Food for thought

Hello readers!

This month ENSA had the great pleasure to hold the TRUSS annual plenary in Santander (Spain). During this 2-days meeting, the members of the TRUSS consortium visited the ENSA’s facilities and saw 3 huge steam generators ready to be shipped to the Beaver Valley NPP. The early stage researchers, we also had the opportunity to attend a presentation of Prof. Castillo on extreme value statistics and to share the progression of our respective research.

Picture taken during my oral presentation

Regarding my personal achievements, I can proudly say that after a general review of the different topics involved, I am able to understand the ENSA methodology and implement it into the ANSYS FE software through ASCII macros. As you can imagine, the numerical analysis of the complex behaviour of racks in seismic conditions is not an easy task. ENSA methodology can be divided into 4 parts. The food for though is summarized hereafter:

I.                    Hydrodynamic masses concept: inertial coupling effects of the surrounding water volume are replaced by the equivalent mass of water vibrating with each rack. The fluid is treated as a virtual extension of the structure changing its effective mass.   

II.                  Thermal analogy: If the fluid is inviscid and incompressible, its velocity field satisfies the Laplace equation (∇²Φ=0), and the associated pressure field can be determined via the Bernoulli equation. Therefore the potential flow shows a complete analogy with the steady-state heat conduction phenomena which can be used to assess the hydrodynamic masses in a cost-effective way.

III.                Dynamic analysis: A simplified wired model maintaining the original natural frequencies is used to perform a cost-effective full transient analysis.

IV.                Stress analysis: Resulting loads and displacements are transferred from the dynamic model to a detailed FE model in order to check local stresses and instabilities.

This knowledge will allow me to automatically generate my own series of numerical outputs. It is a basic starting point, as I count on comparing them with experimental data to evaluate the uncertainties inherent to the methodology.

See you down here soon!