Combining Dynamic Relaxation Method with Artificial Neural Networks to Enhance Simulation of Tensegrity Structures

Abstract

Structural analyses of tensegrity structures must account for geometrical nonlinearity. The dynamic relaxation method correctly models static behavior in most situations. However, the requirements for precision increase when these structures are actively controlled. This paper describes the use of neural networks to improve the accuracy of the dynamic relaxation method in order to correspond more closely to data measured from a full-scale laboratory structure. An additional investigation evaluates training the network during the service life for further increases in accuracy. Tests showed that artificial neural networks increased model accuracy when used with the dynamic relaxation method. Replacing the dynamic relaxation method completely by a neural network did not provide satisfactory results. First tests involving training the neural network online showed potential to adapt the model to changes during the service life of the structure.