If I know more about the attack vectors maybe I can help more or maybe if there is a list of attack vectors for the community to research that would work. But my guess is the concerns are hardware and software vulnerabilities. Software you can solve by formal verification and correct by construction to make it bug free. Hardware it really depends on what you can afford and how determined the adversary is. Economic incentives can also be put into play to improve security. Incentive centered design can take care of that part using smart contracts:
https://www.usenix.org/legacy/event/hotsec06/tech/full_papers/wash/wash.pdfAren't we also talking about emulation of the actual device too? Plug in an emulator pass through behind your TV or Computer or every outlet in your house. Claim to have solar panels and then gain reward of the $20 per SLR price levels that everyone will be buying and vying for pushing the price to $40 then $60 upwards to $400 a coin to compete with BitCoin even though it has to sell to Bitcoin, the selling could actually hurt Bitcoin markets so SLR will have an edge there. Seeing that there will be Billions of dollars of SLR dumping on Bitcoin and lowering the price it gives SLR an opportunity to catch up

That is to say that people will be buying Bitcoin at that point when they could just have an epiphany and use their money to buy Solar and then Claim and then Sell SLR for Bitcoin. At that point we have another problem no one will be buying Bitcoin to sustain the sale of SolarCoin

Or perhaps they don't have to buy Solar and just buy the emulator to plug into every outlet in their house in order to gain the $400 SLR Coins. They still get to use the power they emulate too as a pass through since it only monitors.
So SLR price should sky rocket and Bitcoin price will dive since it has to provide the capitol for the exodus of Bitcoin from the sale of the expensive SLR. Make it happen!
Each device would have to have a cryptographic identifier and you can secure these devices from emulation to a certain extent but the question is at what cost? Even now people can find ways to leech free electricity if they are willing to go through extraordinary effort. What matters is how much fraud is there and is it beyond acceptable levels? At what point is it worth the cost to do something about it?
This is a bit like the botnet problem in Proof of Work mining.