An island of 45 households becomes PLN's first technopark and cuts its cost of supply by a quarter

PT PLN inaugurated the Green Energy Ecosystem Technopark on Pulau Rengit on 22 August 2026. The island sits in Pegantungan Village, Badau District, Belitung Regency, in the Bangka Belitung Islands. It is home to roughly 150 people in about 45 households, most of them fishing families. Reports differ slightly on the household count, with one giving 44 and others 45.

Before the project, the island ran on two diesel generating sets with a combined 100 kilowatts, and electricity was available for 12 hours at night. It now runs for 24 hours.

The cost number is the one to read first

Attention naturally goes to the shift from 12 hours to 24. The figure that carries more weight for policy is what happened to the cost of supply.

The cost of providing electricity on Pulau Rengit, the biaya pokok penyediaan or BPP, fell from Rp16,700 per kilowatt hour to Rp12,300. That is a reduction of roughly 26 percent, achieved while doubling the hours of service.

This combination is unusual and worth pausing on. In most systems, extending service hours raises total cost, because more fuel is burned for more hours. Here the opposite happened, because the hours were extended using solar, storage and wind rather than more diesel. Diesel on a remote island is expensive not primarily because of the fuel price but because of what it takes to deliver that fuel to the island and keep small generating sets running there.

Doubling the service and cutting the unit cost at the same time is only possible when the additional hours come from a different fuel than the original ones.

The project is also credited with avoiding roughly 99.5 tonnes of carbon dioxide a year. On an island this size that number is necessarily small. The ratio it represents is the part that scales.

Four technologies in one small system

The technopark integrates four generating and storage technologies rather than relying on a single source.

The solar component is an agrivoltaic array of 80 kilowatt peak, mounted on elevated structures across roughly 800 to 1,000 square metres, with cultivation continuing on the ground beneath the panels. Battery storage is provided in two units of 100 kilowatt hours each, which carries supply through the night. Wind turbines of both vertical and horizontal axis design are installed, selected to operate at low wind speeds rather than the higher speeds large turbines require. A hydrogen fuel cell completes the set.

The agrivoltaic choice deserves noting on its own. On a small island, usable flat land is limited and already contested between settlement, cultivation and fishing activity. Raising the panels so that crops continue to grow beneath them means the generation does not take land from food. The principle is the same as floating solar on reservoirs and rooftop solar on commercial buildings, which is to place generation on space that is already in use rather than demanding new space.

The reason for combining four technologies on a system serving 45 households is not redundancy for its own sake. Each covers a different failure mode. Solar produces nothing at night and little under heavy cloud. Batteries cover the evening but deplete. Low speed wind can generate when solar cannot, including during overcast and monsoon conditions. A fuel cell provides dispatchable output that does not depend on the weather at all. On a small isolated system with no interconnection to draw on, that layering is what makes 24 hour supply credible rather than aspirational.

A laboratory as much as a power system

Edwin Nugraha Putra, Director of Technology, Engineering and Sustainability at PLN, described Pulau Rengit as the first technopark that PLN has formally operated, and framed the inauguration as a historic moment for the company's work on national electricity self sufficiency. He also said the system would be evaluated to assess the scope for applying and replicating it in other areas.

That evaluative purpose is built into the design. The facility carries research partnerships with Gadjah Mada University on fuel cell development, Sebelas Maret University on batteries, and Bangka Belitung University on agrovoltaic methods. It is intended to function as a research and learning site for integrated green energy systems in archipelagic conditions, not only as a power plant.

The inauguration was attended by Hidayat Arsani of Bangka Belitung, whose office is given differently across reports as provincial governor and as Belitung regent, alongside officials from the Ministry of Energy and Mineral Resources and representatives of the National Energy Council.

The economic effect on the island is visible in the consumption figure. Household electricity use rose from 70 to 172 kilowatt hours. That is not simply people using more light. Round the clock supply makes cold storage possible, which allows fishing households to hold a catch rather than sell it immediately at whatever price is available. The agrivoltaic layout adds cultivation on the same ground the panels occupy.

Sirkularium's read for government and public institutions

Three observations follow for policy and programme design.

The first concerns where renewable energy is most immediately economic. Remote island systems running on diesel have the highest cost of supply in the country, which means they are where clean energy displaces the most expensive alternative. Rengit shows a 26 percent reduction against diesel. Programmes that prioritise the highest cost systems first will deliver larger fiscal savings per unit installed than programmes that begin where grid power is already cheap, and the subsidy line responds directly to that choice.

The second concerns evaluating access by hours rather than by connection. A household counted as electrified while receiving 12 hours of supply is recorded the same way as one receiving 24, yet the economic difference between them is substantial, as the jump from 70 to 172 kilowatt hours indicates. Cold storage, refrigeration and any activity requiring continuity are impossible on a 12 hour system. Measuring service duration alongside connection counts would give a more accurate picture of where electrification is actually enabling economic activity.

The third concerns the technopark model itself. Pairing an operating installation with university research partnerships and an explicit replication assessment costs little more than building the installation alone, and it produces transferable knowledge rather than a single asset. For a country with thousands of inhabited small islands facing broadly similar conditions, that is a sensible way to spend the first project in a series. The question is whether the evaluation is resourced and published in a form other operators can use.

What to watch next is the published evaluation of the Rengit system, whether the cost reduction holds across a full year including the monsoon, and whether the model is replicated at other island sites with the research component intact rather than stripped out for speed.