BESS Degradation Modeling: What It Is and Why It Matters

Battery degradation is the single largest variable in energy storage project economics. A 5% difference in assumed degradation rate can flip a project from a 15% IRR to a single-digit return — or from bankable to non-viable.

Yet many simulation tools still model battery degradation as a simple linear fade: "lose 2% per year, done." In reality, battery degradation is a complex, non-linear function of time, temperature, cycling patterns, depth of discharge, and charge/discharge rates. Getting it wrong means inaccurate financial projections, oversized augmentation budgets, and failed investor confidence.

This guide covers what BESS degradation modeling actually involves, the difference between calendar and cycle aging, how manufacturers provide degradation data, and why 3D interpolation matters for accurate project economics.

What Is State of Health (SOH)?

State of Health (SOH) is the ratio of a battery's current usable capacity to its original rated capacity at beginning of life (BOL). It's expressed as a decimal or percentage:

SOH = Current Usable Capacity / Rated Capacity

A new battery starts at SOH = 1.0 (100%). When SOH drops to 0.8 (80%), the battery is typically considered at end of life (EOL) for most utility and C&I applications. Some applications, like grid frequency regulation, may tolerate lower thresholds.

SOH is not directly measurable during operation — it must be estimated from capacity tests, coulomb counting, or modeled from degradation data. This is where degradation modeling becomes critical.

Calendar Aging vs Cycle Aging

Battery degradation occurs through two independent mechanisms that add together:

Calendar aging — Capacity loss that occurs regardless of whether the battery is cycling. It depends on:

• Time — Linear with calendar age
• Temperature — Accelerated by heat (Arrhenius relationship, roughly 2x degradation per 10°C)
• State of charge — Higher SOC accelerates degradation (storing at 100% SOC is worse than 50%)

Cycle aging — Capacity loss that accumulates with each charge/discharge cycle. It depends on:

• Number of cycles — More cycles = more degradation
• Depth of discharge (DoD) — Deeper cycles cause more wear per cycle
• C-rate — Faster charging/discharging accelerates degradation
• Temperature during cycling — Elevated temperatures worsen cycle aging

A battery in a solar PV firming application might cycle once per day at moderate C-rates, so calendar aging dominates. A battery in frequency regulation might cycle 20+ times per day at high C-rates, so cycle aging dominates. Modeling both correctly is essential.

Round-Trip Efficiency (RTE) Degradation

RTE degrades over time alongside SOH. A battery that starts at 92% DC RTE might drop to 88% over 10 years. This means:

• More energy is lost as heat during each charge/discharge cycle
• Revenue from energy arbitrage decreases year over year
• Augmentation planning must account for both energy capacity loss and efficiency loss

RTE degradation is often correlated with SOH but follows a different curve. Some chemistries (LFP) maintain RTE better than others (NMC) even as SOH declines. A proper model tracks both independently.

How Manufacturers Provide Degradation Data

Major battery manufacturers (CATL, BYD, Samsung SDI, LG, EVE Energy) provide degradation tables in their technical datasheets. These tables typically show SOH retention as a function of:

• Years of operation — at specific temperatures and SOC setpoints
• Cycles at specific DoD — e.g., "6000 cycles to 80% SOH at 25°C, 0.5C, 80% DoD"
• Combined aging curves — some provide both calendar and cycle components

Why Linear Degradation Assumptions Fail

Many tools (and spreadsheets) assume a constant annual degradation rate like 2% per year. This fails for three reasons:

• Non-linear shape — Real degradation follows a "bathtub curve" with faster initial fade, a stable middle period, and accelerated end-of-life fade
• Usage-dependent — A battery cycled 365 times per year degrades differently than one cycled 100 times per year. A linear model can't capture this
• C-rate sensitivity — High C-rate cycling causes disproportionate wear that a flat annual percentage misses entirely

The financial impact is material. A 25-year LCOE calculation using linear degradation can underestimate total energy throughput by 8-15% compared to a manufacturer-table-based model, depending on the cycling profile.

3D Interpolation: Year x C-Rate x Cycles/Day

The most accurate approach to battery degradation modeling uses 3D interpolation from real manufacturer degradation tables across three dimensions:

• Calendar year — time-dependent SOH fade
• C-rate — charge/discharge rate impact on cycle aging
• Cycles per day — cycling frequency effect

This creates a multi-dimensional degradation surface. At each hourly timestep in the simulation, the model reads the current year, the average C-rate over the past cycle, and the cumulative cycles to date, then interpolates SOH and RTE from the manufacturer's actual test data.

The result: a battery that cycles lightly once per day (e.g., solar PV shifting) might show 85% SOH after 15 years, while the same battery cycling aggressively four times per day (e.g., frequency regulation + arbitrage) might hit 80% SOH in 8 years. The difference is captured automatically by the model.

The Financial Impact of Accurate Modeling

To illustrate: consider a 100 MW / 200 MWh LFP battery paired with a 150 MW solar PV farm in California.

• Linear model (2%/yr): SOH = 50% at year 25, total energy throughput = 4.2 GWh over project life
• Manufacturer 3D interpolation: SOH = 74% at year 25, total energy throughput = 5.1 GWh

The linear model underestimates energy throughput by about 18%. This translates to a difference of roughly $0.012/kWh in LCOE and a 3-4 percentage point difference in IRR. For a $200M project, that's millions in valuation difference.

How Energy Optima Models Degradation

Energy Optima simulates battery degradation using manufacturer-specific degradation tables with 3D interpolation across year, C-rate, and cycling frequency. The platform includes 147+ battery models from manufacturers including CATL, BYD, Samsung SDI, LG Energy Solution, EVE Energy, and more — each with its own validated degradation data.

The simulation tracks:

• SOH fade year-by-year across the full project lifetime
• RTE degradation at AC bus and point of connection (POC)
• Battery augmentation planning with configurable SOH triggers
• FAT-to-COD pre-commissioning SOH loss
• Auxiliary power consumption (C-rate x temperature dependent)

The result: financial projections (NPV, IRR, LCOE, payback period) that reflect real battery behavior, not spreadsheet assumptions.