Public defence: German Solorzano Ramirez
German Solorzano Ramirez defended his thesis “Modeling and Simulation of Concrete Structures with Neural Networks” for the PhD in Engineering Science.
German Solorzano Ramirez defended his thesis “Modeling and Simulation of Concrete Structures with Neural Networks” for the PhD in Engineering Science.
Trial lecture: "D- and B- Regions in RC Structures: Range of Applicability".
Ordinary opponents:
Leader of the public defence: Associate Professor Yonas Zewdu Ayele, OsloMet, Norway
Supervisors:
This thesis explores how modern computational techniques can improve the design and analysis of reinforced concrete (RC) structures. The study mainly focuses on two methods: Artificial Neural Networks (ANN) and Genetic Algorithms (GAs).
A key achievement of this research is the creation of a new numerical model using ANN to predict the strength of RC walls. Trained on extensive data from detailed simulations, this model can quickly and accurately estimate the load capacity of these walls. The model is available as an open-source tool that anyone can use. The research also integrates this ANN model into a simplified element for structural analysis, which significantly reduces the time and computational burden needed to simulate complex structures.
Another important part of the study is the development of a genetic algorithm to optimize the design of RC elements. This tool helps engineers find the most cost-effective designs that meet all safety standards, making the design process faster and more efficient.
Overall, this thesis shows how computational intelligence can make building analysis and design more efficient, accurate, and cost-effective. These advancements point to a future where engineering relies more on automation and smart technologies, leading to better, more sustainable buildings.