# Managing More Mine GWh With Less Effort? Let’s Talk About What’s Actually Being Sold Here
Here’s what this really means: battery energy storage systems at mining operations are being deployed faster than anyone can properly manage them. Portfolios are growing faster than the teams and operating models meant to run them. Every added megawatt-hour compounds the underperformance, revenue leakage, and operational mess. That’s the actual problem being described — and frankly, it’s a real one.
Into this gap steps Elysia, the battery intelligence arm of Fortescue Zero, with a product called BESS Copilot. It’s billed as the first technical management solution to translate battery analytics directly into action. Most existing tools tell you what went wrong after the fact. They surface underperformance issues and error signals retrospectively — a post-mortem, not a treatment. Copilot claims to combine cloud-based algorithms with real-time on-premises controls, so it doesn’t just flag problems; it actively corrects them.
Now, before anyone reaches for their wallet, let’s apply some seasoning.
**What the product actually claims to do**
Four things, according to Elysia:
First, recover full availability. Performance diagnostics continuously monitor, diagnose, and prioritise fixes for unexplained output shortfalls. Unexplained shortfalls are the silent tax on every storage portfolio — the kilowatt-hours you paid for but never see.
Second, improve O&M efficacy by automating preventative maintenance and service actions. Fine in principle. The devil, as always, is in whether the automation actually trusts its own diagnostics or just buries operators in alerts they learn to ignore.
Third, capture more revenue by correcting state of charge, state of energy, and telemetry in real time, so inaccurate signals stop distorting planning and dispatch. This is the piece I’d take most seriously. Bad telemetry is not a cosmetic issue. If your SoC is lying to you, your dispatch decisions are garbage, and in markets where storage revenue depends on accurate, timely response, garbage in means real money out. Anyone who’s run an asset-heavy business knows the gap between what the dashboard says and what the asset does is where margins go to die.
Fourth, dispatch power reliably by optimising power allocation across power conversion systems — eliminating the unexpected derating and inverter shutdowns that eat into full-power hours. Inverter shutdowns and derating events are mundane, unglamorous, and expensive. Solving them isn’t sexy, but it’s the kind of operational plumbing that separates assets that hit their pro formas from assets that don’t.
**Where it came from and who’s behind it**
Copilot is the product of Elysia’s acquisition of Zitara, a specialist in on-premises precision controls for utility-scale batteries. The combined outfit claims 125-plus battery engineers, and positions itself as a leading battery intelligence group with a decarbonisation mission spanning grid, road, and heavy industry.
The parent-company backing matters here, and it’s the strongest card in the deck. Fortescue is a global mining operator with roughly US$30 billion in assets, and Copilot is being rolled out across 5 GWh of BESS supporting Fortescue’s own Pilbara Green Grid. That’s not a slide-deck pilot. A vendor eating its own cooking at industrial scale — in one of the harshest operating environments on the planet — is worth something. Plenty of software outfits in this space have never watched their code run next to actual iron ore.
Beyond the internal rollout, the product is live at over 2 GWh of customer sites across the US, UK, and Australia. Still early. Still a young track record. But not vaporware.
**The problem statement, and why I buy it**
Shyam Srinivasan, Elysia’s EVP for battery energy storage, frames it plainly: the BESS industry doesn’t just have a visibility problem, it has an action problem. Asset managers struggle to understand why batteries underperform and to quantify the impact. Even when the root cause is clear, corrective action and warranty-claim evidence are hard to nail down.
This isn’t the first time we’ve seen this circus. Every infrastructure boom follows the same arc: capital floods in, deployment outruns operational capability, and only later does the industry discover that owning assets and managing them profitably are entirely different skills. We watched it with wind, with solar, with telecom fiber. Storage is running the same playbook. Deployment targets are sexy; availability percentages and round-trip efficiency drift are not, and that’s precisely why they get neglected until the revenue shortfall shows up in someone’s quarterly numbers.
The “insight to site” pitch — closing the loop between analytics and real-time control — targets a genuine structural weakness. The analytics layer and the controls layer in most storage operations live in separate worlds, run by separate teams, often on separate contracts. Every handoff between them is a point of delay and error. Merging them is a sensible engineering thesis, not marketing fluff.
**My take**
Here’s the skeptical read: “AI-powered copilot” branding is riding a hype cycle, and I don’t blame a vendor for wearing the fashion of the moment. The question isn’t what it’s called — it’s whether the on-premises control integration actually closes faults faster than a competent operations team with good dashboards and a phone. If it does, the value case is straightforward: recovered availability and corrected telemetry translate directly into dispatchable megawatt-hours, which translate into revenue. Storage economics are thin enough that a few percentage points of availability are the difference between a workable asset and a write-down.
Two practical points for anyone evaluating this space. One: demand performance guarantees tied to measured availability and accuracy improvements, not feature checklists. Software in this category should be paid for out of recovered revenue. Two: ask about the RMA evidence workflow. Warranty recovery on battery systems is a swamp — degradation claims are contested, data is disputed, and operators routinely leave money on the table because they can’t produce defensible telemetry. If Copilot genuinely automates that evidence chain, that alone could justify the ticket, and it’s the kind of unglamorous detail most press releases skip.
The bigger picture: mining is becoming a grid-scale storage customer whether it likes it or not, driven by decarbonisation targets and diesel economics. Pilbara operations running multi-gigawatt-hour storage portfolios is a preview of heavy industry everywhere. The operational tooling catching up to the hardware is overdue, and whoever solves it credibly will have a real business — regardless of what the product is called.
The thesis here is sound. The execution is what matters, and the 5 GWh Fortescue rollout is the honest test. Watch the availability numbers, not the press releases.
Source: https://im-mining.com/2026/11/17/elysia-bess-copilot-managing-more-mine-gwh-with-less-effort/

