When Transpower’s Matt Webb explained the $144 million Waikato Upper North Island Voltage Management project in May 2025, he named the problem directly: “Voltage control in this region is a growing challenge due to the retirement of fossil-fueled generation, a reduction in system inertia.” Over 1,000 MW of thermal generation has been decommissioned in the Waikato and Upper North Island since 2012. Every megawatt took a piece of grid stability with it. What replaces that stability — STATCOMs, synchronous condensers, and increasingly grid-forming battery storage — is the question NZ is now quietly answering.

If you are evaluating a utility-scale battery energy storage system for a NZ project in 2026 or beyond, the grid-forming vs grid-following inverter choice is one of the most consequential decisions you will make. It affects what services the battery can sell and whether the installation contributes to system stability or draws from it.

The core distinction — voltage source vs current source

The difference between grid-forming (GFM) and grid-following (GFL) inverters sounds abstract but decides everything downstream.

• Grid-following inverters act as controlled current sources. They measure the grid’s voltage and frequency, then inject current in synchronisation. If the grid signal disappears, they trip offline. Almost all early solar and battery inverters use this design.
• Grid-forming inverters act as controlled voltage sources. They maintain their own internal voltage phasor — magnitude and frequency set locally — and adjust output in response to grid demands rather than following an external signal.
• AEMO’s working definition: a GFM inverter “maintains a constant internal voltage phasor in a short time frame, with magnitude and frequency set locally by the inverter, thereby allowing immediate response to a change in the external grid.”
• What this enables: because GFM inverters establish rather than follow the grid signal, they provide synthetic inertia, operate in weak grids, respond to voltage disturbances within 15–20 ms, ride through faults, and start a grid section back up from complete blackout.
• The cost gap has closed: Modo Energy’s 2025 NEM analysis found GFM battery cost is now “virtually the same” as grid-following. Additional testing is marginally more expensive; the hardware is not.

Why NZ’s grid needs this capability now

The specific pressures on New Zealand’s grid are documented in Transpower’s own strategic materials.

• The inertia problem, in Transpower’s own words: “New Zealand’s power system currently consists of mostly synchronous generation, with predictable behaviour and controls, resulting in a stable system… [But] an increase in renewable generation is producing a large increase in inverter-based generation resulting in potential challenges to the power system.”
• The thermal retirement pipeline: more than 1,000 MW decommissioned in the Waikato and Upper North Island since 2012. Further retirement of the remaining Huntly Rankine units is under active discussion.
• Demand growth pressure: Transpower forecasts NZ electricity usage will grow ~70% by 2050. Data centres, EV charging, and industrial electrification add load faster than traditional generation can be replaced.
• The 2026 EA pipeline mix: 57% solar, 20% onshore wind, 18% BESS — an overwhelmingly inverter-based generation stack. Every megawatt added at this ratio without grid-forming capability makes the underlying stability problem harder.
• Interim fixes already deployed: Hitachi Energy’s Hamilton Dynamic Reactive Plant (±150 MVAr STATCOM) operational since July 2023. The Otahuhu STATCOM is the second stage. These are dedicated reactive power solutions filling the gap that grid-forming batteries could increasingly cover.

What a grid-forming battery actually delivers

The technical capabilities map directly to services NZ’s grid will need to procure through the rest of the decade.

• Synthetic inertia: traditional synchronous generators provide inertia physically through rotating turbine mass. GFM batteries emulate this response electronically, providing the same stabilising effect during frequency disturbances without moving parts.
• Voltage regulation and reactive power: GFM inverters can inject or absorb reactive power on demand, doing at battery scale what STATCOMs do dedicatedly. For a NZ grid actively investing in reactive power solutions, batteries with GFM capability provide double duty.
• Weak grid operation: GFM inverters operate stably at Short Circuit Ratios as low as 1.25 — meaning they work in grid sections where traditional inverters would trip offline.
• Fault ride-through: GFM systems stay connected during grid faults and inject reactive current within 15–20 milliseconds to support voltage recovery. Grid-following inverters typically disconnect and try to reconnect.
• Black start and island operation: GFM batteries can start a network section back up from complete blackout, and can continue operating a local grid section after separation from the main network.

The Australian precedent for NZ to learn from

Australia is roughly two to three years ahead of NZ on grid-forming deployment, and the trajectory is instructive.

• ARENA funding round: in December 2022, the Australian Renewable Energy Agency committed AU$176 million to eight grid-scale battery projects, each 200–300 MW, all equipped with grid-forming inverter technology.
• Commissioning window: these projects entered grid-forming operation between 2024 and 2026 — the same window as NZ’s own utility-scale pipeline (Glenbrook, Huntly, Ruakākā, Waikato).
• Regulatory recognition: AEMO has moved from studying GFM inverters to specifying voluntary performance requirements. NZ’s EA BESS roadmap is following a similar trajectory but roughly two years behind.
• Implication for NZ: Australian precedent suggests grid-forming capability will move from “nice to have” to “required for connection” within three to five years. Buyers commissioning batteries in 2026 without GFM capability may need to retrofit later.

What this means for battery chemistry choice

Grid-forming is fundamentally an inverter capability, but the underlying chemistry matters for sustaining that capability across a 25-year asset life.

• Cycle life supports GFM services: continuous frequency and voltage regulation means constant small cycles. Vanadium flow’s 15,000–20,000+ cycle rating handles this without degradation; lithium ages faster under the same duty.
• Depth of discharge tolerance: GFM services demand full flexibility across the state-of-charge range. Vanadium flow’s 100% depth of discharge means full capacity is available at any state of charge.
• Response consistency across temperature: GFM inverter response must be reliable across NZ’s temperature range. Vanadium flow operates without active thermal management; lithium derates above 30°C on hot Canterbury days.
• For utility-scale NZ grid-forming deployments, our SPower Series pairs Rongke’s Ushi-validated vanadium flow with grid-forming inverter platforms suited to NZ commercial and utility applications.

Conclusion

Grid-forming battery storage is not an incremental upgrade to conventional BESS. It is the technology that lets inverter-based generation take over the grid stability functions retiring thermal plants used to provide by physics. For New Zealand — with 1,000+ MW of thermal generation already retired in the North Island, further Huntly retirement under discussion, and 57% of new generation coming from solar — the transition to grid-forming capability is not optional. It is the missing piece that turns a renewable-heavy grid from a stability problem into a stability solution. The Australian precedent suggests it will become mandatory for large connections within three to five years. Buyers commissioning batteries now have a choice: build for the grid the country has today, or build for the grid it will have when the asset is mid-life.

For a tailored briefing on how grid-forming capability applies to a specific NZ utility-scale project — including current regulatory positioning and grid-connection considerations — the Zion Technologies team can put together a written summary within 48 hours.