¿Cómo lidias con la carga dinámica y la fatiga en FEA?

Not all loading conditions that vary with time need to be considered as a "dynamic loading". Some loadings vary too slowly compared to the first natural frequency of the structure. These loadings are called "quasi-static" and they can be taken into account with a static solution. Others have a frequency content that is very high and will not excite a structure. Only the loadings that have a frequency content in a range that excite one or some important natural frequencies must be treated as dynamic.

In Dynamic loading, a general thumb rule is - If the highest frequency of excitation remains less than 1/3rd of the system’s lowest natural frequency, such systems are good to be modeled as quasi-static.

Dynamic loading in FEA involves analyzing structures under time-varying loads like vibrations, crucial for applications such as automotive crash simulations. Fatigue analysis predicts the lifespan of structures subjected to cyclic loading, essential in industries like aerospace for assessing the fatigue life of aircraft components. Both dynamic loading and fatigue analysis are vital for ensuring the reliability and safety of engineering designs.

Time step, mesh size and solver settings are knobs you can turn to make sure your fatigue metrics (stress and strain states commonly) are converging to realistic values. This is generally a subsequent step. Load definition comes first, and in FEA, the battle is half won when quality inputs are used/defined.

To model dynamic loading in FEA, there are two paths I follow: (1) defining a load history on time domain (transient). This is practical to do with systems for which we have previous experience with or test data/measurements available. Also, this approach can be computationally expensive, but will be the most accurate. It will allow to obtain stress/strain histories in time domain that can then be used to determine load proportionality, cycle counting and what approach is best (harmonic, random, etc) (2) reduced load histories where time domain loads are processed and represented as discrete load conditions, frequency domain loading (like PSDs, ASDs) or quasi static conditions. This allows to use more efficient superposition methods.

Modeling dynamic loading in FEA involves solving equations of motion to simulate time-varying forces or displacements. Techniques include modal analysis for natural frequencies and mode shapes, and transient analysis to predict dynamic responses over time. You may apply dynamic loads, such as vibrations or impacts, to understand structural behavior. You may use software tools, like SOLIDWORKS Simulation, 3DExperience SIMULIA or Abaqus, specifying loading conditions, time steps, and material properties. Validating results with experimental data ensures accurate dynamic simulations for realistic structural assessments.

I have seen people ignore the effect of notch factor while performing fatigue analysis. While this may help in making the calculations conservative, it may take away the attention from regions of low stress concentration having high nominal stresses.

There are many different design codes that have methods for calculating fatigue. It is best practice to develop a fatigue loading spectrum based on the relevant fatigue code. Many fatigue codes use a method based on Minor’s rule to combine multiple loading scenarios.

To model fatigue in FEA, simulate cyclic loading using stress analysis before. You may define loading conditions and apply loads to represent real-world scenarios. You may Utilize fatigue life prediction methods like S-N curves or damage accumulation approaches. It would help if you considered material properties apply appropriate fatigue material data and perform multiple load cycles to predict potential crack initiation and propagation. You may use specialized FEA software to streamline the process. This helps assess the structural durability and estimate the fatigue life of components accurately.

Validation of complex simulations is best achieved by a detailed review of the testing setup: namely the loads, boundaries, thermal environment, etc. Analysts that understand and investigate how a dynamic test is conducted can increase their digital model accuracy by understanding the test system physics. FEA models can then be tuned in to better align to physical hardware testing.

Optimization with fatigue can be rather tricky. Depending on your application, the design may be subject to a few overload scenarios (storm wind, earthquake, etc.). It is very important to set your optimization parameters correctly so that the design will not fail under any of the extreme load cases. It may be prudent to recheck the design for the extreme load cases after optimization.

The best compliment to accurate fatigue analysis is a good inspection and maintenance program. The idea behind the damage tolerance method is to assume there is a crack of the smallest size with a given inspection method. From there the analyst calculates the crack growth rate during normal use. This analysis is then used to set an appropriate inspection interval. A damage tolerance program integrates fatigue analysis and NDT inspection methods to provide a rational basis for determining effective inspection intervals.