Stacking Ensemble Tree Models to Predict Energy Performance in Residential Buildings

Prediction of Daily Mean PM10 Concentrations Using Random Forest, CART Ensemble and Bagging Stacked by MARS

A novel framework for stacked regression based on machine learning was developed to predict the daily average concentrations of particulate matter (PM10), one of Bulgaria’s primary health concerns. The measurements of nine meteorological parameters were introduced as independent variables. The goal was to carefully study a limited number of initial predictors and extract stochastic information from them to build an extended set of data that allowed the creation of highly efficient predictive models. Four base models using random forest, CART ensemble and bagging, and their rotation variants, were built and evaluated. The heterogeneity of these base models was achieved by introducing five types of diversities, including a new simplified selective ensemble algorithm. The predictions from the four base models were then used as predictors in multivariate adaptive regression splines (MARS) models. All models were statistically tested using out-of-bag or with 5-fold and 10-fold cross-validation. In addition, a variable importance analysis was conducted. The proposed framework was used for short-term forecasting of out-of-sample data for seven days. It was shown that the stacked models outperformed all single base models. An index of agreement IA = 0.986 and a coefficient of determination of about 95% were achieved.

Novel Fuzzy-Based Optimization Approaches for the Prediction of Ultimate Axial Load of Circular Concrete-Filled Steel Tubes

An accurate estimation of the axial compression capacity of the concrete-filled steel tubular (CFST) column is crucial for ensuring the safety of structures containing them and preventing related failures. In this article, two novel hybrid fuzzy systems (FS) were used to create a new framework for estimating the axial compression capacity of circular CCFST columns. In the hybrid models, differential evolution (DE) and firefly algorithm (FFA) techniques are employed in order to obtain the optimal membership functions of the base FS model. To train the models with the new hybrid techniques, i.e., FS-DE and FS-FFA, a substantial library of 410 experimental tests was compiled from openly available literature sources. The new model’s robustness and accuracy was assessed using a variety of statistical criteria both for model development and for model validation. The novel FS-FFA and FS-DE models were able to improve the prediction capacity of the base model by 9.68% and 6.58%, respectively. Furthermore, the proposed models exhibited considerably improved performance compared to existing design code methodologies. These models can be utilized for solving similar problems in structural engineering and concrete technology with an enhanced level of accuracy.

Optimized Support Vector Machines Combined with Evolutionary Random Forest for Prediction of Back-Break Caused by Blasting Operation

Back-break is an adverse event in blasting works that causes the instability of mine walls, equipment collapsing, and reduction in effectiveness of drilling. Therefore, it boosts the total cost of mining operations. This investigation intends to develop optimized support vector machine models to forecast back-break caused by blasting. The Support Vector Machine (SVM) model was optimized using two advanced metaheuristic algorithms, including whale optimization algorithm (WOA) and moth–flame optimization (MFO). Before the models’ development, an evolutionary random forest (ERF) technique was used for input selection. This model selected five inputs out of 10 candidate inputs to be used to predict the back break. These two optimized SVM models were evaluated using various performance criteria. The performance of these two models was also compared with other hybridized SVM models. In addition, a sensitivity evaluation was made to find how the selected inputs influence the back-break magnitude. The outcomes of this study demonstrated that both the SVM–MFO and SVM–WOA improved the performance of the standard SVM. Additionally, the SVM–MFO showed a better performance than the SVM–WOA and other hybridized SVM models. The outcomes of this research recommend that the SVM–MFO can be considered as a powerful model to forecast the back-break induced by blasting.

Factors Influencing Pile Friction Bearing Capacity: Proposing a Novel Procedure Based on Gradient Boosted Tree Technique

In geotechnical engineering, there is a need to propose a practical, reliable and accurate way for the estimation of pile bearing capacity. A direct measure of this parameter is difficult and expensive to achieve on-site, and needs a series of machine settings. This study aims to introduce a process for selecting the most important parameters in the area of pile capacity and to propose several tree-based techniques for forecasting the pile bearing capacity, all of which are fully intelligent. In terms of the first objective, pile length, hammer drop height, pile diameter, hammer weight, and N values of the standard penetration test were selected as the most important factors for estimating pile capacity. These were then used as model inputs in different tree-based techniques, i.e., decision tree (DT), random forest (RF), and gradient boosted tree (GBT) in order to predict pile friction bearing capacity. This was implemented with the help of 130 High Strain Dynamic Load tests which were conducted in the Kepong area, Malaysia. The developed tree-based models were assessed using various statistical indices and the best performance with the lowest system error was obtained by the GBT technique. The coefficient of determination (R2) values of 0.901 and 0.816 for the train and test parts of the GBT model, respectively, showed the power and capability of this tree-based model in estimating pile friction bearing capacity. The GBT model and the input selection process proposed in this research can be introduced as a new, powerful, and practical methodology to predict pile capacity in real projects.

The Effects of Rock Index Tests on Prediction of Tensile Strength of Granitic Samples: A Neuro-Fuzzy Intelligent System

Rock tensile strength (TS) is an essential parameter for designing structures in rock-based projects such as tunnels, dams, and foundations. During the preliminary phase of geotechnical projects, rock TS can be determined through laboratory works, i.e., Brazilian tensile strength (BTS) test. However, this approach is often restricted by laborious and costly procedures. Hence, this study attempts to estimate the BTS values of rock by employing three non-destructive rock index tests. BTS predictive models were developed using 127 granitic rock samples. Since the simple regression analysis did not yield a meaningful result, the development of models that integrate multiple input parameters were considered to improve the prediction accuracy. The effects of non-destructive rock index tests were examined through the use of multiple linear regression (MLR) and adaptive neuro-fuzzy inference system (ANFIS) approaches. Different strategies and scenarios were implemented during modelling of MLR and ANFIS approaches, where the focus was to consider the most important parameters of these techniques. As a result, and according to background and behaviour of the ANFIS (or neuro-fuzzy) model, the predicted values obtained by this intelligent methodology are closer to the actual BTS compared to MLR which works based on linear statistical rules. For instance, in terms of system error and a-20 index, values of (0.84 and 1.20) and (0.96 and 0.80) were obtained for evaluation parts of ANFIS and MLR techniques, which revealed that the ANFIS model outperforms the MLR in forecasting BTS values. In addition, the same results were obtained through ranking systems by the authors. The neuro-fuzzy developed in this study is a strong technique in terms of prediction capacity and it can be used in the other rock-based projects for solving relevant problems.