Introduction

_images/objectives.png

COMPAS TNO is a Python package to find admissible thrust networks in masonry vaulted structures.

The package implements Thrust Network Optimisation, or TNO, within the COMPAS framework. TNO is a modular multi-objective optimisation framework to find admissible thrust networks in vaulted masonry structures. Thrust Networks represent the internal forces in masonry structures as a connected force network contained within the structural geometry. Based on the safe theorem of limit analysis, a structure is safe if at least one thrust network is found within its envelope. This set of compressive, internal forces corresponds to a lower-bound equilibrium solution.

With TNO, multiple particular equilibrium states can be obtained, including the structure’s minimum and maximum horizontal thrusts, its minimum thickness, its vertical and horizontal collapse loads, and the internal stress following support movements. Beyond providing the internal force distribution for these limit states, an indication of the crack patterns is obtained by looking at the points where the network touches intrados and extrados of the vaults.

Exploring network contained within the structural evelope corresponds to solving a constrained nolinear optimisation problem. TNO sets up, run and output the solution networks to these problems.

To use TNO, check the installation guide on the Installation and Getting Started pages to prepare your machine to run TNO.

A tutorial about the package’s elements is provided in Tutorial

A series of simple examples are provided in the Examples section to start using TNO.

Acknowlegments

TNO has been developed during Ricardo Maia Avelino’s doctoral research at the Block Research Group, ETH Zurich funded by the Swiss National Science Foundation SNSF (project grant n. 178953).

If you use COMPAS TNO to your research, please refer to our publications.

Table of Contents