Matrix converters (MCs) have attracted significant interest and found extensive applications across multiple industries owing to their desirable characteristics. These include the capability to produce sinusoidal currents at both input and output, substantial size reduction, and enhanced reliability by minimizing significant passive components. This paper explores the potential of MC technology as a viable alternative to conventional AC-DC-AC converters in industrial applications. It discusses recent advancements in MC structural configurations, modulation/control algorithms, and multiphase structures and control systems. The paper offers an in-depth review of modern industrial uses of MC technology. It also delves into different methods for managing induction motors, particularly the DTC (Direct Torque Control) approach. The study explores the intricacies of DTC and its relationship with SVM. The primary research objective is to examine the performance of an IM when operated with an SVPWM inverter, focusing on harmonic analysis of voltages and currents. Various PWM methods regulate the voltage and frequency supplied to the IM. Sinusoidal Pulse Width Modulation (SPWM) and SVPWM are the two most commonly used 3-phase Voltage Source Inverter strategies. The growing adoption of SVPWM is driven by its ability to reduce harmonic content in voltage and enhance the fundamental output voltage of the IM. Consequently, this study models a DTC-SVM theory-driven IM using MATLAB/SIMULINK to control the speed of induction motors. The following values were calculated for the system: Quality factor=2.236, Damping ratio=4.45, and the cut-off frequency (fc=355.88H).
Modeling and simulation are indispensable when dealing with complex engineering systems. It makes it possible to do essential assessment before systems are built, Cantilever, which help alleviate the need for expensive experiments and it can provide support in all stages of a project from conceptual design, through commissioning and operation. This study deals with intelligent techniques modeling method for nonlinear response of uniformly loaded paddle. Two Intelligent techniques had been used (Redial Base Function Neural Network and Support Vector Machine). Firstly, the stress distributions and the vertical displacements of the designed cantilevers were simulated using (ANSYS v12.1) a nonlinear finite element program, incremental stages of the nonlinear finite element analysis were generated by using 25 schemes of built paddle Cantilevers with different thickness and uniform distributed loads. The Paddle Cantilever model has 2 NN; NN1 has 5 input nodes representing the uniform distributed load and paddle size, length, width and thickness, 8 nodes at hidden layer and one output node representing the maximum deflection response and NN2 has inputs nodes representing maximum deflection and paddle size, length, width and thickness and one output representing sensitivity (∆R/R). The result shows that of the nonlinear response based upon SVM modeling better than RBFNN on basis of time, accuracy and robustness, particularly when both has same input and output data.
Obesity is the excess of body weight relative to height above the desired level as a result of excessive increase in the ratio of body fat mass to lean mass. It causes many health problems due to its negative effects on body systems (cardiovascular system, musculoskeletal system, gastrointestinal system, respiratory system, skin, endocrine system, genitourinary system) and psychosocial status. In this study is aimed to effective detection of the eating and physical condition-based obesity stages using machine learning algorithms. The dataset contains data for the estimation of obesity stages in individuals from Mexico, Peru, and Colombia and is available as open source. There are 2111 records and 17 attributes in the dataset. In the records, obesity stages were categorized as insufficient weight, normal weight, overweight level I, overweight level II, obesity type I, obesity type II and obesity type III. The 10-fold cross-validation method was used to validate the model and the performances of the Support Vector Machine (SVM), Random Forest (RF), and Multilayer Perceptron (MLP) classification algorithms were compared. It has been determined that the highest performance among the algorithms whose performances are compared belongs to the RF Algorithm (95.78%). This paper’s abstract has been presented at the International Conference on Computational Mathematics and Engineering Sciences held in Ordu (Turkey), / 20-22 May. 2022.