Search results

Aluminium alloys can be used as lightweight and high-strength materials in combination with the technology of laser beam welding, an efficient joining method, in the manufacturing of automotive parts. The purposes of this paper are to conduct laser welding experiments with Al2024 in the lap joint configuration, model the laser welding process parameters of Al2024 alloys and use propounded models to optimize the process parameters.

Laser welding of Al2024 alloy has been conducted in the lap joint configuration. Then, the influences of explanatory variables (laser peak power, scanning speed and frequency) on outcome variables (weld width [WW], throat length [TL] and breaking load [BL]) have been investigated with Poisson regression analysis of the data set derived from experimentation. Thereafter, a multi-objective genetic algorithm (MOGA) has been used using MATLAB to find the optimum solutions. The effects of various input process parameters on the responses have also been analysed using response surface plots.

The promulgated statistical models, derived with Poisson regression analysis, are evinced to be well-fit ones using the analysis of deviance approach. Pareto fronts have been used to demonstrate the optimization results, and the maximized load-bearing capacity is computed to be 1,263 N, whereas the compromised WW and TL are 714 µm and 760 µm, respectively.

This work of conducting laser welding of lap joint of Al2024 alloy incorporating the Taguchi method and optimizing the input process parameters with the promulgated statistical models proffers a neoteric perspective that can be useful to the manufacturing industry.

Despite the enormous need to succeed in the urban model, scientists and policymakers should work consistently to create blueprints to regulate urbanization. The absence of coordination between the crucial requirements and the regional strategies of the local authorities leads to a lack of conformance in urban development. The purpose of this paper is to address this issue.

This study intends to manage future urban growth patterns using integrated methods and then employ the results in the genetic algorithm (GA) model to considerably improve growth behavior. Multi-temporal land-use datasets have been derived from remotely sensed images for the years 1990, 2000, 2010 and 2020. Urban growth patterns and processes were then analyzed with land-use-and-land-cover dynamics. Results were examined for simulation and utilization of the GA.

Model parameters were derived and evaluated, and a preliminary assessment of the effective coefficient in the formation of urbanization is analyzed, showing the city's urbanization pattern has followed along with the transportation infrastructure and outward growth, and the scattering rates are high, with an increase of 5.64% in building area associated with a decrease in agricultural lands and rangelands.

The research achieved a considerable improvement over the growth behavior. The conducted research design was the first of its type in that field to be executed to any specific growth pattern parameters in terms of regulating and policymaking. The method has integrated various artificial intelligence models to monitor, measure and optimize the projected growth by applying this design. Other research on the area was limited to projecting the future of Amman as it is an urbanized distressed city.

The amount of features in handwritten digit data is often very large due to the different aspects in personal handwriting, leading to high-dimensional data. Therefore, the employment of a feature selection algorithm becomes crucial for successful classification modeling, because the inclusion of irrelevant or redundant features can mislead the modeling algorithms, resulting in overfitting and decrease in efficiency.

The minimum redundancy and maximum relevance (mRMR) and the recursive feature elimination (RFE) are two frequently used feature selection algorithms. While mRMR is capable of identifying a subset of features that are highly relevant to the targeted classification variable, mRMR still carries the weakness of capturing redundant features along with the algorithm. On the other hand, RFE is flawed by the fact that those features selected by RFE are not ranked by importance, albeit RFE can effectively eliminate the less important features and exclude redundant features.

The hybrid method was exemplified in a binary classification between digits “4” and “9” and between digits “6” and “8” from a multiple features dataset. The result showed that the hybrid mRMR +  support vector machine recursive feature elimination (SVMRFE) is better than both the sole support vector machine (SVM) and mRMR.

In view of the respective strength and deficiency mRMR and RFE, this study combined both these methods and used an SVM as the underlying classifier anticipating the mRMR to make an excellent complement to the SVMRFE.