Structural optimisation of concrete isolated foundations using Generative Design

Foundations represent crucial parts for almost all our worlds’ infrastructures and structures. However, optimisation of these foundations seems to be deprioritised, likely due to the highly consumed time and resources for developing the working design. The potential benefits for reducing the cost and emissions levels from optimisation of concrete isolated foundations, creates an opportunity to implement a new design environment based on Generative Design methodology. In this thesis a comprehensive literature review is conducted to get an understanding of how Generative Design is implemented into structural design and industrial optimisation nowadays. The literature review forms the basis for a new framework to implement Generative Design Optimisation for concrete isolated foundations. This is done by establishing an objective function with the goal of maximising the structural utilisation. The framework is built using Grasshopper software with Galapagos in order to benefit from the aid of the genetic algorithm optimisation process. The framework is extensively tested on two real-life cases, supplied by Sweco with permission from Statnett and Bane NOR. A total of 100 evolutionary runs are performed, 50 for each case. The approach shows acceptable results for the fitness values, cost, and emission, though it is also clear that there is a discrepancy between what is considered optimal. All the cases show that an optimal fitness value based on the utilisation alone, does not necessarily mean that the cost and emissions are optimal. The reason behind this difference between the fitness value and cost and emissions, is discussed to be a result of the used input parameters and their restrictions, while also being a potential weakness of the objective function. A statistical analysis is also performed on the gathered data and can form a basis for further analysis. In overall, the proposed strategy using the implementation of Generative Design to optimise concrete isolated foundations can be an effective tool for design engineering. Future research suggested to focus on weighted inputs to the objective function, the implementation of a multi-objective function and even the implementation of machine learning. Keywords: Concrete isolated foundations; Optimisation; Generative Design; Galapagos; Objective function; Utilisation