Underground spaces have become increasingly important in recent decades in metropolises. In this regard, the demand for the use of underground spaces and, consequently, the excavation of these spaces has increased significantly. Excavation of an underground space is accompanied by risks and many uncertainties. Tunnel convergence, as the tendency for reduction of the excavated area due to change in the initial stresses, is frequently observed, in order to monitor the safety of construction and to evaluate the design and performance of the tunnel. This paper presents a model/equation obtained by a gene expression programming (GEP) algorithm, aiming to predict convergence of tunnels excavated in accordance to the New Austrian Tunneling Method (NATM). To obtain this goal, a database was prepared based on experimental datasets, consisting of six input and one output parameter. Namely, tunnel depth, cohesion, frictional angle, unit weight, Poisson’s ratio, and elasticity modulus were considered as model inputs, while the cumulative convergence was utilized as the model’s output. Configurations of the GEP model were determined through the trial-error technique and finally an optimum model is developed and presented. In addition, an equation has been extracted from the proposed GEP model. The comparison of the GEP-derived results with the experimental findings, which are in very good agreement, demonstrates the ability of GEP modeling to estimate the tunnel convergence in a reliable, robust, and practical manner.

This study investigated the different thicknesses of TiO₂ photoanode films and the effect of surface plasmon resonance (SPR) of Ag-TiO₂ nanocomposites on the current-voltage (I–V) performance of dye-sensitized solar cells (DSSC). The TiO₂ layer was deposited using the doctor blade technique and the thickness of the TiO₂ films was controlled by using a different number of Scotch tape layers. The silver nanoparticles (AgNP) were synthesised using a chemical reduction method and the concentration of sodium citrate as a reducing agent was varied from 4 to 12 mM to study the effect of citrate ion on the size of the nanoparticles. Ag-TiO₂ nanopowder was prepared by adding pure anatase TiO₂ powder into AgNP colloidal solution. The mixture was left to dry for 24 h to obtain Ag-TiO₂ powder for paste preparation. The three-layer Scotch tape, with thickness of 14.38 µm, achieved a high efficiency of 4.14%. This results showed that three layers was the optimal thickness to improve dye loading and to reduce the charge recombination rate. As for the Ag-TiO₂ nanocomposites, 10 mM of AgNP, with a mean diameter of 65.23 nm and high efficiency of 6.92%, proved that SPR can enhance the absorption capability of dye and improve the photon-to-electron generation.

Alzheimer's disease (AD) is an irreversible neurological disorder that causes the gradual decline of one's cognitive abilities, and thus, early detection is significant to slow down its deterioration. Magnetic resonance imaging (MRI) images have been commonly used to diagnose AD. Furthermore, deep learning techniques such as the Convolutional Neural Network (CNN) are utilized to assist the diagnosis due to the complexity of MRI's analysis. However, CNN models require large datasets for training and have a challenging nature for model optimization. Thus, Transfer Learning, an emerging method that can improve the performance of the deep learning model by eliminating the need for training from scratch, is introduced. This paper will propose a Transfer Learning-based EfficientNet-B0 model to classify MRI brain images for AD diagnosis. The MRI images are obtained from Alzheimer's Disease Neuroimaging Initiative (ADNI) database; only axial-plane images are used in this study. As a result, the multi-class classification of MRI images into AD, MCI, and NC classes using a Transfer Learning-based model resulted in a training accuracy of 98.93% and a validation accuracy of 87.17%. These results evidenced the significance of Transfer Learning in improving model performance.

Fuzzy logic controller (FLC) is an attractive alternative to existing classical or modern controllers for designing the challenging Non-linear control systems as it provides a heuristically method by If-Then Rules which resembles human intelligence. This paper presents the design and analysis of a FLC for the outer loop and a PD controller for the inner loop of a ball and beam system. Additionally, a classical PID controller based on ITAE equations for the beam position is also designed to compare the performance of both types of controllers. Results reveal that FLC found to give better transient and steady state results compare to classical PID.