There are two new studies ongoing with the aim to investigate the feasibility of generating solar energy from solar panels borne by satellites. I can see the reasoning here: at the top of the atmosphere, the amount of the Sun's radiation falling onto the Earth amounts to about 1,400 W/m^2 (Watts per square meter); however, the air-molecules absorb various wavelengths of light and so at any point at ground level, the intensity of radiation is attenuated. In northern Europe a reasonable estimate is around 150 W/m^2, somewhere quite sunny it is about 350 W/m^2 and in the Sahara Desert the energy is not far short of 1,000 W/m^2. Hence, out in space the full complement of the Sun's power might be harvested.
In this particular concept being explored, a network of orbiting satellites would serve as energy transducers, capturing the abundance of energy from the Sun and sending it down to Earth in the form of a laser or microwave beam. (Mmmm... I wonder what the Health and Safety people would have to say about that?) . The original notion originated in the mind of Peter Glaser, a scientist working at Arthur D. Little in Cambridge, Massachusetts in the late 1960's. No one is in any doubt that there are difficulties and drawbacks attending this particular type of technology, which smacks of "Star-wars", and it is debatable just what the real economic benefit would be in practice.
The challenges of Space Solar Power (SSP) are various, but while concluding that the launching costs were too high, NASA (in 1995) decided to take a fresh look at the technology, possibly to attract financial and public support for its future activities. Since then interest in SP has accumulated, and a special study group at the National Research Council (NRC) has singled out several technological advances, which might be worth following-up on: (a) improvements in the efficiency of solar panels and the fabrication of light-weight panels, (b) progress in wireless transmission on Earth, notably in Japan and Canada, (c) robotics, deemed essential to run such an SSP assembly, has demonstrated significant improvements in terms of manipulators, machine vision systems, hand-eye coordination, task-planning and reasoning.
The panel concluded that significant breakthroughs are necessary before SSP technologies could produce energy in a cost-competitive way compared with Earth-based power generation. The final success of generating power on Earth from satellites depends "critically on 'dramatic reductions' in the cost of transportation from Earth to geosynchronous orbit (i.e. the altitude at which the orbiting period of the satellite is the same as that of a point on the Earth's surface, meaning that from an Earth-bound perspective, the satellite stays in the same place at all times... you wouldn't want it drifting off would you?!"). It further concluded the need for ground-based demonstrations of point-to-point wireless power transmission, and also the 'desirability' of ground-to-space and space-to-ground demonstrations... i.e. potential "Star-Wars" weapons.
We shall see, and I would rule nothing out as the world clutches ever more desperately at means to avoid drowning in the "Oil Dearth" era.
"Bright Future for Solar Power Satellites," by Leonard David. http://www.space.co./businesstechnology/technology/solar_power