Commercial Energy Storage
Commercial Energy Storage covers cabinet and container battery systems for commercial and industrial sites, where they shave peak demand, shift load into cheaper hours and hold critical circuits alive when the grid drops. Every system is specified twice: usable capacity in kWh and continuous power in kW. Most units are high voltage LiFePO4 racks with an integrated BMS and EMS, connected through three-phase inverters and metered by 3-phase smart meters. Brands include BYD, Dyness, Pylontech, Soluna and Victron Energy. Guide
- Reduce energy expenses by storing electricity during off-peak hours.
- Enhance grid independence with backup power during outages.
- Integrate renewables like solar and wind for cleaner operations.
- Improve energy management with smart, scalable storage.
Ideal for factories, warehouses, microgrids, and utilities, our high-performance systems ensure seamless energy resilience. Explore top-tier brands and find the perfect storage solution for your commercial needs.
Product sorting
List of products
Related storage categories
- Batteries for photovoltaics
- High voltage batteries
- Hybrid inverters
- Three-phase inverters
- Smart meters
- Monitoring
Commercial storage: kWh, kW and the tariff it is paid back by
A commercial energy storage system is a battery block, a power conversion system and a controller in one enclosure, sized to a site's load profile rather than to its roof. Two numbers define it and they are chosen separately: usable capacity in kWh, which sets how long the system can hold a load, and continuous power in kW, which sets how much load it can hold at once. Their ratio is the C-rate. A 215 kWh cabinet rated 107 kW runs at 0.5C and can discharge for roughly two hours; the same cabinet at 215 kW runs at 1C for about one hour.
Cells are almost universally LiFePO4 (lithium iron phosphate), chosen for thermal stability, a flat discharge curve and a cycle life in the thousands. Around them sit an integrated BMS for cell balancing and protection, an EMS that decides when to charge and discharge against the tariff and the meter reading, liquid or forced-air thermal management, and a fire detection and suppression package. Stocked brands include BYD, Dyness, Pylontech, Soluna and Victron Energy, usually paired with a photovoltaic array on the same connection point.
FROM THE FIELD
"Before quoting a single kWh, ask the client for twelve months of quarter-hour interval data from the meter. The demand charge is set by a handful of fifteen-minute peaks a year, not by annual consumption, and once you can see them you usually find the site needs far more kW and far fewer kWh than the first guess. Sizing from the annual energy bill alone is how people end up with an expensive battery that never touches the peak."
Ladislav Proc · Photovoltaics specialist, ONSA Plus
How to choose a commercial energy storage system
Commercial storage is bought against a business case, not against a datasheet. The site's metered load profile, its tariff structure and its reserved capacity decide the size; the enclosure, the cooling and the grid code decide what can actually be installed and approved.
Work through it in this order: load profile and tariff, then power in kW for the peaks you want to cut, then capacity in kWh for how long those peaks last, then enclosure, safety and grid connection, and only then the brand and the warranty terms.
Show the full guide Close the guide
Why kWh and kW are sized separately
Capacity and power are independent specifications. Capacity in kWh is how much energy the battery holds; power in kW is how fast it can be pushed in or pulled out, and it is limited by the power conversion system and by the cells' allowed C-rate. A site with short, sharp peaks needs high power and modest capacity. A site running a long evening shift on stored solar needs the opposite.
- C-rate is power divided by capacity. 0.5C means a full discharge over two hours, 1C over one hour, 0.25C over four.
- Usable capacity is what counts, not nominal. Depth of discharge and reserved backup capacity both cut into it.
- Continuous versus peak power differ. Check how long a unit will sustain its boost rating before derating, and at what ambient temperature.
- Round-trip efficiency of a modern LiFePO4 system with its conversion stage typically lands around 88 to 92 %. Every stored kWh returns slightly less than a kWh.
Peak shaving and demand charge reduction
On most European commercial tariffs a large part of the bill is not energy but capacity: a charge on reserved capacity or on the highest quarter-hour demand recorded in the billing period. Storage discharges into those peaks so the meter never sees them, which lowers the billed maximum without changing what the site actually consumes. The EMS watches the metering point in real time and releases power the moment measured demand crosses a set threshold.
- Size from interval data, ideally a year of quarter-hour readings. Peak magnitude sets kW, peak duration sets kWh.
- Response speed matters. A system that reacts in a few hundred milliseconds catches a compressor or press start; a slower one does not.
- Reserved capacity can be renegotiated downwards with the supplier once storage reliably caps demand, which is often where the larger saving sits.
Load shifting, arbitrage and solar self-consumption
Load shifting charges the battery when electricity is cheap and discharges it when it is expensive. Under a spot or block tariff this is straightforward arbitrage; the spread has to cover round-trip losses and the cycle cost before it earns anything. Paired with a photovoltaic array the same mechanism raises self-consumption: midday surplus that would otherwise be exported at a low price is held back for the afternoon and evening load.
- Daily cycling is what pays back arbitrage. A system that cycles once a day earns roughly 365 spreads a year, so the cycle life budget has to allow it.
- Export limitation is often a condition of the grid connection. Storage absorbs what the limiter would otherwise curtail.
- EV charging is a growing use case: a battery buffers a wallbox bank so a site can add fast charging without upgrading the connection.
Tip from practice
Do not let one system chase every application at once. Peak shaving needs headroom held in reserve, backup needs a reserved state of charge, and arbitrage wants to run the battery empty. Set the EMS priority explicitly and reserve a fixed share of state of charge for whichever function actually carries the business case, then let the rest trade.
Backup power, transfer time and UPS-grade supply
Backup is a separate design question from savings. A standard grid-tied system disconnects on grid failure and does nothing. Backup requires a changeover point, an islanding-capable conversion system and clear knowledge of which circuits are protected. Transfer time is the discriminator: a typical islanding transfer takes 10 to 20 ms, which most machinery tolerates, while genuinely uninterruptible loads such as servers, medical equipment or continuous process control need an online double-conversion path with no break at all.
- Define the backup circuits on a separate distribution board. Backing up the whole site is usually far more expensive than backing up what matters.
- Check the surge rating. Motor and compressor inrush can be several times running current and it is the inverter, not the battery, that has to supply it.
- Black start with photovoltaics lets an islanded site recharge from the array during a long outage. Confirm the inverter supports it.
Cells, cycle life and what the warranty actually covers
LiFePO4 cycle life is quoted as a number of cycles to a remaining capacity, typically 6000 or more cycles to 70 or 80 % of original capacity, measured at a stated depth of discharge and temperature. Commercial warranties are usually written either in cycles or in total energy throughput in MWh, whichever limit is reached first, alongside a calendar term. Throughput warranties suit heavily cycled arbitrage sites; cycle counts suit predictable single-cycle operation.
- Read the conditions: guaranteed capacity at end of term, ambient temperature range, maximum C-rate and required connectivity for remote monitoring.
- Thermal management is the main lever on real degradation. Liquid-cooled cabinets hold cells in a tighter band than air-cooled ones under sustained high C-rate.
- Modularity lets capacity be added later. Check whether the manufacturer permits mixing production batches on one string, because many do not.
Enclosure, fire safety and grid connection
Cabinets from roughly 50 to 400 kWh sit outdoors on a prepared slab or indoors in a plant room and suit most single-site commercial projects. Containers from several hundred kWh upwards are used for industrial loads, microgrids and generation sites. Both need a fire strategy: cell-level detection, aerosol or gas suppression, off-gas and smoke sensing, pressure relief and a documented separation distance from occupied buildings and boundaries.
- IP rating and ambient range govern outdoor siting, along with noise from fans or chillers near neighbours.
- Grid connection needs DSO approval before installation. Expect a connection application, grid code compliance evidence for the conversion system, protection settings and a commissioning test.
- Metering and communication: a 3-phase smart meter at the connection point feeds the EMS, and remote monitoring is normally a warranty condition.
- Civil works: foundation, access for a crane or forklift, cable routing and earthing are often the long pole in the schedule, not the battery lead time.
Application comparison
| Application | What drives the sizing | Typical shape of the system |
|---|---|---|
| Peak shaving | Height and length of quarter-hour demand peaks | High kW, modest kWh, 0.5C to 1C, fast EMS response |
| Load shifting and arbitrage | Tariff spread and number of cycles per day | High kWh, moderate kW, 0.25C to 0.5C, throughput warranty |
| Solar self-consumption | Midday surplus from the photovoltaic array | Capacity matched to daily export, hybrid or coupled conversion |
| Backup and islanding | Critical load in kW and required autonomy in hours | Reserved state of charge, changeover point, surge headroom |
| EV charging buffer | Charger power against the site connection limit | High kW in short bursts, capacity sized to a charging session |
Swipe the table to the left
Frequently asked questions about commercial energy storage
How do I size a commercial battery system?
Start from twelve months of quarter-hour interval data from the site meter. The height of the demand peaks sets the power rating in kW, and how long those peaks last sets the usable capacity in kWh. Annual consumption alone is not enough to size either.
Why are kWh and kW quoted separately?
They describe different limits. kWh is how much energy the battery holds, kW is how fast it can charge or discharge. Their ratio is the C-rate: a 200 kWh system rated 100 kW runs at 0.5C and sustains full power for about two hours.
Can storage reduce my demand charges?
Yes. The EMS monitors the connection point and discharges the moment measured demand crosses a threshold, so the meter never records the peak. Once demand is reliably capped, reserved capacity can often be renegotiated downwards, which is usually the larger saving.
What round-trip efficiency should I expect?
A modern LiFePO4 system including its conversion stage typically returns 88 to 92 % of the energy put in. Losses come from the cells, the power conversion system and the auxiliary load of cooling and controls, which matters most in hot conditions.
Cabinet or container: which suits my site?
Cabinets of roughly 50 to 400 kWh suit single commercial sites and can stand outdoors on a slab or indoors in a plant room. Containers from several hundred kWh upwards suit industrial loads, microgrids and generation sites where floor space and crane access exist.
Will the system keep my production running during an outage?
Only if it is specified for backup. Grid-tied systems disconnect on grid failure. Backup needs an islanding-capable conversion system, a changeover point and defined backup circuits, plus enough surge headroom for motor and compressor inrush current.
Is a battery the same as a UPS?
No. A typical islanding transfer takes 10 to 20 ms, which machinery and lighting tolerate but sensitive IT does not. For UPS-grade supply you need an online double-conversion path with no break, either as a dedicated unit or as a mode of the storage system.
How long does a LiFePO4 commercial system last?
Manufacturers typically warrant 6000 cycles or more to 70 to 80 % of original capacity, at a stated depth of discharge and temperature. Calendar life is usually 10 to 15 years. Thermal management and average C-rate are the main levers on real degradation.
Is the warranty in cycles or in throughput?
Both are used, often together with whichever limit is reached first. Cycle counts suit predictable single-cycle operation; energy throughput in MWh suits heavily cycled arbitrage. Check the guaranteed remaining capacity, the temperature range and any monitoring requirement.
Do I need DSO approval to install one?
Yes. Any storage system connected to the network needs a connection application and approval from the distribution system operator before installation, including grid code compliance evidence for the conversion system, agreed protection settings and a commissioning test.
System sizing, the achievable saving and the connection conditions depend on your measured load profile, tariff structure and the approval of the distribution system operator.
About ONSA Plus
Why commercial operators across Europe work with us
ONSA Plus is a European B2B distributor of photovoltaics, battery storage, heat pumps and EV charging stations. In our own EU warehouses we keep over 1,500 products from 25+ brands, including Huawei, Solis, SolaX and Dyness.
We are not just a trading company. We design and service energy systems ourselves, so we answer questions on peak shaving, storage sizing and grid connection from engineering practice, not from a catalogue. You handle a warranty claim directly with us and you buy per job.