Why storage decides whether your renewables are useful

A renewable plant without storage produces energy when the resource is available. An industrial site consumes energy when the process demands it. Storage is what reconciles those two facts, and choosing it badly is expensive in a way that is difficult to correct later.

The most common framing — how many megawatt-hours do we need — skips the question that actually determines the technology. Two installations with identical energy capacity but different duration requirements should be built from different technologies.

Five technologies, five different jobs

Battery energy storage (BESS)

Fast response, short duration. Batteries excel at frequency support, peak shaving and smoothing a renewable plant’s output over minutes and hours. Where the requirement is to respond within seconds and ride through short gaps, nothing else competes on responsiveness. Where the requirement is to shift energy across days, the economics deteriorate quickly.

Thermal storage

Where the end use is heat — or steam driving a turbine — storing heat directly is far cheaper per unit than converting to electricity, storing electrons and converting back. This is the reasoning behind molten-salt storage in concentrated solar-thermal plants, and it is what allows CSP to overcome the greatest weakness of solar thermal systems: it makes the output dispatchable after sunset.

Hydrogen storage

Long duration and seasonal shifting, with a property none of the others share: hydrogen is usable as a feedstock as well as a fuel. For a site that already consumes hydrogen in its process, storage and supply become the same investment.

Mechanical and pump-turbine storage

For very long duration, pumped storage hydropower remains the most established technology in the world, and where topography permits it, it outperforms batteries on both duration and asset life. It should be evaluated as a storage asset sized against the grid balancing it is being asked to do, rather than as a generation project that happens to store energy.

Start from the load, not the generation

The sequence that produces good answers is consistent: characterise the load profile, characterise the generation profile, and look at the gap between them. The shape of that gap — how deep, how long, how often — selects the technology far more reliably than any comparison of headline costs.

• Gaps measured in seconds to minutes, recurring frequently: batteries.
• Gaps measured in hours, where the end use is heat or steam: thermal.
• Gaps measured in days to seasons: hydrogen, or pump-turbine where the site allows.
• Grid-scale balancing with a suitable head and reservoir: pumped storage.
• A combination, which is the honest answer on most industrial sites.

Storage is rarely a standalone project

In practice storage arrives attached to something else — a solar plant that needs to serve an evening load, a wind farm facing curtailment, a fossil plant being upgraded, an industrial site trying to reduce peak demand charges. Treating it as a separate procurement usually produces a system that is correctly sized in isolation and poorly matched in context.

Because we engineer all five storage technologies alongside solar, CSP, wind, hydropower, geothermal and waste-to-energy, the storage decision is made inside the generation project rather than deferred to a later one. That also means the recommendation is not constrained by a single product line — if the right answer for your load is thermal rather than batteries, there is nothing pushing us toward the more expensive option.

One question worth asking early

Before sizing anything: what does an outage actually cost you? On many industrial sites the answer reframes the whole project. If a short interruption is merely inconvenient, a modest battery may be sufficient. If it stops a production line or spoils a batch, the duration requirement — and the technology — changes completely.