The dream of space photovoltaics will become a reality. Elon Musk: Build a space AI computing center. Photovoltaic tycoons: The same solar panel can generate 7 to 10 times more electricity in space than on Earth.

In the vast expanse of space, huge arrays of photovoltaic panels silently spread out in low Earth orbit. Tens of thousands of solar panels are neatly arranged in a strict geometric formation, resembling an endless "wheat field" floating in the vacuum.

This kind of space photovoltaic scenario seems likely to be seen in the near future. Recently, Elon Musk proposed a plan to deploy 100GW (gigawatts) of photovoltaic capacity in space each year, instantly triggering a buzz around the space photovoltaic concept in the A-share market.

Recently, Li Xiande, the chairman of JinkoSolar, stated in his New Year's speech that for the same solar panel, the average power generation in space is 7 to 10 times higher than that on Earth. It overcomes the bottlenecks of intermittency and degradation, and in principle, it can operate and generate electricity continuously.

Gao Jifan, the chairman of Trina Solar, also mentioned in his New Year's speech that in the new year, Trina Solar will accelerate the mass production and commercialization of perovskite technology, and kick-start a new era of space photovoltaics and interstellar computing power.

Both big names in the photovoltaic industry are optimistic about space photovoltaics. Perhaps, this seemingly science - fiction concept of space photovoltaics is not as far - fetched as it seems. On the last day of 2025, Fan Bin, the chairman of GCL - OPV, and Wen Yanjie, the founder of Guangyin Technology, were respectively interviewed by reporters from NBD. Both perovskite manufacturers believe that space photovoltaics have great potential, but they are still restricted by factors such as the cost of commercial space transportation.

The Potential of Space Photovoltaics

At the beginning of the new year, space photovoltaics are becoming a new market hot - spot.

So, what advantages does space photovoltaics have over ground - based photovoltaics? Currently, ground - based photovoltaics can achieve grid - parity. However, due to the intermittent nature of power generation, energy storage systems are necessary to meet electricity demand.

In contrast, the power generation of space photovoltaics is much more regular. Ground - based power plants mainly rely on the peak sunlight at noon. How does space photovoltaics generate electricity?

Fan Bin, the chairman of GCL - OPV, told NBD reporters: "Taking low - orbit satellites as an example, they orbit the Earth every few dozen minutes, experiencing alternating light and dark periods. Therefore, space photovoltaics also require energy storage, but its energy - storage cycle is very stable and doesn't need to consider factors like weather. On the ground, we need to have extra capacity to deal with extreme changes, which is not necessary for space photovoltaics."

The reason why Elon Musk wants to build a large - scale space photovoltaic system is the growing energy demand of his Starlink project. China is also starting to build its own low - orbit satellite constellations.

CITIC Construction Investment Securities believes that low - orbit satellites and satellite internet constellations are currently the core application scenarios for space photovoltaics. China's low - orbit satellite constellation construction is unleashing huge market potential. Currently, six giant constellation projects have been planned, including communication backbone constellations such as "State Grid" and "G60 Qianfan Constellation", as well as commercial constellations in niche fields like "Geely Future Mobility Constellation" (for vehicle - networking positioning), "Tianqi" (for Internet of Things data), "Honghu - 3" (for broadband communication), and "Three - Body Computing Constellation" (for on - orbit computing power). The total planned number of satellites exceeds 50,000.

Fan Bin estimated: "For example, a satellite launched by SpaceX requires several square meters of photovoltaic panels. Assuming an efficiency of 30%, with 300 watts per square meter, a single satellite may need one or two kilowatts. Ten thousand satellites would then need more than ten megawatts."

Even when calculated based on 50,000 satellites, the demand is relatively small compared to ground - based photovoltaics. However, Fan Bin also said: "Satellites don't use a large amount of photovoltaic panels, but due to the need for extremely lightweight and ultraviolet - blocking structural designs, the value per watt is relatively high. Take crystalline silicon as an example. A crystalline silicon component on the ground costs only 0.7 yuan per watt, while a space - grade crystalline silicon component costs dozens of yuan per watt."

In the future, with the construction of space data centers, the demand for space photovoltaics will grow exponentially. Wen Yanjie, the founder of Guangyin Technology, told NBD reporters: "Space computing power requires a large amount of energy, and the sun is an excellent nuclear fusion device. We can absorb solar energy and convert it into electricity to supply space computing power. After the computing center finishes its calculations, it only needs to transmit the results back to Earth."

Regarding the reason for deploying space computing power, Wen Yanjie said: "Firstly, the space in space is almost infinite; secondly, most of our energy comes from the sun. Why not deploy the computing center in space to directly obtain energy from the sun, which is more efficient; in addition, ground - based computing centers need various cooling methods, such as air - cooling and liquid - cooling. However, the temperature in space is very low, so the demand for cooling is less, and the overall energy consumption is reduced."

Due to the many advantages of space data centers, both domestic and international parties are actively planning and investing in their construction.

Domestically, the Beijing Space Data Center plans to deploy a gigawatt - level system in the dawn - dusk orbit at an altitude of 700 to 800 kilometers, advancing in three stages: from 2025 to 2027, a computing power constellation with 200KW/1000POPS will be built to achieve the application goal of "space data processed in space"; from 2028 to 2030, the second - phase project will be promoted to realize the commercialization of "ground data processed in space"; from 2031 to 2035, satellite mass - production and on - orbit docking will be completed, and a large - scale cluster will be established.

Internationally, Elon Musk has proposed the concept of a space AI computing center, planning to deploy 100GW to 500GW - level solar AI satellites using Starship rockets.

Can Perovskite Break Through?

Currently, space photovoltaics mainly use gallium arsenide and crystalline silicon products. In addition, there is also a small amount of perovskite product application. Perovskite is regarded as the most promising technological path and the future of space photovoltaics.

CITIC Construction Investment predicts that in the short term (from 2024 to 2027), triple - junction gallium arsenide batteries will dominate high - value communication satellites, deep - space exploration, and other scenarios; in the medium term (from 2026 to 2030), P - type HJT (heterojunction) batteries have better radiation resistance and lightweight performance among existing mass - production technologies and are expected to gradually penetrate into short - term low - orbit missions; in the long term (after 2028), perovskite tandem batteries will accelerate breakthroughs due to their high specific power advantage.

So, what are the advantages of perovskite?

Fan Bin said: "Gallium arsenide is very expensive, and its cost has not been effectively reduced for a long time. Therefore, it is very likely that space photovoltaics will shift to perovskite in the future."

He added: "Reusable rockets have significantly reduced the satellite launch cost. Against this background, the expensive gallium arsenide batteries stand out. The price of gallium arsenide batteries per square meter is estimated to be between 200,000 and 300,000 yuan, and the price per watt is about 1,000 to 2,000 yuan, which is more than 1,000 times that of ground - based crystalline silicon batteries."

In fact, apart from the price factor, gallium arsenide outperforms crystalline silicon batteries in many aspects such as light - to - electricity conversion efficiency and radiation resistance.

Fan Bin said: "Gallium arsenide has almost no other drawbacks except for its high cost. The drawback of crystalline silicon is that its light - to - electricity conversion efficiency is not very high, and the conversion efficiency of gallium arsenide is more than 20% higher than that of crystalline silicon. For satellite applications, high efficiency is very necessary. In addition, gallium arsenide has strong radiation resistance. Crystalline silicon batteries have a low tolerance for impurities, so they will degrade quickly in space. Gallium arsenide batteries degrade significantly less than crystalline silicon in space."

Perovskite combines the advantages of gallium arsenide in terms of light - to - electricity conversion efficiency and radiation resistance. Fan Bin said: "Currently, the light - to - electricity conversion efficiency of perovskite tandem cells in the laboratory has approached 35%. At the same time, perovskite is a thin - film material, and its radiation resistance in space will be similar to that of gallium arsenide. Moreover, perovskite is relatively inexpensive."

Wen Yanjie also said: "On the one hand, perovskite can achieve a very high photoelectric conversion efficiency; on the other hand, perovskite batteries are relatively light and can be made flexible. In addition, perovskite also has very strong radiation resistance."

The demand for space photovoltaics is vast, and there are high - quality products like perovskite gradually maturing. However, some industry insiders told NBD reporters: "Space photovoltaics must withstand extreme temperature differences of 300 degrees, strong radiation, and atomic oxygen, which pose a severe test to the battery life. To achieve the ambitious scale of 100GW per year, the component cost alone may reach hundreds of billions of dollars, not including the sky - high launch, construction, and on - orbit maintenance costs. Space photovoltaics is worth imagining. As the ultimate complementary solution for ground - based energy, its commercialization may become a reality with the development of commercial spaceflight and new battery technologies."

In fact, both Fan Bin and Wen Yanjie believe that the launch cost of commercial spaceflight will affect the future development of space photovoltaics. Only when the satellite launch cost is low enough can the dream of space photovoltaics become a reality.

This article is from the WeChat public account "NBD". Author: Zhu Chengxiang, Editor: He Xiaotao. Republished with permission.