Solar Energy in the Caucasus and Central Asia: The Grid Will Decide the Pace
Solar development is accelerating across the region. The harder work now lies in grid modernization, industrial integration, energy storage and getting projects connected—not merely announced.
Solar Energy in the Caucasus and Central Asia: The Grid Will Decide the Pace
By Lela Kurtanidze, CEO of ASTORIOS Holding Inc.
I have lost count of the meetings where forty minutes were spent discussing modules, inverters, and electricity yield, followed by five minutes on the connection point.
In the Caucasus and Central Asia, that order should usually be reversed.
A solar project can have excellent irradiation, competitive equipment, and an interested investor. None of it matters if the proposed substation has no available capacity, the transmission line is already constrained, or the connection conditions remain uncertain when construction is supposed to begin.
This is becoming the defining issue for renewable energy development across the region.
A Grid Built for a Different Model
Georgia, Armenia, and Azerbaijan, together with Kazakhstan, Uzbekistan, Kyrgyzstan, Tajikistan, and Turkmenistan, inherited electricity systems designed largely during the Soviet period. Those networks were not poorly conceived. They supplied cities, mines, factories, and heavy industrial operations for decades.
They were, however, built for a different model: relatively few large generating stations, centralised system control, and electricity moving in predictable directions toward major consumers.
Solar power changes that pattern. So do rooftop generation, private power plants, battery energy storage systems, and industrial facilities that want to generate electricity behind the meter. Power flows become less predictable. Consumers can become generators. A factory's changing production schedule can affect how much solar power its internal network can absorb at any particular moment.
Parts of the regional infrastructure are now being asked to perform functions that were never part of their original design.
That does not mean solar development must wait until every transmission line and substation has been reconstructed. It does mean that grid modernisation must run alongside the new generation. Modern protection systems, digital substations, SCADA, accurate forecasting, transparent connection studies, and clearly defined rules for battery energy storage are no longer secondary matters.
The Countries Are Moving at Different Speeds
Armenia
Armenia entered distributed solar earlier than most of its neighbours. By May 2025, installed solar capacity had exceeded 870 MW, with household installations accounting for approximately 485 MW. Solar generation had grown from almost nothing in 2015 to nearly 1 TWh in 2024.
That early growth created local experience and a real market. It also exposed the next layer of work: transmission rehabilitation, smarter distribution management, and the ability to accommodate a much larger number of generators.
Azerbaijan
Azerbaijan has taken a more centrally coordinated route. The 230 MW Garadagh Solar Power Plant was commissioned in 2023 with a dedicated 330 kV connection substation. More recently, Azerbaijan constructed the digital 330 kV Shafag substation specifically to integrate the 240 MW Shafag Solar Power Plant in Jabrayil.
There is a practical lesson in that. Generation and evacuation infrastructure were treated as parts of the same project, rather than dealing with the grid after the solar plant had already been developed.
Georgia
Georgia is at another stage. Smaller solar installations are no longer unusual, but the move into industrial-scale and utility-scale solar is only now gathering real momentum.
Georgia's electricity system benefits substantially from hydropower, but hydro production is seasonal, while demand - particularly industrial demand - does not always follow the same pattern. IRENA's 2025 assessment identified significant untapped solar and wind resources, while also pointing to grid integration, financing, governance, and local capacity as areas requiring further work.
On paper, Georgia's solar project pipeline is expanding quickly. In practice, the market will be decided by which projects have viable land, credible grid access, and customers or off takers able to use the electricity.
Industrial Self-Consumption
Industrial self-consumption deserves particular attention here.
A factory with a substantial daytime load may not need to export all - or any - of its solar generation. Electricity can be produced at the industrial site and consumed inside the facility. This reduces dependence on external supply and avoids adding uncontrolled generation to an already constrained connection point.
That sounds straightforward until the project reaches the commissioning stage.
At ASTORIOS, we encountered this directly while delivering the First Light Solar Power Plant for Rustavi Azot in Georgia. First Light is an approximately 24 MWp industrial self-consumption project covering 29 hectares, with 34,424 photovoltaic modules, 100 string inverters, and four substations.
It was engineered as a zero-export plant: the electricity generated is used by the industrial consumer rather than being pushed into the national grid.
The difficult part was never proving that the modules could generate electricity. The difficult part was making a large variable power source behave correctly inside an operating chemical plant.
Industrial loads do not remain perfectly stable. Production units stop and restart. Substations may be taken out of service. Consumption changes according to the factory's operating requirements - not according to the solar forecast.
When part of the factory load disappears, the solar plant must recognise the change and reduce generation fast enough to prevent backflow. That involves active power control, reliable communications, protection coordination, accurate metering, and an understanding of the customer's electrical system that goes well beyond installing photovoltaic equipment.
A small alteration in factory operation can matter more on that day than another decimal point in the annual yield model.
First Light is relevant because it shows one way of proceeding where grid-export capacity is limited. Industrial self-consumption and zero-export control allow substantial solar generation to be installed without waiting for the wider network to absorb the full plant output. The project was also prepared for future BESS integration to support load stability, peak management, and resilience.
This approach will not suit every industrial consumer. A business needs adequate daytime consumption, usable electrical infrastructure, and accurate load data. A factory operating intermittently cannot be modelled in the same way as a continuous industrial process.
Still, for manufacturers in Georgia and across the wider region, industrial solar offers a credible route to lower electricity exposure and reduced carbon intensity while national grid upgrades continue.
Central Asia Is Moving at a Much Larger Scale
Kazakhstan
Kazakhstan ended 2025 with 162 renewable energy facilities totalling approximately 3.5 GW, including 49 solar power plants. Renewables accounted for around 7% of national electricity generation.
Kazakhstan's geography makes transmission impossible to treat as an afterthought. Strong solar or wind resources may be located far from the major load centres. The commercial viability of a project, therefore, depends not only on production cost, but also on where the electricity must travel and what network reinforcement is required.
Uzbekistan
Uzbekistan has accelerated more aggressively. During 2025, it commissioned five solar plants totalling 1,413 MW and ten battery energy storage facilities totalling 1,245 MW. The same programme included 1,614 MVA of new substation capacity and approximately 420 kilometres of distribution networks.
That last point matters. Uzbekistan has not treated solar generation, BESS, and network infrastructure as entirely separate conversations. The speed creates execution risks, but the direction is correct: storage and grid investment have to accompany generation.
Kyrgyzstan, Tajikistan, and Turkmenistan
Kyrgyzstan and Tajikistan enter the discussion from a hydropower-heavy position. Solar can complement hydro production, particularly where water availability and seasonal demand do not align. It can also support remote loads and reduce pressure on constrained networks. Turkmenistan remains early in its solar development, although work is underway on a national solar roadmap, pilot projects, and local technical capacity.
Regional electricity trade may eventually become as important as individual national programmes. Today, cross-border electricity trade accounts for only about 3% of Central Asian demand. The World Bank's REMIT programme is intended to establish the region's first formal electricity market, strengthen transmission, and support large-scale renewable energy integration.
Where Projects Succeed or Fail
For developers, investors, and industrial consumers, the practical questions remain fairly basic.
- Where is the connection point?
- What happens when the consumer's load changes?
- Who is responsible for the grid study?
- Is the proposed BESS solving a measurable problem?
- Who will operate the plant after handover?
- And are the connection conditions firm enough to justify construction expenditure?
These questions are less attractive in a presentation than installed megawatts and carbon-reduction forecasts. They are also where projects succeed or fail.
My view is straightforward: I would rather commission 10 MW that the network and consumer can use reliably than announce 100 MW that never leaves the spreadsheet.
The Caucasus and Central Asia do not need more theoretical solar potential. They need connected plants, modernised infrastructure, and people capable of operating both.
