Shreyas Rajendra Hole

Dr. Shreyas Rajendra Hole


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
9 December 2021 in Energy Harvesting and Systems
Verified
Journal Article
Closed access
SHShreyas Rajendra Hole
Shreyas Rajendra Hole
AGAgam Das Goswami
Agam Das Goswami
Photovoltaic (PV) systems have paved their way into general households due to their high efficiency, low deployment cost and huge power savings. These advantages combined with Government incentives further assist in wide-scale adoptability of the solar powered systems. PV systems generate direct current (DC) outputs, which needs to be converted into alternating current (AC) via inverters. The efficiency of inverter design decides the overall efficiency of the PV system, which allows effective utilization of the solar power for feeding to grid or for local usage. In order to design effective inverter models, a large number of electrical configurations are designed by researchers over the years. These include, stand-alone inverters, grid-tie inverters, battery backup inverters and hybrid inverters, each of which are further divided into multiple sub-categories. Each of these sub-categories have a different application, for instance, string-converters are used for moderate power applications up-to 150k W, while central converters are used for high power applications above 80k W, etc. Apart from power capabilities, these designs vary in terms of efficiency of conversion, usability, cost, etc. Due to so many parametric variations, effective selection of these converters for a given PV application becomes ambiguous. In order to reduce this ambiguity, the underlying text statistically evaluates performance of some of the most efficient PV converter models, and compares them on the basis of power capabilities, accuracy of conversion, converter used, control model used, etc. This review will assist researchers and system designers to select the best suited models for their given applications, and thus reduce the time needed for efficient PV inverter system design. This text also recommends future research which can be adopted for improving efficiency of these systems.

28 February 2025
Verified
Conference Paper
Closed access
SHShreyas Rajendra Hole
Shreyas Rajendra Hole
VKVinothkumar Kolluru
Vinothkumar Kolluru
Shreekant SalotagiShreekant Salotagi
Shreekant Salotagi
YCYagnesh Challagundla
Yagnesh Challagundla
SMSudeep Mungara
Sudeep Mungara
SRSriramkumar R
Sriramkumar R
The latest evolution in power systems, is ‘smart grids'that offers real time monitoring, control features as well as effective management of renewable energy sources. Nonetheless, with the increase in the system complexity, the chances for insta- bility also increase and thus it is crucial to forecast the stability accurately for the safety of the grid. Weather it is in terms of Gradient Boosting Tree Algorithm or any other traditional algorithm, these machine learning models are successful in achieving the objectives of the task but they lack a forthcoming metric to assist in high stake situations, uncertainty quantification. The paper is devoted to the analysis of the reliability of smart grid systems based on Bayesian Neural Networks (BNN) taking into account both the reliability of predictive modeling and forecasting uncertainty. The smart grid stability augmented dataset was used to benchmark the efficacy of using BNNs or BNNs combined hybrid models, one with XGBoost, for feature extraction and the other with LightGBM. The present study shows that the BNN used with the above mentioned smart grid system achieved an accuracy of 92% with precision of 0.89, recall of 0.91 and F1 of 0.90. These metrics are appending with the help of the additional features of BNN where they are capable of addressing the key questions around uncertainty and thus the metrics confirming the predictions are quite strong in terms of real-time grid stability evaluation. This article encapsulates a detailed study of BNN architecture, the role of uncertainty in machine learning models and the natural application of such models in the design of dependable and fault-tolerant smart grids.

18 May 2023 in Energy Harvesting and Systems
Verified
Journal Article
Open access
SHShreyas Rajendra Hole
Shreyas Rajendra Hole
AGAgam Das Goswami
Agam Das Goswami
This paper presents a new approach to improve the performance of Zeta converters, which are commonly used in cost-sensitive circuits to manage unregulated power supply. The converters are designed to produce positive output voltages based on input voltages, and they use a buck controller to power a PMOS-based FET for high-side control. Compared to other converters, such as SEPIC, Zeta converters are smaller and more scalable for micro applications due to the use of coupled inductor circuits. The performance of Zeta converters is heavily influenced by the ratings of their passive components. To optimize component rating choices, researchers have developed several pattern analysis models. However, these models often require context-specific ratings and lack a parameter selection method for continual reconfigurations, making them difficult to deploy in practice for different use cases. To address these limitations, the authors propose a hybrid soft computing methodology for passive component selection in multiple load Zeta converters. The proposed approach combines Particle Swarm Optimization (PSO) to determine initial component ratings and Grey Wolf Optimization (GWO) to improve conversion efficiency, output gain, and Total Harmonic Distortion (THD). This is achieved by modeling a fitness function that incorporates output metrics and optimizes them incrementally for real-time deployments. The results show that the suggested methodology can reduce THD by 6.5 %, increase conversion efficiency by 3.4 %, and maintain a gain improvement of 1.5 % across numerous use cases. These improvements make the model suitable for real-time use applications. Overall, the proposed approach provides a promising solution to the challenges of passive component selection in Zeta converters, which can lead to more efficient and cost-effective power management in various circuits.

8 November 2022 in Energy Harvesting and Systems
Verified
Journal Article
Closed access
SHShreyas Rajendra Hole
Shreyas Rajendra Hole
AGAgam Das Goswami
Agam Das Goswami
The output of Solar Panels is directly dependent on the intensity of direct Sunlight that is incident on the panels. But this efficiency reduces due to shadow effects for rooftop-mounted panels. These shadows can come from other solar panels, nearby buildings, or high-rise structures. It is possible to optimize Maximum Power Point Tracker (MPPT) controllers, which draw the most power possible from PV modules by forcing them to function at the most efficient voltage to increase the output of solar panels even while they are in the shade. Thus, the MPPT analyses the output of the PV module, compares it to the voltage of the battery, and determines the best power the PV module can provide to charge the battery. It then converts that power to the optimum voltage to allow the battery to receive the maximum level of currents. Additionally, it can power a DC load linked directly to the battery. Existing shadow detection and MPPT control models are highly complex, which increases their computational requirements, thereby reducing the operating efficiency of the solar panels. This text discusses a novel Saliency Map-based low-complexity shadow detection model for Solar panels to overcome this issue. The proposed model initially extracts saliency maps from connected Solar panel configurations and evaluates the background for the presence of shadows. Based on the intensity shadows, the model tunes MPPT parameters for optimal voltage & current outputs. Due to this, the model can maximize Solar panel output by over 8.5%, even under shadows, making it useful for various real-time use cases.

6 February 2024 in Science and Technology for Energy Transition
Verified
Journal Article
Open access
PDF file
SHShreyas Rajendra Hole
Shreyas Rajendra Hole
AGAgam Das Goswami
Agam Das Goswami
With the increasing demand for renewable energy, solar power has emerged as a promising option for sustainable power generation. However, the effectiveness and efficiency of solar power systems rely on the ability to effectively manage their performance, making it essential to develop efficient control models. This paper proposes a novel ensemble predictive control model for solar deployments using bio-inspired optimizations to improve load-connected solar deployments’ performance. The proposed model integrates multiple control devices, including Maximum Power Point Tracker, Proportional-Integral-Derivative, Proportional-Integral, and Fuzzy Logic Controllers, to selectively control the solar Photovoltaic systems. The proposed model incorporates a predictive control operation utilizing an LSTM-GRU (Long Short-Term Memory-Gated Recurrent Unit) with the VARMA (Vector Auto-Regressive Moving Average) model, which can accurately predict the future power generation of the solar system. This feature can facilitate efficient energy management and increase the system’s performance for different use cases. Implement a SEPIC (Single Ended Primary Inductor Capacitor) converter design to improve the system’s overall efficiency levels. To validate the effectiveness of the proposed approach, the author conducted experiments using real-world data and compared the proposed results with other control strategies. The results demonstrate that the ensemble predictive control model based on bio-inspired optimizations outperforms the existing control models regarding accuracy, efficiency, and stability levels. The proposed model has the potential to significantly improve the performance of load-connected solar deployments, offering a more practical approach to solar power generation. The combination of predictive control operations with bio-inspired optimizations can facilitate the design of sustainable energy systems with higher efficiency and accuracy.

8 July 2025 in Healthcare
Verified
Journal Article
Open access
PDF file
VKVinothkumar Kolluru
Vinothkumar Kolluru
SHShreyas Rajendra Hole
Shreyas Rajendra Hole
ASAjeeb Sagar
Ajeeb Sagar
ACAdvaitha Naidu Chintakunta
Advaitha Naidu Chintakunta
JRJeevaraj R
Jeevaraj R
Shreekant SalotagiShreekant Salotagi
Shreekant Salotagi
Background: Antimicrobial resistance (AMR) in Neisseria gonorrhoeae is an escalating global health challenge, affecting over 82 million individuals each year. The increasing resistance to commonly used antibiotics such as azithromycin, ciprofloxacin, and cefixime hinders timely and effective treatment, primarily due to the delayed detection of resistant strains. Methods: To overcome these limitations, a hybrid machine learning (ML) and deep learning (DL) framework was developed using a dataset comprising 3786 N. gonorrhoeae isolates. The dataset included clinical metadata and phenotypic resistance profiles. The preprocessing steps involved handling 23% data sparsity, imputing 31 skewed columns, and applying resampling and harmonisation techniques sensitive to data skewness. A predictive pipeline was constructed using both clinical variables and genomic unitigs, and a suite of 33 classifiers was evaluated. Results: The CatBoost model emerged as the top-performing ML algorithm, particularly due to its proficiency in handling categorical data, while a three-layered neural network served as the DL baseline. The ML models outperformed genome-wide association study (GWAS) benchmarks, achieving AUC scores of 0.97 (ciprofloxacin), 0.95 (cefixime), and 0.94 (azithromycin), representing a 4–7% improvement. SHAP analysis identified biologically relevant resistance markers, such as penA mosaic alleles and mtrR promoter mutations, validating the interpretability of the model. Conclusions: The study highlights the potential of ML-driven approaches to enhance the real-time prediction of antimicrobial resistance in N. gonorrhoeae. These methods can significantly contribute to antibiotic stewardship programs, although further validation is required in low-resource settings to confirm their generalisability and robustness across diverse populations.