broad use of renewable energy in the power grid can provide environmental and economic benefits; nevertheless, renewables are highly stochastic in nature, with variability and uncertainty. New management with adequate planning and operation in the power system is necessary to address the challenges incorporated with the penetration of renewable energy. The primary aim of this review is to provide a comprehensive overview of power system flexibility, including appropriate definitions, parameters, requirements, resources, and future planning, in a compact way. Moreover, this paper potentially addresses the effects of various renewable penetrations on power system flexibility and how to overcome them. It also presents an emerging assessment and planning of influential flexibility solutions in modern power systems. This review’s scientific and engineering insights provide a clear vision of a smart, flexible power system with promised research direction and advancement.

1. Introduction

The future of our energy lies in renewable energy due to the limited opportunity of fossil fuels in future. Renewable energy is only the secure source of meeting the ever-increasing energy demand imposed by growing technologies. In addition to the depletion of fossil fuels, global warming has become a major concern throughout the world. Moreover, with the seventeen sustainable development goals (SDG) shown in Figure 1, the United Nations’ Agenda 2030 envisions a progressing and more livable future for global society [1]. Among them, SDG 7 aims for affordable, reliable, sustainable, and clean energy for everyone with more penetration of renewable energies [2]. SDG 13, on the other hand, urges more action to combat climate change and its consequences on the globe by cutting carbon emissions and expanding the penetration of renewable energy sources [3]. According to the United Nations’ Agenda 2030, SDG 7 (affordable and clean energy) and SDG 13 (climate change mitigation) are closely related and complementary to each other [4]. To achieve these United Nations (UN) goals and ensure sustained modern civilization, there is a global trend to reduce our dependency on fossil fuels, including petroleum, coal, and natural gas, because of their negative effects on the environment and limited availability. Burning fossil fuels harms the environment. For example, fossil fuel-based power plants contribute around 43% of carbon emissions worldwide [5].

For modern societies, climate change is a tremendous existential threat and must be handled enthusiastically and promptly; otherwise, the trace of humans on this planet earth will be jeopardized badly. According to the Paris Agreement, using affordable, sustainable, and clean energy sources is necessary to prevent a rise in the global mean temperature of more than 2 degrees Celsius. The shift from fossil fuel to renewable energy is a smart transition to cope with this agreement [6], and solar and wind are the two major sources of energy being domesticated around the world [7]. Some other renewable energy sources (RES), for example, tidal energy [8] and geothermal energy [9], have the capability to contribute to one hundred percent of demand in some parts of the world. For instance, countries like Denmark and the USA plan to achieve the goal of 80% dependency on RES by 2050 [10].

However, during transmission and distribution, renewable energy presents unique challenges in the demand–supply chain. For instance, due to their reliance on natural circumstances, these resources are extremely unpredictable. They share four characteristics because of their dependency: dispatchability, uncertainty, intermittent, and volatility. Abrupt variations in solar radiation cause power-generating failure in a traditional photovoltaic system. The amount of sunlight available, the intensity of the irradiation, and the length of the day all have a significant impact on solar-based photovoltaic (PV) power output. These are caused by erratic weather patterns and the sun’s yearly location, which makes the power that is harnessed erratic and sporadic [11]. Wind-based power sources experience the same issue [12]. Compared to solar energy, wind power is more erratic and sporadic. The speed and quality of the air have a major impact on wind systems’ power quality. Intermittency and volatility in the power supply are incorporated into any airspeed randomness [13]. The variability of tidal power harvesting is a significant concern during power grid integration, even though it is less erratic and intermittent than solar and wind power [14]. If there is no backup power generation or storage, the abrupt failure of power generation brought on by the erratic and unpredictable nature of solar, wind, and tidal sources will cause the entire power grid to shut down. Conversely, resources such as geothermal, hydropower, and biomass have relatively high dispatchability but moderate volatility, intermittency, and unpredictability [15]. The extensive integration of renewable energy sources such as photovoltaic, wind, and tidal into the power grid results in non-dispatchability due to intermittency and generation unpredictability (refer to

Table 1. Comparative variable characteristics of variable renewable energy resources (VRES). Data are collected from the following references: [16,17,18,19].

Characteristics of VRES	Photovoltaic	Wind	Geothermal	Hydropower	Biomass	Tidal
Volatility 1	3	3	1	1	1	2
Intermittency 1	3	3	1	1	1	2
Uncertainty 1	3	3	1	1	1	2
Dispatchability 1	1	1	3	3	3	2

¹ 3 denotes high intensity; 2 denotes mild intensity; 1 denotes low intensity.

for analysis on the variabilities of renewable power penetration with no storing system) [16].

Consequently, these four characteristics of renewable energy sources jeopardize the dependability of the power system and give birth to a significant issue known as inflexibility in electric power systems. Some standard gauges of power system inflexibility are less supply than demand (especially during peak periods), the sudden curtailment of RES in the power grid, zonal stability violations, fickleness, negative trading fares, and so on [20].

Power system flexibility (PSF) commonly reinforces the whole system to be capable of compensating the real-time supply–demand mismatches [21] and plays a role in evolving affordable, reliable, sustainable, and clean power systems. Therefore, it initiates the resulting need to increase the flexibility of the system to avoid inconveniences like power failure in the power system and to increase the dependability of the power system. As PSF is an important factor in the utility power grid, understanding different aspects of the PSF drivers, resources, and requirements will be useful for researchers, academicians, scientists, and engineers of smart power systems to further penetrate the conventional power grid with VRES.

1.1. Definition of Power System Flexibility

To promote widespread comprehension and the adoption of the flexibility concept in the power system, careful consideration of semantics is necessary. Therefore, it is crucial to consider the fundamental meaning of the words and assess how well they convey the ideas. The semantic definition of “flexibility” is the capacity to adapt promptly to new circumstances or situations. In other words, flexibility suggests ways to manage this uncertainty even while future situations may be unpredictable. A system with great flexibility can react quickly to variations in the demand and supply chain. The PSF is defined as “the extent to which a power system can change the production and consumption of electricity in response to variability and uncertainty, predicted or otherwise” by the International Energy Agency (IEA) and the North American Electric Reliability Corporation (NERC) [22]. In various literature, there are several definitions of power system flexibility, some of which include uncertainty and ambiguity, and some include a clear point of view (please refer to

Table 2. Selected definitions of power system flexibility used by various organizations and research groups.

Source	Definition
A. González-Dumar et al., 2024	“Power system flexibility is the ability to handle differences between supply and load and can be quantified to measure the effects of renewable energies on power systems [23].”
H. Energy, 2024	“The ability of power systems to cope with variability and uncertainty at all times [24].”
International Smart Grid Action Network (ISGAN), 2019	“Power system flexibility relates to the ability of the power system to manage changes [25].”
IEA, 2019	“The ability of a power system to reliably and cost effectively manage the variability and uncertainty of demand and supply across all relevant timescales, from ensuring instantaneous stability of the power system to supporting long-term security of supply [26].”
European Federation of Local Energy Companies (CEDEC), 2018	“Flexibility is defined as the modification of generation injection and/or consumption patterns, on an individual or aggregated level, often in reaction to an external signal, in order to provide a service within the energy system or maintain stable grid operation [27].”
International Renewable Energy Agency (IRENA), 2018	“The capability of a power system to cope with the variability and uncertainty that VRE (variable renewable energy) generation introduces into the system in different time scales, from the very short to the long term, avoiding curtailment of VRE and reliably supplying all the demanded energy to customers [28].”
Hsieh and Anderson, 2017	“Flexibility is the capability of the power system to maintain balance between generation and load under uncertainty [29].”
Zhao et al., 2016	“Ability of a system to respond to a range of uncertain future states by taking an alternative course of action within acceptable cost threshold and time window. Flexibility is an inherent property of a system [30].”
Electric Power Research Institute (EPRI), 2016	“The ability to adapt to dynamic and changing conditions, for example, balancing supply and demand by the hour or minute, or deploying new generation and transmission resources over a period of years [31].”
B. Drysdale et al., 2015	“The degree of flexibility, i.e., the ability of a load to vary in response to an external signal with minimal disruption to consumer utility, varies between load categories [32].”
Eurelectric, 2014	“The modification of generation injection and/or consumption patterns in reaction to an external signal (price signal or activation) in order to provide a service within the energy system [33].”
H. Holttinen et al., 2013	“Ability to accommodate the variability and uncertainty in the load-generation balance while maintaining satisfactory levels of performance for any time scale [34].”
International Council on Large Electric Systems (CIGRE), 1995	“The ability to adapt the planned development of the power system, quickly and at reasonable cost, to any change, foreseen or not, in the conditions which prevailed at the time it was planned [35].”

).

Therefore, most of the definitions reflect the main point: “a flexible power system adheres to economic constraints and reacts rapidly to unexpected changes in demand and supply by ramping down production when demand is low and increasing it when demand increases during peak scheduled or unforeseen events”.

1.2. Mathematical Formulations of PSF

For the mathematical formulation and assessment of PSF, variable supply from generators and demand are two crucial factors that need to be granted. Other essential factors are time duration, ramping rate, reserve minimum and maximum capacities, etc. For long-term planning and measurement, an insufficient ramping resource expectation (${exp}_{irr}$) is an index that gives probabilistic expectations based on traditional generation criteria [36]. For ${exp}_{irr}$, the stepping-down PSF of each unit can be defined as follows:

$${PSF}_{t,i}=R_{up}(1-\left(1-{ϵ}_{t,i}\right).T_i)$$

(1)

where $R_{up}$ is the generator’s ramp-up, ${ϵ}_{t,i}$ is the time series operational stage, and $T_i$ is the start-up time of the generator. Following Equation (1), the time series PSF can be defined as:

$${PSF}_t^{system}={\sum }_i{PSF}_{t,i}$$

(2)

The probability of insufficient ramping resources, $P_{irr}$, can be defined as follows:

$$P_{irr(t,i)}={\beta }_i({LR}_{t,i}-1)$$

(3)

where ${\beta }_i$ is the PSF distribution, and ${LR}_{t,i}$ is the net load ramp. The insufficient ramping resource expectation, ${exp}_{irr}$, is the summation of $P_{irr(t,i)}$ over the entire time (0, T). So,

$${exp}_{irr\left(i\right)}={\sum }_0^T{P}_{irr(t,i)}$$

(4)

Along with ${exp}_{irr}$, another important PSF indicator is the normalized PSF index (NFI). Since the power system’s overall PSF is the weighted sum of the PSF of each generation unit, we must first evaluate each generation unit’s PSF before measuring the NFI [37]. Typically, each generation unit can change the reserve simply by changing the ramp-up rate and each upward capacity and the ramp-down rate and each downward capacity. The NFI for each unit can be defined as follows:

$${NFI}_i={\displaystyle \frac{\frac{1}{2}[P_{max}(i)-P_{min}(i)]+\frac{1}{2}[Ramp(i)]}{P_{max}(i)}}$$

(5)

where Ramp(i) is the average of down and up ramps of unit i. $P_{max}\left(i\right)$ is the maximum capacity of unit i, and $P_{min}\left(i\right)$ is the minimum capacity of unit i. Based on Equation (5), the overall NFI of a system can be calculated as follows [38]:

$${NFI}^{system}={\sum }_i{\displaystyle \frac{P_{max}(i)}{{\sum }_i{P}_{max}\left(i\right)}}({NFI}_i)$$

(6)

The flexibility assessment tool (FAST) can measure the PSF and identify the available PSF resources. After measurement, this tool also evaluates the power system’s PSF needs and sources. On the other hand, as they are more obvious criteria in the power system than flexibility, inflexibility indicators are frequently utilized. Measuring inflexibility can determine how much flexibility we need to insert into the power system. Some common inflexibility measurements are supply–demand imbalance, supply curtailment, load changes, and VRES generation imbalances. Another empirical way to measure PSF is flexibility charts. Flexibility charts present generation-based PSF more clearly.

2. Review Methodology

A systematic review approach has been taken to answer how power systems can adapt to high levels of renewable energy integration. In this study, to deepen our understanding and to align renewable energy with SDG, all the related documents of PSF found on the Scopus and Web of Science database based on the following categories are selected:

• Definition and Concept of PSF;
• Large-Scale Penetration Effects of VRES;
• Chronological Requirements of PSF;
• PSF Resources;
• PSF Planning.

All the papers appear with the term power system flexibility considered in the first screening. Among the 4000 papers, 3500 were removed due to duplication, ineligibility, and inappropriateness by the automation web tool Rayyan, which is becoming popular with the science community. Figure 2 is a standard flow diagram created by PRISMA to strictly stick to a systematic review that has been followed in this study [39]. Among 500 papers, 300 papers were removed from the list by reading titles, abstracts, introductions, and conclusions. A total of 50 documents were not retrievable and excluded from the list. Finally, 130 papers were considered for writing the review article by extracting information and have been included in this article.

The sustainable transition of power systems with maximized flexibility is envisioned as a revolutionary process with three important phases, and this paper presents an outline of the policies, technical advancements, and institutional frameworks required to make it possible. To support and circulate this transition, herein, a systemic review is presented. First, we present the various effects on PSF of large-scale penetration of variable renewable energy resources. In the following section, a reader will find a thorough overview of PSF with suitable definitions, parameters, requirements, resources, and future planning in a concise manner. Section 5 presents an hawk-eye review of various flexibility resources, such as conventional units, demand, power grid side resources, market design, energy storage, smart and microgrids, etc. Section 6 explores the emerging assessment and planning of influential flexibility solutions in modern power systems. The unique addition of this work is that it offers a comprehensive view of the power system flexibility as well as a more thorough analysis of the various issues and how they interact in the power system to cope with renewable and sustainable transitions.

3. Large Scale Penetration Effects of VRES

Due to the intermittent and unstable nature of VRES energy generation, successful production requires a sharp, steep slope and numerous starts. Coupled with variable loads, this penetration of VRES introduces operational and planning difficulties to the conventional units of the power system [37]. The power system variability plays its role stage by stage. In different stages, the variability requirement is different. For example, the net load variability does not significantly affect long-term resource planning and operation but plays a vital role in daily operation [40]. Some control mechanisms need to be employed for very short-term response intervals due to the instantaneous changes in the RES penetration. Low-voltage-ride-through (LVRT) [41], active and reactive power, operational voltage, and ramping slope [42] are some typical control strategies for very short-term response intervals. The response time scale has a significant impact on the operational PSF. For short-term time intervals (seconds to minutes), the frequency control and storage reservoir are enough to ensure PSF in the power system. In contrast, with a rapid ramping rate for a medium-term period (minutes to hours) and for a long-term time interval (hours to days), future planning flexibility is crucial (Figure 3) [34].

During the substantial scale penetration (almost 100%) of RES generation, the generation from the baseload plants must stop or reduce due to the sudden surges. However, if the RES generation is suddenly reduced, the baseload plants must be re-dispatched to meet the flexibility conditions. This dispatching and re-dispatching cycles constitute a concerning problem since most of the baseload plants are coal-, oil-, gas-, or nuclear fuel-based, and the start-up time is considerably very high [42]. Moreover, the excessive re-dispatching cycles introduce mechanical problems to the components of generators, such as corrosion, erosion, wear and tear, and metallic fatigue due to pressure and friction [43]. As a result, the estimated lifespan of the plants is shortened, and maintenance and fuel costs rise.

To cope with the UN’s agenda for 2030 and fulfill SDG goals requirements, a rapid transition from carbon fuel to RES is expected in the baseload technologies with further flexibility requirements than current conditions [44]. In addition, smart and innovative transmission lines are also needed during planning and operation with more flexibility. In the present power transmission scenario, electricity from the generation site to the load site is transmitted via transmission lines. However, the RES generation site could be anywhere. Unlike conventional power plants, RES’s power plants need to be installed at a specified site based on weather, environment, and energy availability locations. This installation location of RES plants introduces additional transmission distances that require an increase in the voltage of the generation end to compromise the transmission loss and ensure proper voltage at the consumer end. On the contrary, if the VRES power is dispersed over a large area with heavy loads, it will reduce the overall variabilities of the generation, which is somewhat helpful for operational planning [45]. Large-scale VRES penetration requires an optimal transmission line topology to minimize the losses. However, this will cause a serious disturbance problem to the overall power system and introduce reduced performance (Figure 4) [46].

Power system stability is the most important aspect for ensuring secure and uninterrupted operation. The capacity to restore the system’s stability as quickly as feasible after being subjected to any physical disruption is the official definition of the power system’s stability [47]. The system’s inertia is one of its key characteristics contributing to stability. Power system inertia is the ability to retain its current state and phase in the absence of incoming power. Higher inertia indicates the power system’s ability to maintain the current state to an extent to which the system’s re-dispatch does not disturb the loads. On the other hand, less inertia makes the power system more susceptible to frequency variations [48]. In conventional power plants, the generator has a freewheel attached at the rotor’s end, providing enough inertia to sustain within a specific time frame. RES power plants, on the contrary, do not add to power system inertia since the generator is not connected to a free-wheeling mechanical system. Most of the power conversion in RES plants is power electronics-based. For example, PV systems produce electrical power using semiconductor devices and cannot contribute inertia to the whole power system [48]. Moreover, this PV generation is DC type and needs to be converted to AC through power electronics. These power electronics do not attribute any particular functionality that can provide or create necessary inertia to the power system. Hence, the overall power system inertia is decreased; consequently, it reduces the power transmission frequency. Therefore, this non-inertial contribution of RES influences the power system’s transient stability, which is present in conventional generators with more pronounced rotor oscillations [49].

3.1. Effects of Photovoltaic (PV) Penetration on Power System Stability

The rising penetration of PV in power grids raises stability problems. In PV generation, semiconductor solar cells transform solar energy into electrical energy. The PV generation is DC in nature and largely depends on sunlight’s availability, hence weather conditions. The unpredictable weather conditions increase the variability of the PV generation. Regardless of penetration from the rooftop or very large PV plants, the variable nature of PV power causes the power system to be unstable and hence non-flexible due to volatility, intermittency, uncertainty, and non-dispatchability. Moreover, the absence of inertia in PV generation makes the power system less controllable. Due to the PV penetration’s unpredictability, the voltage, frequency, and active and reactive power become unstable. The analysis of PV reactive power and voltage responses demonstrates that overvoltage issues are brought on by widespread PV penetration in the transmission line [50]. Moreover, the PV-equipped power system’s static var compensators (SVCs) cause high transient overvoltage. This has been the case since SVCs’ protracted, low-speed entrance of reactive power into the power system after the fault was removed.

The transmission line busbars show overvoltage during 20% or more than PV penetration. It also causes higher voltage drops after the fault during transient events [51]. Another study shows that the distributed PV system has much-improved voltage stability compared to centralized power systems [52]. Numerous studies demonstrate that the widespread adoption of PV affects the power system’s transient stability in both favorable and unfavorable ways. The degree of PV penetration, power system types, power system topology, and the location of the PV generation are only a few of the variables that will determine the nature of the impacts—whether positive or negative [53]. Due to the voltage and frequency instability, very high PV injection could severely influence transient stability. On the other hand, distributed power systems have a more positive impact than centralized power systems. The distributed power system with supervisory control can enhance the transient stability to a greater extent, making the whole generation and transmission more flexible. On the other hand, the centralized power system with robust reactive power and voltage control regulates the power system’s dynamic stability [52].

The inertial effects of PV generation are another crucial factor in the power system [54]. As noted earlier, the generation of a PV power system is DC and does not show power system inertia during operation. The non-inertial condition brings stability concerns during any sudden failure and frequency fluctuations. Equipped with a voltage source inverter (VSI), the as-generated PV power is converted to AC and fed to the networks. This VSI integrates virtual inertia at the point of common coupling (PCC) end to synchronize the voltage waveforms generated by the PV system with those of the electrical power system. This synchronization aligns voltage and frequency on the same grid level; otherwise, the system will not be able to sustain itself.

In terms of frequency stability, it is shown that the increase of PV penetration level in the power systems decreases the frequency stability [55]. Equipped with automatic generation control (AGC), the high PV penetration with the power system tunes the typical generators’ output power, quickly damps the frequency oscillation, and reduces the overshoot [56]. In addition, the penetration of PV also challenges the present circuit protection system as the fault current profile due to high PV penetration being different, which would lead to a demand for new breakers, fuses, and relay protection systems [57]. It was also described in some of the findings that high PV penetration increases the harmonics of the distribution network, as the PV system itself introduces nonlinear loads to the network. The super harmonics and inter-harmonics rising from the PV plant might create a resonance that cannot be neglected, as the PV distribution system stays closer to the load [58].

3.2. Effects of Wind Power on Power System Stability

Wind power harnessing consists of wind turbines to collect the power from wind and power management equipment to provide flexibility in the power system. Since rotating turbines exert power from the wind, it is AC in nature but still shows variability due to the weather conditions. Due to the variable nature, such as volatility, uncertainty, intermittency, and non-detachability, wind power integration hampers power system stability considerably and needs to be managed accordingly. Otherwise, the power system becomes non-stable and less flexible.

Generally speaking, the storage requirement and wind power generation provide extra flexibility in the supply–demand balance. The two major factors of voltage stability are wind power penetration levels and power plant location [59]. The low penetration levels ensure better voltage stability. The power system needs enough reactive power for voltage stability. When used in a wind power system, a doubly fed induction generator (DFIG), which is a type of wind turbine generator commonly used in wind power systems due to its good performance in both efficiency and flexibility, is unable to provide as much reactive power as a synchronous generator (SG). Furthermore, DFIG is unable to produce enough short-circuit current. So, upon power system failure, the voltage stability in DFIG is worse than it is in SG. Additionally, the squirrel cage induction generator (SCIG) behavior of the DFIG turbines during transient events causes them to use reactive power, lowering the voltage stability limit [60]. Because of wind turbines’ nature, voltage stability situations in low-level wind penetration are poorer than in traditional power systems.

Contrarily, high-level wind injection can also improve voltage stability. Additionally, more reactive power is available in the power system thanks to the high injection of wind power [61]. According to research, the closeness of a fault has a negative effect on the transient and frequency stabilities of the power system. It took place because when the wind turbines are partially loaded, the active power decreases, and the reactive power increases during the crowbar protection [62]. To investigate system power stability, the Ayodele study team looked at the effects of elements such as power dispatch, wind farm location, and wind power penetration level [63]. Their investigation shows that wind power favors local area modes and that the DFIG wind turbines produce better damping in the inter-area mode. Additionally, the distance between the generating and user ends has little bearing on the stability of the power system. In terms of frequency stability, wind penetration exhibits better frequency stability than PV generation since the turbines of the wind power system can accommodate a better response to the load conditioning [62]. The DFIG turbines incorporate emulated inertial response by utilizing stored rotational energy [64], hence improving frequency response.

3.3. Effects of Hydropower on Power System Stability

In hydropower plants, the gravitational potential energy is converted to rotational mechanical energy to rotate the generators to produce electricity. A hydraulic turbogenerator is fed to the system to do this conversion. The hydraulic turbogenerator operates with the synchronicity of the power grid to maintain the power system’s frequency. In this regard, the turbogenerator rotates at a constant rotational speed. If there is any frequency oscillation due to sudden load changes or other causes, the governor tries to reduce that effect and ensures power system stability. The penstock size, hydro-inertia, and governor settings are a few crucial hydro system variables that regulate the entire generation process and maintain voltage and frequency [65]. Though hydro-based power generation mimics conventional combustion-based power generation, it has variability to some extent. Hydropower generation also depends on weather conditions, reservoir size and capacity, the height of the reservoir, etc. This variability brings some uncertainty and non-dispatchability to the power system. Although there is less uncertainty than solar and wind power, non-dispatchability is still a significant issue. So, hydropower still has some instability and non-flexibility to some extent and is relatively controllable more quickly than others.

3.4. Effects of Tidal Power on Power System Stability

Harnessing tidal energy from the ocean surface or river stream is another renewable energy source to combat climate change and accommodate affordable clean energy. The regular periodicity of tides makes them generally more predictable [66]. However, tides largely depend on the weather conditions such as sunlight, temperature, wind speed, water density, etc. Therefore, despite being low compared to others, it nevertheless exhibits some unpredictability and introduces some voltage and frequency instability as a result of the power system’s insufficient inertia [8]. The tidal generator is a specially equipped turbine similar to wind turbines that harness energy from the tidal gust. Since the tidal is more predictable, the rotational speed of the turbines can be easily controlled to synchronize with the frequency of the generated power.

3.5. Effects of Geothermal Power on Power System Stability

Unlike PV and wind energy, geothermal energy has more advantages since it provides reliable and stable power to the power grid. Moreover, geothermal has more inertia, enhanced voltage and frequency stability, and lower intermittency than others [67]. The functionality of geothermal is more like combustion-based power generation, in which steam with very high pressure is used as a governor to rotate the turbine. The same mechanism is used to produce electrical power in geothermal, but the governor (steam) is produced from renewable sources with zero carbon footprints. Since the governor is the same, the controllability is the same as with a conventional power generator. Geothermal generation shows low volatility, uncertainty, intermittency, and high dispatchability. So though geothermal is a renewable type of penetration, it still introduces a slight instability in the power system [9]. In terms of flexibility, since geothermal power generation depends on the geo-depth’s temperature gradient, it may show any abrupt changes. It may cause sudden failure of the power system. To overcome this flexibility problem, a storage system becomes handy.

3.6. Effects of Biomass Power on Power System Stability

Biomass is a plant-derived fuel that can directly be used in the combustion chamber to produce electricity. Like conventional thermal power plants, biomass is directly burnt to produce high-pressure steam [68]. This as-generated steam is then used as a governor to rotate the turbine. Since biomass-based power’s overall characteristics are precisely the same as conventional rotor-based power systems, the stability issue is more controllable in traditional ways. The governor determines the transient nature and can be regulated if there are any disturbances and sudden changes in the demand side. Though similar to power generation, biomass fuel has some advantages over conventional hydrocarbon fuels, as hydrocarbons are limited in nature, and biomass is potentially unlimited, with fewer carbon footprints than traditional fuels [69].

4. Power System Flexibility

PSF in a conventional power system is maintained through robust generation planning and large-scale power storage. Large-scale power storage, commonly called reserve, stores power and delivers energy to the power grid during generation or transmission failures. Some generation plants do not have any flexibility characteristics, for example, nuclear power plants. This is because nuclear power plants operate at baseload conditions. The power grid connected with nuclear generation will show inflexibility and will fail to balance supply–demand as quickly as possible. This will cause a sudden blackout, and to resolve this problem, it is necessary to integrate any additional large-scale storage system that will balance the power system during any abrupt changes. For example, to address the flexibility issue with nuclear power plants, the USA is expanding pumping-based hydropower plants (PHP) [70]. In this association, the PHP stores a portion of power in the means of gravitational potential energy from the nuclear unit and releases power by employing a hydro-generator when power grid demand is increased.

Due to the widespread insertion of VRES, maintaining PSF in contemporary power systems is a significant problem. Features like variability and the uncertainty of VRES provide different ramping patterns in the power system during transmission and distribution, which necessitates greater PSF inevitable [71]. Three key characteristics are used to describe PSF at the generation end: the output range of absolute power (MW), the ramping rate (MW/min), and the continuity of energy level (MWh) (Figure 5) [29]. The absolute output power range is the difference between installed power and the minimum amount of power needed to keep the power system stable while operating. The power system’s PSF grows as the output power range widens. The ramping rate, on the other hand, indicates how quickly the power system may vary its output within a specific amount of time; the higher the ramping rate, the more flexible the power system is. Energy level continuity relates to the power system’s operational duration without any disturbances.

Over this duration, the supply unit provides fixed power to the system for a certain period. If the source generation has a longer supply duration without any disturbances based on the demand, the source is more flexible. In a power system, the PSF is devoured by the variability of loads, forecasting errors, power grid outages, and the penetration of VRES [73]. Based on the task force of integration and variable generation of “the North American Electric Reliability Corporation (NERC)”, there are three main features of the power system that are responsible for assessing PSF requirements [74]:

• Net load changes;
• Time interval of load changes;
• Ramping events frequency.

These features are equally important since they can distinguish between unpredicted ramps from predicted ramp events and provide the necessary flexibility to the power system using resource adjustment [75]. These PSF resources are crucial for supply–demand adjustments and keeping the whole power system in work without any outages. Potential flexibility sources, actual flexibility sources, flexibility reserves, and flexibility reserves in the power market are the four main categories into which these PSF resources may be divided. Potential flexibility sources are physical resources that can be used but cannot be controlled or observed. Contrarily, usable sources—those that can be controlled and observed—represent the useful percentage of potential sources. A portion of reserve resources known as flexibility reserves can be acquired in the electricity market.

4.1. Chronological Requirements of Power System Flexibility

The widespread integration of renewable sources into power systems necessitates greater flexibility needs. The more renewable penetration needs more flexibility. It is reported that if the renewable penetration exceeds 30%, the flexibility requirement highly increases, especially for PV generation [76]. For scenarios of supply made entirely of renewable resources predicted for 2050, the generation (mixed with nominal capacity) must be twice as great as maximum demand [77]. A recent report by the Electric Reliability Council of Texas (ERCoT) reported that if the renewable penetration is more than 1 GW, the flexibility requirement needs to be recalibrated. A total of 14.5 GW penetration of PV in ERCoT necessitates a ramp increment by 135% of 1 h ramp rates and 30% in 3 h ramp requirements [78]. The utilization of renewable energy sources in power systems therefore results in increased degrees of flexibility needs. It is crucial to understand how the temporal impacts of renewable resources affect the flexibility of the power system in order to detect and fix flexibility problems. Time sequence effects are the effects’ durability in terms of flexibility. The penetration of renewable energy in flexibility requirements has four distinct time sequence effects: super short-term, short-term, mid-term, and long-term requirements (Figure 6).

4.1.1. Super Short-Term Requirements

Sometimes, variability in renewables is instantaneous. In that case, some millisecond-range control systems, such as low-voltage-ride-through, frequency control, reactive power control, supervisory control through data procurement, voltage control, ramp rate control, power electronic control, and dynamic modeling of power plants, have been proposed and investigated to deal with these instantaneous variations on renewable generation [79]. The power system’s transient stability is the foremost important thing in the case of dynamic modeling. Smart and microgrid incorporation can fulfill super short-term flexibility requirements.

4.1.2. Short-Term Requirements

Reserve capacity elevation, responsiveness of inertia and frequency, ramp capabilities, and lowest generation limits are the most frequent short-term impacts of flexibility requirements [80]. For quick generation availability and to compensate for the demand uncertainties, proper reserve allocation is essential for reliable power systems. Due to their unpredictable generation and bad forecasting, the power injection from renewable energy sources causes uncertainty and intermittent in the power systems. So, it is crucial to provide a conventional reserve in the power system for power balance during operation and management. On the other hand, the ability of renewable resources to support reserves largely depends on their ramp capability. Hence, along with reserve increasing, the arrangement of ramp capability is important to soothe the variability and uncertainties caused by renewable resources. The issue arises from the requirement of conventional generators to be able to modify their set points in response to the hourly fluctuations in the net demands. As a result, the demand change’s slope must be less than the total ramp capacity of traditional generators. In that case, generating units with more dispatchability are employed to follow the loads in response to demand variabilities. For the short-term effects, the proper regulation services are the only response option for net demand variation. Regulation services compensate for any frequency deviation and/or area control errors (ACE). The power distribution between loads would deviate if there were any regulation services lacking. In this situation, the market price and energy price would be different, affecting the market with long-term duration. Thus, the reserve regulation and accompanying flexibility would be affected as a result of the variability and uncertainty of renewable generation because any divergence from renewable needs to be compensated appropriately. The minimum power constraint of generators is another short-term effect on flexibility requirements. Consider a situation where demand is fulfilled by renewables and conventional resources are idle; this situation is highly unlikely since the turn-on and turn-off times are so high for renewable resources. If there is any sudden drop in renewable generation, a rapid starter is needed in the power system to prevent sudden failure. This rapid starter is so costly and thus not economical. In addition, consecutive turning on and off would also be expensive in terms of fuel, labor, administration, repairs, and maintenance. The large penetration of inverter-based power such as PV, wind, tidal, etc. decreases the power system’s total inertia. This will affect the system’s frequency response. In conventional power systems, if any disturbance occurs between the demand–supply chain, the traditional generators change the inertia accordingly and thus adjust frequency deviations. During significant renewable penetration, conventional generators have limited load sharing. In this situation, the power system inertia and frequency response might not be sufficient to trigger the frequency load shedding relays.

4.1.3. Mid-Term Requirements

Since the large-scale penetration of renewables causes the conventional unit’s rapid turn-on/off cycling, traditional units, such as hydro and open-cycle gas turbines, can cope with frequent cycling. Still, other conventional units that are designed for baseload conditions cannot be sustained without any damage. Frequently cycling causes large-scale damage to the plant’s components and increases maintenance requirements. Some common plant-related cycling damages are thermal shock, corrosion, erosion, fatigue, and heat decay. This cycling-based damage may cause power system failure and force any unexpected outage. On the other hand, this cycling effect will reduce the power system’s efficiency and will make the power system less economical due to raised maintenance costs [81].

4.1.4. Long-Term Requirements

The global community is switching from fossil fuel-based power generation to low-carbon baseload generation units like nuclear, geothermal, PV, wind, etc. to attain a sustainable society. These generation units have low ramp capability and limited flexibility, which will prevent rapid integration with power grids [82]. In this case, the less flexible units will become less desirable, and fast reaction units with high flexibility will become more prevalent despite increased energy costs. Based on this inference, some typical signs of an inflexible power system include but are not limited to [83]:

• Imbalanced demand supply will lead the frequency deviations and load losses;
• Reduction of renewables due to transmission constraints;
• Rapid errors in area control brought on by changes to the power exchange schedule. It highlights the inadequate methods used for power wheeling and trading;
• Negative market prices;
• Transmission line capacity shortages limited peak units, and a lack of demand responses are the causes of price volatility.

To ensure enough flexibility in the power systems, the power system planning and operating management are trying to find better solutions with the desired level of flexibility.

4.2. PSF Classification Based on System Needs

To maintain the power flow smooth and continuous in the power system, it is crucial to maintain a stable transfer frequency, global energy supply, and stable power grid voltages and transfer capacities locally. It has been seen that the PSF solutions and resources can be found locally in the power system and regulating these local resources can make the whole power system flexible and sound. Using the four power system parameters, PSF may be locally maintained over an entire power system. Specifically, flexibility in voltage, flexibility in power, flexibility in energy, and flexibility in transfer capacity (Figure 7).

4.2.1. Flexibility for Power

If the supplied power and the power sought by the loads are equal, a fully functional power system is flexible. In a power grid, transmission and distribution typically take place via alternate current (AC) methods. Flexibility for power in an AC transmission system refers to keeping the power system frequency within a predetermined range to avoid frequency instability for brief periods (the range is seconds to an hour). Traditionally, this is accomplished by synchronically managing the active power of the generators.

4.2.2. Flexibility for Energy

Supply and demand energy equilibrium implies future demand scenarios where flexibility requirement is the securement of future energy demand for the short to long-term periods (hours to several years). Energy flexibility is ensured for the long-term perspective by stockpiling raw materials (fuels) for plants or using hydro reservoirs to store energy for the future outlook. Maintaining energy flexibility in the power system for the long term is somewhat complex and requires robust optimization due to variable forecasting and future load scenarios. The scheduled exact timing of maintenance of the baseload thermal units provides energy flexibility and ensures energy availability during high demand.

4.2.3. Flexibility for Transfer Capacity

In an AC transmission system, the structural topology and capacity of the power grid play a vital role. The remedial actions, which are real-time operation or anticipated topology changes during use, provide cost-free flexibility in the power system. Over short to medium-term periods, this topological flexibility maintains the supply–demand scenarios during increased peak demands, peak supply, and variability of loads.

4.2.4. Flexibility for Voltage

For power system stability and improved power quality, the power grid voltages must be maintained within a pre-defined level throughout the power system. In contrast to frequency stability, voltage stability is greatly impacted by the reactive power of the generating units. The modern power system’s increased distributed generation causes operational variances and bidirectional power flow, which raises the risk of voltage instability. To maintain the voltage within pre-defined levels, urgent flexibility solutions must be provided within the power system. Ancillary services from distributed generators, storage medium, and demand-side smart responses are the paramount flexibility solutions for voltage stability over short-term periods.

4.3. PSF Dimension

From a technical standpoint, the PSF is necessary to account for power system uncertainties and any variations in demand at the generation site. However, according to the economics, offering PSF entails additional costs. As a result, a trade-off between PSF and extra expenses must be taken into account. The PSF would be accessible through regulation processes and might be offered at a marginal cost for a set period based on risk management criteria [84]. According to the discussion above, four crucial dimensions of PSF could be categorized: response time, control function, future uncertainty, and cost. The first three dimensions are technical norms, and the last one is economic. Other PSF dimensions could be resource location and their proximity to the power system [85].

4.3.1. Response Time

The response time length measures how quickly the power system can correct any deviations and return to the default condition. The time it takes to respond can range from a few seconds to months. The power system determines how long it will take, and depending on how quickly it responds, it can execute different levels of flexibility [34]. Short response times are related to flexibility over the short term; the time span could range from a few minutes to several hours. A large interval, on the other hand, is connected to a long response time. It might concentrate on long-term management and planning over a few months, including generational pairing, legislation policies, and altered consumption patterns. A power system that lacks short-term flexibility has enough long-term flexibility for the following cycles. For instance, a power system may have sufficient capacity to handle long-term load growth but cannot tolerate daily demand fluctuations. The precise calculation of the reaction time dimension is therefore essential for analyzing PSF, as is evident from the prior discussion.

4.3.2. Control Functions

A series of corrective processes often referred to as “control function” should be used to ensure proper operation over a given time response. It is obvious that this collection of control functions depends most on reaction time. The six typical corrective functions in power systems for various periods include AGC, load flow, unit commitment (UC) trading scheduling, short and long-term outage coordination, production, and anticipated investments [86]. Each corrective function corresponds to the specific time intervals and does particular actions to correct the inflexibility in the power system. As depicted in Figure 8, the power system operator is equipped with individual control processes in each specific time interval. The extended set of control functions implies that more operational choices should allow the operator to handle unwanted events during operation.

4.3.3. Future Uncertainty

Lack of comprehensive knowledge about the power system’s future state is referred to as uncertainty. Power system planning and operation are continually hampered by power system uncertainty. The major indicators of power system uncertainty are the likelihood of power system outages, load forecasting mistakes, weather forecasting errors, generation forecasting errors in VRES, and market pricing. The degree of uncertainty in the power system suggests how much flexibility should be added to manage it. The decision-maker always chooses the specific amount of uncertainty concerning the objective interval to reflect the risk calculation.

4.3.4. Response Costs

Response cost is always associated with market variation and uncertainty. These aspects of the response cost are directly related to the power system’s control process, hence flexibility. To reduce the response cost, the power system is made more flexible. The power system incorporates some specially designed control process units that can respond to uncertainty promptly. On the other hand, to minimize flexibility-related costs, the power system’s marginal cost is sometimes considered system flexibility. For instance, the control process has no restrictions and related expenses if the marginal cost is significant. In contrast, some control mechanisms declare themselves to be uneconomical and should be removed from the power system if the marginal cost is minimal. Thus, for proper power system operation and planning, the PSF and response cost should be traded off to maximize the power system’s flexibility and minimize costs.

5. PSF Resources

There are frequently two main forms of flexibility in power systems: one is physical flexibility, and the other is structural flexibility. Physical flexibility is the capacity of the power system to physically respond to alterations in demand and supply. Structural flexibility exploits physical flexibility through operational and marketing procedures. There are nine types of common flexibility resources that fall into either physical or structural categories (

Table 3. Flexibility resources of power systems [86].

Resources	Physical	Structural
Resource diversity	√
Demand end management	√	√
Power grid side flexibility	√	√
Market design		√
Robust control over VERs	√	√
New ancillary services		√
Energy storage	√
Smart grid and microgrid	√	√
Sectoral integration		√

). If a resource’s flexibility provision happens by physical characteristics, then it is classified as physical or structural. Some resources inherently provide flexibility through physical means, such as conventional units, energy storage units, etc. Other resources require the power system to be physically flexible while also adhering to operational and market standards. These resources are categorized as both physical and structural.

5.1. Demand Side Management

In conventional power systems, achieving demand–supply equilibrium is essential, and it is accomplished by regulating the supply from generation units. On the other hand, for large-scale renewable energy penetration, demand-side control provides low-cost flexibility solutions [38]. Demand-side management could mobilize the end uses, which will enable variations in demand regularly, increase demand flexibility, and integrate with efficiency opportunities [87]. Electric vehicles and the heating sectors are only two examples of emerging demand-side technologies that will be crucial to the flexibility of the power system. Electric vehicles comprise Li-ion battery storage which could serve as a distributed energy storage proportional to demand capacity [88]. On the other hand, heat pumps could also be a storage if combined with a thermal storage system [89]. Moreover, targeted incentives and behavioral changes through education to the customers are another idea that will help to soothe the peak demand and introduce some flexibility to the power grid. Integrated demand-side management (IDSM) might also help by sending real-time notifications to the consumer to adjust loads based on the system condition. This idea will work, as the loads these days are controllable by the consumer through the internet cloud when they are not at home. Hence, these new demand-side technologies could bring more flexibility to the power system.

5.2. Market Design

Market design and structure control the cost and capability of a power system with high VRES. A traditional power system that primarily relies on zero variable cost resources has different driving forces of market transactions. In this power system, the wholesale value is pretty low because of the reliance on zero variable cost energy. In terms of power reliability from variable resources, power grid services will dominate the market trading, reflecting the flexibility increase. However, this power grid expansion will result in higher risk for the producers of renewable energy, who will need more substantial monetary backing or alternative means to spur growth than in low-risk circumstances [90]. In this case, it may be imperative to incorporate flexible resources into the market design and structure to guarantee reliability even though they may not be competitive on an energy basis [91]. The market’s affordable services will enable participation from a range of resources, including conventional units, renewables, distributed system level generation, demand-side resources, and energy storage systems (Figure 9). Additionally, the market might be able to trade in real time, which would improve forecasting precision and enable adaptability to fluctuations. To allow huge participation from a wide range of flexibility sources, the market should extend to the regional and national borders [92]. Reaching out across national borders will need additional institutional adjustments and a combination of new power markets. This will increase additional transmission construction and the interconnectedness of disparate markets. In addition, a capacity market that will always ensure enough production during peak hours is another component that can be added to the market design.

5.3. Robust Control over VRES

Many market designs encourage renewable generation assuming the prevailing traditional generation in the power system will provide stability and reliability in terms of frequency control, voltage control, and balancing services. During large-scale renewable penetration, this power grid stability should come from the renewable generation itself. This creates a great transition in operation by devaluing the power grid support services that maximize renewable generation. Machine learning and artificial intelligence can also optimize and maintain the operation of VRES by predicting production and managing energy storage. Markets and technology will both alter as a result of this transition. For instance, the structure of the production credits needs to be changed to account for penalties for low production compared to demand. It is necessary to incorporate additional payments for more power grid support services [93].

5.4. Resource Diversity

The cost per unit of energy is the key factor when considering new renewable resource sites. Based on this economic premise, new sites are being constructed without consideration for the quality of the electricity or the relative timing of the resources in relation to other resources. For example, wind farms are developing in sites where the wind speed is abundant, and transmission access is relatively easy. As a result, the resource stations are clumped in a specific site and have similar ramp events. This aggregates overall variability and increases non-flexibility in the power systems [94]. In some areas, the site is divided into some zones hoping for the spreading of resource stations with different weather regimes with low spatial correlation. This affects the sites negatively with poor forecasting dependency. To get around this, one should establish a boundary where the output of new resources will correlate with that of old resources in order to provide financial incentives to situate them in areas where there is less correlation. Additionally, incentivizing distributed generation also creates some diversity in site selection that could be considered in collaborative planning. However, in order to retain low transmission infrastructure, the benefit of geographic diversity must be appropriately balanced. Evaluation accuracy, as well as well-planned site development, will make the resource diversity economically viable.

5.5. Energy Storage

In a power system where renewable energy sources are the predominant generation sources, energy storage is essential. Storage systems can provide energy during spans of peak demand and store excess energy generated during spans of low demand. Storage units also can smooth the power distribution and management during any fluctuations. It can also ensure power systems control stability and reliability. Energy storage is a fantastic tool for the power system’s flexibility in that regard. In power systems, the main focus is developing technologies that can consume electric power during low demand and deliver it during peak demand. Storage units can integrate at any point of the power system or all three points. The storage units on the main input side could be coal heaps, hydro reservoirs, etc. Storage options for the power grid side include batteries, pumped storage, etc. On the user’s end, they might be ice chests, electric cars, hot water tanks, etc. Nowadays, the supply from the controlled generators is cleverly adjusted to maintain the demand–supply balance at the primary input side. Since there is a renewable mix to a large extent, the power balancing role of conventional generators is largely compromised. In this situation, alternative solutions are needed to meet the long-term demand and provide extra control capability to the power system. This long-term energy storage, for example, hydro storage, commonly named bulk storage, can provide long-term flexibility. The challenge associated with bulk storage is costly.

Some smart and innovative power storage ideas [95] could be to use renewable power to synthesize biofuels and use these biofuels in conventional means [96]. This would be an affordable and clean storage system since biofuel has fewer carbon footprints. For example, synthesized hydrogen and methane could be future flexible solutions for power generation. Methane could be used in traditional facilities and kept in natural gas storage. Though this seems not economical, high renewable assisted methane will reduce overall costs and make the storage and second-round power generation profitable.

5.6. Efficient Power Grid Side Transmission Lines

The mass population will accept an updated development in the power transmission system if the renewable penetration is very high along with an integrated strategy. A strategic plan and implementation are needed for compensation benefits for affected people by new infrastructure, improved coordination for the building process of outs, robust coordination of electricity and fuels infrastructure, balanced mechanisms of finance for cross-border extension, and orchestrated operation.

Two difficult concerns are the social acceptability of transmission infrastructure for renewable distribution in outlying locations and for the continental scale backbone transmission across vast geographic areas. An alternate approach might simply enable dynamic capability assessment and control of additional power flows to improve the efficiency of the current transmission system. Power flow control tools like phase shifters, high voltage direct current (HVDC) lines, and other flexible AC transmission systems (FACTS) tools can be used in that circumstance to redirect power around dense lines [97]. Another option is to enhance the capability of transmission lines by introducing high-performance conductors and control and measurement technology to enable network infrastructure availability. On the other hand, the geospatial specificity of electric market costs could be another aspect of the efficient use of power grid-side transmission. The copper plate zone is a large area where network characteristics are invisible to the market participants and hold uniform market costs. This condition may lead to congestion and can be resolved by the dispatchability of generation units.

5.7. Smart and Microgrid Systems

For stable operation, grid-type power systems largely depend on controlling a few components, such as generators, switches, transformers, reactors, etc. Operations are made more complex by the large-scale power injection of VRES, for example, solar, wind, hydro, geothermal, biomass, energy storage, etc. These VRES incorporate variability and instability in power systems operations. Automation in power systems with high self-controllability enables two-way communication and operation. Due to this variability and instability, the VRES penetration of the power systems faces some vital challenges when controlling and communicating with a sheer number of components. A new paradigm system infrastructure has emerged, commonly named “Smart Grid or Microgrid”, to address the previously mentioned issues [98]. For a number of stakeholders, including customers, system operators, and the massive power system, a smart grid may successfully optimize the performance of VRES [99]. To give the power system significant flexibility, coordinated operation between aggregators and transmission and distribution operators is equally crucial. Additionally, the distribution system operators need a change in regulation and operation to fully activate VRES flexibility [100].

5.8. New Ancillary Services

Ancillary services are mandatory to prevent overload in the power system over all time scales, maintain voltage and frequency within limits, and balance demand and supply. Since the traditional power systems are phasing out due to distributed energy resources (DERs) ancillary services are becoming more relevant to the modern power grid to allow the swift transition from the conventional power grid system [101]. A few typical auxiliary services to consider include load following, frequency response, and inertial response. The stability, efficacy, and security of the power system are ensured by these auxiliary services. In traditional power systems, the ancillary services come from synchronous generators by coupling the electro-mechanical power of the power grid and spinning rotors [102]. These synchronous generators provide power system stability by ensuring the generator end’s controlled torque.

The load following method incorporates the demand variations in the power system while maintaining the supply–demand chain. In hourly energy markets, the operators provide a load following service by economically dispatching conventional units. The necessary ramp capability may be more important than traditional units during demands for widespread adoption of VRES. While reserve storage may be dedicated for deployment during emergencies or very high generation losses, load following ancillary service could be the best alternative solution during normal operating conditions.

Reserving frequency response is currently not considered an auxiliary service in the power market, despite being identified in power system studies. Frequency and inertia response are two essential auxiliary services that must be taken into consideration in inverter-based power integration, for example, renewables, storage, etc., when there is no inertia [103]. The development of technology has made it possible for RESs, storage, and responsive loads to supplement conventional generators by providing steady frequency response. In this case, the frequency response would not be granted as a requirement of the power system network, but rather as an additional service to fulfill the power system’s demand for an acceptable response.

In a conventional power system, the traditional generators provide system inertia from the kinetic energy of spinning rotors. The more the system inertia, the more the power system can sustain stability. The large-scale penetration of VRES lacks inertia or no inertia at all. In that instance, the power system will experience severe frequency variations that it cannot withstand. The power system frequency response clearly decreases under certain stress conditions, such as low demand, VRES penetration, or a lack of synchronous generators [104]. On the other hand, modern resources like storage and DERs can offer an inertial response with reliable control systems. The market’s inertial supply or the network’s prerequisites have piqued the curiosity of the research community. However, there is a debate about whether the inertial response could be obtained from the required network connection conditions.

5.9. Surplus Generation and Sectoral Integration

If VRES fulfills all the demands, there would be a situation where VRES could provide extra power, creating an abundant situation. This will necessitate bulk energy storage or bypass the extra power to other economic sectors. This bypassing is another flexibility resource that could enhance the energy supply’s security. However, it brings some additional challenges, such as converting energy from one form to another with low conversion efficiency. However, this conversion might offer an additional form of storage like producing green hydrogen through electrolysis, enhancing the power system’s flexibility. This sectoral bypass suggests that the market mechanisms need to be updated to allow consumers to make the most of the best energy sources.

6. PSF Planning

The world is transitioning from fossil fuel-based electricity generation to renewable energy to achieve the UN’s 2030 objectives. A significant yet coordinated worldwide initiative has been undertaken to deliver affordable and clean energy to all while addressing climate change. This transition to a sustainable energy future, predominantly characterized by renewables, anticipates an increase in the share of renewable energy resources from 30% in 2021 to 85% by 2050 [90]. In 2023, renewable electricity capacity additions were predicted at 507 GW, about 50% greater than in 2022, driven by sustained policy support across over 130 countries, resulting in a notable shift in the global development trajectory. For the next few years, the growth of solar and wind will be consistent, as the generation costs of these two powers are cheaper than the other fossil fuel-derived powers due to the policies of most countries [105].

In 2022, photovoltaic generation constituted 60% of global renewable energy, succeeded by wind and hydropower. The present share of renewables in the USA is 21% and is projected to increase by 2050 (Figure 10). Projections indicate that photovoltaics will constitute 51% of all renewable energy in the USA by 2050, with wind energy at 31%, hydroelectric power at 12%, and other sources at 6%. Similar conditions are observed globally, including in the European Union, Asia, South America, and Africa [106]. The electricity generation from renewables is stochastic in nature, with high intermittency and variability. These variabilities could range from very short-term time duration (millisecond) to very long-term time duration (over the year). Figure 10 (left side) shows the USA case study’s renewable generation variabilities. From this depiction, it is clear that the VRES generation varies from time to time and is highly unpredictable. On the other hand, the user end complexity is ever-growing due to the increase of variable loads.

Coupled with variable loads, this penetration of VRES introduces operational and planning difficulties to the conventional units of the power system [37]. The power system variability plays its role based on stage by stage. From generation to end users, there are multiple stages that need proper assessment and planning. Since the variability differs from short-term time duration to long-term duration like hours to days, the planning stages would be i. generation and transmission stages for flexibility adequacy, ii. a day ahead stage for VRES forecasts, iii. intraday stage for generation rescheduling with forecasting, and iv. medium to the long-term stage for seasonal change-related variabilities. Along with stages, some new approaches, like updated power systems and market operations, regulations, and public policy, are also required in case of a new assessment of the transition from conventional to renewables.

The large-scale penetration of VRES has raised serious reliability and stability issues in contemporary power systems. A power system can be reliable and stable if it is flexible enough during demand–supply imbalances and has adequate generation and reserve capacities. Conventional power systems lack appropriate regulatory provisions and have an unsmart market design to cope with the generation variabilities. So, during large-scale penetration of VRES, the traditional power system fails to provide reliability and stability. In that case, a comprehensive flexibility assessment and planning with proper utilization is inevitable for regulators and operators [107]. Some newly assessed practices, for example, integrating new operational protocol with a conventional power system, retrofitting existing generating assets, and making the transmission line extended and smart, can provide enough flexibility.

On the other hand, additional reservoir and demand side response can provide flexibility; however, they need some aggressive policy interventions and market mechanisms. In that sense, designing and integrating suitable flexibility resources with appropriate policy infrastructure needs long-term power system assessment and planning. On the other hand, short-term assessment and planning could focus on smart transmission infrastructure, exploit reservoirs and storage in immediate need, and retrofit fossil fuel-based plants. To provide enough flexibility, proper assessment and planning should be investigated and validated from proper modeling and utilization in real-time scenarios. The modeling should interlink various sectors and geospatial-based models and conditions [108].

Proper planning and utilization are key considerations of regulators and power system operators to make the power system flexible enough. Though the optimization of PSF is somewhat complex and spans various phases, proper planning can enhance the flexibility of the power system, thereby improving its cost-benefit ratio. For PSF planning, the first step is the proper estimation of flexibility and inflexibility in the current power system. Good estimation can help to plan and forecast the current and future flexibility requirements effectively. If there is any sudden deficit of flexibility in the power system, the regulators and operators should increase the investment in the flexibility resources sectors; otherwise, VRES generation curtailment is inevitable. On the other hand, the pressing need for PSF could influence stakeholders and decision-makers to see alternative PSF resources differently when weighing their cost-effectiveness and gestation period.

Generally, the present flexibility is fulfilled in the case of future PSF planning, and then proper planning is carried out. During PSF planning, some important factors need to be considered, such as VRES curtailment, reserve margin, generation costs, level of over/under generation, network capacity, etc. [109]. After considering these factors, proper PSF planning involves three critical steps:

• Present PSF evaluation;
• PSF deficiency response;
• Assessment of future PSF requirements (Figure 11).

During existing PSF evaluation, two essential evaluations are generation cost modeling and network analysis. Under generation evaluation, the level of curtailment, excess production, and reserve margin adequacy need to be assessed innovatively. On the other hand, planners and operators need to evaluate voltage and frequency violations, capacity assessment to recover the uncertainty of VRES generation, sufficient power system inertia, and power systems overload assessment for network analysis. After the existing PSF evaluation, planners and operators need to respond to the deficiency from the first planning stage using the least cost technique if there is still a flexibility gap. In this stage, they need to unlock the existing PSF and invest in new assets. After investing in new assets, planners and operators need to implement demand management schemes. Planners and operators go to the final stage of future PSF requirements if there is no flexibility gap after the existing PSF evaluation. At this stage, first, planners have to optimize the VRES plant locations, optimize the VRES generation mix with conventional units, estimate VRES generation based on areas and policy goals, and predict future net demand. After optimization and prediction, planners have to implement future capacity expansion. In this step, they study and optimize the future technologies of traditional power resources based on net demands. Then inject extra PSF resources with advantages evaluation. Suppose there are any long-term operability issues after extra PSF injection. In that case, they must repeat VRES generation optimization and inject additional PSF resources until long-term operability issues are minimized.

PSF resource planning considers several variables, including the degree of over/under generation, backup margin, VRES generation and management, and generation costs. This kind of planning can be carried out by estimating future demands and evaluating resource combinations to preserve reliability and minimize costs while considering risk preferences and regulatory restrictions. Achieving and sustaining adequacy, or the amount of capacity of the resource to supply 100 percent depending on the demands, is the expected goal of resource planning. Regarding VRES, the “net load” idea can be used to plan PSF Netload, which is calculated as (actual load minus variable generation)/hour and represents the need that conventional generators must satisfy hourly if all VRES are employed. For example, Figure 11 indicates how the VRES (wind generation) can impact the power systems’ operations. In this case, shorter peaks represent fewer hours of operation for conventional generators. This could have an impact on cost recovery and obstruct long-term supply security. Steeper ramps indicate a faster increase/decrease in power dispatch. Lower turn-down indicates the dispatchable generation needs to turn down to low levels as soon as possible but may rise again quickly [19]. Other VRES like solar show similar characteristics. If the PSF requirement is not fulfilled, the power system will face reliability and financial consequences like load reduction, VRES curtailment, scheduled area power balance deviations, frequency and voltage instability, and price volatility.

During PSF planning, the regulators and operators should act accordingly; otherwise, the reliability and economic consequences would be tremendous. The regulators and operators have various physical and institutional interventions to provide flexibility in the power system. The physical interventions include storage, conventional generation, active power control of variable generation, load-side storage response such as electric vehicles and heat storage, smart transmission lines with high capacity, etc. On the other hand, institutional interventions help extract the best possible flexibility from physical intervention. Some common institutional interventions are broad balancing areas, innovative market design for fast dispatch, accurate VRES forecasting, and smart or microgrid implementation [92]. The appropriate PSF assessment and planning are summarized in Figure 12 based on the intervention types and associated economic costs [83].

7. Conclusions

Conventional power systems always incorporate an appropriate level of flexibility so that the regulators and operators can provide prompt balance during demand–supply imbalances. But this balance during the large-scale penetration of VRES like solar, wind, etc. becomes hampered due to their volatility, intermittency, and uncertainty. The large-scale penetration of VRES injects a high inflexibility in modern power systems that needs to be addressed accurately; otherwise, the power system may fail and bring some economic consequences. A power system can be flexible if it holds fast to monetary restraints and reacts rapidly to unexpected changes in the demand–supply chain. It can slow down power production when demand is low and ramp up when demand spikes during peak scheduled or unpredictable events. So, the quick response of the power system during any fluctuations in the demand–supply chain is an important factor in making the power system alive all the time. In that sense, an appropriate level of flexibility injection in the power system is crucial regarding reliability and resiliency.

To cope with the UN’s vision 2030, a renewable and sustainable energy transition is inevitable. Though this transition can provide affordable and clean energy and combat climate changes, it brings some unavoidable issues in power systems, particularly during generation and transmission. Inflexibility is the most serious of these issues since RESs, like solar and wind energy, are so variable. Considering this, in this review, we first elaborately discuss the large-scale penetration consequences of VRES. Due to their volatility, intermittency, and uncertainty in the VRES, the power system becomes insecure and unreliable. To address these issues in the power system, flexibility is the term that defines and describes which adjustments need to be incorporated into the power system. The “flexibility” is coined relatively in modern times, and a dim view of it makes it difficult for the stakeholders and power system policy planners to address this issue appropriately. In this sense, this review provides the proper definition and explanation of power system flexibility from various points of view.

More flexibility is required because of the widespread adoption of renewable energy sources in power systems. These requirements, along with flexibility dimensions, are incorporated to bring accurate mathematical formulations of flexibility. A proper mathematical formulation would help the policy planner simulate and model the power system accordingly, which could improve its security and reliability and make the power system more economical. In that case, some new metrics are coined to qualify and quantify flexibility for instantaneous and long-term planning. When choosing suitable flexibility metrics, many factors must be considered, such as the generated energy portfolio, traditional and renewable energy mix, transmission line constraints, and institutional and environmental constraints. A new transition to renewable generations also requires a new power transmission and management policy. This necessitates a new development and deployment of overall methods with more flexible resources and policies. A smart grid policy with high flexibility can make the power system more secure for users and investors. Some key policy factors are as follows:

• Demand-side proper framework to allow aggregation;
• Flexible power markets for a shift from wholesale to power grid support services, removal of price caps and suitable price signals for flexibility, curtailment of inflexible resources, design of robust power grid modeling, etc.;
• Power grid supportive renewable generations with smartness, such as adequate generation estimation, barriers removal, and proper forecasting;
• Diverse VRES generation in order to reduce aggregated variability;
• Energy reservoir for sudden need and supply smoothening;
• Smart and extended transmission lines in order to cover huge geographic areas;
• Smart/microgrid formation with enhanced communication and robust control infrastructure with a large institutional shift to accommodate community-based local distribution;
• Robust control mechanisms with high virtual inertia in inverter-based AC power generation;
• Other electrification sectors, for example, electric vehicles, heat storage, etc., for surplus power utilization.

Operating power systems with flexibility (especially VRES-based) are technologically viable but institutionally complex. A great leap from fuel-based generation to renewables strongly suggests a brand-new assessment of the power system with an appropriate level of flexibility is inevitable. For a holistic flexibility assessment, the policymakers and operators should account for the interaction between complex institutions and newly available technologies. Proper planning with various flexibility solutions and smart market designs for both resources and transmission systems can make the power system more secure and reliable both in economy and technology.