Assistant Professor, Department of Electronics & Communication Engineering Vel Tech Rangarajan Dr Sagunthala R&D Institute of Science and Technology
Predictive analysis of wind power using bi-directional permutation enhanced LSTM-RNN on SCADA dataset Sridhar S., Shamanth M. Hiremath, Neelamsetti Kiran Kumar, Rahul S.G., Yuvaraja Teekaraman, Amirthalakshmi T.M. Array, 2026 The transition to sustainable energy positions wind power as a key renewable solution. As demand grows, wind turbines are deployed across diverse terrains. However, wind’s stochastic nature and environmental variability complicate power forecasting, affecting grid stability. The study leverages data-driven techniques to enhance wind power forecasting using high-resolution SCADA system time-series data. Key operational parameters include wind speed, rotor speed, generator speed, nacelle orientation, ambient temperature and power output. A comparative analysis evaluates traditional machine learning models—Linear Regression, Decision Trees, Random Forests, Gradient Boosting and Support Vector Machines—against deep learning models like Long Short-Term Memory (LSTM) networks and a novel Recurrent Neural Network (RNN) architecture. The core contribution is an optimized Bidirectional LSTM-RNN model with permutation layers and attention. These layers capture long-range dependencies and nonlinear interactions in wind data. The structure improves long-range dependency capture and nonlinear interaction modeling. Bidirectionality enables learning from both past and future time steps, while attention mechanisms highlight critical temporal features. Experimental results demonstrate the proposed model’s superior performance, achieving a Mean Absolute Error (MAE) of 0.0994 and Root Mean Square Error (RMSE) of 0.1390, significantly outperforming traditional models (e.g., Random Forest: MAE 86.44, RMSE 220.30) and basic LSTM models (MAE 14.48, RMSE 15.27). Robust cross-validation confirms its ability to generalize across different temporal segments. Feature importance analysis improves interpretability, supporting informed decision-making in wind farm operations. The framework is scalable, modular and well-suited for real-time forecasting applications. The work presents a reliable deep learning model for wind power forecasting, enabling intelligent, data-driven energy management in modern power systems.
Aquaculture Fish Disease Detection Using a Hybrid Dual-Branch Convolutional Neural Network Avinaash Arjun V, Rahul S G, T M Amirthalakshmi Iccids 2026 9th International Conference on Computational Intelligence in Data Science, 2026 Aquaculture has been widely used in the world in terms of its contribution to the global food security but fish diseases have been a major challenge to sustainable production. These diseases need to be detected early and accurately to reduce the massive death rates and financial losses. Deep learning-based methods of image classification in fish disease detection are investigated in this paper. The dataset employed to test four models, namely ResNet50, EfficientNetB0, MobileNetV2, and the proposed Hybrid Dual-Branch CNN, combining ResNet50 with MobileNetV2, consisted of 697 images of seven fish diseases. According to the experimental outcomes, ResNet50 and EfficientNetB0 demonstrated the highest accuracy of 95.55% and 94.26, respectively, and MobileNetV2 had the least accuracy of 66.00. The proposed Hybrid Dual-Branch CNN performed better than all the other models in the two-class classification scenario with an accuracy of 96.84 and a macro-F1 score of 0.97. This enhanced performance is due to the complementary nature of feature extraction in the two networks where ResNet50 is more efficient in deep structural and textural features and MobileNetV2 is specialized in lightweight shape and colour patterns. These results highlight the promising nature of hybrid architectures in strong and scalable fish disease diagnostics in aquaculture.
Scalable Hybrid Ensemble Models for Crop Recommendation in Precision Farming Contexts Avinaash Arjun V, Rahul S G, T M Amirthalakshmi Iccids 2026 9th International Conference on Computational Intelligence in Data Science, 2026 Precision agriculture requires proper recommendation of crops, which makes it possible to make decisions that maximize yields and resource utilization. The conventional approaches to crop selection are not always effective in such circumstances as soil erosion and changing climate. This paper provides a machine learning model that forecasts the most appropriate crop, which is determined by the data-driven classification of soil nutrients (N, P, K), climate (temperature, humidity, rainfall) and soil pH of 22 types of crops. At the beginning, standard models (Decision Tree, Random Forest, LightGBM) are tested. The paper then suggests a composite stacking ensemble whereby LightGBM, CatBoost, Random Forest, Multi-Layer Perceptron, and XGBoost are used as base models, and Logistic Regression is used as a metalearner. The validation and training were based on a publicly available crop recommendation dataset. The findings indicate that the hybrid ensemble system achieves greater prediction accuracy—99.55% F1-score, recall, and precision—which outperforms the baseline classifiers. The above results point to promising hybrid ensemble learning as a highly scalable and robust solution to aid agricultural decisions in various real-world scenarios.
DDoS Detection Using ML Algorithm Ashok Kumar Nanda, S. Harshavardhan Reddy, S. Rahul, V. Karthikeshwar Lecture Notes in Electrical Engineering, 2026
Robust Federated Learning for Non-IID Medical Images with Trust-Guided Aggregation A. Allan, Sachin Ramesh, Rahul S G, Amirthalakshmi T M Iccids 2026 9th International Conference on Computational Intelligence in Data Science, 2026 Federated learning (FL) allows training medical professionals cooperatively without exchanging patient information. Nevertheless, the heterogeneity in the real world tends to cause client drift and unjust models. Our suggested FL framework is comprised of confidence-guided FL, dynamic regularization (FedDyn), local batch normalization (FedBN), and label-aware aggregation. Our approach, which is validated on MedMNIST benchmarks and the HAM10000 (high-resolution) dataset, is much more stable and fair than baselines. Through extensive ablation experiments, FedDyn is shown to avoid accuracy decline, and our trust-based aggregation can drop inter-client variance by a factor of more than 50, providing fair diagnosis between different hospitals.
Hybrid Deep Learning-Based EEG BCI for Accurate Mental Intention Recognition Amirthalakshmi.T.M, Kishore N, Sreyas Sanil Kumar, Rahul S G Iccids 2026 9th International Conference on Computational Intelligence in Data Science, 2026 Brain-Computer Interface (BCI) systems are changing the manner in which severe motor or speech impaired persons communicate, so that thoughts may be directly converted to actions. This paper describes a cheap, real-time EEG-based BCI that are able to identify three major mental intentions that are yes, no, and help. We have an integrated EEG acquisition system using a mixed deep learning model—Convolutional Neural Networks (CNN) for spatial network features, Recurrent Neural Networks (RNN) for time patterns and a Support Vector Machine (SVM) for final classification. This combination captures both the where and when of brain signals, which deals with the drawbacks of less complex and single-model methods. Experiments demonstrate that our system attains more than 95% classification and is useful in daily life, as it can be run in real time. Combining low costs with high technology but simple to use machine learning approaches, this work shows that there is potential to move towards available and viable assistive communication tools.
Transfer Learning Models for Robust Cardamom Leaf Disease Detection Dasari Naga Vinod, Rahul S G, Vijay Kumar Gowda B N, P. Dharmesh, A. Mohan, Ch. Siva Krishna Reddy Proceedings of the 2026 International Conference on AI Driven Smart Systems and Ubiquitous Computing Icauc 2026, 2026 Known in the trade as the "Queen of Spices," cardamom (Elettaria cardamomum) is a very important economic crop for India. But, there is a large-scale issue with leaf diseases like Phyllosticta leaf spot and Blight, which indeed do great damage to the quality and yield of the crop. Presently, for disease detection, farmers do the field work, which is a very time and labor-intensive process is also very prone to human error. This study proposes an AI-based automated system that identifies and classifies cardamom leaf conditions using the MobileNetV2 model and Transfer Learning. The system uses different Deep Learning techniques to find leaf diseases accurately. A custom dataset was prepared using publicly available data from Kaggle containing three types of cardamom leaves: Healthy, Blight, and Phyllosticta. Each image was processed through a specific pipeline consisting of resizing to 224×224 pixels, normalization, and data augmentation techniques like rotation, zooming, flipping, and brightness adjustment. The MobileNetV2 architecture was fine-tuned on ImageNet for this domain. The lower-level features obtained from convolution layers were kept intact by freezing them, and the additional regularization layers, dense and dropout, were added for training efficiency. This was done on Google Colab with the use of TensorFlow and GPU acceleration (2.17+). The model reported average training, validation, and testing accuracies of 96.4%, 94.8%, and 94.2%, respectively. The system was robust and reliable, supported by a confusion matrix and other evaluation metrics, which included precision (93.5%), recall (92.7%), and F1-score (93.1%).
Hybrid Firefly-Genetic Algorithm for Accurate Heart Disease Detection in Wearable Healthcare Systems P. Nalayini, K.R.Ram Victoria, Rahul S G, S.Padhma Vinodhini, R. Manikandan, Ramesh R 2026 International Conference on Communication Computing and Emerging Technologies Ic3et 2026, 2026 This article presents a new hybrid optimization method- Hybrid Firefly-Genetic Algorithm (HFGA) which is unique to wearable health care systems to provide more accurate predictions of heart disease and at the same time offer minimal computational load. The suggested approach combines the exploration power of the Firefly Algorithm with the powerful capabilities of the Genetic Algorithms in terms of evolution so that the best set of features and classifier parameter tuning will be accomplished. HFGA is a way of balancing local and global search in an adaptive manner to prevent premature convergence and guarantees that there are high quality subsets of the features of multi-sensor wearable data. The wearable architecture is implemented with a lightweight machinelearning classifier optimized by HFGA that allows inference in real time. Benchmark medical data and simulated wearable streams experimental assessment shows higher prediction accuracy, lower features dimensionality and better stability than traditional metaheuristics. The results present HFGA as a potential approach to the next-generation wearable health monitoring and the early cardiac risk evaluation. 96.8% was achieved accuracy by this model.
FPGA-Based CNN-LSTM System for Audio Signal Classification M.R. Ezilarasan, G. Kavitha, Rahul S G, Hangjun Che, Xiangguang Dai, Yuming Feng, Man-Fai Leung 2026 14th International Conference on Intelligent Control and Information Processing Icicip 2026, 2026
Automated Aquaponics System with AI-Based Plant Health Monitoring for Ocimum Basilicum Rahul S G, Avinaash Arjun V, Kalpana Devi P, T M Amirthalakshmi, Logeswari Panneerselvam, Talari Sofiya Rheema 2025 International Conference on Next Generation Computing Systems Intelligent System for Sustainable Development Icngcs 2025 Conference Proceedings, 2025
IOT BASED SMART AND ECONOMIC GREENHOUSE MONITORING AND AUTO-TUNED CONTROL SYSTEM FOR RURAL FARMING Journal of Theoretical and Applied Information Technology, 2024
IoT-Based Traffic Control System for Emergency Vehicles Kavitha P, K. Pavan Teja, Rahul S G, K. Sri Charan, K. Rama Krishna Reddy, V. Sai Charan Reddy International Conference on Intelligent Algorithms for Computational Intelligence Systems Iacis 2024, 2024
Scrapetalk: Chatbot Conversation with Web Scraped Insights Sriram G, Rahul S, Adaline Suji R, Priyanka Nallusamy, Nivitha K, Arumuga Arun R 2024 International Conference on Integration of Emerging Technologies for the Digital World Icietdw 2024, 2024
IoT-Enhanced Automatic Child Detection System for Cars Kavitha P, K. Pavan Teja, Rahul S G, Naga Ramakrishna Chikkam, Deepika Andra, Venkannababu Tanakala 5th International Conference on Electronics and Sustainable Communication Systems Icesc 2024 Proceedings, 2024
Remote Health Prediction System Using Machine Learning S. Rahul, P. Yakaiah, Bittu Kumar, S. Monika, Dusamulla Sampreeth, Kolupula Sai Kiran Advancements in Science and Technology for Healthcare Agriculture and Environmental Sustainability A Review of the Latest Research and Innovations Proceedings of the International Analytics Conference Iac 2023, 2024
Solar Powered Wireless Power Transmission System for Electric Vehicles Rahul S G, G. Srinivasa Sudharsan, Vijay Anand Kandaswamy, Modium Jaswanth Reddy, Gaddipati Sri Harsha, Bijjula Lahari Proceedings of the 8th International Conference on Communication and Electronics Systems Icces 2023, 2023
Smart Social Distancing Robot for COVID Safety S. G. Rahul, Velicheti Sravan Kumar, D. Subitha, Seeram Sai Sudheer, Amruthavalli Archakam, M. Nikhileswara Sri Venkat Lecture Notes in Mechanical Engineering, 2023
Contactless Fog based Handwash Kit for COVID Safety Rahul S G, Neelamsetti Kiran Kumar, Sri Lekha Y, Velicheti Sravan Kumar, Subitha D, Siripireddy Venkateswarulu 4th International Conference on Circuits Control Communication and Computing I4c 2022, 2022
Model Based Cardiac Control System for the Left Heart Using LabView S G Rahul, R Chitra, Seeram Sai Sudheer, Palevla Venkata Naga Ravi Teja, Amruthavalli Archakam, Jaswanth Reddy Modium 3rd IEEE International Virtual Conference on Innovations in Power and Advanced Computing Technologies I Pact 2021, 2021
Analysis of Baroreflex Function in Cardiovascular Variability Model Chitra R, Rahul S G, Amruthavalli Archakam, Sai Pramitha Meesala, Jaswanth Reddy Modium, Jagadish Kumar Pakalapati 3rd IEEE International Virtual Conference on Innovations in Power and Advanced Computing Technologies I Pact 2021, 2021
Soldier Strap for Health Monitoring and Tracking A Proposed Solution Rahul S G, Rajnikant Kushwaha, Sayantan Bhattacharjee, Agniv Aditya, K. Somasekhar Reddy, Durri Shahwar 3rd IEEE International Virtual Conference on Innovations in Power and Advanced Computing Technologies I Pact 2021, 2021
Adaptive controller to minimize position disturbances of tool pin while joining aluminium metal matrix composites by friction stir welding International Journal of Mechanical Engineering and Technology, 2018
Prediction of electricity load using artificial neural network for technology tower block of VIT university International Journal of Applied Engineering Research, 2017