Catastrophe in elastic tensegrity frameworks (Q2101702)

This paper deals with the application of numerical nonlinear algebra to the study of catastrophes in the context of elastic mechanical systems. One can say that, in a given system, a catastrophe happens when the stability of an equilibrium state is lost, causing the system to move into a new dynamical regime and ultimately to a different equilibrium state. The systems that are considered in this papers are the elastic tensegrity frameworks. They are configurations made of rigid bars and elastic cables that can be depicted as graphs endowed with a parameter space whose small changes can cause a catastrophe (Definition 1). Bar and elastic cables are distinguished by an energy function which is based on Hook's law. The stable equilibrium positions are determined by minimizing the energy function, hence by an optimization study. The parameters are of two types: the control parameters, whose values can be chosen, and the internal parameters, whose values are determined by those of control parameters and by the conditions that are imposed. The authors of this paper ``use numerical nonlinear algebra to calculate all equilibrium positions, in contrast to the more widely used iterative methods (e.g. Newton-Raphson) which can only find one solution at a time, with no guarantees on finding them all''. Also a method to compute all catastrophes is given. In Section 2 an accurate description of elastic tensegrity frameworks is presented. Then, in Section 3, the posed questions are described by means of algebraic tools, under some added hypotheses with the intention to simplify the exposition. In Section 4 the authors show how to use this algebraic description in order to solve particular problems, such as: given real values for specific control parameters, compute all the local minima of the considered optimization problem; given a subset \(\Omega\) of values for control parameters, compute the positive-dimensional semialgebraic subset \(\mathcal C_{\Omega}\subset \Omega\), called the catastrophe set (Definition 7). Considerations about the difficulties and the efficiency of the computations are discussed. Some hints are also given about dedicated packages that can be used to make computations. The reader can find nice examples throughout the paper, until a final section of conclusions.