AlKadhim Journal for Computer Science

Incipient Fault Protection using Artificial Intelligence Techniques

2024-12-05T22:53:04+03:00

In the field of incipient fault protection, various sources can cause failures, such as lightning, switching transients, mechanical imperfections, and chemical breakdown. To guard against these errors, Buchholz relays and pressure relief devices have been utilized. However, in recent years, preventive health measures have gained more attention. One popular approach is the implementation of the Dissolved Gas Analysis (DGA) system, which detects incipient faults by analyzing the gases dissolved in the transformer oil. In this context, the use of artificial neural networks (ANN) and artificial neural networks combined with expert systems (ANNEPS) has shown promise for power transformer protection against incipient faults using DGA. Power transformers, especially large oil-filled ones, are commonly subjected to DGA for identifying and diagnosing early-stage faults. By analyzing the dissolved gases and employing interpretation systems, such as (ANNEPS), unexpected failures can be prevented. The objective of this research is to identify internal problems within transformers, and an (ANN) structure has been specifically developed for this purpose. The ANNEPS approach combines the outputs of ANN and expert systems to ensure rapid and accurate identification of various types of transformer failures. By comparing the results of both computational methods, a reliable assessment can be made, enhancing the effectiveness of incipient fault protection strategies. Overall, the combination of (DGA) and advanced techniques like (ANN) and (ANNEPS) provides a robust approach to detect and prevent incipient faults in power transformers. These methods offer improved accuracy and promptness in identifying transformer failures, ultimately contributing to the reliability and efficiency of power systems.

Hybrid PSO-Bagging Approach for Efficient and Accurate Network Anomaly Detection

2025-03-23T02:07:43+03:00

The surge in internet usage has triggered a substantial increase in network attacks, raising serious cyber security concerns. Fog computing, which enhances cloud computing by providing low-latency services to mobile users, is particularly susceptible to these threats due to its proximity to end users and limited computational resources. Traditional Intrusion Detection Systems (IDS) designed for conventional networks may not directly apply to fog computing environments, where the ability to process and analyze large volumes of data efficiently is crucial. This paper presents a novel approach for network anomaly detection within fog environments, utilizing a Particle Swarm Optimization (PSO) -based Wrapper feature selection method combined with the Bagging technique. By applying this methodology to the NSL-KDD dataset, our approach effectively reduces computational complexity and improves the accuracy of intrusion detection models. The proposed system demonstrates superior performance compared to existing methods, achieving an impressive 98.3% accuracy and a low false positive rate of 1.5%. These results underscore the potential of the PSO-Bagging framework to enhance the security of fog computing systems, offering a robust solution to the growing problem of network attacks in distributed computing environments.

A Comprehensive Framework for Quality Assurance of Generative AI Text

2024-12-01T00:31:17+03:00

This paper presents a comprehensive framework for the quality assurance (QA) of text outputs generated by artificial intelligence (AI) models. The framework incorporates multiple metrics to evaluate the generated text, including grammar and spelling correctness, relevance to the prompt, and linguistic diversity. The proposed method employs the Python library language for grammatical error detection, TF-IDF vectorization coupled with cosine similarity for relevance assessment, and NLTK for measuring lexical diversity. By integrating these metrics, the framework provides a robust mechanism to ensure the generated text meets the desired quality standards. This approach is demonstrated through a sample implementation in Python, which can be easily extended and customized for various applications in generative AI.