COVID-19, a highly infectious respiratory disease a used by SARS virus, has killed millions of people across many countries. To enhance quick and accurate diagnosis of COVID-19, chest X-ray (CXR) imaging methods were commonly utilized. Identifying the infection manually by radio imaging, on the other hand, was considered, extremely difficult due to the time commitment and significant risk of human error. Emerging artificial intelligence (AI) techniques promised exploration in the development of precise and as well as automated COVID-19 detection tools. Convolution neural networks (CNN), a well performing deep learning strategy tends to gain substantial favors among AI approaches for COVID-19 classification. The preprints and published studies to diagnose COVID-19 with CXR pictures using CNN and other deep learning methodologies are reviewed and critically assessed in this research. This study focused on the methodology, algorithms, and preprocessing techniques used in various deep learning architectures, as well as datasets and performance studies of several deep learning architectures used in prediction and diagnosis. Our research concludes with a list of future research directions in COVID-19 imaging categorization.

The article presents the new power conversion for parallel-operated wind energy conversion systems. It has been formulated by new multilevel inverter (MLI) topologies with reduced switch counts, lowered conduction losses and a very good output voltage spectrum. The wind energy conversion systems included permanent magnet synchronous generator (PMSG), a diode bridge rectifier, a conventional boost converter and a novel multilevel inverter connected to an isolated load. The power conversion utilizing a novel multilevel DC to AC converter has been proven for its better efficiency, voltage utilization and power quality. The integration of wind energy conversion systems has been explored in MATLAB Simulink, and the hardware setup does the authentication of the MLI structure.

Designing the control structures of fractional order PID controllers has proven to be effective in providing adaptability in set point tracing the performance of a nonlinear cruise control system. Wheel rolling resistance, wind drag force, and road gradient are incorporated into the design to better describe the system under consideration and to show how the nonlinear cruise control system behaves. This study presents a comparative investigation using simulation between control structures such as fractional order proportional–integral–derivative, fractional order integral minus proportional derivative, and fractional order proportional integral minus derivative. By preserving integral error indices as the goal function, a genetic algorithm is used to improve the controller gain parameters and fractional scaling values. To prevent integral windup conflicts and derivative boost issues, both traditional fractional order structures and adaptive neuro-fuzzy-based fractional order structures were used to create the adaptive cruise control system. The FO ANFIS PD plus I controller for the cruise control system exceeds the competition in servo and regulatory difficulties.