There is two PLC's that lose programming at the same time a few times a month. There are feed by separate transformers on the same utility service. The utility service is on the same service as variable speed wind generator farm and can be fed from two different utility substations. The only other way that these PLC's are connected is through the communication fiber optic lines and obviously earth ground. Any Suggestion would be appreciated.

Some PLCs use battery backed memory (no indication what units you have) and a power glitch that affects both units would cause that problem. Not sure there are many other events that would cause that impact to separate PLCs at the same time.

Which plc's are you using?

I ran into that problem before with 1 PLC loses its program. So if 2 PLC (in the same cabinet) lose their program at the same time, I am pretty sure that is your problem. You have a huge electromagnetic noise surge inside your cabinet. That happens when you have a dry contact opens on a heavy inductive load. So you can cure the problem by connecting appropriate surge protection across that load.

Are the PLCs running the identical programs? If so do the PLC use any commands that 'mirror' or copy registers between each PLC to the opposite PLC? Are you sure the program is 'gone', or is the PLC simply no longer running? I have several years of experience running dual fully redundant PLC in a substation environment, and to my knowledge we never had both PLCs fail or stop running simultaneously. Some differences, ours were DC powered from a station 24 VDC battery, and all 125VDC controls were optically isolated from the PLCs (inputs and outputs).
Some things to consider, is there something in the PLC code itself that is causing the problem, for example as specific set of active inputs causes a branch to a non-existent subroutine? Since you are using AC power, can you install a UPS to eliminate noise, spikes, and short interruptions in the AC supply? Many PLCs have 'hidden' connections, for example the inputs or outputs will share a common connection that is not obvious from the instruction manual or connections. So you may have another connection between the PLCs you are not aware of. Are all of the inputs and outputs isolated from each other?
Check configuration switch settings, as mentioned previously, most PLCs have a battery to retain the program, there is typically a set of switches that tell the PLC what to do when power is lost and restored, these may be set wrong.
Finally, add some minor programming that does not do anything (a counter, blinking light, etc.) to only ONE of the PLC and see if this changes the operation of both failing at the same time.
Just my thoughts, good luck.

Thanks for the responses. These PLC's are in separate buildings separated by approximately a 1/2 mile. One of the PLC's is in a cabinet with several identical drives that do not fail. Like I said before the only thing that connects them together is Utility power feed at 24,900V. They are feed by separate transformers and the only thing that I can think of is that if they had a sag in there voltage to a specific amount could that cause to PLC's to fail. Another engineer is running with the Idea that it is Solar Flares, because the last time it happened there were high numbers of solar flares. Not sure if that is the right way to go.

I have never seen a solar flare able to interfere, as the required cable length to get a suitable signal is typically miles long. However, this brings up an interesting idea. Are your input, output, and power supply cables shielded? For input and output leads, I would expect single point grounding at the PLC. Check your PLC ground connections, at least one manufacturer had very specific, and somewhat confusing grounding requirements. If you have multiple grounds, this may be part of your problem.

Also, when you say the programming is "lost", is it truly gone from memory, or is it simply no longer running? If it is not running, look into noise getting into the PLC via any common (both PLCs) conductive connections, including ground. Put one of the PLCs temporarily on a UPS with the ground ISOLATED from all remote grounds, and see if this helps.

More information about your installation and facility may help, as well as any info if you have tried previous suggestions.

Dave,
We have mentioned to the customer that they should have shielded cable with proper grounding. We have proposed to fix ground loop issues that they currently have. Yes they have a single point ground and the manufacture requires that the motor ground goes to the drive then the drive is grounded to the bus. We have also proposed line reactors on their VFDs to reduce the systems THD (Current THD 80%). Now we wait and see what they say. Thank you for the response.