The Zwietering Correlation is a good start for determining an off-bottom conditions for solids suspension.  A number of other studies have found the correlation to be quite adequate, although other methods of expressing the relationship are easier to use.  Of course large tanks, 500,000 gallons, are further from the test results and predictions will be less accurate than for smaller tanks, but the correlation has been successfully used to design large tanks.
 
A bigger problem than particle size may be particle density.  Most of Zwietering's tests were with sand and glass, with densities in the mid range (sp.gr. 2 to 3).  Low density polymer particles and high density (sp. gr. > 6.0) mineral particles may not be as well represented.  A wide distribution of particle sizes, also needs to include a wide difference in densities.  A rough guide used in the mineral processing industries is to design for the 85%ile particle, but they plan to dig out the tank every few years as part of routine maintenance.  If you don't want any settled particles, you probably need to design for the largest, densest particle, with the emphasis on densest.
 
Uniform suspension is a completely different story.  The Zwietering correlation only tells you if you can get everything off the bottom.  If the particles settle slowly it takes only a modest increase in mixing intensity for uniform suspension.  If the particles settle rapidly, the amount of mixing intensity must increase by a large amount.  In the case of very rapidly settling particles and moderately tall tanks, uniform suspension my be nearly impossible.
 
In most cases, with the complexities you describe, the only way of determining how much power or mixing intensity you need will require testing, with geometric similarity, followed by scale-up.  Not all characteristics of solids suspension are handled by simple correlations.