How Much Storage Capacity Pays Off? Before Asking the Price, Look at Your Factory's Load Curve

A factory's electricity "personality" is written all over its load curve. Two plants can each use several hundred thousand kWh a month and look nothing alike: one draws a nearly flat line, the other a camel hump with morning and evening peaks and a dip in between. Whether a storage system is sized correctly comes down to this curve. Below, we break down the typical curve shapes and explain how to read them—and how to put them to work.

**Type 1: Continuous and steady — essentially a flat line**

Chemical plants, aluminum smelters, cement plants, and some metallurgical operations fall into this category. Equipment runs around the clock, power draw stays high, and the daily load curve is close to a flat line. In metallurgy, the daily load factor typically reaches about 90%, with minimum load very close to maximum load.

What does this mean for storage? The room to maneuver is narrow, but demand management still has value. Because load is steady, there's no obvious "peak window" for the battery to discharge into, so peak-valley arbitrage depends mainly on the time-of-use tariff rather than on the plant's own load swings. That said, if the plant is billed on maximum demand, storage can discharge in small amounts during any period when load ticks slightly higher, shaving that "highest point" down a little. The reduction is small—but the monthly demand charge is the demand rate multiplied by that highest point, so every bit shaved is money saved.

**Type 2: Daytime production — up in the morning, down at night, one clear peak**

This is the pattern for a large share of ordinary manufacturers. Load climbs quickly after startup, holds high through the day, and drops off sharply when production stops in the evening. For single-shift light industry, the daily load factor typically runs 0.5–0.7, with a minimum load factor of only 0.2–0.4.

This is one of the curve types commercial and industrial storage likes best. The daytime production window lines up closely with the tariff's peak periods, so the battery can charge during valley hours and discharge into the daytime production peak—a clean arbitrage path. One caveat: if the plant's daytime peak isn't very pronounced—a gentle hill rather than a sharp mountain—the demand-reduction benefit is diluted, and capacity should be sized conservatively so you don't pay for batteries you never use.

**Type 3: Two- or three-shift operations — two or more peaks a day**

Three-shift plants show two to three peaks in the daily load curve, usually after shift changes and meals, as load falls during breaks and climbs again when work resumes. Two-shift operations fall between one- and three-shift patterns, with daily load factors typically around 0.7–0.9.

Storage strategy here has to be precise down to the shift level. You can't simply "charge by day, discharge by night." You need to identify which peak coincides with the highest tariff and which valley with the lowest. If the morning peak falls in a flat-rate period and the evening peak in a peak-rate period, the battery's discharge window should target the evening peak—not be spread evenly across both.

**Type 4: Volatile — up and down, hard to predict**

Make-to-order operations and some machining shops fit here. Equipment starts and stops frequently, and the load curve shows sharp sawtooth or random swings. Machine tools inherently produce short spikes when the spindle accelerates, periodic cycling from coolant pumps, and standby load between cuts.

This curve poses the biggest challenge for storage sizing. Capacity can't be based on an "average peak"—it has to reflect how long the highest instantaneous spike lasts. If a spike lasts only ten or fifteen minutes, a relatively small battery can discharge at the critical moment and hold demand down. If spikes are frequent and unpredictable in duration, the dispatch strategy needs smarter forecasting and control logic; otherwise the battery may discharge when it isn't needed and run dry when it is.

**Type 5: Night production — low by day, high by night, an inverted curve**

At night-shift-dominated plants, the shop floor is quiet during the day and demand peaks from evening into the night. Solar output at midday is completely out of sync with this load—power generated during the day goes unused, and by the time it's needed at night, solar output has fallen to zero.

In this scenario, storage isn't optional—it's essential. If the plant has solar, storage must capture surplus midday PV output for nighttime production; otherwise the solar investment loses much of its value. If there's no solar, the logic is purely valley charging and peak discharging—but note that night production itself may already cover part of the valley period, which can actually shrink the charging window.

**The bill and the curve: result versus process**

Your electricity bill tells you how much power you used this month, what you paid, your peak-to-valley ratio, and your demand charge. That's the result.

The load curve tells you when that power was used, at what rate, and according to what production rhythm. That's the process.

A "maximum demand: 680 kW" line on a bill might trace back to fifteen minutes one morning at ten o'clock when equipment started up all at once—or to a single high-power machine running alone on a night shift. Without the curve, you only know how much you were charged. With it, you know which moment you paid for—and whether you can bring it down by adjusting production scheduling or adding storage.

**The right order for sizing storage**

Look at the load curve first, then set PCS power and battery capacity. Not the other way around.

Here's an example. A plant wants storage, and a vendor proposes 400 kW/800 kWh. Pull the load curve, and you find the plant already carries 450 kW of baseload during valley hours at night—equipment never fully stops; lighting, air compressors, and some lines keep running. Its daytime peak is only 700 kW, meaning the plant's existing maximum demand is roughly 700 kW.

Now the battery charges at 400 kW during the night valley. Nighttime grid draw = existing 450 kW + 400 kW charging = 850 kW—higher than the daytime 700 kW. So the maximum demand recorded this month is no longer the daytime 700 kW but the 850 kW "manufactured" by storage charging. Demand charges are billed on maximum demand; pushing demand from 700 to 850 adds a monthly cost that eats up a big chunk of the arbitrage savings.

This is the classic trap of "size first, look at the curve later": charging at night is itself a load, and it stacks on top of existing valley-hour load. When valley load is already substantial, adding charging power can create a new demand peak higher than the daytime peak. If capacity is sized without reading the curve first, a bigger battery and higher charging power can actually push the demand charge up.

**What we do**

Enerbond doesn't lead with a standard cabinet—we lead with your load curve. You provide recent electricity bills and a stretch of 15-minute interval load data. We analyze how well your peak-valley periods line up with tariff periods, how much maximum demand can realistically be cut, and whether charging would create a new demand spike during valley hours. Then we recommend storage capacity matched to your actual production rhythm. Read the curve correctly, and storage is an asset. Read it wrong, and you've just bought another piece of equipment that eats electricity bills.