Electrical Power System

Chapter 15
Sustainable and Future Power Systems

Post Graduate Program in Electrical Engineering
Electrical Power System • Post Graduate

Sustainable and Future Power Systems

Introduction

Power systems are undergoing a structural transition from centrally dispatched, fossil-fuel-dominated networks toward low-carbon systems characterized by renewable generation, energy storage, power-electronic converters, flexible demand, distributed resources and extensive digital control. Sustainability in this context is broader than decarbonization. A sustainable power system must be environmentally responsible, technically secure, economically viable, socially equitable and resilient to disturbances.

Future systems must coordinate resources across time scales ranging from converter switching in microseconds to transmission planning over decades. They must also reconcile competing objectives: low emissions, affordability, reliability, resilience, land and material constraints, cybersecurity and public acceptance. This chapter develops the analytical foundations required to evaluate and design such systems.

Learning Objectives

After studying this chapter, readers should be able to:

Explain the technical, environmental, economic and social dimensions of sustainable power systems.
Formulate network-constrained generation, storage and transmission planning problems.
Apply correct power-balance and storage-energy equations.
Analyze frequency security in systems with declining synchronous inertia.
Interpret voltage-stability limits using AC power-flow equations and Jacobian analysis.
Compare grid-following and grid-forming converter controls.
Evaluate renewable integration using reliability, flexibility, emissions and economic metrics.
Assess regulatory, ethical, cybersecurity and governance implications of future grids.

1. A Five-Layer View of Future Power Systems

A future power system can be represented by five interacting layers. The layers exchange energy, control commands, measurements, prices and environmental information in both directions.

Five bidirectionally interacting layers of a sustainable power system

Figure: Five bidirectionally interacting layers of a sustainable power system

Figure: Five distinct layers connected by bidirectional physical, informational and institutional interactions.

Primary resources constrain conversion technologies, while conversion choices affect resource extraction and emissions. Networks determine whether energy can reach consumers. Digital controls coordinate physical assets, but their design is influenced by market rules and regulation. Consumer behavior and public policy, in turn, alter investment and operation throughout the physical system.

2. Sustainability and Performance Metrics

2.1 Environmental performance

Operational emissions measure pollutants released during plant operation. Life-cycle assessment additionally includes extraction, manufacturing, construction, fuel processing, decommissioning and recycling. A system-level life-cycle emissions intensity is

I_LC = \frac{\sum_k E_k}{\sum_t W_t}

where E_k is the life-cycle emissions assigned to stage k, typically in kg CO₂-equivalent, and W_t is delivered electricity in kWh or MWh. Numerator and denominator units must be selected consistently.

Low operational emissions do not imply zero environmental impact. Renewable and storage projects can create land-use conflicts, mineral demand, biodiversity impacts and end-of-life waste. Sustainability assessment should therefore consider multiple indicators rather than carbon alone.

2.2 Economic performance

The levelized cost of electricity is

LCOE = \frac{\sum_t=0^T\frac{I_t+O_t+F_t+C_t}{(1+r)^t}}
{\sum_t=0^T\frac{E_t}{(1+r)^t}}

where I_t is investment expenditure, O_t operating and maintenance cost, F_t fuel cost, C_t other applicable costs, E_t generated electricity, r the real discount rate and T the project life.

LCOE is useful for comparing technologies under consistent assumptions, but it does not represent system value. It omits location, dispatchability, correlation with demand, network reinforcement, balancing services and capacity adequacy unless these are explicitly added. System planning must therefore use chronological and network-constrained models.

2.3 Reliability, adequacy and resilience

Important metrics include:

Loss-of-load probability (LOLP): probability that available supply cannot meet demand.
Expected energy not served (EENS): expected unserved electrical energy, normally in MWh/year.
Effective load-carrying capability (ELCC): additional load that can be supported by a resource while maintaining a specified adequacy criterion.
System average interruption duration index (SAIDI): average annual interruption duration per customer.
Resilience: ability to anticipate, withstand, adapt to and recover from high-impact disturbances.

Nameplate capacity is not firm capacity. The adequacy contribution of wind, solar, storage or demand response depends on weather, load correlation, outage patterns, storage duration and the existing resource portfolio.

2.4 Equity and energy justice

A transition can be technically efficient but socially unsustainable. Distributional analysis asks who pays for network upgrades, who receives clean-energy benefits, which communities host infrastructure and whether vulnerable consumers can access flexibility programs. Procedural justice requires transparent consultation and meaningful participation in siting and policy decisions.

3. Renewable Variability and Flexibility

Wind and solar output are variable, uncertain and spatially correlated. Variability describes actual changes over time; uncertainty describes imperfect knowledge of future output. Forecast errors generally decrease nearer to real time, making intraday markets and updated dispatch valuable.

Flexibility is the ability to accommodate changes in net demand while maintaining security. Sources include:

dispatchable generation with adequate ramp rates;
short- and long-duration storage;
demand response and managed electric-vehicle charging;
geographically diverse renewable portfolios;
stronger interconnection and transmission;
renewable curtailment when economically justified;
sector coupling with heat, transport and hydrogen.

Curtailment is not equivalent to unserved demand. Renewable curtailment is available generation deliberately not used. Load shedding or unserved demand is consumer demand not supplied. These require distinct variables, costs and reliability interpretations.

4. Network-Constrained Planning

4.1 Multi-period stochastic formulation

A representative planning model chooses investment variables x and scenario-dependent operating variables y_s:

\min_{x,y_s}\; C^inv(x)+\sum_s \pi_s C^op(y_s)

Here \pi_s is the probability of scenario s, with \pi_s \ge 0 and \sum_s\pi_s=1. Operating cost may include fuel, start-up, emissions, curtailment penalties and the value of lost load. Integer variables are required for discrete line construction, unit commitment or mutually exclusive operating states.

For a lossless DC approximation, the nodal balance at bus i, time t and scenario s can be written as

\sum_{g\in G_i}p_gts+\bar r_its-k_its+p^dis_its-p^ch_its-L_its+\ell_its
=\sum_{j\in\Omega_i}B_ij(\theta_its-\theta_jts)

where:

p_gts is dispatchable generation in MW;
\bar r_its is available renewable power in MW;
k_its is renewable curtailment in MW;
p^dis_its and p^ch_its are storage discharge and charge powers in MW;
L_its is demand excluding storage charging, in MW;
\ell_its is unserved demand in MW;
B_ij is the DC line coefficient in MW/radian on the adopted base;
\theta_i is bus voltage angle in radians;
the right-hand side is net power exported from bus i.

The separate bounds

0\le k_its\le \bar r_its, \qquad 0\le \ell_its\le L_its

ensure that curtailment cannot exceed renewable availability and unserved demand cannot exceed demand. Storage charging appears on the demand side of the balance and must not be omitted when demand is defined independently of charging.

DC branch flow and thermal limits are

F_ijts=B_ij(\theta_its-\theta_jts), \qquad |F_ijts|\le F^max_ij

The DC model assumes voltage magnitudes near 1 per unit, small angle differences, predominantly inductive lines and negligible active-power losses. It is useful for large-scale planning but cannot establish voltage feasibility, reactive-power sufficiency or dynamic stability.

4.2 Storage dynamics

For time step duration \Delta t in hours,

e_t=e_t-1+\eta_chp^ch_t\Delta t-\frac{p^dis_t\Delta t}{\eta_dis}

where e_t is stored energy in MWh, \eta_ch and \eta_dis are charging and discharging efficiencies, and powers are in MW. Constraints include

E^min\le e_t\le E^max,\quad 0\le p^ch_t\le P^ch,max,\quad 0\le p^dis_t\le P^dis,max

A terminal-energy condition is needed to prevent artificial depletion at the planning horizon. Simultaneous charging and discharging can be prohibited with binary variables. A continuous relaxation is valid only when the objective and prices make simultaneous operation suboptimal and no other constraint rewards it.

Worked Example: Storage Energy Accounting

A battery initially stores 40 MWh. It charges at 10 MW for 2 hours with \eta_ch=0.90, then discharges at 12 MW for 1.5 hours with \eta_dis=0.90. Its final energy is

e=40+(0.90)(10)(2)-\frac{(12)(1.5)}{0.90}=38\;\text{MWh}

The charging operation adds 18 MWh, while supplying 18 MWh externally during discharge removes 20 MWh internally.

4.3 Security-constrained planning

A credible plan must remain within limits following specified contingencies. Under the deterministic N−1 criterion, the system is tested after the outage of each designated single element. Security constraints may include post-contingency line loading, voltage, generator redispatch, reserve and transient performance.

Planning models commonly use DC security constraints for computational tractability, followed by AC power flow, contingency analysis and dynamic simulation. Ignoring network constraints can place renewable or storage assets where their output is frequently trapped by congestion.

Uncertainty can be treated using stochastic programming, robust optimization or chance constraints. Stochastic models require probabilities; robust models protect against a defined uncertainty set; chance-constrained models limit the probability of violating selected constraints. Their results are only as credible as the scenarios, uncertainty sets and correlation assumptions used.

5. Voltage Stability in Converter-Rich Networks

5.1 AC power-flow foundation

For bus admittance element Y_ij=G_ij+jB_ij, the injected active and reactive powers are

P_i=V_i\sum_jV_j[G_ij\cos(\theta_i-\theta_j)+B_ij\sin(\theta_i-\theta_j)]

Q_i=V_i\sum_jV_j[G_ij\sin(\theta_i-\theta_j)-B_ij\cos(\theta_i-\theta_j)]

These equations assume a balanced sinusoidal steady state represented by positive-sequence phasors. They do not describe fast electromagnetic transients or unbalanced distribution operation.

Linearization produces the power-flow Jacobian:

\begin{bmatrix}\Delta P\\\Delta Q\end{bmatrix}
=
\begin{bmatrix}
\partial P/\partial\theta & \partial P/\partial V\\
\partial Q/\partial\theta & \partial Q/\partial V
\end{bmatrix}
\begin{bmatrix}\Delta\theta\\\Delta V\end{bmatrix}
=J\begin{bmatrix}\Delta\theta\\\Delta V\end{bmatrix}

Near a saddle-node voltage-collapse point, J becomes singular under the modeled controls and bus types. A small singular value or large reactive-power sensitivity can indicate proximity to instability, but no single index is universally reliable. Generator reactive-power limits, tap changers, load recovery and converter current limits can change the governing equations and produce limit-induced instability before the theoretical nose point.

5.2 Continuation power flow

Continuation power flow introduces a loading parameter \lambda:

P_L=P_L0+\lambda\Delta P_L, \qquad Q_L=Q_L0+\lambda\Delta Q_L

A predictor–corrector method traces the solution through regions where ordinary Newton–Raphson power flow becomes ill-conditioned. The maximum feasible \lambda defines a voltage-stability margin for the specified load-growth and generation-participation pattern. Because the margin depends on that pattern and on contingency assumptions, it is not a universal system constant.

Voltage-secure planning should represent reactive capability curves, switched compensation, inverter current limits, transformer taps and post-contingency conditions. Dynamic studies are additionally required when motor loads, protection, converter controls or delayed voltage recovery are significant.

6. Frequency Stability and Reduced Inertia

For a coherent equivalent system with synchronous inertia constant H in seconds, active-power base S_base in MW, nominal frequency f₀ in Hz and frequency-sensitive damping D_f in MW/Hz, the small-signal swing relation is

\frac{2HS_base}{f₀}\frac{d\Delta f}{dt}
=\Delta P_m-\Delta P_e-D_f\Delta f

All power terms are in MW. The coefficient on the left is dimensionally consistent with MW when multiplied by d\Delta f/dt in Hz/s. If powers are instead expressed in per unit, S_base is omitted and damping must also be expressed consistently in per unit.

Immediately after a disturbance, before damping and governor response become material,

\left.\frac{d\Delta f}{dt}\right|_0^+
=\frac{f₀\Delta P}{2HS_base}

where \Delta P=\Delta P_m-\Delta P_e in MW. A generation loss gives negative \Delta P and therefore negative rate of change of frequency (RoCoF).

Worked Example: Initial RoCoF

A 50 Hz system has S_base=1000\;MW, H=3.5\;s and suddenly loses 120 MW of generation. Neglecting immediate damping,

RoCoF=\frac{50(-120)}{2(3.5)(1000)}=-0.857\;\text{Hz/s}

This is the initial coherent-system estimate. The later frequency nadir depends on governor response, load damping, converter response, protection and network-dependent oscillations.

For several coherent synchronous machines, stored kinetic energy is aggregated through \sum_iH_iS_i. Non-synchronous renewable capacity should not be assigned a mechanical inertia constant unless its control and energy source actually provide a quantified inertial response.

7. Grid-Following and Grid-Forming Converters

A grid-following converter uses a phase-locked loop to estimate an existing grid angle and inject controlled current. It performs well in strong grids but may interact adversely with weak-network impedance, other phase-locked loops and current limiters.

A grid-forming converter establishes an internal voltage angle and magnitude. A common steady-state droop representation is

\omega=\omega₀-m_p(P-P^*), \qquad V=V₀-n_q(Q-Q^*)

where m_p and n_q are droop coefficients, and starred quantities are set points. A generic virtual-synchronous dynamic law is

M_v\frac{d\Delta\omega}{dt}=P^*-P-D_v\Delta\omega

where M_v and D_v are controller parameters with units selected so every term is power. This is a control law, not proof that the converter possesses physical rotating inertia. Sustained response requires an adequate battery, DC source or renewable energy margin.

All converter controls are constrained by semiconductor current:

\sqrt{i_d²+i_q²}\le I_max

During faults, active- and reactive-current commands cannot both be maintained if their vector magnitude exceeds I_max. Current-priority logic, saturation and anti-windup design therefore materially affect stability. Electromagnetic-transient simulation is often required for weak grids, fast controls, fault ride-through and interactions among multiple converters.

8. Sector Coupling and Emerging Resources

Electrification links power systems with transport, buildings and industry. Managed electric-vehicle charging can shift demand away from peaks, while uncontrolled coincident charging may increase distribution congestion. Heat pumps can provide thermal flexibility when buildings or heat stores permit temporal shifting.

Power-to-hydrogen can absorb electricity over long periods, but round-trip electricity efficiency through electrolysis, storage and reconversion is substantially lower than battery efficiency. Hydrogen is consequently more attractive for feedstock, high-temperature processes, seasonal storage and difficult-to-electrify applications than for routine short-duration electricity arbitrage.

Other developments include long-duration storage, advanced geothermal generation, high-voltage direct-current grids, dynamic line ratings, virtual power plants, vehicle-to-grid services and small modular reactors. Their value depends on system context rather than technological novelty alone.

9. Applied Case: Congestion-Aware Renewable Planning

Background

This is an illustrative three-bus planning case. Bus 1 contains a thermal generator with marginal cost $60/MWh and emissions of 0.7 t CO₂/MWh. Bus 2 has 80 MW of available renewable generation. Bus 3 has a 120 MW load. A line from bus 1 to bus 3 is limited to 70 MW, while the bus 2–bus 3 line is limited to 50 MW. Losses and reactive power are neglected for this screening calculation.

Problem and analysis

Renewable energy is assumed to have zero marginal cost. The existing bus 2–bus 3 line can deliver only 50 MW, so 30 MW must be curtailed. The remaining 70 MW of demand is supplied by the thermal generator:

P_R=50\;MW,\quad k=30\;MW,\quad P_G=70\;MW

The hourly thermal cost is

70\times60=\$4{,}200/\text{h}

Approach

A proposed reinforcement raises the bus 2–bus 3 transfer limit to 80 MW. Renewable output can then reach 80 MW, reducing thermal production to 40 MW. The new operating cost is $2,400/h, giving a saving of $1,800/h.

If this congestion pattern occurs for 2,500 hours per year, the gross annual operating saving is

1{,}800\times2{,}500=\$4.5\;\text{million/year}

For an annualized reinforcement cost of $3.2 million, the simplified net benefit is $1.3 million/year. Annual avoided emissions are

(30)(2{,}500)(0.7)=52{,}500\;\text{t CO₂/year}

Outcome and limitations

The screening analysis supports further evaluation of the reinforcement. However, it is not an investment decision by itself. A full study must include all chronological hours, renewable uncertainty, losses, AC voltage feasibility, N−1 security, construction lead time and possible alternatives such as demand flexibility or storage at an electrically useful location. Storage placed behind an already saturated export interface may be unable to relieve peak-hour congestion unless it charges during uncongested periods or is located on the load side.

Learning points

Renewable capacity has limited value if the network cannot transport its output.
Curtailment and unserved demand represent fundamentally different conditions.
Investment decisions require chronological system value, not LCOE alone.
DC optimization should be followed by AC and dynamic security assessment.

10. Digitalization, AI and Cybersecurity

Advanced metering, phasor measurements and distributed sensors increase situational awareness. AI methods can support forecasting, anomaly detection, equipment diagnostics and preventive maintenance. Nevertheless, learned models may fail under extreme events that differ from training data. Operational deployment requires uncertainty estimates, independent validation, explainability appropriate to the risk, fallback controls and monitoring for model drift.

Greater connectivity also enlarges the cyberattack surface. Security should include network segmentation, authentication, least-privilege access, secure updates, encryption where appropriate, incident response and recovery testing. Safety-critical protection should not depend solely on remote cloud services. Cybersecurity must be designed into procurement and architecture rather than added after commissioning.

11. Governance, Regulation and Professional Practice

Markets designed around energy production alone may inadequately reward flexibility, inertia-like response, reactive power, black-start capability or firm capacity. Market products should specify measurable performance without favoring a technology unnecessarily. Poorly designed subsidies can encourage generation in congested locations or shift costs unfairly to consumers unable to adopt distributed technologies.

Regulators must coordinate transmission planning, distribution modernization, interconnection standards and resource adequacy. Clear responsibility is required for converter models, data quality and compliance testing. Engineers have a professional obligation to disclose model assumptions, uncertainty and limitations, particularly when decisions affect public safety, affordability or vulnerable communities.

Recommended practice includes:

co-optimizing generation, storage and networks;
preserving chronological and spatial detail;
validating DC results with AC and dynamic studies;
testing extreme-weather and common-mode failures;
using transparent assumptions and sensitivity analysis;
incorporating lifecycle impacts and equitable cost allocation;
retaining human oversight of safety-critical automated decisions.

12. Challenges and Research Directions

Major unresolved challenges include multi-timescale co-simulation, stability assessment for systems dominated by grid-forming converters, probabilistic treatment of climate-dependent resources, scalable AC expansion planning, interoperability among vendor controls, and valuation of resilience. Research is also needed on recyclable storage materials, sustainable mineral supply chains and market arrangements that coordinate millions of distributed devices without compromising privacy or security.

The transition cannot be reduced to replacing one generator technology with another. It requires coordinated evolution of conversion, networks, controls, markets and institutions.

Chapter Summary

Sustainable power systems integrate environmental responsibility with reliability, affordability, resilience and social legitimacy. Renewable variability can be managed through diversified resources, storage, flexible demand, interconnection and improved forecasting. Network-constrained planning must explicitly distinguish renewable curtailment, storage charging and unserved demand while respecting line, energy and security constraints.

DC planning models provide scalability but require subsequent AC voltage and dynamic validation. Voltage stability depends on nonlinear power-flow behavior, reactive limits and converter current constraints. Frequency equations must use consistent per-unit or MW quantities; in MW form, the inertia term includes the active-power base. Grid-forming converters can support weak and low-inertia systems, but their response is limited by current and available energy. Successful transition ultimately depends on sound governance, cybersecurity, ethical practice and coordinated investment across all five system layers.

Key Terms

Adequacy: Ability of available resources to meet aggregate demand under specified conditions.

Curtailment: Deliberate reduction of available generation, usually because of congestion or economic conditions.

Effective load-carrying capability: Additional load supportable by a resource at a fixed reliability criterion.

Expected energy not served: Expected quantity of demand that cannot be supplied over a stated period.

Flexibility: Ability to respond to changes in net demand and system conditions.

Grid-following converter: Converter that synchronizes to an externally established voltage waveform.

Grid-forming converter: Converter that establishes an internal voltage angle and magnitude reference.

Inertia constant: Stored synchronous kinetic energy at rated speed divided by machine rating.

Life-cycle assessment: Evaluation of impacts across extraction, manufacture, operation and end-of-life stages.

LCOE: Discounted lifetime cost per discounted unit of generated electricity.

Load shedding: Intentional disconnection or non-supply of demand to preserve system security.

N−1 criterion: Requirement that specified limits remain satisfied after any designated single contingency.

Continuation power flow: Predictor–corrector method for tracing nonlinear power-flow solutions as loading changes.

Resilience: Ability to anticipate, withstand, adapt to and recover from severe disturbances.

RoCoF: Time derivative of system frequency following a power imbalance.

Sector coupling: Coordinated interaction of electricity with transport, heat, industry or fuels.

Stochastic programming: Optimization using scenarios with assigned probabilities.

Voltage stability: Ability to maintain acceptable bus voltages following a disturbance or load change.

Review Questions

Short Answer Questions

Distinguish renewable curtailment from unserved demand.
Why is LCOE insufficient for system-level resource planning?
What assumptions underlie the DC power-flow approximation?
Define ELCC and explain why it differs from nameplate capacity.
What physical constraint limits converter active and reactive current during faults?

Long Answer Questions

Explain the five-layer architecture of a sustainable power system and the bidirectional interactions among its layers.
Develop a network-constrained planning formulation including generation, renewables, storage, curtailment and load shedding.
Compare grid-following and grid-forming converter controls, including their behavior in weak networks.
Discuss voltage-stability assessment using AC power flow, Jacobian analysis and continuation power flow.
Critically evaluate the environmental, economic, social and cybersecurity challenges of large-scale power-system decarbonization.

Analytical / Application Questions

A 60 Hz, 2,000 MW system with an inertia constant of 4 s loses 150 MW. Calculate the initial RoCoF, neglecting damping and primary response.
Formulate the storage-energy and nodal-balance constraints for a renewable-powered microgrid in which demand excludes battery charging.
Design a study sequence for evaluating a proposed renewable-rich transmission expansion, progressing from investment optimization to AC and dynamic security validation.

Multiple Choice Questions with Answers

Question 1

A 50 Hz system has H=3.5\;s, S_base=1000\;MW and loses 120 MW of generation. Neglecting immediate damping, what is the initial RoCoF?

A. −0.429 Hz/s

B. −0.857 Hz/s

C. −1.714 Hz/s

D. +0.857 Hz/s

Correct Answer: B. −0.857 Hz/s

Question 2

A battery initially contains 40 MWh. It charges at 10 MW for 2 h with 90% efficiency and then discharges at 12 MW for 1.5 h with 90% efficiency. What is its final stored energy?

A. 36 MWh

B. 38 MWh

C. 40 MWh

D. 42 MWh

Correct Answer: B. 38 MWh

Question 3

When demand is defined independently of storage operation, how should battery charging appear in a nodal power balance?

A. As additional supply

B. As renewable curtailment

C. As additional demand

D. As unserved load

Correct Answer: C. As additional demand

Question 4

Which condition is commonly associated with a saddle-node voltage-collapse point in the modeled AC power-flow equations?

A. The power-flow Jacobian becomes singular

B. All line angles become zero

C. System frequency becomes exactly nominal

D. Active-power losses vanish

Correct Answer: A. The power-flow Jacobian becomes singular

Question 5

Which statement correctly distinguishes renewable curtailment from unserved demand?

A. Both represent consumer load that is disconnected

B. Curtailment is unused available generation; unserved demand is load not supplied

C. Curtailment is always caused by equipment failure

D. Unserved demand increases renewable availability

Correct Answer: B. Curtailment is unused available generation; unserved demand is load not supplied

Question 6

In the illustrative three-bus case, reinforcing the renewable export line reduces thermal generation by 30 MW. At $60/MWh, what is the hourly operating-cost saving?

A. $500/h

B. $1,200/h

C. $1,800/h

D. $3,600/h

Correct Answer: C. $1,800/h

Question 7

What does effective load-carrying capability measure?

A. The maximum instantaneous output of a generator

B. Additional load supportable while maintaining a specified adequacy criterion

C. The annual energy consumed by storage losses

D. The reactive-power rating of a transmission line

Correct Answer: B. Additional load supportable while maintaining a specified adequacy criterion

Question 8

Which feature most directly characterizes a grid-forming converter?

A. It must follow an externally measured voltage angle

B. It establishes an internal voltage angle and magnitude reference

C. It can operate without a current limit

D. It always contains a rotating mass

Correct Answer: B. It establishes an internal voltage angle and magnitude reference

Question 9

Why must large-scale DC planning results be checked using AC analysis?

A. DC models cannot represent investment costs

B. DC models cannot include renewable scenarios

C. DC models do not establish voltage and reactive-power feasibility

D. DC models necessarily prohibit transmission expansion

Correct Answer: C. DC models do not establish voltage and reactive-power feasibility

Question 10

Under the deterministic N−1 criterion, a planned system is tested for its ability to withstand which condition?

A. Simultaneous loss of every generator

B. Any designated single-element contingency

C. Only forecast errors in renewable output

D. A permanent doubling of demand

Correct Answer: B. Any designated single-element contingency

Electrical Power System
Chapter 15 of 15

Chapter 15 Sustainable and Future Power Systems