Roger Temam writes in SOME DEVELOPMENTS ON NAVIER-STOKES EQUATIONS IN THE SECOND HALF OF THE 20th CENTURY:

A remarkable property of the Navier-Stokes equations is that they are one of the very few (if not the only) nonlinear equations in mathematical physics for which the nonlinearity is derived from mathematical argument (just chain rule differentiation) and not from physical modelling.

What are the other "very few" PDE possessing this property?