In this research, we present an efficient implementation of machine learning (ML) models that forecast the mechanical properties of basalt fiber-reinforced high-performance concrete (BFHPC). The objective of the present study was to predict compressive, flexural, and tensile strengths of BFHPC through ML techniques and propose some correlations between these properties. Moreover, the modulus of elasticity (ME) values and compressive stress–strain curves were simulated using ML techniques. In this regard, three predictive algorithms, including linear regression (LR), support vector regression (SVR), and polynomial regression (PR), were considered. LR, SVR, and PR were utilized to forecast the compressive, flexural, and tensile strengths of BFHPC, and the PR technique was employed to simulate the compressive stress–strain curves. The performance of the models was also determined by the coefficient of determination (R2), mean absolute errors (MAE), and root mean square errors (RMSE). According to the obtained values of R2, MAE, and RMSE, the performance of PR was better than other types of algorithms in estimating the compressive, tensile, and flexural strengths. For example, R2 values were 0.99, 0.94, and 0.98 in predicting the compressive, flexural, and tensile strengths using PR, respectively. This shows the higher accuracy and reliability of the PR technique compared with other predictive algorithms. Finally, we concluded that ML techniques can be appropriately applied to assess the mechanical characteristics of BFHPC.

/pdf/prediction-of-the-mechanical-properties-of-basalt-fiber-2q50neoy.pdf

Prediction of the Mechanical Properties of Basalt Fiber Reinforced High-Performance Concrete Using Machine Learning Techniques

The application of robots is growing in most countries, occupying a relevant place in everyday environments. Robots are still affected by errors due to their limitations, which may compromise the final performance. Accurate trajectories and positionings are strict requirements that robots have to satisfy and may be studied by the inverse kinematic (IK) formulation. The IK conventional numerical techniques are computationally intensive procedures, focusing on the robot joint values simultaneously and increasing the complexity of the solution identification. In this scenario, Machine Learning strategies may be adopted to achieve effective and robust manipulator’s IK formulation due to their computational efficiency and learning ability. This work proposes a machine learning (ML) sequential methodology for robot inverse kinematics modeling, iterating the model prediction at each joint. The method implements an automatic Denavit-Hartenberg (D-H) parameters formulation code to obtain the forward kinematic (FK) equations required to produce the robot dataset. Moreover, the artificial neural network (ANN) architecture is selected as a structure and the number of layers in combination with the hidden neurons per layer is defined by an offline optimization phase based on the genetic algorithm (GA) technique for each joint. The ANN is implemented with the following settings: scaled conjugate gradient as training function and the mean squared error as the loss function. Different network architectures are examined to validate the IK procedure, ranging from global to sequential and considering the computation direction (from end-effector or from basement). The method is validated in the simulated and experimental laboratory environment, considering articulated robots. The sequential method exhibits a reduction of the mean squared error index of 42.7–56.7%, compared to the global scheme. Results show the outstanding performance of the IK model in robot joint space prediction, with a residual mean absolute error of 0.370–0.699 mm in trajectory following 150.0–200.0 mm paths applied to a real robot.

/pdf/machine-learning-sequential-methodology-for-robot-inverse-2rox8bc0.pdf

Machine Learning Sequential Methodology for Robot Inverse Kinematic Modelling

The present study is focused on the method of designing historical structures through data analysis and soft programming. The current research aimed to find the design method of the old columns of the structure and compare it with modern design methods. The case study was Goharshad Mosque (1400 AC) located in Iran, Mashhad. A correlation matrix was created to find the relationship between structure parameters by data mining in Python. The results indicate that the modern design method was more reliable than the old method due to the safety factor, but some parameters such as loading calculation in the historical method and the modern method were the same with more than 70% similarity. But the results of the coefficient of determination show that the loading of the R2 results was more than 0.44 and the area of the columns was more than −0.5. The modern and old design has a big engineering gap. Finally, the current study shows the old structure design method and compares it with the new design method. 

https://www.tandfonline.com/doi/pdf/10.1080/23311916.2023.2220499?needAccess=true&role=button

Historical structure design method through data analysis and soft programming


Purpose
The purpose of this research paper is to investigate and model the fused deposition modelling (FDM) process to predict the mechanical attributes of 3D printed specimens.


Design/methodology/approach
By exploiting the main effect plots, a Taguchi L18 orthogonal array is used to investigate the effects of such parameters on three mechanical attributes of the 3D printed specimens. A radial-based integrated network is then developed to map the eight FDM parameters to the three mechanical attributes for both PEEK and PEKK. Such an integrated network maps and predicts the mechanical attributes through two consecutive phases that consist of several radial basis functions (RBFs).


Findings
Validated on a set of further experiments, the integrated network was successful in predicting the mechanical attributes of the 3D printed specimens. It also outperformed the well-known RBF network with an overall improvement of 24% in the coefficient of determination. The integrated network is also further validated by predicting the mechanical attributes of a medical-surgical implant (i.e. the MidFace Rim) as an application.


Originality/value
The main aim of this paper is to accurately predict the mechanical properties of parts produced using the FDM process. Such an aim requires modelling a highly dimensional space to represent highly nonlinear relationships. Therefore, a radial-based integrated network based on the combination of composition and superposition of radial functions is developed to model FDM using a limited number of data points.


/pdf/the-development-of-a-radial-based-integrated-network-for-the-3g3fse3m.pdf

The development of a radial based integrated network for the modelling of 3D fused deposition

Fused deposition modelling (FDM) is simulated in this research work to anticipate the characteristics of 3D printed parts. Eight FDM variables are studied for Polyether-ether-ketone (PEEK) and Polyether-ketone-ketone (PEKK). A set of experiments based on Taguchi L18 is carried out. An interval type-2 fuzzy logic system (IT2$F$LS) is then established to extract the relationships among the materials and the various FDM parameters and the characteristics of the 3D printed parts. Based on the results obtained by a set of experiments, the IT2$F$LS can extract the relationships and predict the characteristics of the printed specimens with very good predictive performances. In addition, the IT2FLS outperformed the type-1 fuzzy logic system with an overall improvement of approximately 8%.

Ahmad Alsoussi

Papers

The development of a radial based integrated network for the modelling of 3D fused deposition

An Interval Type-2 Fuzzy Logic System for the Simulation of Fused Deposition