Journal of Online Engineering Education

Analysis of Hybrid Meta Heuristic Optimization Based MPPT Controller for Improved Operational Efficiency of Solar PV System

2024-08-12T15:16:02+00:00

The incorporation of sophisticated control methodologies is essential. The goal of this work is to optimise the performance of solar PV systems through the design and development of a hybrid duty cycle controller based on the Grey Wolf Optimizer-Cuckoo Search Algorithm (GWO-CSA). The main goal is to maximise power point tracking (MPPT) in a variety of environmental settings, which will increase the system's overall efficiency and dependability. The suggested hybrid GWO-CSA algorithm makes use of the cuckoo bird's brood parasitism and the social hierarchy and hunting behaviour of grey wolves to provide a reliable and effective search mechanism for the ideal duty cycle. The shortcomings of traditional MPPT approaches are addressed by this unique methodology, which improves convergence speed, accuracy, and responsiveness to sudden changes in temperature and sun irradiation. MATLAB/Simulink simulation simulations were performed to verify the effectiveness of the hybrid GWO-CSA controller. Traditional MPPT methods including Particle Swarm Optimisation (PSO), Incremental Conductance (IC), and Perturb and Observe (P&O) were evaluated using the performance metrics. The outcomes show that the hybrid GWO-CSA controller continuously beats the traditional techniques, obtaining faster reaction times and greater energy conversion efficiency. Furthermore, the hybrid GWO-CSA algorithm demonstrated enhanced stability and resilience, reducing power fluctuations and guaranteeing dependable functioning in the presence of partial shade and further environmental disruptions. The application of this cutting-edge control approach in solar photovoltaic systems has the potential to greatly improve their operational effectiveness, hence augmenting the sustainability and financial feasibility of solar energy solutions. To sum up, the hybrid GWO-CSA based duty cycle controller offers a viable way to raise the solar PV systems' operational efficiency. The construction of more robust and efficient renewable energy systems is facilitated by this research, which sets the way for future developments in intelligent control techniques.

Design and Development of Hybrid Meta Heuristic Optimization Based Duty Cycle Controller for Improved Operational Efficiency of Solar PV System

2024-08-12T15:19:15+00:00

Analysis and Design of SVM Based Brain Tumor Classification and Detection Technique

2024-08-27T11:21:29+00:00

Brain cancers can be detected using the Automatic Support Intelligent System, which utilises both a neural network and a fuzzy logic system. Both the diagnosis and treatment of brain cancers are made easier because to this technology. Finding a tumour in the brain is difficult due to the elusive nature of brain tumour cells. There remains a considerable challenge in automated medical image segmentation, which has attracted attention from researchers in recent years. Research in this area will centre around segmentation of MRI brain images (MRI). A classification problem is what we're approaching here, and we're looking for ways to distinguish between regular pixels and those that aren't. Support Vector Machine (SVM) classification is one of the most often used methods for this purpose. In the experiment, a dataset of gliomas of varied forms, sizes, and intensities will be employed. The brain serves as the central processing unit for the body. It is possible for a tumour to cause mortality if it is not discovered early enough. Magnetic Resonance Imaging (MRI) is superior to other imaging modalities when it comes to determining the tumor's size and determining its grade. MRI does not produce any harmful radiation. For the time being, there is no automated method for determining the grade of the tumour. This study demonstrates how MRI data can be used to segment and classify brain tumours. It's a helpful tool for clinicians to use when putting together therapy or surgery plans. Classifying tumours as benign or malignant requires the use of a support vector machine (SVM).

Leveraging Large Language Models (LLMs) and Advanced Machine Learning Techniques: A Comprehensive Review

2024-12-23T06:15:42+00:00

Large Language Models (LLMs) such as OpenAI's GPT, Google's BERT, and others have transformed the field of natural language processing (NLP) and artificial intelligence (AI) in recent years. These models, built on deep learning and transformer architectures, demonstrate remarkable capabilities in understanding, generating, and interacting with human language. This review paper explores the significance of LLMs, their underlying architectures, training strategies, and the integration of advanced machine learning (ML) techniques. Furthermore, we analyze their applications, ethical considerations, challenges, and future prospects. The goal is to provide a comprehensive understanding of how LLMs leverage modern ML techniques to push the boundaries of AI and NLP.

Design Simulation and Analysis of Intelligent Malware Detection Using Machine Learning Approach

2025-02-19T08:02:35+00:00

With the rapid advancement in cyber threats, malware detection has become an essential task in securing information systems. Traditional signature-based detection methods have become increasingly ineffective due to the evolving nature of malware. The advent of machine learning (ML) offers a promising alternative by enabling systems to identify and classify unknown malware based on patterns in their behaviors. This paper presents the design, simulation, and analysis of an intelligent malware detection system using machine learning techniques. Various machine learning algorithms, including supervised and unsupervised approaches, are evaluated for their effectiveness in malware detection. The results indicate that machine learning provides a robust and adaptive solution to combating modern malware threats.