You have several different problems here, and - in truth - you need also to specify "what" accuracy (and tolerance!) - you really want.
And what expense you are willing to pay (schedule, cost, personnel, more delays, more adjustments, more instrumentation, much, much, much more adjustment methods and equipment construction time, etc.) for your entire project. All of your decisions (what instrument to use, where is your baseline, how did you determine your baseline and WHERE did THAT baseline come from, are going to come from your original requirements. Both of you are writing in a pressure vessel (NOT "rotating high speed equipment" forum, so I'm goin g to assume both of your are going to be 9had have been) connecting expensive but immobile pressure vessels to some other very expensive but immobile pressure vessel or pump, probably via a large number of pipe and pump and pressure vessel flanges.
When high-speed rotating high-mass equipment is the constraint, the "pipe-to-generator" or "pipe-to-turbine-flange" alignment is (almost) meaningless, compared to the rigid 1/1000 of an inch alignment of the entire bolted mass of HP-steam-turbine-rotor-flange-to-LP-rotor-to-LP-rotor-to-LP-rotor-to-generator-rotor-flange. And the turbine manufacturer's pay tens of thousands of dollars to shim and move the entire 200 foot long, multi-thousand ton around on very carefully constructed concrete pillars very carefully constructed inside of billion dollar power plants to allow them to do that. When a rotor or turbine casing or turbine shell is removed and replaced during outages, even more thousands are paid to laser-align all back to those 1/1000 specifications, even on "little" gas turbines only generating 150 or 200 Megawatts. An out-of-balance nuclear rotor at 1800 rpm can (and has!) destroyed the entire power plant. (Before lasers, micrometers and ultra-tight piano wire "tightline" measurement runs were used.)
But note the "solution": the power plants align rotors-rotors very, very carefully and at great expense, but then fabricate the pipe to "fit where the flange ends up at." The nuclear steam generators and pressure vessels and reactor pumps (or steam boilers and HP feed pumps) are built and located carefully of course, but with rulers and plumb bobs and surveying instruments and (ultimately) cranes and lines on concrete. Still, the pipes in between are built in place and welded up with pipefitters and welders making use of their "butt-weld gaps" and "final flange makeup" to prevent steam leaks. Ain't very glorious, is it?
But that is not specifically your problem, is it?
But that is not specifically your problem, is it?
You have a very expensive pressure vessel, probably very expensively built from iron ore and rust into "precisely" located flange points and bolt hole foundation plates on supports.
What part of the location problem is immediately critical to you?
Definition of the Location of the project reference point on the ground w/r to the rest of the refinery (?)?
Theorectical Location of the specific PV centerline point (CL AND elevation AND vertical axis) within the new construction of the plant?
Physically locating that theoretical CL on a concrete foundation?
Physically locating the mounting bolt holes w/r to the actual (or theoretical) location of that PV on the concrete?
Physically drilling the mounting bolts/anchor bolt pours into the foundation w/r to the actual drilled holes in the steel plates?
Lifting and locating the PV onto those anchor bolts?
Aligning it to perpendicular and to rotation after placement? (As mentioned above) the grating within a PV tower needs to be accurate as well, which cannot occur if the tower is not truly vertical. A 150 foot tower, out of plumb by only 0.1 degree, moves the top flange sideways by 3 inches.)
Or is the concern specifying a tolerance and accuracy for locating individual PV flanges to each other on that PV theoretical CL and elevation point?
Pipes for installation into existing plants for connecting any as-built equipment are normally ordered pre-fabricated to "plan", but left with one end for the field to cut, fitup and weld out.
Assuming you have an existing plant, you need to specify a formal surveyed point that everybody on the project agrees to as a "project reference datum" (x, y, AND z). THEN, you need to ensure ALL companies building equipment and designing eqpt and foundations and towers or supports within your project ALSO locate that reference datum on THEIR plans and drawings. AFTER these other design companies have that reference datum located, THEN they can begin drawing their specialized pipes, eqpt, plumbing, power or storage and service building dimensions. (Only IF and AFTER a common datum has been established, you can start assuming that most of the stuff will fit together when it arrives on site.)
On past projects, where no datum has been agreed to, I have had to do that manually by measuring from the middle of steel beams and from the center of columns to figure out column lines, then plotting those lines outside the building through holes in the walls to draw lines on concrete pads to place transformers and piping with carpenter squares and measuring tapes. (And yes, everything did "fit" once the competing companies did get a reference point on the concrete. No, they had already started fabrication in several cases; and yes, they were charged for their rework and scrap. Whose fault? The original project manager.)
A surveyor will do the same, his "perfect" PV centerline will be drawn within 1/16 of an inch of where he believes the true reference is. The bolt holes for anchors will be drawn on that concrete pad to within a 1/16 (maybe) of that theoretical CL. The actual holes will be drilled in that concrete within 3/32 of those theoretically-drawn bolt hole centers. The PV will be lifted up, and placed around those anchor bolts, then aligned as carefully as possible to be plumb and vertical, BUT at that time (when the PV is in mid-air and coming down), neither the anchor bolts nor mounting plates can be moved any more: At best, you can only hope that the bolts go into the holes. But a carpenter's level is going to be their tool. Pry bars and chainfalls are their "persuaders" to pull a PV around as it gets close. Yo hope sledge hammers or forklifts are not needed. Shims under various anchor plates is their method of aligning it to position and holding it in place. Grout is going to be inserted to fill the gap between pad and concrete.