Advanced Defense Science and Technology
Year:2023 | Volume:13 | Issue:3 (پیاپی 49)|Papers Count:6

The feeding of off-grid AC loads, like military loads or defense systems is of high importance. This paper proposes a novel structure for Switched-Capacitor based Multi-Level Inverter (SCMLI) that produces a high quality (low THD) 7-level output-voltage waveform for supplying off-grid AC loads. The proposed inverter requires only single DC-source and can boost the input voltage up to three times, at the output port. The boosting capability is very crucial at PV applications. The reduced device count, high quality (low THD) output voltage, and natural voltage balancing of capacitors are other main merits of suggested inverter. The proposed inverter can effectively supply any load type, including pure resistive, resistive-inductive and pure inductive loads. The short discharging interval of capacitors leads to reduced capacitors' voltage-ripple, high voltage quality, low voltage-ripple losses and better inverter efficiency. Comparison results approve the superiority of proposed converter over existed counterparts. Also, the correct operation of proposed inverter during different operational conditions, has been validated by simulation (performed in MATLAB-Simulink software) and experimental analysis.

Human face recognition by drones is necessary for various applications, such as surveillance, search and security. The previous methods for face recognition are highly sensitive to limitations such as height, angle and distance from the face. In this article, a new approach for face detection and identification by deep learning is presented. The proposed method is done in three steps. In the first step, images are zoned with the selective search algorithm. In the second step, a deep network is proposed for box refinement operation to identify the target boxes with high accuracy and speed. Actually, a two-class classification problem is performed by deep learning to locate faces. In the third step, the localized images are trained to the proposed deep network to perform face recognition. In the architecture of the proposed method, the properties of widely used deep networks are used in combination, and a quantitative comparison of the proposed method with new methods in terms of computational complexity shows that training the proposed model requires less execution time than other methods. In addition, the evaluation of the proposed method on the DroneFace dataset shows that for different distance and height from the target, the proposed method has an average face recognition rate of 75. 9 and an average face recognition rate of 84. 6. Therefore, the proposed method has higher accuracy and efficiency than the new methods in this field and can be used for surveillance and security applications.

The most important limitations for wireless underwater telecommunication systems are multipath, fading, attenuation, and extensive interference in underwater channels. In these channels, due to wide changes in the signal-to-noise ratio, transmitting and receiving in fixed modulation mode does not have the desired performance. For this reason, in this paper, the use of turbo coding to encode an adaptive modulation system based on orthogonal frequency division multiplexing (OFDM) is introduced for the first time, which reduces the transmission error and also achieves high efficiency. In the proposed system model, the optimal system was achieved by selecting appropriate parameters for communication in the underwater acoustic environment and also choosing proper modulation modes, which decreases the bit error rate by 16% compared to other systems introduced so far for underwater acoustic communication.

Robotics has been widely used to develop industries and reduce human injuries, and by adjusting the controller to achieve proper speed and accuracy, improvement of the performance of these robots and reducing the number of injuries can be achieved. Until now, controllers have often used the governing equations of direct and inverse kinematics, aiming to control the position of the final actuator of the arm. Difficulty to solving direct and inverse kinematics equations, errors in solving equations, lack of user-friendly environment, inflexibility in decision-making and the volume of calculations are among the problems of existing robotic control systems. In this article, the lifting robot is modeled by two control methods (Classic and Fuzzy) with 4 degrees of freedom, in which four parts of the arm are checked by PID, PD and PI controllers and Matlab-Simulink is used as a tool for testing the robot's movement characteristics. It was observed that the PD controller, despite not having a high percentage, leads to a steady state error in most cases. While, the PI controller provides a favorable settling time in most cases, it has a higher increase percentage than PID. Finally, the results showed that the fuzzy PID controller provides better responses than the classical PID controller and the classical and fuzzy PI and PD controllers.

Progressive failure in the structure due to the occurrence of accidental actions and the chain effect of the failure, leads to the failure of a wide range of the structure or even the total collapse of the structure. Power transmission lines are one of the integral components in the power system. Failure of power transmission line structures is one of the main problems of power companies in different regions of the world. In this research, the progressive failure behavior of a case study of 230 kV power transmission line in Fars province has been investigated. In the present study, two methods of linear dynamic analysis and nonlinear dynamic analysis have been used to identify damage with the alternative load path method. The finite element model of the passing tower (DC0), the end tower (DC90) and the conductors of the 230 kV power transmission line have been modeled in SAP2000 software. In this research, the behavior of progressive failure due to the removal of the base of power transmission towers has been investigated and evaluated. The results showed that the passing and end masts are stable under gravity load due to the failure of one leg. Also, the results showed that due to the removal of one pillar in the progressive failure analysis, the end masts have higher resistance than the passing masts.

In the present paper, the performance of multi-arched doors against explosion has been investigated using modeling in the LSDYNA finite element software. The use of multi-arch structure, in addition to reducing the weight and volume of materials and as a result reducing the cost of providing and building the door, causes a more uniform distribution of stress and reduces the concentration of stress in comparison with common doors with a smooth surface. The parameters investigated in this research include the number of arcs, the height of the arc, the weight of the explosive, the effect of far-field explosions, the overall dimensions of the door, the materials used in the construction of the door, and the type of hardener used to increase its resistance. The numerical modeling results show that by increasing the number of arcs from 1 to 5 arcs, the maximum displacement created in the center of the door decreases by about 86%. Increasing the height of the arch from 5 to 10 and 20 cm, based on the number of different arches, the maximum displacement of the center of the door decreases between 5 and 32%. Among the different methods of strengthening the door, the use of multi-arch method increases the resistance. This method creates this resistance by consuming less steel and is more economical.