Hybrid Renewable Systems, Microgrids and Resilience

Introduction

Electricity systems based on a single generating technology are increasingly being supplemented by hybrid renewable systems. A hybrid system combines two or more energy resources—such as photovoltaic (PV), wind, batteries, diesel generators, and flexible loads—with power-conversion equipment and supervisory control. When these resources serve a defined electrical area and can operate independently of the utility, the arrangement is commonly called a microgrid.

Hybrid systems are valuable because renewable resources are variable, demand changes with time, and remote communities may have weak or no grid access. Storage, dispatchable generation, demand response, and intelligent controls can coordinate these differences. The engineering task is not simply to install the largest possible renewable capacity. It is to select and operate equipment so that reliability, cost, emissions, power quality, safety, and resilience are jointly acceptable.

A microgrid may operate grid-connected, islanded, or transition between both modes. Its design must therefore address energy balance, voltage and frequency regulation, protection, communications, black start, fuel logistics, cybersecurity, and restoration after disturbances.

Learning objectives

After completing this chapter, the reader should be able to:

1. Explain the architecture and operating principles of AC, DC, and hybrid AC/DC microgrids.
2. Model PV, wind, storage, diesel generation, and controllable loads at an appropriate planning level.
3. Formulate energy-management and capacity-optimization problems.
4. Distinguish grid-connected, islanded, and resynchronization operation.
5. Evaluate reliability, resilience, economics, emissions, and rural-electrification performance.
6. Explain how sector coupling and digital control expand microgrid flexibility.
7. Identify practical, ethical, regulatory, and cybersecurity requirements.

1. Hybrid renewable system architecture

A hybrid renewable system contains four functional layers:

• Sources: PV arrays, wind turbines, small hydro, biomass, or diesel generators.
• Flexibility resources: batteries, thermal storage, pumped storage, hydrogen systems, and controllable loads.
• Power conversion and distribution: inverters, rectifiers, transformers, switchgear, protection, and conductors.
• Supervisory control: measurement, forecasting, optimization, dispatch, alarms, and human interfaces.

A simple planning-period power balance is

where all terms are powers in kW at time interval t. The subscripts PV, W, D, L, imp, and exp denote photovoltaic generation, wind generation, diesel generation, load, grid import, grid export, and curtailed renewable power. Positive battery terms are defined here as discharge P_dis and charging P_ch is placed on the demand side. This equation assumes that losses have either been included in the individual terms or represented separately. It is an instantaneous or interval energy-balance model, not a complete power-quality model.

Energy over an interval of duration ∆t hours is E = P∆t, measured in kWh. For variable resources, the interval should be short enough to represent ramps and constraints. Hourly models are useful for preliminary planning; minute-level or sub-second models are needed for inverter controls, frequency stability, and protection.

PV and wind representation

A planning model may estimate PV output as

where P_PV,r is rated PV power in kW and 0 ≤ f_PV(t) ≤ 1 is a normalized weather- and temperature-dependent availability factor. A more detailed model uses irradiance and temperature, but the normalized form is often adequate for capacity studies.

Wind power depends strongly on wind speed. A turbine power curve can be represented by

where v is wind speed, v_ci is cut-in speed, v_r is rated speed, v_co is cut-out speed, and P_r is rated power. The middle expression is an approximation; actual turbine curves are supplied by manufacturers and include control and air-density effects.

The capacity factor of a generator is

where E_annual is annual electrical energy and P_r is rated power. Capacity factor is not the fraction of time a generator operates at rated power; it is an energy-normalized measure.

Battery model

A simple battery state-of-charge model is

where SOC is expressed as a fraction from 0 to 1, η_ch and η_dis are charging and discharging efficiencies, E_nom is nominal battery energy capacity in kWh, and powers are in kW. The model assumes constant efficiencies and neglects temperature, rate, ageing, and reserve effects. Practical operation imposes

and limits on charging and discharging power. Simultaneous charging and discharging should normally be prohibited through controller logic or binary variables in an optimization model.

The usable energy between specified limits is approximately

This is not the same as lifetime throughput. Battery degradation depends on temperature, time, depth of discharge, current, and chemistry, so a life-cycle model should include replacement or degradation cost.

Diesel backup

Diesel generators are dispatchable but produce fuel cost, emissions, noise, and maintenance requirements. A common planning approximation is

where F_D is fuel use, P_D is generator output, and a and b are empirically determined coefficients. Units must be stated by the modeler, such as litres per hour and kW. A minimum loading constraint may be required because prolonged operation at very low load can cause poor efficiency and wet stacking. Start-up time, minimum up/down time, ramp rate, and reserve capability should also be represented when operational decisions are important.

2. AC, DC, and hybrid microgrids

AC microgrids

AC microgrids distribute power through an alternating-current bus, commonly at low voltage. They integrate naturally with conventional motors, transformers, protection systems, and utility interconnection equipment. Their challenges include synchronization, reactive-power management, frequency control, and fault coordination. Inverters may operate in grid-following mode when a stiff voltage source exists, or grid-forming mode when they must establish voltage and frequency.

DC microgrids

DC microgrids are attractive where many resources and loads are inherently DC: PV modules, batteries, data equipment, LED lighting, electric vehicles, and electronic converters. Avoiding repeated AC/DC conversion can improve efficiency. However, DC fault interruption, arc suppression, grounding, voltage standardization, and protection coordination require specialist design. DC systems do not have AC frequency, but converter controls must regulate voltage and share current safely.

Hybrid AC/DC microgrids

A hybrid microgrid contains interconnected AC and DC subgrids linked by a bidirectional interlinking converter. PV and batteries may connect to the DC bus, while motors and legacy loads connect to the AC bus. The converter exchanges active power and can support voltage and frequency. Its rating must account for expected transfers, overload duration, fault behavior, and islanding transitions.

3. Energy-management strategies

The energy-management system (EMS) operates above fast converter controls. Its functions include forecasting, scheduling, reserve allocation, dispatch, tariff response, fault handling, and reporting.

A rule-based strategy may follow this sequence:

1. Serve critical loads.
2. Use available renewable generation.
3. Charge storage when renewable power is surplus, subject to SOC limits.
4. Discharge storage during deficits, retaining a reserve.
5. Start diesel generation when storage cannot maintain reserve or power limits.
6. Shed or shift noncritical loads only according to agreed priorities.
7. Import or export through the utility connection when permitted.

Rule-based control is transparent and robust, but it may not minimize total cost under complex tariffs or forecasts. Optimization-based control can use linear programming, mixed-integer programming, dynamic programming, or model-predictive control.

A representative objective over periods t is

Here J is total modeled cost; c terms are monetary coefficients; y_D is a diesel start indicator; C_cap and C_rep represent annualized capacity and replacement costs. The optimization is subject to power balance, equipment limits, SOC equations, minimum generator loading, reserve requirements, and network constraints. A cost-minimizing solution is not automatically socially optimal: critical services, equity, environmental impacts, and resilience may require constraints or additional objective terms.

Forecast uncertainty can be addressed by reserve margins, scenario-based stochastic optimization, or robust optimization. Model-predictive control repeatedly solves a short future horizon, implements only the first control action, then updates the forecast. This reduces sensitivity to forecast error but requires reliable measurement, communications, and computational resources.

4. Grid-connected and islanded operation

In grid-connected operation, the utility often establishes voltage and frequency. The microgrid may exchange real and reactive power, reduce peak demand, provide ancillary services, or operate under a tariff. Anti-islanding protection is required so that an unintentional island does not energize a supposedly de-energized utility line.

In islanded operation, local resources must establish and maintain electrical conditions. Grid-forming inverters can set voltage magnitude, frequency, and phase reference. Droop control is a common decentralized method:

where f and V are controlled frequency and voltage, f₀ and V₀ are references, P and Q are measured active and reactive power, and m_P and n_Q are droop coefficients. Droop permits multiple sources to share load without high-speed communication, but it allows steady-state deviations and requires secondary control for restoration.

A planned transition to islanding should include detection, separation, load and generation assessment, grid-forming mode activation, reserve verification, and stable reconnection. Resynchronization requires acceptable differences in voltage magnitude, frequency, and phase angle before the breaker closes. Uncontrolled reconnection can produce severe currents and mechanical stress.

5. Reliability and resilience

Reliability generally concerns the probability of adequate supply under expected component failures and operating conditions. Resilience is broader: it includes preparation, resistance, adaptation, continuity of critical functions, and rapid recovery after high-impact disturbances such as storms, floods, fires, cyberattacks, or fuel-supply interruption.

Common reliability indicators include:

where N_i is the number of customers affected by event i, U_i is interruption duration, λ_i is interruption frequency, and N_T is total customers served. SAIDI has units of time per customer per period; SAIFI has interruptions per customer per period. These indices describe service performance but do not fully capture the importance of hospitals, water systems, or communications facilities.

A resilience-oriented design may include:

• critical-load segmentation and prioritized shedding;
• islandable feeders and black-start capability;
• redundant communication and control paths;
• spare inverters, transformers, and protection devices;
• elevated or hardened equipment in hazard zones;
• fuel diversity and minimum fuel reserves;
• cyber access control, logging, backups, and incident procedures;
• drills and restoration plans involving operators and emergency agencies.

A useful planning metric is expected energy not served (EENS), the expected unserved energy over specified failure scenarios. It is calculated by weighting the unserved energy in each scenario by its probability. The value of resilience is context-dependent: one kWh of avoided interruption may be far more valuable at a medical clinic than at a discretionary load.

6. Sizing and optimization considerations

Sizing should begin with measured or carefully estimated load profiles. Aggregate annual energy is insufficient because peak demand, evening ramps, motor starting, seasonal variation, and critical-load requirements determine equipment ratings.

A practical workflow is:

1. Define service territory, voltage levels, reliability target, critical loads, and operating modes.
2. Collect load, solar, wind, temperature, fuel, tariff, and outage data.
3. Select a time resolution and planning horizon.
4. Develop resource and component models with documented assumptions.
5. Optimize candidate capacities and dispatch.
6. Test extreme weather, forecast errors, equipment outages, and fuel constraints.
7. Check power flow, short circuit, harmonics, protection, and stability in detailed software.
8. Compare life-cycle cost, emissions, resilience, and social outcomes.
9. Validate the design through commissioning tests and staged islanding trials.

The levelized cost of electricity is useful for comparing energy technologies, but it can mislead when storage, network reinforcement, backup, or reliability value differs among alternatives. Net present cost is often more appropriate for a microgrid because it includes capital, replacement, fuel, operation, maintenance, and residual values over a planning horizon.

Worked example: preliminary rural microgrid dispatch

Consider an illustrative one-hour interval. A village has a 120 kW load. PV produces 70 kW and wind produces 20 kW. The battery can discharge at up to 40 kW and has sufficient SOC. A diesel generator can provide the remainder.

The renewable supply is

P_RE = 70 + 20 = 90 kW.

The deficit is

P_deficit = 120 − 90 = 30 kW.

Because the battery can discharge 40 kW, the EMS supplies the entire deficit from the battery. Diesel output is therefore 0 kW, assuming no minimum-generation or reserve constraint requires it to run. Energy delivered by the battery during the hour is 30 kWh. If discharge efficiency is 90%, the stored-energy reduction is

30 ÷ 0.90 ≈ 33.3 kWh.

If the battery has 200 kWh nominal capacity and SOC decreases from 80%, its new SOC, neglecting charging and other constraints, is approximately

0.80 − 33.3 ÷ 200 = 0.6335,

or 63.4%. If a 70% minimum reserve had been required, the battery could not have supplied the full deficit; the diesel would need to contribute at least the amount necessary to preserve that reserve. This example demonstrates why dispatch decisions depend not only on instantaneous power but also on energy capacity and reserve policy.

7. Rural electrification and sector coupling

For rural electrification, the design must reflect local livelihoods and payment ability. Productive loads—irrigation, refrigeration, milling, welding, telecommunications, and small industry—can improve economic sustainability but may create large motor-starting currents or evening peaks. Tariff design should be transparent and avoid excluding low-income households. Community participation is important for siting, maintenance, payment collection, and conflict resolution.

Sector coupling links electricity to other energy services. Examples include:

• electric cooking replacing biomass or LPG;
• water pumping scheduled during renewable production;
• thermal storage for refrigeration or water heating;
• electric vehicles as flexible loads or, where permitted, distributed storage;
• electrolyzers producing hydrogen when surplus electricity is available;
• desalination and wastewater treatment coordinated with renewable output.

Sector coupling can reduce curtailment and increase useful energy, but it may increase demand, conversion losses, safety requirements, and operational complexity. Hydrogen systems require particular attention to ventilation, leakage detection, materials, pressure systems, and applicable codes.

8. Case study: illustrative islandable health-centre microgrid

Background

An isolated health centre serves a population of approximately 4,000 people. Its refrigeration, lighting, communications, and medical equipment require continuous electricity. A diesel generator currently operates many hours each day, but fuel delivery is unreliable during the rainy season.

Problem

The centre needs reliable power, lower fuel consumption, and the ability to maintain critical services during several cloudy days. The design must also tolerate failure of a single major component.

Analysis

A load audit separates critical loads—vaccines, communications, emergency lighting, and selected medical equipment—from deferrable loads such as water pumping and laundry. Solar measurements indicate strong daytime resource but seasonal variation. The design team models hourly weather and load data, battery SOC, diesel minimum loading, fuel availability, and a critical-load reserve. It evaluates normal operation, prolonged cloud, inverter outage, battery outage, and communications failure.

Approach and solution

The selected architecture is an AC microgrid with PV inverters, a bidirectional battery inverter, an appropriately sized diesel generator, automatic transfer and isolation equipment, and a manual local control fallback. The battery inverter is grid-forming during islanded operation. PV is curtailed when the battery is full and demand is low. Water pumping is shifted to high-solar periods. The diesel generator starts when SOC reaches its reserve threshold or when predicted renewable power is inadequate. Critical loads are placed on a protected feeder; noncritical loads are shed first.

The EMS records energy, fuel, outages, alarms, and maintenance status. Operators receive training in safe isolation, manual generator operation, and recovery from communications loss. Spare fuses, inverter components, and battery monitoring equipment are stored locally.

Outcome

In the planning simulation, renewable energy supplies most annual energy, while diesel remains essential for prolonged low-resource periods and emergency reserve. The key benefit is not merely fuel reduction: the protected feeder maintains critical services during grid or generator interruptions. Final performance must be verified through commissioning, field measurements, and post-installation evaluation rather than assumed from simulation alone.

Learning points

1. Critical-load definition is as important as generation sizing.
2. Storage power and storage energy are separate design constraints.
3. Resilience requires procedures, spares, training, and communications—not only hardware.
4. Local operators should be involved from design through maintenance.

9. Challenges, governance, and contemporary practice

Hybrid microgrids face uncertainty in weather, demand, equipment ageing, tariffs, and policy. Inverters can interact through control loops, producing instability or harmonics if models and settings are poorly coordinated. Protection designed for one-way fault current may fail when inverter-based sources contribute limited or controlled fault current.

Interconnection rules should specify voltage and frequency limits, anti-islanding, reconnection, power quality, grounding, metering, and protection responsibilities. Ownership and dispatch authority must be clear when a microgrid contains utility, private, community, and public assets. Data from smart meters and health or household services should be collected proportionately, protected, and used with informed governance.

Artificial intelligence can improve short-term forecasting, anomaly detection, predictive maintenance, and demand classification. It should not replace engineering validation. AI models may fail under unusual weather or equipment conditions, and opaque decisions are inappropriate where safety or equitable service is involved. Human override, secure data pipelines, model monitoring, and fail-safe fallback controls are essential.

Best practice is to use traceable assumptions, calibrated component models, independent protection review, hardware-in-the-loop or staged testing where feasible, and clear acceptance criteria. Designers should distinguish simulation results from measured field performance and should report uncertainty rather than presenting a single forecast as fact.

Chapter Summary

Hybrid renewable systems coordinate variable renewable sources, storage, dispatchable generators, and flexible loads. Their fundamental planning requirement is an energy and power balance subject to equipment, reserve, operational, and network constraints. PV and wind are modeled from resource profiles or power curves; batteries require SOC, efficiency, power, energy, and degradation considerations; diesel generators provide dispatchability but incur fuel, emissions, and maintenance costs.

AC microgrids integrate conventional equipment effectively, DC microgrids suit DC-native sources and loads, and hybrid AC/DC systems combine both through an interlinking converter. Grid-connected systems rely partly on the utility reference, whereas islanded systems require grid-forming capability, reserve, coordinated controls, and safe reconnection. EMS strategies range from transparent rules to optimization and model-predictive control.

Reliability measures expected service performance, while resilience addresses preparation, continuity, adaptation, and recovery from severe disturbances. Rural microgrids must consider productive use, affordability, local governance, critical services, and maintenance capability. Sector coupling adds flexible services such as pumping, thermal storage, electric mobility, and hydrogen, but also introduces new safety and control requirements. Sound practice combines optimization with detailed electrical studies, cybersecurity, regulatory compliance, community participation, and field validation.

Key Terms

Hybrid renewable system: An energy system combining multiple generation or flexibility technologies.

Microgrid: A bounded electrical system containing loads and distributed resources that can operate connected to or separate from a utility.

Islanded operation: Operation electrically separated from the main grid.

Grid-forming inverter: A converter that establishes or regulates voltage and frequency references.

Grid-following inverter: A converter that synchronizes to an existing voltage and frequency waveform.

Energy-management system: Supervisory system that monitors, schedules, and dispatches resources.

State of charge: Available battery energy expressed relative to a specified reference capacity.

Reserve: Capacity or energy intentionally held to respond to uncertainty or contingencies.

Demand response: Deliberate change in electricity use in response to operational or economic signals.

Curtailment: Reduction of available generation because it cannot be used or exported.

Capacity factor: Annual energy divided by rated power multiplied by annual hours.

Droop control: Decentralized control that changes frequency or voltage reference with power flow.

Black start: Restoration of a power system without relying on an external energized supply.

Resynchronization: Controlled matching of voltage, frequency, and phase before reconnection.

Reliability: Ability to provide adequate service under defined expected conditions.

Resilience: Ability to prepare for, withstand, adapt to, and recover from disturbances.

Sector coupling: Coordination of electricity with transport, heating, water, fuels, or industrial services.

Critical load: Load whose interruption has unacceptable safety, health, or societal consequences.

Expected energy not served: Probability-weighted unserved electrical energy across failure scenarios.

Review Questions

Short Answer Questions

1. What are the four functional layers of a hybrid renewable system?
2. Why must a battery model distinguish power capacity from energy capacity?
3. State one important difference between AC and DC microgrids.
4. What is the purpose of grid-forming control during islanded operation?
5. Define resilience in the context of a microgrid.

Long Answer Questions

1. Explain how PV, wind, batteries, diesel generation, and controllable loads can be coordinated by an EMS.
2. Compare AC, DC, and hybrid AC/DC microgrids in terms of applications, conversion, control, and protection.
3. Discuss the assumptions and limitations of an hourly hybrid-microgrid optimization model.
4. Explain the sequence and engineering requirements for a safe transition from grid-connected to islanded operation and back.
5. Discuss why rural electrification projects require social, economic, and institutional analysis in addition to electrical design.

Analytical / Application Questions

1. A load is 180 kW, PV supplies 100 kW, wind supplies 35 kW, and a battery can discharge 30 kW. Determine the minimum diesel output if no load shedding is allowed. State the additional information needed to verify whether the battery can sustain this output.
2. Design an EMS priority sequence for a health-centre microgrid with vaccine refrigeration, lighting, water pumping, and laundry. Justify the reserve and load-shedding decisions.
3. Propose a resilience assessment for a coastal microgrid exposed to storms, fuel-delivery interruption, and communications failure. Identify scenarios, metrics, equipment measures, and nontechnical measures.

Multiple Choice Questions with Answers

Question 1

Which resource primarily provides dispatchable backup in the hybrid system described?

A. PV array

B. Wind turbine

C. Diesel generator

D. Uncontrolled load

Correct Answer: C. Diesel generator

Question 2

In the battery model, what does η_dis represent?

A. Discharging efficiency

B. Nominal battery voltage

C. Maximum discharge power

D. State-of-charge limit

Correct Answer: A. Discharging efficiency

Question 3

A microgrid operating without the utility grid generally requires which capability to establish voltage and frequency?

A. Grid-forming inverter control

B. Higher PV temperature

C. Passive anti-islanding only

D. A larger energy meter

Correct Answer: A. Grid-forming inverter control

Question 4

If a load is 120 kW and renewable generation is 90 kW, the deficit before storage or diesel contribution is:

A. 30 kW

B. 90 kW

C. 120 kW

D. 210 kW

Correct Answer: A. 30 kW

Question 5

Which statement correctly describes capacity factor?

A. The fraction of customers interrupted annually

B. Annual energy divided by rated power times annual hours

C. The battery’s usable energy divided by nominal voltage

D. The ratio of reactive power to apparent power

Correct Answer: B. Annual energy divided by rated power times annual hours

Question 6

What is the main purpose of retaining battery reserve in an islanded microgrid?

A. To increase curtailment

B. To respond to uncertainty and contingencies

C. To eliminate all diesel maintenance

D. To prevent any renewable generation

Correct Answer: B. To respond to uncertainty and contingencies

Question 7

Which load would normally receive the highest priority in the illustrative health-centre case?

A. Laundry

B. Decorative lighting

C. Vaccine refrigeration

D. Optional workshop equipment

Correct Answer: C. Vaccine refrigeration

Question 8

What must be matched before controlled reconnection of an islanded microgrid?

A. Fuel colour and battery chemistry

B. Voltage magnitude, frequency, and phase

C. Annual energy and capacity factor only

D. PV module temperature and wind speed

Correct Answer: B. Voltage magnitude, frequency, and phase

Question 9

A principal advantage of sector coupling is that it can:

A. Make all conversion losses disappear

B. Provide flexible uses for surplus electricity

C. Remove the need for protection systems

D. Guarantee zero equipment failures

Correct Answer: B. Provide flexible uses for surplus electricity

Question 10

Which metric measures average interruption duration per customer over a period?

A. SAIDI

B. SAIFI

C. SOC

D. EENS only

Correct Answer: A. SAIDI