Siddaraj U

Mr. Siddaraj U

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
SUSiddaraj U
Siddaraj U
UYUdaykumar R. Yaragatti
Udaykumar R. Yaragatti
NHNagendrappa Harischandrappa
Nagendrappa Harischandrappa
The microgrid is a group of smaller renewable energy sources (REs), which act in a coordinated manner to provide the required amount of active power and additional services when required. This article proposes coordinated power management for a microgrid with the integration of solar PV plants with maximum power point tracking (MPPT) to enhance power generation and conversion using a hybrid MPPT method based on particle swarm optimization-adaptive neuro-fuzzy inference system (PSO-ANFIS) to acquire rapid and maximum PV power along with battery energy storage control to maintain the stable voltage and frequency (V-f) of an isolated microgrid. In addition, it is proposed to provide active and reactive power (P-Q) regulation for the grid connected. The approach used provides more regulation due to the least root mean square error (RMSE), which improves photovoltaic (PV) potential extraction. The comparison results of the PSO-ANFIS and P&O controllers of the MPPT and the controller of the energy storage devices combined with the V-f (or P-Q) controller of the inverter all show effective coordination between the control systems. This is the most important need for contemporary microgrids, considering the potential of changing irradiance in the grid following mode, the grid forming mode under an island scenario, and back-to-grid synchronization. With the test model, the islanded and grid-islanded-grid connected modes are investigated separately. The results demonstrate conclusively that the proposed strategies are effective. To run the simulations, MATLAB and SimPowerSystems are utilized.

6 February 2026 in Scientific Reports
Verified
Journal Article
Open access
PDF file
SUSiddaraj U
Siddaraj U
UYUdaykumar R. Yaragatti
Udaykumar R. Yaragatti
LPLakhshman Rao S. Paragonda
Lakhshman Rao S. Paragonda
STSwathi Tangi
Swathi Tangi
To maximize photovoltaic (PV) energy extraction, this study proposes a novel hybrid maximum power point tracking (MPPT) method that combines artificial neural networks (ANNs) with particle swarm optimization (PSO). The ANN–PSO controller is integrated within a PV-battery microgrid system and enables efficient tracking of the maximum power output while minimizing oscillations. The MPPT unit operates alongside a droop-controlled inverter to coordinate the power flow between the PV array and battery energy storage system (BESS), supporting dynamic transitions between grid-connected and islanded modes. The controller monitors the current and voltage at the point of common coupling (PCC) to ensure reliable power delivery. By enhancing the accuracy of irradiance estimations, the ANN–PSO approach improves tracking responsiveness and overall system efficiency. The proposed strategy is benchmarked against traditional and intelligent MPPT techniques, including Perturb & Observe, ANFIS-PSO, and PSO-SMC, under varying irradiance and load conditions. The performance is validated through detailed MATLAB/Simulink simulations. The results demonstrate superior tracking performance and faster, more stable microgrid operation, highlighting the controller’s potential for efficient renewable energy integration. This work supports the advancement of intelligent, autonomous energy systems and contributes to the development of resilient, grid-interactive solar microgrids. This research supports SDG 7: Affordable and Clean Energy by enhancing the efficiency and integration of solar energy systems, and it aligns with SDG 9: Industry, Innovation, and Infrastructure through the development of advanced control strategies for smart power grids.

16 March 2026 in PLOS ONE
Verified
Journal Article
Open access
PDF file
PPPriyanka Panchal
Priyanka Panchal
VPVishal Polara
Vishal Polara
SUSiddaraj U
Siddaraj U
ABAbdullah Baz
Abdullah Baz
SPShobhit K. Patel
Shobhit K. Patel
This paper provides a novel deep learning model for captioning of images by using an advanced vision transformer architecture with a powerful LLM. Proposed models show a significant improvement over traditional CNN-RNN hybrids and existing transformer-based approaches by integrating a unique cross-attention mechanism that enables deep alignment between linguistic context and visual features. We show the superiority of our proposed architecture through extensive evaluation on different datasets like MSCOCO, Flickr30K, and NoCaps. The proposed model consistently shows good performance for leading methods such as GIT, BLIP-2, and CoCa across a comprehensive suite of metrics. On the MS COCO dataset, the BLEU-4, METEOR, and CIDEr scores of proposed models are equal to 0.495, 0.390, and 1.32, respectively. In this paper, we have critically analyzed the key challenges of this field, like enhancing caption diversity, ensuring robust multimodal alignment, and mitigating inherent biases. By providing a new performance level, the proposed model provides a source of reference for the next generation of image captioning systems. The results show the efficiency of our fusion strategy and facilitate the development of techniques that use models that can produce more precise, contextually rich, and human-like image depictions. This work supports SDG 9 (Industry, Innovation, and Infrastructure) by advancing multimodal AI systems, and SDG 4 (Quality Education) by enabling intelligent and accessible image understanding technologies.

14 January 2026 in Scientific Reports
Verified
Journal Article
Open access
PDF file
DLDawa Chyophel Lepcha
Dawa Chyophel Lepcha
BGBhawna Goyal
Bhawna Goyal
ADAyush Dogra
Ayush Dogra
VVVivek Vullikanti
Vivek Vullikanti
JMJ. Albert Mayan
J. Albert Mayan
SUSiddaraj U
Siddaraj U
Underwater images typically suffer from poor visibility, low contrast, and severe color distortion caused by wavelength-dependent absorption and scattering of light. These degradations not only reduce visual quality but also affect subsequent analysis and interpretation in marine and robotic imaging applications. To address these challenges, this study presents an efficient underwater image enhancement (UIE) framework that integrates color balancing, morphological residual processing, and gamma correction to achieve natural color restoration and structural enhancement. Initially, an adaptive color compensation strategy corrects the imbalance in red and blue channels, followed by morphological residual processing that refines fine textures while suppressing unwanted noise. The enhanced outputs are then fused through an adaptive multiscale fusion process guided by optimized weight maps to preserve both global illumination and local detail. A final gamma correction step ensures perceptually balanced contrast and brightness. The proposed method requires no training data or prior depth estimation making it computationally efficient and robust for real-time applications. Extensive experiments conducted on multiple benchmark underwater datasets demonstrate that the proposed approach consistently outperforms 22 state-of-the-art UIE techniques in both qualitative and quantitative assessments. The method achieves superior results in terms of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), underwater image quality measure (UIQM), and underwater color image quality evaluation (UCIQE) metrics, confirming its capability to restore realistic colors, enhance visibility, and preserve fine details. The proposed framework provides an effective and lightweight solution for practical underwater imaging enhancement. This work supports SDG 14 (Life Below Water) by enhancing underwater imagery for marine monitoring, SDG 9 (Industry, Innovation and Infrastructure) through an efficient real-time imaging framework, and SDG 12 (Responsible Consumption and Production) by enabling accurate underwater inspection that promotes sustainable resource use.

21 January 2026 in Scientific Reports
Verified
Journal Article
Open access
PDF file
ATAyyarao Tummala
Ayyarao Tummala
GKG. Indira Kishore
G. Indira Kishore
ADArk Dev
Ark Dev
SUSiddaraj U
Siddaraj U
This paper proposes a Kabaddi Game Optimizer (KGO), a sport-inspired metaheuristic for accurate solar photovoltaic (PV) modelling. KGO models Kabaddi strategies (Dubki and Akraman), adaptive weights and weak-player replacement to balance exploration and exploitation. Its performance is first validated on the CEC 2017 benchmark set against seven well-known optimizers, where KGO consistently attains the best average rank. KGO is then coupled with the Newton–Raphson method to estimate parameters of single, double, and triple diode PV models and a PWP-201 PV module. Using RTC France cell and module data, KGO achieves RMSE values of 7.729857E−04, 7.43146E−04, 7.3771E−04, and 2.0529E−03 for the single-, double-, and triple-diode models, and the PV module, respectively, demonstrating accurate, robust, and fast PV parameter estimation.

5 March 2026 in Scientific Reports
Verified
Journal Article
Open access
PDF file
GAGhazala Ansari
Ghazala Ansari
RSRupali Singh
Rupali Singh
SKSachin Kumar
Sachin Kumar
RKRavi Kumar
Ravi Kumar
SUSiddaraj U
Siddaraj U
In controlled dynamic laboratory conditions representative of real-world variability, the gases are usually mixtures rather than individual components. In chemometrics and electronic-nose (E-nose) systems, accurate concentration is a key challenge. Standard algorithms like support vector machine (SVM), k-nearest neighbors (KNN), and shallow multi-layer perceptron (SMLP) have limited feature extraction and poor generalization ability when faced with overlapping responses from sensors, as well as strong cross-gas interactions. While deep learning methods offer enhanced performance, they generally rely on large labeled datasets and cannot inherently maintain robustness in mixed gas scenarios. In order to overcome this limitation, the paper presents a hybrid approach of 3D convolutional neural networks (3DCNNs) with semi-supervised generative adversarial network (SGAN). The 3DCNN component captures spatiotemporal dynamics of sensor array responses, while the SGAN improves generalization under limited labeled data by generating realistic synthetic samples. Experimental results demonstrated that the model obtains a classification accuracy of 99.10%, which is higher than SVM (93.80%), KNN (92.60%), and SMLP (95.30%). These results indicate that the model can be well used for high-precision gas mixture analysis. This work supports SDG 9 (Industry, Innovation and Infrastructure) through advanced AI-based gas sensing, SDG 11 (Sustainable Cities and Communities) by enabling accurate air-quality monitoring, and SDG 3 (Good Health and Well-Being) by improving detection of hazardous gas mixtures that impact public health.