You said "has been achieved in an armed conflict". That's an exaggeration because 'armed conflict' is so very broad, and includes bloodless wars. You've qualified it now with "has been achieved in long term occupation by peacekeepers." I agree that if you narrow the definition enough, you can make it be correct.
(I should have included the Falklands War as a war with few innocent/non-combatant civilian deaths compared to military deaths. Argentina lost 649 people as part of their military force, the UK 255, and three Falklands civilians died, from friendly fire.)
There's little need to look towards Soviet occupation for a comparison. We did long term occupations of the Dominican Republic (1916-1924) (http://www.dtic.mil/dtic/tr/fulltext/u2/a491390.pdf is the best document I found on the occupation; most other references are to the 1965-1966 occupation) and Haiti (1915-1934), which was publicly described as a mission to 're-establish peace and order'. We also controlled the Philippines from 1902 to 1946. There was insurgent opposition in all three places. By comparison, the US response in Iraq is indeed much better than General Smith's order to "KILL EVERY ONE OVER TEN" as part of the American atrocities in the Philippines.
In any case, when you talk about long-term peacekeeper forces, you need to include things like the UN peacekeepers on Cyprus, who have helped prevent sectarian conflict between ethnically Greek and Turkish Cypriots. And done so with very few casualties. (Go ahead and add another qualifier, in that the UNFICYP has only 1,000 people.)
But put all that aside. When you write "this rate of collateral damage is far better than has ever been achieved in an armed conflict" ... how do you know? How do you measure collateral damage? Whose numbers do you trust, and why?