Drew is right about checking the system file integrity for Windows. While you haven't picked up on bad sectors using RAID-0, it is not uncommon for some minor problem to take place that throws a RAID-0 completely offline. I would highly suggest using this method to check and see if Windows system files were damaged. The chkdsk indicates very clearly that while no bad sectors were detected (this would mean parts of the drives are unusable), a system crash, power failure, or some other I/O error created problems with the file system itself. (Remember, that the file system is an artificial construct that simply existed on the physical hardware).
Any weird noises coming from the drives must be viewed as a sign of potential failure. Verify the image backups and make sure that if the system goes down nothing will be lost. While the array status is still good, you are seeing data indicating to you that at one point an error did occur with both drives. However, out of some luck, the RAID-0 is still in tact. How long that remains that way is unknown. You need to consider carefully the next steps you take.
I would immediately backup everything using an image backup utility.
I would run sfc /scannow in the command prompt.
I would check the S.M.A.R.T status on each drive before replacing unless they are making unusual noises.
On the type of Intel RAID array you have, it is possible to preserve the array back and forth between motherboards on occasion. I have accomplished this by moving a few drives over from one Gigabyte board with a ICH controller to a different model and it picked up on the RAID-0. This primarily because you are striping the drives and not really creating a redundancy, as far as I can tell. Still I would not count on it, and if you replace both drives, you should be prepared to fully restore a complete image backup of the system.
Image backups can be created with Windows Backup (although somewhat unreliably), but are certain to be created properly with Acronis True Image Home or Paragon products.
The noise and the error state seem to be the major problems here. How the RAID is still functioning when the drives posted errors is unusual, and likely the result of a lot of luck in this situation. With RAID-0 there is no redundancy to save a drive that has an error. If the error occurred simultaneously on both drives that could explain why there is no significant data loss or loss of the RAID stripe. Remember that in order for RAID-0 striping to exist properly, both drives must be in constant operation to split the data as it is written and one can not afford to be disconnected off the ICH for even a second when these writes are taking place.
I would also look at moving away from RAID at this time, if an upgrade is to be performed. Consider VelociRaptor 10,000RPM drives, hybrid SSDs, or pure SSD drives. Do not run them in RAID and avoid the problem of lost parity altogether. The need for constant high speed performance will be negated when you start to see how good the newer drives are at performance. At this point in time the VelociRaptor drives are overpriced in comparison to their counterparts (Hybrid SSD and pure SSD drives). For a replacement you may want to avoid the headache of RAID unless you are willing to make image backups all the time.
The installation of Windows 7 to restore your files will not take place as you think. You should have a recovery disc and external media to store a complete image of your data and run this disc to bring your data back onto the new hardware. Installing Windows itself and then trying to bring programs and files back will not work entirely: you'd be able to get files off the backup but you would lose software and licensing and that stuff would need to be re-installed. This is why image backup is usually the best way of handling the situation, or a clean install.
If you've had the noise since you bought the drives, its quite possible the acoustics are just bad on the drives. But it looks like something happened that disconnected both of the drives briefly, perhaps even for a millisecond, which caused error detection on the RAID post screen. This should prompt an immediate full image backup. Remember: FULL IMAGE BACKUP and not just a backup of critical files or folders.