Dynamic Relaxation Analysis

A number of different analyses can be performed on the same data file using the dynamic relaxation solver. The basic analysis option is the single increment option. Using this option will generate forces and deflections in the model resulting from the imposed loads specified in a single analysis case.

This is similar to linear elastic static analysis except that material and geometrical non-linear effects will be considered. Also the post processing of results is different from linear static analysis as the principle of superposition is invalid in non-linear analysis. Therefore results from single increment analysis should not be combined and a different approach to load capacity factors and incremental loading is needed.

In order to match the loading from a linear analysis and combination cases the loading for a nonlinear analysis should be factored and combined. Thus the loading may be specified as 1.4L1+1.6L2. It is also possible to use a combination case such as 1.6C1. In this case the loading implied by that combination will be used.

Note that each dynamic relaxation run relates to a single analysis case and that results should not be combined post analysis. Combinations of loads and factored analysis cases should be run through the solver. If the task is form-finding with no imposed loading, a blank load case can be input. Note that in the case of incremental load options, the solver generates analysis cases for each load stage.

Non-linear analysis can be used to show the change in stiffness of the whole model as loading increases. This can be used to investigate near-collapse large deflection behaviour, plastic behaviour, and buckling.

Automatic load increment, and Individual component buckling analysis options are offered to investigate these issues. With these options, the imposed loading specified is used to define the loading pattern. Analysis cases are generated by the program using factored versions of the load pattern.

As the solver is iterative, non-convergence can occur. This may be due to inappropriate damping, or an inadequate number of iterations. However it will also occur if the model is a mechanism (possibly due to a modelling error), or is approaching model capacity. Non-convergence problems should be investigated. With incremental loading options, the solver monitors the convergence rate for each analysis. If an initial run at low load converges the program stores the information about the convergence process. This initial run shows that the model is not inherently unstable. If convergence becomes significantly harder to achieve as the load increases it is likely that the model has reached capacity.

Note that the model capacity is dependent on the simplifications used in the modelling process (particularly material properties, and node positions), and may not have any relation to the ultimate capacity of the real structure defined by codes of practice.

Another use of the nonlinear solver is for Form-finding. This process will transform a data file with approximate geometry and prestress loading into a data file with an accurate 3 dimensional form in equilibrium with the boundary conditions and prestress. Selecting form finding analysis options tells the solver to use form finding properties for the elements where these are defined, and to create new data representing the final equilibrium geometry and prestress. This new data overwrites the original data.