The project had obvious flaws.
In an interview with China Daily, Wu said the satellite’s mirror-like exterior would reflect sunlight down to Earth, creating a glow about eight times brighter than the moon. The artificial moon, which he said would orbit about 500 kilometers above Earth, could save $174 million in electricity from streetlights.
For one thing, you can’t place a satellite over a Chinese city. Period. Orbital dynamics do not work that way. In low Earth orbit (500 kilometer altitude) the satellite moves at greater than 7000 meters a second, while the Earth rotates at only 460 meters per second. That is a large mismatch. You would only be able to see the satellite from Chengdu for a few minutes at a time, and probably no more than every few nights.
OK, so, let’s not believe that the press release is accurate and try to design the orbit and satellite. You could put the satellite at 35,786 km altitude, and in that case, it goes around the Earth at about the same speed the Earth rotates. But there are two problems. First, the mirror is over 35,000 km away, a second, it still can’t stay put. You see, every satellite has to orbit around the center of the Earth, not some arbitrary point on the surface of the Earth.
Here is what happened when Japan tried to put a satellite over Japan. The satellite does indeed spend time over Japan. But also over Australia
So, by using multiple geosynchronous satellites, you could have a least one of them over China on a given night. On the other hand, it doesn’t have to be over China. If it sat at the equator, at that distance, it could still reflect toward China.
But what about the 35,000 km thing? Well, assuming you have to illuminate no more than your city, let’s say a 20 km radius, you need about a half milliradian beam radius from your mirror. But, you can’t do it. Why? The Sun is has a divergence full angle, seen from the Earth of about 8.5 milliradians. That means that at a distance of 35,000 km you are going to illuminate at a bare minimum a 300 km diameter circle on the ground. There is no zoom lens bull s**t that you can invoke to fix the problem. (The reason is probably worth several Quora questions to answer it in detail, but it violates some well known laws in optics.) If you want to have light 10 times as bright as the full Moon light, the mirror will need to provide 1/40,000 of a “sun” as we call it in the space business. A “sun” is a solar constant of 1366 watts per square meter, about 500 watts of which is visible light.
That means that you need a mirror in space that has 1/40000 the area of the 300 km diameter spot that is being illuminated, or 1/200 the diameter. That is about 1.5 km. I will just note that the only mirror larger than 2.4 meters to be launched is the JWST, which has taken 15 years longer than planned and the cost has been capped by Congress at more than 16 times the original budget. (And it has not been launched yet.)
A 1500 meter (roughly a mile) diameter mirror is not feasible at this time. A mirror closer to the Earth would be a reasonable size, but would only be in the night sky for a few minutes, and would not be visible from China on every orbital pass. We have looked at inflatable mirrors, but they have not gone beyond TRL 2 or TRL 3. I spent a lot of time doing risk management and Technical Readiness Levels (TRL), and some time with Program Readiness Levels (PRL) and Manufacturing Readiness Levels (MRL). The Air Force, the Army, and NASA got together to assign these readiness levels to avoid starting programs that had little to no chance of success.
These are problems that even a moderately experienced space company would have been able to foresee and makes me wonder whether this was ever a real project at all. When I first read of it in 2018, it only took about 5 minutes to prove that it could not possibly work without a group of 1500 meter diameter mirrors. Actually, one mirror over the equator would work, but it would still be 1500 meters in diameter.
Note: the meteor crater is the size of the needed mirror in space.
Footnotes