Battery Degradation and State of Health in Grid Storage

Battery degradation is the gradual loss of useful storage capability. In grid storage, it affects delivered energy, power capability, round-trip efficiency, reserve confidence, thermal margin, warranty compliance, and safety. State of health is the engineering estimate of how much capability remains relative to a defined reference condition.

The important point is that state of health is not a single universal number. A battery may retain most of its energy capacity while losing high-power capability because internal resistance has increased. Another battery may pass a short discharge test while no longer meeting a long-duration backup requirement at low temperature. Good storage engineering therefore treats state of health as a set of capability limits, not as a decorative dashboard percentage.

Principle

The useful principle is:

Battery dispatch consumes not only stored energy but also lifetime margin; state of health must therefore be part of the operating envelope.

A grid battery is not adequately described by beginning-of-life nameplate capacity. The design and operating review should state:

1. current available energy;
2. current power capability;
3. state-of-charge estimation confidence;
4. temperature and thermal derating limits;
5. cycle and calendar aging assumptions;
6. warranty limits for depth of discharge, throughput, C-rate, temperature, and dwell time;
7. evidence used to update the state-of-health estimate.

Without these details, a dispatch schedule can look feasible in an energy spreadsheet while consuming too much lifetime, violating reserve assumptions, or pushing the system outside its warranty envelope.

Capacity Fade

Capacity fade is the loss of charge or energy that the battery can store and deliver under specified conditions. A simple capacity-based state of health is:

where Q_{available} is measured or estimated available capacity and Q_{reference} is the reference capacity, often beginning-of-life rated capacity or a commissioning-test value.

For energy-based review:

The chosen reference must be stated. A battery at 88\% state of health relative to beginning-of-life nameplate capacity may have a different practical meaning than 88\% relative to commissioned usable energy at a specific temperature and discharge rate.

Capacity fade can result from active-material loss, lithium inventory loss, side reactions, electrolyte degradation, solid-electrolyte interphase growth, mechanical cracking, current-collector corrosion, cell imbalance, and repeated operation outside preferred conditions. The exact mechanisms depend on chemistry, cell design, manufacturing quality, thermal management, and duty cycle.

Resistance Growth and Power Capability

Degradation is not only capacity loss. Internal resistance often increases with age, cycling, temperature exposure, and cell imbalance. Higher resistance creates more voltage drop and heat at the same current.

A simplified ohmic loss estimate is:

where I is current and R is internal or path resistance. If resistance rises, the same power command may create more heat, more voltage sag, lower efficiency, and earlier current limiting.

For high-power grid services, resistance growth can be more restrictive than capacity fade. A battery may have enough stored energy for a service but fail to deliver the required active power at low SOC, low temperature, high temperature, or aged condition. This is why power tests, thermal tests, and inverter-current limits matter during acceptance and periodic validation.

Calendar Aging

Calendar aging occurs with time, even when the battery is not cycling. It is influenced by temperature, state of charge, chemistry, cell voltage, storage duration, and manufacturing condition.

For many battery systems, long dwell at high SOC and high temperature accelerates aging. This matters for grid storage because reserve services and backup services often ask the asset to remain ready for long periods. A battery held near full SOC for emergency support may age differently from a battery cycled daily through a moderate SOC window.

Calendar-aging review should ask:

• What SOC is required for readiness?
• How long does the asset dwell at high SOC?
• What temperatures occur during standby?
• Does the control system relax SOC when the service does not require full readiness?
• How is readiness balanced against long-term capacity retention?

The best operating policy is not always maximum SOC. It is the SOC policy that satisfies the service while preserving lifetime and safety margin.

Cycle Aging

Cycle aging is associated with charge and discharge. It depends on depth of discharge, C-rate, temperature, SOC range, dwell time between charge and discharge, charge termination, and rest periods.

A simple equivalent-full-cycle estimate is:

where E_{throughput} is the sum of charged and discharged energy at a defined boundary and E_{rated} is the reference energy capacity. This is only a rough accounting method. Two operating profiles with the same equivalent full cycles can produce different degradation if one uses high C-rate, deep cycling, high temperature, or long high-SOC dwell.

Cycle aging should be tied to the dispatch service. Frequency response may involve many shallow movements. Energy arbitrage may involve deeper daily cycles. Backup may involve rare deep events plus long standby. Renewable smoothing may create frequent partial cycles. Each service consumes lifetime differently.

Temperature and Thermal Coupling

Battery degradation is strongly coupled to temperature. High temperature can accelerate chemical side reactions and aging. Low temperature can reduce power capability, increase resistance, limit charge acceptance, and raise the risk of damaging operation if charging is not controlled properly.

Thermal management therefore affects state of health in two ways:

1. it controls immediate power and safety limits;
2. it changes the long-term degradation rate.

Thermal review should include ambient extremes, cooling-system failure, blocked filters, coolant degradation, fan or pump failure, rack imbalance, repeated high-power events, and emergency operation. A battery warranty or dispatch model is weak if it assumes a uniform cell temperature that the installed system cannot maintain.

SOC Window and Depth of Discharge

The selected SOC window is one of the most important operating decisions. A wide SOC window increases usable energy per event, but it can increase degradation and reduce reserve flexibility. A narrow window preserves margin but may require more installed capacity.

Usable energy for routine dispatch can be screened as:

where E_{current} reflects current degraded capacity, f_{reserve} protects nonroutine energy, f_T accounts for temperature derating, and f_{availability} accounts for unavailable blocks or maintenance assumptions.

This equation is deliberately conservative. It forces the engineer to separate current state of health from SOC limits and from reserve policy. A project that uses beginning-of-life nameplate energy for every dispatch calculation will overstate capability as the asset ages.

State-of-Health Estimation

State of health is estimated from evidence. Common inputs include capacity tests, coulomb counting, open-circuit voltage behavior, impedance measurements, voltage response under load, thermal behavior, cell imbalance, model residuals, historical throughput, temperature history, and manufacturer battery-management data.

No single measurement is perfect:

• capacity tests can be intrusive and may require controlled operating conditions;
• coulomb counting can drift;
• voltage-based estimates depend on chemistry and rest state;
• impedance indicators depend on temperature and SOC;
• model-based estimates are only as good as the model and calibration data;
• field data can be biased by incomplete or inconsistent dispatch records.

For engineering decisions, the estimate should include uncertainty. A battery with reported 25\ \text{MWh} available capacity and large uncertainty may require a larger reserve than one with the same estimate and strong validation evidence.

Dispatch with Degradation Constraints

Degradation-aware dispatch treats lifetime as a constrained resource. The controller should not optimize only short-term energy revenue, demand reduction, or service availability. It should also respect operating envelopes that preserve long-term capability.

Useful dispatch constraints include:

1. minimum and maximum SOC;
2. maximum charge and discharge power as a function of SOC and temperature;
3. protected reserve for emergency services;
4. maximum daily or monthly throughput;
5. maximum time at high SOC;
6. temperature limits and cooling availability;
7. C-rate limits;
8. warranty limits for cycles, throughput, and operating state;
9. state-of-health uncertainty margin.

The economic dispatch problem should include a degradation cost or lifetime constraint. Otherwise, the controller may accept every profitable-looking cycle while eroding future service capability.

Warranty Envelope

Battery warranties commonly define limits on energy throughput, cycle count, capacity retention, operating temperature, SOC window, C-rate, availability, maintenance, and data reporting. The warranty envelope is an engineering boundary, not only a commercial clause.

Operating outside the envelope can create several problems:

• measured capacity may fall below guaranteed levels;
• reserve assumptions may no longer be valid;
• performance guarantees may be voided;
• safety and thermal margins may shrink;
• root-cause analysis after an event becomes harder;
• dispatch profitability may be overstated.

The operating system should therefore record the variables that matter to the warranty. If temperature, SOC, throughput, alarms, and control overrides are not retained, the project may not be able to prove compliant operation.

Reliability and Safety Implications

Degradation can create safety and reliability risk. Capacity fade reduces service duration. Resistance growth increases heat. Cell imbalance can push weak cells closer to voltage limits. Sensor drift can hide abnormal behavior. Cooling degradation can turn a normal dispatch into a thermal event.

Failure-mode review should include:

• loss of emergency reserve because current capacity is lower than assumed;
• high-power dispatch rejected by battery management limits;
• accelerated heating during aged operation;
• underestimated recharge time after emergency discharge;
• SOC estimate drift after long standby;
• cell-string imbalance reducing usable capacity;
• protection trips during stacked services;
• warranty envelope breach during market dispatch.

The asset should fail predictably. Operators should know whether degraded state of health reduces duration, power, allowed SOC window, reserve commitment, or service priority.

Validation Evidence

State-of-health claims should be validated with measured evidence. Useful checks include:

Claim	Evidence
Available energy remains above requirement	Capacity test or reconciled dispatch event at the service boundary.
Power capability remains adequate	Power test at representative SOC, temperature, and aged condition.
SOC estimate is trustworthy	Reconciliation between metered energy, SOC change, and capacity test results.
Degradation model is credible	Updated model residuals, measured capacity trend, and temperature-throughput history.
Warranty envelope is respected	Logged SOC, temperature, C-rate, throughput, alarms, and operating modes.
Reserve is protected	Dispatch records showing post-event stored energy above reserve threshold.

Validation should be repeated because state of health changes. A commissioning test proves initial capability, not lifetime capability.

Practical Engineering Workflow

A practical battery-degradation workflow is:

1. Define the required service and measurement boundary.
2. State beginning-of-life capacity, current capacity basis, and end-of-life requirement.
3. Define SOC window, reserve, temperature range, and charge/discharge limits.
4. Estimate calendar and cycle aging for the intended duty cycle.
5. Include state-of-health uncertainty in sizing and dispatch rules.
6. Track energy throughput, temperature, SOC dwell, alarms, and control overrides.
7. Reconcile dispatch events with metered energy and SOC change.
8. Update the degradation model from measured capacity and power tests.
9. Revise dispatch limits when state of health changes.

This workflow keeps degradation tied to decisions. It avoids treating state of health as an after-the-fact report that arrives only when the battery no longer meets its service.

Common Mistakes

A common mistake is using one state-of-health percentage for every decision. Capacity state of health, power capability, resistance, thermal margin, and safety state are related but not identical.

Another mistake is assuming that shallow cycles are always harmless. Frequent shallow cycling, high C-rate corrections, high temperature, or long high-SOC dwell can still consume lifetime. The relevant question is not whether a cycle is small, but whether the full duty pattern fits the validated degradation model.

A deeper mistake is dispatching against beginning-of-life assumptions. The grid service is delivered by the battery that exists today, not by the battery specified in the procurement table. State of health must be updated, validated, and reflected in operating limits.

Battery degradation is manageable when it is measured, modelled, and built into dispatch. It becomes a project risk when it is hidden behind nameplate capacity, optimistic revenue schedules, or unverified dashboard values.