Quick answer: Designing, setting and commissioning microgrid protection to established practice is engineering — the outcome is determinable in advance. A core R&D activity may exist in the narrower case where no established basis predicts whether a protection or islanding-transition strategy can work at all, typically because inverter-limited fault current defeats conventional discrimination, and the answer comes only from a systematic progression of work conducted to generate new knowledge. Activities associated with complying with statutory requirements or standards are separately excluded from being core.
18 August 2026 — this article describes the current rules. The 2026-27 Federal Budget announced R&DTI changes proposed to apply to income years starting on or after 1 July 2028; those changes are not yet law.
A microgrid controller closes and opens one breaker. Everything difficult about a microgrid sits in what the rest of the network is supposed to do in the second either side of that.
This article is about that second: the protection scheme, the anti-islanding behaviour of the inverters already on the site, and the transition into and out of island mode. It is not about sizing solar and storage, and not about EV charging or vehicle-to-grid — both have their own articles in our Insights, as does the separate question of dispatch, forecasting and market-bidding software. Those are asset and algorithm questions. This one is a fault-behaviour question, and it lands on a different part of the statute.
The Statutory Test the Activity Has to Meet, Stated in Full
Eligibility under the R&D Tax Incentive is assessed activity by activity, not project by project. Under s 355-25(1) of the Income Tax Assessment Act 1997, core R&D activities are experimental activities whose outcome cannot be known or determined in advance on the basis of current knowledge, information or experience, but can only be determined by applying a systematic progression of work that is based on principles of established science and that proceeds from hypothesis to experiment, observation and evaluation, and leads to logical conclusions; and that are conducted for the purpose of generating new knowledge, including new knowledge in the form of new or improved materials, products, devices, processes or services (business.gov.au; ITAA 1997).
Two limbs do unusual work here. The established-science limb asks something narrower than it first appears: it requires that the particular systematic progression of work be based on principles of established science, not merely that established science exists somewhere in the field. Power-system fault analysis, symmetrical components and machine dynamics give a protection investigation a settled body of principle to build such a progression on, so the limb is capable of being satisfied here — but whether it is satisfied turns on how the specific progression is designed, conducted and recorded, and that has to be self-assessed activity by activity.
The limb that bites is the first: most outcomes can be determined in advance from fault studies, standards, relay manuals and vendor engineering. (AusIndustry guidance frames that first limb by asking whether a competent professional in the field could have determined the outcome in advance; the phrase "competent professional" is guidance language and does not appear in s 355-25(1) itself.) Difficulty, cost and novelty to the company are not the test.
Why Island-Mode Protection Is Where an Unknown Can Genuinely Sit
Conventional distribution protection is built on a physical fact: a synchronous grid delivers fault current several times rated current, so time–overcurrent devices can be graded by magnitude and time, and the device nearest the fault operates first. Three things about an inverter-fed island remove that fact:
Fault current becomes a control decision, not a machine property
A grid-forming inverter may limit its fault current to a relatively small amount above rated current, with the magnitude and duration depending on the inverter and its current-limiting control. In a radial island fed by a single such source, the current at the incomer and at the faulted feeder can be nearly identical, because the source sets the current rather than the fault loop impedance. Whether any grading margin exists at all is a property of that inverter's control firmware, and available vendor information may not provide the fault-behaviour characterisation needed for a particular grading study.
Anti-islanding and intentional islanding pull in opposite directions
Grid-connected inverter energy systems must detect an unintentional island and disconnect, and the inverters already on a site implement that in firmware to the connection standards. A microgrid asks the same devices to stay connected through an intentional island. Whether a transition can be shaped so standard-compliant devices ride through it — rather than each vendor's loss-of-mains logic acting on the transient and taking generation offline exactly when the island needs it — no datasheet answers, because the standards define behaviour under their own test conditions, not under this transition.
Seamless transfer and resynchronisation are timing problems with real physics
Open transition costs an outage; a closed or seamless transition requires the grid-forming source to take the voltage and frequency reference within a window measured in milliseconds, and reconnection requires a synchronism check inside a defined voltage, phase-angle and frequency envelope. Where the achievable window on a given plant has no reliable predictive basis and can only be found by instrumented trial, an experimental activity may exist.
None of this makes protection work R&D by default: most of it is a fault study, a grading calculation and a settings sheet — established methods on a new site, outcome determinable in advance. See what does not qualify.
Where Standards Work Sits — and What s 355-25(2)(f) Actually Says
Section 355-25(2) lists activities that are not core R&D activities, and paragraph (f) is directly relevant:
"(f) activities associated with complying with statutory requirements or standards, including one or more of the following: (i) maintaining national standards; (ii) calibrating secondary standards; (iii) routine testing and analysis of materials, components, products, processes, soils, atmospheres and other things"
Three consequences follow:
It excludes those activities from being core, not from the claim: An activity that cannot be core may still be a supporting R&D activity. But s 355-30(2) applies a higher bar to exactly this kind of work: where an activity (a) is an activity referred to in s 355-25(2), or (b) produces goods or services, or (c) is directly related to producing goods or services, it is a supporting R&D activity only if it is undertaken for the dominant purpose of supporting core R&D activities. Each limb is tested against the particular activity, not against the site: a microgrid generating electricity does not by itself make every activity on it one that produces goods or services. Where the compliance activity is of a kind referred to in s 355-25(2), limb (a) applies; where a commissioning activity produces, or is directly related to producing, goods or services, limbs (b) or (c) may also apply.
It is not triggered merely because a standard exists in the field: The exclusion attaches to activities associated with complying with a statutory requirement or standard — which may include type testing, witness testing, certification, connection testing or periodic verification where those activities are required under legislation or by a regulator acting under legislation. Where a standard states a required outcome but no established means of achieving it on a particular plant, whether the work of finding that means is itself an activity associated with complying is a question of fact, not a settled carve-out — and "associated with" is broad wording. Some development work directed at meeting a standard will fall inside paragraph (f). It is assessed activity by activity, on the facts, and it is the company that self-assesses.
The exclusion does not depend on the result being predictable: It is tempting to say a test with a published method and a published pass criterion has a known outcome, but that is wrong: the unit under test may fail, and the result of a compliance test is genuinely not known before it is run. What puts that work outside core is paragraph (f) itself — the activity is undertaken to demonstrate compliance with a statutory requirement or standard, so it is an activity associated with complying with it, whatever the test returns. Whether a given activity falls there is still assessed on its own facts.
A Worked Example (Hypothetical and Illustrative Only)
Invented to show where the boundary falls. It is not a real project and says nothing about whether any actual claim would be accepted.
A regional South Australian precinct operates an 11 kV embedded network with three feeders, a 1.2 MW solar array on grid-following inverters, a 1 MW / 2 MWh battery on a grid-forming inverter and a standby diesel set. The intent is to island on loss of supply and keep the precinct running.
Baseline and target, fixed before energisation: Grid-connected, the distributor's fault level gives about 6 kA at the 11 kV bus and the existing inverse-time overcurrent relays grade with a 300 ms margin. Islanded, the battery inverter is the only source and is limited to about 1.2 times its rated 52 A — roughly 63 A, two orders of magnitude below the grid-connected case. The recorded target: for a three-phase and a single-phase-to-earth fault at the mid-point of each of the three feeders in island mode, the faulted feeder's device must operate and the incomer must not, within 400 ms, with a discrimination margin of at least 150 ms, and the island must survive with frequency within ±1 Hz and voltage within ±10%.
Held and varied: Held: battery state of charge 55–65%, inverter firmware v4.2, 120 kW resistive load per feeder, diesel offline, ambient 18–26 °C, the same fault-applying switch and arc-resistance emulation. Varied: the protection strategy, fault type and fault location. Eighteen cases, three repeats each.
The approach that failed, and what it ruled out: Configuration A was a second relay settings group with the pickup lowered from 1.25 to 0.35 times feeder rating and the time multiplier reduced — the strategy most protection engineers would try first. For a three-phase fault at feeder 2, the feeder relay operated at 480 ms and the incomer at 505 ms: a 25 ms margin against a 150 ms requirement, and on two of six repeats the incomer operated first and collapsed the whole island for a fault on a single feeder. Repeating with a five-times-higher CT burden gave the same times, which ruled out measurement error and CT saturation as the cause. What the failure ruled out was larger than a settings sheet: because the measured current at the incomer and at the faulted feeder differed by under 2%, no combination of pickup and time multiplier on a magnitude-graded principle can discriminate in this island — the source, not the fault position, sets the current.
The sequence that followed: Configuration B, voltage-restrained overcurrent, discriminated for bolted three-phase faults but not for the earth-fault cases: measured residual current was about 2 A, below any practical pickup, which located that failure in the island's earthing arrangement rather than in the relay logic. Configuration C, negative-sequence directional, failed for a reason not stated in any datasheet — the inverter held balanced current through the limit and the measured negative-sequence quantity was 0.03 pu, so there was nothing to polarise from. Configuration D, a communications-assisted blocking scheme over the existing fibre with a time-delayed backup, held a 210–260 ms margin across all eighteen cases with a maximum clearance of 290 ms, and the island survived every downstream fault.
Result and boundary: The knowledge produced was the eliminated principles as much as the surviving one. The earth-fault discrimination question was not answered — it was framed as a separate follow-on investigation into the earthing arrangement. For this example, the candidate experimental activity is documented from the formulation of the hypothesis and experimental approach through to the instrumented trials, evaluation and recorded conclusion. Outside it: installing the relays and fibre, the type tests, the distributor's witness test, periodic protection verification, and the microgrid operating normally afterwards. On these facts the fault-injection programme is a candidate core activity, subject to the full facts — not an eligibility outcome, which the company self-assesses.
What Is Generally Unlikely to Be Core on Those Facts
Activity
Why
Distributor's connection and witness testing
May fall within paragraph (f) where undertaken to comply with a statutory requirement or standard
Periodic protection verification on a maintenance schedule
May fall within paragraph (f)(iii) where it is routine testing and analysis rather than experimental investigation
Where an RSP Fits
An RSP is a scientific or technical service provider registered in specific fields that a company can engage to conduct R&D activities on its behalf (business.gov.au). On a microgrid, that work is upstream of commissioning: framing the technical unknown, designing the fault-injection programme and specifying instrumentation to the resolution the measure requires, so the experimental record exists while the work happens rather than being reconstructed from a commissioning log afterwards (record keeping).
There is also a threshold point for smaller sites: qualifying expenditure incurred to a non-associate RSP may still be taken into account in determining R&D tax offset entitlement where total notional deductions are below the usual $20,000 threshold, provided the services are within a research field for which the RSP is registered, and using an RSP does not guarantee eligibility — you still self-assess. Offset rates and the two tiers are covered in our article on the refundable and non-refundable offset; see also claiming R&D under $20,000, R&D for renewable energy and R&D Tax Incentive in Adelaide.
Talk to Ignition Research before the island transition is first attempted, while the trial can still be separated from commissioning. As a Registered Research Service Provider at Lot Fourteen in Adelaide, we design and conduct the experimental programme and produce the technical record while the work is happening. We are not a registered tax agent: your company self-assesses and remains responsible for its own claim. Get in touch.
Frequently Asked Questions
Q: Is designing microgrid protection an eligible R&D activity?
A: Generally not, as a core R&D activity. A fault study, a grading calculation and a settings sheet apply established methods with an outcome a competent protection engineer can determine in advance. A core activity may exist in the narrower case where no established basis predicts whether any discriminating scheme can work in the island — for example where a single current-limited inverter is the only source — and that can only be determined by a systematic progression of instrumented trials conducted to generate new knowledge. You self-assess.
Q: Why does inverter fault current break protection coordination in island mode?
A: A synchronous grid supplies fault current several times rated current, which is what lets time–overcurrent devices grade by magnitude and time. A grid-forming inverter instead limits its current to close to rated and holds it there, so in a radial island the current measured upstream and downstream of a fault can be almost identical. Where that is so, the fault magnitude no longer indicates fault location, and grading by magnitude cannot discriminate.
Q: Is anti-islanding testing eligible for the R&D Tax Incentive?
A: Testing an inverter's anti-islanding function against a standard's published test procedure is generally unlikely to be a core R&D activity on those facts, subject to the activity's own facts and the statutory tests: s 355-25(2)(f) excludes activities associated with complying with statutory requirements or standards from being core. That does not depend on the result being predictable — the unit may fail the test — it depends on the activity being associated with demonstrating compliance with a statutory requirement or standard. Investigating how existing standard-compliant inverters behave through an intentional island transition, where that behaviour has no established basis, is a different activity assessed on its own facts.
Q: Can seamless transfer development be claimed as R&D?
A: It is assessed as an activity, not as a deliverable. Configuring a controller's documented transfer modes applies a product as designed. Determining whether a defined transfer window can be held on a specific plant — where simulation demonstrably could not predict it and the answer comes from instrumented trials against a hypothesis recorded in advance — may be a core R&D activity. Whether the surrounding commissioning work qualifies as supporting depends first on whether it is directly related to a core R&D activity and, where s 355-30(2) applies, whether it is undertaken for the dominant purpose of supporting that core activity.
Sources & Further Reading
legislation.gov.au — Income Tax Assessment Act 1997 — Div 355, incl. ss 355-25 and 355-30
Related: R&D for renewable energy · what does not qualify · what an RSP is · claiming R&D under $20,000 · refundable vs non-refundable offset · R&D Tax Incentive in Adelaide · more Insights
This article is general information from a Registered Research Service Provider about the R&D Tax Incentive. It is not tax, legal or financial advice; eligibility depends on your circumstances and you should self-assess and seek your own advice.
Thinking about a project like this?
If you're weighing up an AI, software or technical improvement project and can't tell yet whether it's implementation or research, start with a quick read on where it sits.