Reliability and Prediction of Embedment Depth of Sheet pile Walls Using Hybrid ANN with Optimization Techniques

PCA-Based Hybrid Intelligence Models for Estimating the Ultimate Bearing Capacity of Axially Loaded Concrete-Filled Steel Tubes

In order to forecast the axial load-carrying capacity of concrete-filled steel tubular (CFST) columns using principal component analysis (PCA), this work compares hybrid models of artificial neural networks (ANNs) and meta-heuristic optimization algorithms (MOAs). In order to create hybrid ANN models, a dataset of 149 experimental tests was initially gathered from the accessible literature. Eight PCA-based hybrid ANNs were created using eight MOAs, including artificial bee colony, ant lion optimization, biogeography-based optimization, differential evolution, genetic algorithm, grey wolf optimizer, moth flame optimization and particle swarm optimization. The created ANNs' performance was then assessed. With R² ranges between 0.7094 and 0.9667 in the training phase and between 0.6883 and 0.9634 in the testing phase, we discovered that the accuracy of the built hybrid models was good. Based on the outcomes of the experiments, the generated ANN-GWO (hybrid model of ANN and grey wolf optimizer) produced the most accurate predictions in the training and testing phases, respectively, with R² = 0.9667 and 0.9634. The created ANN-GWO may be utilised as a substitute tool to estimate the load-carrying capacity of CFST columns in civil engineering projects according to the experimental findings.

Hybrid Ensemble Model for Predicting the Strength of FRP Laminates Bonded to the Concrete

The goal of this work was to use a hybrid ensemble machine learning approach to estimate the interfacial bond strength (IFB) of fibre-reinforced polymer laminates (FRPL) bonded to the concrete using the results of a single shear-lap test. A database comprising 136 data was used to train and validate six standalone machine learning models, namely, artificial neural network (ANN), extreme machine learning (ELM), the group method of data handling (GMDH), multivariate adaptive regression splines (MARS), least square-support vector machine (LSSVM), and Gaussian process regression (GPR). The hybrid ensemble (HENS) model was subsequently built, employing the combined and trained predicted outputs of the ANN, ELM, GMDH, MARS, LSSVM, and GPR models. In comparison with the standalone models employed in the current investigation, it was observed that the suggested HENS model generated superior predicted accuracy with R² (training = 0.9783, testing = 0.9287), VAF (training = 97.83, testing = 92.87), RMSE (training = 0.0300, testing = 0.0613), and MAE (training = 0.0212, testing = 0.0443). Using the training and testing dataset to assess the predictive performance of all models for IFB prediction, it was discovered that the HENS model had the greatest predictive accuracy throughout both stages with an R² of 0.9663. According to the findings of the experiments, the newly developed HENS model has a great deal of promise to be a fresh approach to deal with the overfitting problems of CML models and thus may be utilised to forecast the IFB of FRPL.

Modelling Compression Strength of Waste PET and SCM Blended Cementitious Grout Using Hybrid of LSSVM Models

Nowadays, concretes blended with pozzolanic additives such as fly ash (FA), silica fume (SF), slag, etc., are often used in construction practices. The utilization of pozzolanic additives and industrial by-products in concrete and grouting materials has an important role in reducing the Portland cement usage, the CO₂ emissions, and disposal issues. Thus, the goal of the present work is to estimate the compressive strength (CS) of polyethylene terephthalate (PET) and two supplementary cementitious materials (SCMs), namely FA and SF, blended cementitious grouts to produce green mix. For this purpose, five hybrid least-square support vector machine (LSSVM) models were constructed using swarm intelligence algorithms, including particle swarm optimization, grey wolf optimizer, salp swarm algorithm, Harris hawks optimization, and slime mold algorithm. To construct and validate the developed hybrid models, a sum of 156 samples were generated in the lab with varying percentages of PET and SCM. To estimate the CS, five influencing parameters, namely PET, SCM, FLOW, 1-day CS (CS_1D), and 7-day CS (CS_7D), were considered. The performance of the developed models was assessed in terms of multiple performance indices. Based on the results, the proposed LSSVM-PSO (a hybrid model of LSSVM and particle swarm optimization) was determined to be the best performing model with R² = 0.9708, RMSE = 0.0424, and total score = 40 in the validation phase. The results of sensitivity analysis demonstrate that all the input parameters substantially impact the 28-day CS (CS_28D) of cementitious grouts. Among them, the CS_7D has the most significant effect. From the experimental results, it can be deduced that PET/SCM has no detrimental impact on CS_28D of cementitious grouts, making PET a viable alternative for generating sustainable and green concrete. In addition, the proposed LSSVM-PSO model can be utilized as a novel alternative for estimating the CS of cementitious grouts, which will aid engineers during the design phase of civil engineering projects.