Solar photovoltaic (PV) systems stand out as a promising solution for generating clean, carbon-free energy. However, traditional solar panel installations often require extensive land resources, which could become scarce as the population grows. To address this challenge, innovative approaches are needed to maximize solar power generation within limited spaces. One promising concept involves the development of biological tree-like structures housing solar panels. These “solar trees” mimic the arrangement of branches and leaves found in natural trees, following patterns akin to phyllotaxy, which correlates with the Fibonacci sequence and golden ratio. By adopting an alternative 1:3 phyllotaxy pattern, three solar panels can be efficiently arranged along the stem of the solar tree structure, each rotated at a 120-degree displacement. Optimizing the performance of solar trees requires effective maximum power point tracking (MPPT), a crucial process for extracting the maximum available power from solar panels to enhance the overall efficiency. In this study, a novel metaheuristic algorithm called horse herd optimization (HHO) is employed for MPPT in solar tree applications. Moreover, to efficiently manage the generated power, a cascaded buck–boost converter is utilized. This converter is capable of adjusting the DC voltage levels to match the system requirements within a single topology. The algorithm is implemented using MATLAB and embedded within a Raspberry Pi Pico controller, which facilitates the generation of pulse-width modulation (PWM) signals to control the cascaded buck–boost converter. Through extensive validation, this study confirms the effectiveness of the proposed HHO algorithm integrated into the Raspberry Pi Pico controller for optimizing solar trees under various shading conditions. In essence, this research highlights the potential of solar tree structures coupled with advanced MPPT algorithms and power management systems to maximize solar energy utilization, offering a sustainable solution for clean energy generation within limited land resources.

Fault diagnosis and detection are essential for ensuring the dependability and operational efficiency of solar photovoltaic (PV) systems. This research introduces an innovative machine learning-based fault diagnosis and detection methodology implemented on a 33 kW solar PV system located at P S R Engineering College, Sivakasi. The comprehensive study leverages five years of real-time operational data, encompassing 23,000 data instances that capture seven distinct fault categories: sensor faults, inverter faults, converter faults, grid irregularity faults, PV array mismatch faults, open circuit faults, and maximum power point tracking (MPPT) controller faults. Six machine learning (ML) models were rigorously evaluated to determine the most effective fault detection approach: the random forest classifier, decision trees, support vector machine (SVM), K-nearest neighbors (KNN), logistic regression and the naive Bayes classifier. Performance metrics, including accuracy, precision, and standard deviation (SD), were systematically analysed. The random forest classifier emerged as the best model, achieving an exceptional accuracy of 98.7% with a remarkably low SD score of 0.003; consequently, it was selected as the primary fault prediction algorithm. The implementation utilized the Python programming language and successfully validated the model through comprehensive test case scenarios. Additionally, a sophisticated prototype of a smart PV system was developed, which demonstrated advanced fault detection, data collection, and diagnostic capabilities via NodeMCU ESP32 controllers and an integrated sensor array. The key innovative features of the prototype include real-time monitoring of the PV panel voltage, current, and temperature; automated data logging to Google Sheets via the IFTTT server; an autonomous water spray mechanism for panel dust removal to optimize system efficiency; intelligent camera activation for capturing and documenting abnormal parameter conditions; and comprehensive fault diagnosis and reporting capabilities. The proposed approach significantly enhances the capabilities of PV system operators and maintenance personnel by enabling timely, accurate fault identification and localization, ultimately improving system reliability and operational performance. This technically sophisticated methodology represents a substantial advancement in predictive maintenance strategies for solar photovoltaic infrastructure, offering a robust framework for comprehensive system health monitoring and management.

Today’s security attacks on computer networks are becoming more complex and severe, which has prompted security researchers to use a variety of machine learning techniques to safeguard the information and reputation of their clients. Detecting network infiltration has long been a difficult task. Machine learning advancements have raised the way for improving intrusion detection systems (IDS). This development has led to intrusion detection’s integration into network security. Using supervised machine learning techniques, intrusion detection has attained great detection accuracy. However, it is unlikely that a machine learning (ML) classifier will be able to correctly identify all attacks, particularly obscure ones.An approach based on deep learning is presented for more precise intrusion detection. This review article presents an extensive survey and classification of deep learning-based intrusion detection techniques with an emphasis on these approaches. The main background ideas about the IDS architecture and several machine and deep learning approaches are initially presented. Then, it categorizes these schemes based on the many types of methodologies each one employs. It explains how accurate intrusion detection is achieved through the use of machine and deep learning networks. The researched IDS frameworks are then fully analysed, with final thoughts and suggested directions for the future underlined.

During the past two decades, the world has confronted many pandemic disease outbreaks. Ebola, severe acute respiratory syndrome, Middle East respiratory syndrome, and, recently, coronavirus disease (COVID-19) have had a massive global impact in terms of stress on local and global human health, economic destruction, and, above all, damage to usual human life. Analyzing past similar infections will help in drawing inferences such as maintaining social distancing, herd immunity, and vaccinating massively to go forward beyond this pandemic. The development of a forecasting model of COVID-19 infectious disease spreading rate plays a vital role in the future preparation of hospital facilities, such as setting up isolated wards, oxygen cylinders, and ventilators, etc., for future patients by the government. Also, the forecasting technique and model is in immediate need for us to understand and face the effect of this and future pandemics. The main objective of this work is to develop an intelligent model based on deep learning for forecasting or estimating COVID-19 future spreading rate in terms of confirmed, recovered, and deceased cases of 85 days in 4 states in India and India overall. Deep learning neural networks, a kind of machine learning technique, are a powerful tool to predict the future because of their nature of discovering complex nonlinear dependencies. A deep learning long short-term memory (LSTM) network, which is explicitly designed for learning long-term dependencies, is utilized in this work. Hence, one can predict 1 day ahead to any number of (up to 400) days ahead by using this model. To evaluate the performance of the deep learning forecasting model and to endorse its forecasting accuracy, the criteria of mean absolute error, mean square error, root mean square error, mean absolute percentage error, and Ro are used. The results of the proposed deep learning–based LSTM model are validated by statistical analysis and graphical analysis. Moreover, the proposed model exhibited superior forecasting accuracy.