marshall wrote:
I'm trying to understand the connection between the mathematical definition of tensor and the definition physicists and fluid dynamicists use. I don't think I'm quite there yet. I usually prefer the definitions mathematicians devise in terms of precision and clarity, but they can be lacking in intuitive motivation.
In physics tensors are linear relations between vector quantities that preserve their geometric meaning under different choices of coordinate bases. Physical vectors have the same meaning as first order tensors and higher order tensors are linear geometrically preserving relations between two lower order tensors. My problem is the notation physicists use leads one to confuse the multi-dimensional arrays of numbers that represent tensors with tensors themselves. Tensors are really something more than just indexed multi-dimensional arrays.
You know about linear transformations, right? A linear transformation can qualify as a rank-2 tensor provided that it obeys the requisite transformation rules. Physicists generally work with tensors of rank 2. Tensors of higher rank are generally encountered more frequently in (pure)differential geometry and general relativity.
I guess I'd like to understand precisely how the definition Ruveyn gave is equivalent to the definition used for continuum/fluid mechanics. I'm not sure how stress tensor and strain tensor are linear transformations