- Cycle and calendar aging depend on operating conditions; one universal €/MWh degradation cost is not physically exact.
- A useful optimizer compares the marginal value of an action with its incremental lifetime cost and future opportunity cost.
- Warranty compliance is a constraint, not a complete economic model; augmentation and end-of-life value also matter.
Battery aging has more than one cause
Calendar aging occurs as time passes, influenced by chemistry, temperature and state of charge even when the battery is idle. Cycle aging is associated with charge/discharge use and depends on throughput, depth of discharge, C-rate, temperature, SOC window and the shape of the duty cycle. The mechanisms interact.
NREL notes that lifetime models need to consider operating windows, charge/discharge rates, storage environment and cycling patterns. That is why “one cycle costs X” is an approximation. Two equivalent-full-cycle totals can produce different aging when one involves shallow operation near moderate SOC and the other repeated deep cycles or high-temperature, high-rate events.
The marginal cycle decision
A simple optimizer can convert expected lifetime consumption into a variable cost:
For a throughput approximation:
This prevents a battery from cycling through small spreads that look positive before lifetime cost. But degradation is not the only opportunity cost. Discharging now can remove SOC from a higher-value later interval or a reserve obligation. The optimizer needs both lifetime and intertemporal value.
Three practical modeling levels
| Approach | Strength | Limitation |
|---|---|---|
| Flat throughput cost | Transparent, fast and easy to use in linear optimization. | Can miss SOC, temperature, depth and rate dependence. |
| Piecewise-linear cost | Represents different depths, regions or marginal wear while remaining solver-friendly. | Requires calibration and careful accounting to avoid double counting. |
| Physics/empirical health model | Can reflect chemistry and duty-cycle interaction more closely. | Data-intensive; may be too complex to place directly inside every dispatch solve. |
A common production pattern uses a tractable cost inside the optimizer and a more detailed offline model to recalibrate it. Dispatch history and temperatures are periodically replayed through the health model; the marginal cost or operating envelope is then updated.
A model should distinguish energy throughput from cycle count. Equivalent full cycles are useful reporting shorthand, but a rainflow or depth-aware method may better characterize variable cycling. The selected method should align with the warranty and financial model.
Warranty is a boundary, not the full answer
OEM warranties may constrain years, throughput, equivalent full cycles, SOC, temperature, availability, retained capacity or approved use. Operating inside those limits reduces contractual risk, but it does not prove the chosen dispatch maximizes project value.
The economic value of degradation also depends on augmentation. If modules will be added in year seven, incremental early fade can accelerate cash outflow. If project revenue falls later because spreads cannibalize, preserving capacity may be less valuable than assumed. End-of-life, salvage, capacity obligations and financing covenants can all change the marginal cost.
Degradation-aware optimization needs operational feedback
The optimizer should consume available power and energy rather than infer them solely from nameplate capacity. BMS-reported limits, validated capacity tests, temperature and operating mode update the feasible envelope. If measured round-trip efficiency or auxiliary load changes, the threshold spread changes too.
Commands still pass through BMS and PCS protections. The optimizer may avoid an economically weak high-C-rate action, but it does not replace cell-level protection. EMS and edge controls enforce the approved schedule inside current site limits, while monitoring systems record the duty cycle needed for health and warranty evidence.
Questions for an optimization or revenue study
- Which chemistry, OEM warranty and usable AC envelope are modeled?
- Does cost vary with throughput, depth, SOC, temperature or C-rate?
- How is calendar aging represented when the battery is idle?
- Are capacity fade, efficiency fade and power fade distinguished where material?
- How do augmentation and replacement timing affect marginal value?
- Does the backtest carry health forward instead of resetting the battery each year?
- Are maintenance, downtime and BMS derates included?
- Can the model explain why it rejected a positive gross spread?
For investment decisions, run several degradation assumptions and report both gross trading contribution and net contribution after lifetime cost. Precision without calibration is false confidence; transparent sensitivities are more useful.
Worked example: when a positive spread should be rejected
Illustrative calculation. Assume a complete charge/discharge action buys 1 MWh and later returns 0.88 MWh after losses. If the purchase price is €50/MWh and sale price is €70/MWh, gross cash flow is €61.60 − €50 = €11.60 before fees, auxiliaries and degradation. A €20 headline spread is therefore not €20 of margin.
If transaction and auxiliary costs are €2 and the modeled incremental lifetime cost is €8, only €1.60 remains before imbalance and opportunity cost. The optimizer should reject the action if keeping SOC headroom has more than €1.60 of expected future value or if uncertainty makes the downside unacceptable.
The numbers are not a benchmark. They show why efficiency must be applied to energy quantities and why degradation belongs beside fees. Different efficiency boundaries—DC rack, PCS AC terminal or revenue meter—will change the calculation.
Separate state of health from available capability
State of health is an estimate, often tied to retained capacity or resistance. Available power and energy are current operating limits. A battery at 90% SOH can be temporarily derated by temperature; a battery with lower capacity may still deliver rated power for a shorter duration. The optimizer should use current limits and retain SOH for planning and degradation calibration.
Validate the degradation model
Reconcile modeled and measured throughput, temperatures, capacity tests and BMS limit history. Compare predicted fade with observed fade over sufficiently long periods. Track model version and uncertainty. If the warranty method and internal economic model differ, maintain a bridge between them so operating teams can explain both compliance and value.
Frequently asked questions
What is battery degradation-aware optimization?
It is dispatch optimization that includes incremental lifetime consumption and health constraints alongside market revenue, losses and operational limits.
How much does one BESS cycle cost?
There is no universal number. It depends on installed and replacement economics, warranty, chemistry, depth, rate, temperature, SOC, augmentation and the definition of a cycle.
Is throughput a good degradation model?
It is a useful transparent approximation, especially for optimization, but may need piecewise or offline correction for depth, SOC, temperature and C-rate effects.
Should a battery avoid all degradation?
No. Degradation is often the cost of earning revenue or delivering a service. The objective is to consume lifetime where expected value exceeds that cost and associated risks.
Does the BMS perform degradation optimization?
The BMS estimates health and enforces equipment limits. A commercial optimizer uses those limits and an economic model to decide whether an allowed action is worthwhile.
Sources and methodology
Sources were reviewed on 7 September 2026. Rules, participation requirements and product specifications can change; verify the current documents for the relevant asset and market before making a commercial decision.
- NREL — Battery lifespan research — Lifetime drivers and health-aware controls.
- NREL — Predicting battery capacity under varied conditions — SOC, temperature, depth of discharge, duty cycle and C-rate dataset.
- IEA — Batteries and Secure Energy Transitions — Storage technology and energy-system context.
Test how degradation changes the schedule.
Compare throughput, warranty and augmentation assumptions against the market strategy and actual operating envelope.
This article is technical and commercial information, not legal, regulatory, investment or trading advice.