The first operation to be performed at the level of analysis of the results is the verification of the quality of the solution obtained, using appropriate numerical indicators (eg: the discontinuity of voltage between adjacent elements); in particular, it is necessary to check that the mesh is sufficiently dense at the points where tension concentrations occur. If it has been necessary to simulate particularly complex structures (with contact elements, non-linearity, etc.), it is also necessary to program an experimental verification, at least for some particularly significant points.
Once the “goodness” of the model has been ascertained, the analysis of the results can be started; finite element models produce a considerable mass of data, so it is a priority to be able to synthesize them effectively.
A first synthesis consists in replacing the tensor that describes the stresses / deformations in each node of each element making up the model, a scalar quantity, using the concept of “equivalent” tension / deformation, that is equally onerous from the point of view of structural strength. The type of equivalent voltage / deformation formulation must be chosen according to the properties of the material (fragile or ductile) and the type of stress (static, fatigue, impulsive).
Once the most critical areas of the structure have been identified from the point of view of the extent of the deformations and therefore of the tensions, ie the areas where, under the constraints and hypothesized loads, there could be breaks or in the case of the bone tissue of necrosis and or resorption , it is often useful to return to the tensor to understand which type of stress (bending, twisting, cutting, traction or compression) is prominent.
In biomechanics, the danger of a given stress level often does not depend so much on its value in form, but on how much it deviates from its respective physiological value and its distribution: a peak of tension implies in fact a greater effort of adaptation by the bone, as defined by the remodeling laws.