To size a battery for time-of-use (TOU) optimization, calculate the total energy load you need to shift out of peak pricing windows, then divide that figure by the battery’s usable depth of discharge to find the minimum required capacity. A well-sized TOU battery covers your peak-hour consumption without over-investing in storage you will rarely use. The sections below walk through every factor that shapes that calculation, from load profiling to solar integration to cost-effectiveness thresholds.
What loads and usage windows actually drive TOU battery sizing?
The loads that drive TOU battery sizing are the high-draw appliances and systems that operate during your utility’s peak pricing hours, typically late afternoon through early evening. Identifying which loads fall inside those windows, and how much energy they consume per hour, is the starting point for any accurate battery sizing calculation.
Begin by pulling interval data from your utility meter or energy monitoring system. You want at least 30 days of 15-minute or hourly consumption data so you can see real demand patterns rather than averages. From that data, isolate the hours your utility classifies as peak or on-peak, and sum the energy consumed during those windows on a typical weekday.
The loads that matter most are those that are both large and difficult to shift in time. HVAC systems, industrial process equipment, EV chargers, and commercial refrigeration all tend to dominate peak consumption. Lighting and small plug loads contribute less but still add up across a large facility. Loads that can be rescheduled to off-peak hours through demand management controls should be subtracted from your battery sizing target, since the battery only needs to cover what cannot be moved.
Peak pricing windows vary significantly by utility tariff. Some utilities define peak as a four-hour window; others stretch it to eight hours or apply different rates on weekends. Always size against the longest and most expensive window your tariff includes, because that is the scenario that determines whether the investment pays off.
How do you calculate the usable capacity needed for TOU shifting?
To calculate usable battery capacity for TOU shifting, multiply the average hourly load during peak windows by the number of peak hours, then divide by the battery’s usable depth of discharge (DoD). This gives you the minimum nameplate capacity required to cover your peak-hour energy demand without over-discharging the battery.
The formula looks like this:
- Determine your average peak-hour load in kilowatts (kW)
- Multiply by the number of peak hours to get peak energy demand in kilowatt-hours (kWh)
- Divide by the usable DoD of the battery chemistry you are specifying (typically 0.80 to 0.90 for lithium iron phosphate)
- Add a buffer of 10 to 15 percent for efficiency losses during charge and discharge cycles
For example, if your facility draws an average of 50 kW during a four-hour peak window, your peak energy demand is 200 kWh. With a lithium battery at 90 percent usable DoD and a 10 percent efficiency buffer, you need roughly 245 kWh of nameplate capacity to fully cover that window.
Round-trip efficiency also matters. Most lithium battery systems operate at 90 to 95 percent round-trip efficiency, meaning some energy is lost in the charge and discharge process. Factor this into your calculation by dividing the required output energy by the round-trip efficiency before sizing the system. Ignoring this step leads to undersized batteries that fall short during real-world operation.
What’s the difference between sizing for TOU versus sizing for backup power?
Sizing for TOU optimization and sizing for backup power are fundamentally different objectives. TOU sizing is about shifting a predictable volume of energy from peak to off-peak periods on a daily cycle. Backup sizing is about sustaining critical loads for an unpredictable duration during a grid outage. The two goals lead to very different capacity requirements and discharge strategies.
TOU sizing: daily cycling around a known schedule
A TOU battery charges during off-peak hours when electricity is cheap, then discharges during peak hours to avoid high-rate consumption. The sizing target is the energy volume needed to cover peak windows, and the battery is expected to complete one full charge-discharge cycle per day. Because the schedule is predictable, you can size precisely to your load profile without large safety margins. The battery does not need to hold reserve capacity for emergencies.
Backup sizing: duration and critical load coverage
A backup battery must sustain a defined set of critical loads for a minimum number of hours or days without grid support. Sizing depends on which loads are deemed critical, how long the outage scenario lasts, and whether any on-site generation like solar can recharge the battery during the event. Backup systems are typically sized with significant reserve margins because the duration of an outage is unknown. This often results in larger nameplate capacity than a pure TOU system would require.
When a project needs both TOU optimization and backup capability, the battery must be sized to satisfy the more demanding of the two requirements, which is almost always the backup scenario. The TOU function then operates within whatever capacity the backup reserve does not consume.
How does solar generation affect TOU battery sizing?
Solar generation directly reduces the battery capacity needed for TOU shifting by charging the battery during daylight hours at zero marginal cost, reducing or eliminating the need to charge from the grid during off-peak periods. In many commercial and industrial installations, solar generation also offsets a portion of peak-hour load directly, which shrinks the energy volume the battery must cover.
The key variable is the timing relationship between solar output and peak pricing windows. In markets where peak pricing occurs in the late afternoon and early evening, solar generation typically ramps down just as peak rates begin. This means the battery must be charged earlier in the day from solar, then held in reserve to discharge during the peak window. The battery’s state of charge at the start of the peak window becomes a critical design parameter.
Accurate yield simulation is essential here. A solar system that looks adequate on an annual energy basis may fall short on cloudy winter days when generation is low and the battery cannot fully charge before peak hours begin. Sizing the battery for TOU optimization in a solar-plus-storage system requires modeling generation and load together across seasonal and weather variations, not just annual averages.
We built Virto.MAX specifically to handle this kind of integrated analysis. The web-based tool simulates solar energy production against real load profiles, helping engineering and sales teams evaluate how much storage is actually needed to achieve TOU shifting targets across different seasons and weather conditions, without manual spreadsheet work.
What tools or software are used to size a TOU battery accurately?
Accurate TOU battery sizing requires tools that can model load profiles, solar generation, tariff structures, and battery performance together in a single simulation environment. Spreadsheet-based approaches can handle simple cases, but they introduce significant error risk as system complexity grows.
The most widely used software categories for TOU battery sizing include:
- Energy simulation platforms such as PVsyst, which model solar generation and can be combined with load data to evaluate storage dispatch strategies
- Battery sizing and dispatch optimizers that model charge and discharge cycles against time-of-use tariff schedules, accounting for round-trip efficiency, degradation, and seasonal variation
- Utility interval data tools that process 15-minute meter data to build accurate load profiles for sizing inputs
- Integrated solar design platforms that combine layout, yield simulation, and engineering outputs in one environment
For engineering teams working on commercial and utility-scale projects, the most efficient approach is software that connects yield simulation directly to the engineering design workflow. Our solar design platform integrates PVsyst-compatible outputs with automated engineering calculations, reducing the manual data transfer that typically introduces errors between simulation and design stages.
Regardless of the tool used, the quality of the input data determines the accuracy of the sizing output. Interval load data, accurate irradiance data for the site location, and the specific tariff schedule from the utility are non-negotiable inputs. Generic assumptions in any of these areas will produce a battery size that either underperforms or overspends.
When does TOU battery sizing stop being cost-effective?
TOU battery sizing stops being cost-effective when the value of energy arbitrage, the savings from shifting consumption out of peak pricing windows, is smaller than the annualized cost of the battery system including installation, financing, and maintenance. This threshold depends on the spread between peak and off-peak electricity rates, the number of peak hours per day, and the battery’s usable cycle life.
The rate spread is the most important variable. A tariff with a peak-to-off-peak price difference of less than a few cents per kilowatt-hour rarely justifies the capital cost of a commercial battery system on arbitrage value alone. Markets with large rate spreads, often driven by demand charges or high peak rates, produce much stronger economic cases.
Cycle life and degradation also set a ceiling on cost-effectiveness. A lithium iron phosphate battery rated for 4,000 to 6,000 cycles at 80 percent DoD will deliver roughly 10 to 15 years of daily cycling before capacity degrades to the point where replacement is needed. If the annualized arbitrage savings do not recover the system cost within that window, the project does not pencil out on TOU economics alone.
Several factors can push a borderline project across the cost-effectiveness threshold:
- Stacking TOU savings with demand charge reduction, which targets peak demand in kilowatts rather than energy in kilowatt-hours
- Grid services revenue from frequency regulation or demand response programs
- Incentives or tax credits that reduce the effective capital cost of the battery
- Rising electricity rates over the project lifetime that improve payback as years pass
If you are evaluating whether a TOU battery makes sense for a specific project and want to model the numbers accurately, speaking with our team can help you work through the simulation and engineering inputs before committing to a system size.
Frequently Asked Questions
How often should I recalculate my TOU battery sizing as my energy loads change?
You should revisit your TOU battery sizing at least once a year, or whenever a significant load change occurs — such as adding EV charging stations, expanding production equipment, or onboarding new HVAC systems. Interval data from your utility meter will reveal whether your current battery is still covering peak windows effectively or whether capacity has become mismatched with actual demand. If your utility also updates its tariff structure or shifts peak pricing windows, that alone can change your optimal battery size even if your loads stay the same.
What happens if I undersize my TOU battery and it runs out of charge before the peak window ends?
If your battery depletes before the peak pricing window closes, your facility will automatically draw the remaining energy from the grid at peak rates, eliminating the savings you were counting on for that portion of the window. Repeated deep discharges that push the battery beyond its rated depth of discharge can also accelerate capacity degradation, shortening the system's usable life. To avoid this, always include the 10–15% efficiency buffer in your sizing calculation and monitor state-of-charge data during the first few weeks of operation to confirm real-world performance matches your simulation.
Can I use a TOU battery to reduce demand charges at the same time as shifting energy costs?
Yes, and stacking these two value streams is one of the most effective ways to improve the economics of a commercial battery system. Demand charges are billed on your highest 15-minute or 30-minute peak demand interval in a billing period, so a battery that discharges during those spikes can significantly reduce that charge independent of TOU energy arbitrage. The key is configuring your battery management system to prioritize demand peak shaving during the highest-draw moments within the peak window, rather than spreading discharge evenly across all peak hours — a distinction that requires careful dispatch programming or intelligent energy management software.
What battery chemistry is best suited for daily TOU cycling, and does it affect sizing?
Lithium iron phosphate (LFP) is the dominant chemistry for commercial TOU applications because it combines high cycle life (4,000–6,000+ cycles), thermal stability, and a usable depth of discharge of 80–90%, all of which directly improve the economics of daily cycling. Compared to older lead-acid or NMC chemistries, LFP's higher usable DoD means you need less nameplate capacity to deliver the same usable energy, which reduces upfront cost. When sizing, always confirm the DoD rating and cycle life warranty from your battery manufacturer, since these figures vary between products and directly affect both your capacity calculation and your long-term cost-effectiveness model.
How do I account for battery degradation when sizing for a 10–15 year project lifetime?
Battery capacity degrades over time — typically 20–30% over the rated cycle life for LFP systems — meaning a battery sized precisely for today's load will underperform in later years. A common approach is to size the system to meet your peak-hour energy target at end-of-life capacity, which means specifying a nameplate capacity that is 20–25% larger than the minimum required today. Alternatively, some project models plan for a partial capacity augmentation mid-life rather than over-sizing from day one, which can be more capital-efficient depending on financing structure and projected electricity rate escalation.
Is interval meter data always available, and what should I do if I can't access it?
Most commercial and industrial utility accounts in the U.S. and many other markets have smart meters that record 15-minute interval data, and utilities are generally required to provide this data upon request — sometimes through an online portal, sometimes via a formal data request. If interval data is unavailable for a new facility or a site that hasn't yet been metered, you can build a proxy load profile using equipment nameplate ratings, operational schedules, and comparable facility benchmarks from the same industry sector. While proxy profiles introduce more uncertainty than real meter data, they are a valid starting point for preliminary sizing — just plan to validate and refine the battery configuration once actual interval data becomes available after the first billing cycle.
What are the most common mistakes engineers make when sizing a TOU battery for the first time?
The most frequent mistakes are using monthly energy averages instead of interval-level load data, ignoring round-trip efficiency losses, and failing to model seasonal variation in both load and (for solar-paired systems) generation. Another common error is sizing only for current loads without accounting for planned load growth over the project lifetime, which leads to a system that becomes undersized within a few years. Finally, many first-time sizing exercises overlook the tariff details — specifically whether peak rates apply on weekends, holidays, or only on weekdays — which can significantly change the number of annual cycles the battery needs to perform and therefore the cost-effectiveness calculation.
Related Articles
- What is C-rate and how does it affect battery sizing?
- What safety codes apply to battery storage wiring?
- Why are solar panel layouts different from what gets built?
- What makes bill of materials generation complex for solar projects?
- How do you validate solar system performance before construction?
This content was generated with the help of AI — it may contain mistakes