@Christian Wolinski You are conflating two distinct definitions of constant (noun):

1. A symbol which for the purposes of a specific problem or solution process is not to be considered a variable or an independent variable.
2. A symbol representing a specific number or other fixed, well-defined mathematical entity.

When you speak of redeclaring constants in the context of a solver, you are using the first definiton. However, the predefined (yet modifiable) global constants refers to the second definition. There's no need to mix the two. Maple lets you create your own environment variables to your heart's content. So, if you want to use _EnvConstants for the solver purpose that you describe (which I agree is potentially useful), you may do so. 

@ondrejkubu The only part of that process that doesn't seem totally automatic to me is choosing which independent variable to declare in the identity clause. Let me know if you have any trouble with that. 

@Christian Wolinski I do think that there should be a built-in type var equivalent to And(name, Not(constant)). I get tired of typing that. 

Thank you for the additional information about your problem. Now I understand it, and I'm about to post a new Answer below.

@AzU  You've done a despicable thing by deleting your question after I helped you. It's like you spit in my face. I suspect that not many people here would be willing to help you if they suspected that you were going to delete the question. 

@ActiveUser Again, the example that you show is not a permutation. It doesn't have any representation in disjoint cycle form. When a permutation is applied to a list, the result is a list with the same elements as the original, each occuring the same number of times as in the original; only their order may be different.

Neither list that you show is itself a permutation, nor are those lists permutations of each other.

@AzU I have your problem solved, but I'm not at a computer right now. I'll post it soon. But you really only need to do a few simple things to get there: 

1. Change Pel to P[el].
2. Get rid of the units, like I said before.
3. Add the numeric option to dsolve.

@AzU Ah, so T[a](t) is constant, not really a function of t. That makes things much easier. But what about P[el](t)? Certainly you'll get no response from dsolve without specifying it also.

Do you have an initial condition, for example T[ovn](0) = 400?

Although it is possible to include units directly in the problem specification, I think that you should avoid that for the time being, until you get a bit more experience with Maple. Just use compatible and consistent units and keep track of them mentally.

The Question area (this area) is the appropriate place for this. I deleted your duplicate of this from the Post area.

@Rouben Rostamian  Vote up. Excellent graphic presentation of an Answer. Nice abstraction through use of [...]. Nice use of vertical and horizontal shading to show direction of integration (this, plus some arrows, is how I teach double integrals).

@28th You're welcome. I'd guess that it's nearly certain that that solution is a transcendental number other than a rational multiple of Pi, and it has no symbolic representation other than "it is what it is". Proving things like that for specific numbers is notoriously difficult, yet it's easy to prove that almost every (a.e.) real number is transcendental. 

@Kitonum Order matters. He wants sublists, not subsets. But I don't know if contiguousness (contiguity?) matters.

@nm Please stop deleting Answers that have been responded to! Can't you see that it makes the whole thread hard to read?

@Kitonum There are two cases that are trickier to handle, both of which occur in the OP's set. Those are types Not(`+`) = 0 (i.e., a single term) and anything = Not(0). That's why my two-line procedure below took me 45 minutes to write.