Mayur Rele

Mr. Mayur Rele

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2018
Verified
Conference Paper
Closed access
KDKushal Rashmikant Dalal
Kushal Rashmikant Dalal
Mayur ReleMayur Rele
Mayur Rele
A threat can be anything that causes potential damage to the network system. These threats can turn out to be a attack to the system. Threat may occur in any forms like viruses, outright attack, and phishing attack from hackers to gain information. Such attacks put a user's system and also business system at risks. Cyber security aims at the protection of system from attacks like unauthorized network access, intrusions attacks etc. This paper presents a novel architecture model based on machine learning for the prediction of Cyber security malware that requires execution in a sandbox environment. In order to prevent the attackers from infiltrating the system Machine Learning approach is adapted.

3 October 2023 in Frontiers in Energy Research
Verified
Journal Article
Open access
SASadaquat Ali
Sadaquat Ali
Lahcen El IysaouyLahcen El Iysaouy
Lahcen El Iysaouy
MLMhammed Lahbabi
Mhammed Lahbabi
YBYounes BOUJOUDAR
Younes BOUJOUDAR
SASultan J. Alharbi
Sultan J. Alharbi
SNStephanie Ness
Stephanie Ness
Introduction: The utilization of solar energy in large-scale photovoltaic arrays has gained immense popularity on a global scale. However, shadows in the array lead to significant reductions in power output and create multiple power peaks in the P-V characteristics. To address this issue, the Total Cross Tie (TCT) interconnection pattern is commonly employed to minimize mismatch loss. Additionally, physical relocation methods have proven effective in dispersing shadows.Method: In this context, the Magic Square View (MSV) offers a physical rearrangement of PV modules within a TCT scheme, effectively scattering shadows across the entire photovoltaic array.Results: Simulation results confirm the MSV efficacy in enhancing the PV array’s output power under various Partial Shading Conditions (PSCs) patterns. Four PSCs patterns (Short and Wide, Long and Wide, Long and Narrow, and Short and Narrow) are considered and compared to the TCT and the recently validated Competence Square (CS) techniques. The MSV method is vital in improving the PV array’s power output, especially when confronted with Long and Wide shading patterns. The outcomes demonstrate that adopting the MSV configuration leads to a substantial increase of 33.78% and 29.83% in power output for LW and SW shading patterns, respectively, compared to the TCT setup. Even under LN and SN shading patterns, there is a notable power enhancement, achieving a remarkable 25.15% increase for the LW shading pattern compared to the TCT, surpassing enhancements achieved by SuDoKu, DS, and CS methods, which improved by 20.5%, 18.2%, and 21.6%, respectively. Overall, the MSV configuration presents a promising solution for enhancing the performance of photovoltaic arrays under shading conditions.

3 September 2023
Verified
Conference Paper
Closed access
Mayur ReleMayur Rele
Mayur Rele
DPDipti Patil
Dipti Patil
This paper investigates the use of transfer learning in MRI-based brain tumor detection. The goal is to develop a reliable and efficient model for accurately classifying brain tumors, addressing the challenging task of interpreting complex and variable tumor characteristics. By leveraging transfer learning, the model can benefit from prior knowledge and improve accuracy. The study emphasizes the significance of early diagnosis and treatment planning in enhancing patient outcomes. In this research, the proposed model is trained on a large dataset of MRI images and fine-tuned using transfer learning to improve its accuracy and efficiency. The performance of the model is evaluated using precision, recall, F1-score, and support metrics, and an average accuracy of 97% is achieved. The proposed model can assist medical professionals in making accurate diagnoses and improve patient outcomes.

1 August 2024 in International Journal of Electrical and Computer Engineering (IJECE)
Verified
Journal Article
Open access
PDF file
SNStephanie Ness
Stephanie Ness
YBYounes BOUJOUDAR
Younes BOUJOUDAR
AAAyman ALJARBOUH
Ayman ALJARBOUH
Lahcen El IysaouyLahcen El Iysaouy
Lahcen El Iysaouy
MAMohamed AZEROUAL
Mohamed AZEROUAL
Mayur ReleMayur Rele
Mayur Rele
The existing battery management systems perform many functions, such as simply monitoring the battery's voltage, current, and temperature for the most basic and compensating energy imbalances between battery cells for the most advanced systems. In this last example, the function balancing helps protect the battery from obtaining a better lifespan. However, these systems with such functions remain complex because they involve techniques specific to power electronics and energy conversion. The number of components, implementation complexity, and cost increased. The work presented in this paper fits directly into this context. The main objective is to provide a solution to the problem of battery management and careful pack cell balancing. The proposed system aims to balance the battery pack cells based on the intermediate state of charge by charging or discharging the imbalanced cell. The implementation of the proposed control strategy was for a battery pack composed of five cells under MATLAB/Simulink.

22 August 2023
Verified
Conference Paper
Closed access
Mayur ReleMayur Rele
Mayur Rele
DPDipti Patil
Dipti Patil
This study illustrates how machine learning (ML), deep learning (DL), and anomaly-based techniques can be combined for network intrusion detection. The objective is to improve intrusion detection systems’ capability to identify known and novel network threats. While Random Forest is used for classification and feature selection in the ML portion, Convolutional Neural Networks (CNN) capture nuanced spatial patterns in network traffic data in the DL portion. Statistical analysis techniques are used to identify anomalies. Integrating the two approaches, the hybrid method enhances detection accuracy while minimizing false positives. Using real-world datasets, we demonstrate that the proposed method outperforms contemporary machine learning (ML), deep learning (DL), and anomaly-based methods. This hybrid method may greatly facilitate the proactive identification of hazards in the constantly evolving field of intrusion detection. It may be necessary to conduct additional research on optimization strategies and the hybrid model’s potential network-wide applicability.

4 September 2024
Verified
Book Chapter
Closed access
PDF file
Mayur ReleMayur Rele
Mayur Rele
AJAnitha Julian
Anitha Julian
DPDipti Patil
Dipti Patil
UKUdaya Krishnan
Udaya Krishnan