Luis Santos

Luis Santos

Auto-generated profile, based on publicly available information

Luis Santos

School of the Built Environment and Architecture, London South Bank University, United Kingdom

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 February 2024 in Engineering Applications of Artificial Intelligence
Verified
Journal Article
Open access
Ikram AbarkanIkram Abarkan
Ikram Abarkan
MRMusab Rabi
Musab Rabi
FFFelipe Piana Vendramell Ferreira
Felipe Piana Vendramell Ferreira
RSRabee Shamass
Rabee Shamass
VLVireen Limbachiya
Vireen Limbachiya
LSLuis Fernando Pinho Santos
Luis Fernando Pinho Santos

17 August 2022 in Structural and Multidisciplinary Optimization
Verified
Journal Article
Open access
PDF file
LSLuis Santos
Luis Santos
BIBassam Izzuddin
Bassam Izzuddin
LMLorenzo Macorini
Lorenzo Macorini
When subjected to bending loads, sandwich panels are highly efficient structural components with the potential to achieve substantial weight reduction. A successful design methodology for sandwich panels should aim at maximising this potential for weight reduction while considering the various possible failure mode in a simple yet accurate manner. This paper investigates the application of steel sandwich panels as two-way deck systems. Near-optimal designs for all-steel Rectangular Honeycomb Core Sandwich Panels (RHCSPs) under general out-of-plane loading are achieved using a gradient-based optimisation method. The method relies on continuously optimising the design limit state constraints while the response constraints are considered a priori in the analysis stage using simplified analytical assessment. Plate bending solutions and sandwich bending solutions are used as alternatives to estimate the internal stresses on each layer of the sandwich panel under out-of-plane loads, where comparisons are made between these two analysis methods in terms of computational efficiency and accuracy. The internal stresses are then used to formulate design limit state equations for each relevant failure mode, including material yielding, plate buckling and deformation control. The Method of Moving Asymptotes is used for the optimisation of RHCSPs, considering the limit states as the constraints of the optimisation problem and weight as the objective function to be minimised. The proposed methodology for simplified assessment is verified against detailed nonlinear finite element models for optimal design solutions. The implications of the results of the proposed optimisation strategy on the development of a systematic design methodology for RHCSPs are also highlighted, making specific reference to critical failure modes.

24 May 2024
Verified
Conference Paper
Open access
LSLuis Santos
Luis Santos
Today, an engineer's success is largely based on their skill at employing Finite Element Analysis (FEA), a standard engineering modeling tool to numerically assess the behavior of structures. So far, improvements in computational power and FEA element formulations have supported significant advancements in this field. FEA can be used to provide accurate and fast answers to many engineering problems. However, FEA models are still associated with prohibitive time costs when solving complex scenarios. Unlike traditional computer algorithms, Machine Learning, by minimizing pre-established loss functions, can ascertain patterns and make predictions without being explicitly programmed for a given task. This paper presents initial results on improving structural analysis tools by harnessing the power of Machine Learning algorithms, namely Deep Artificial Neural Networks (ANNs) and Physics-Informed Neural Networks (PINNs), with the final objective of substituting or accelerating FEA in accurately predicting the structural behavior within an engineering domain. It is found that ANNs can accurately predict the stress, strain, and displacement maps for a cantilevered rectangular plate under a concentrated load. ANNs demonstrate that, with a large training dataset, efficient and accurate predictions can be achieved although only for specific problems. These algorithms could be enhanced with the governing differential equations to improve convergence and speed. PINNs show that, despite being less efficient than FEA, the method shows more generalization potential than ANNs. Both results show the potential for machine learning algorithms to enhance traditional computational mechanics methods.