http://210.212.227.212:8080/xmlui/handle/123456789/362 Sun, 17 May 2026 00:00:56 GMT 2026-05-17T00:00:56Z http://localhost:8080/xmlui/bitstream/id/0cfefaf1-8d81-40ac-872c-939496454c24/94e30183-02e2-4ffa-bdc9-4170eeef636b.jpg http://210.212.227.212:8080/xmlui/handle/123456789/362 http://210.212.227.212:8080/xmlui/handle/123456789/376 CHARGE SCHEDULING OF PLUG-IN ELECTRIC VEHICLES IN DC NANOGRID Vaishnavi, Vijayan; Dr. Mohammed, Mansoor O The rapid increase in environmental issues lead to the development of Plug-In Electric Vehicle (PEV). PEVs are environmentally friendly and have a lower operating cost. However, the large scale deployment of PEVs can cause several detrimental effects on power distribution systems. To address these issues, scheduled charging schemes for PEVs are formulated. Charging of PEVs using renewable energy sources can reduce the demand on the grid. However, renewable energy sources are intermittent and require grid connection for reliable power supply. The development of smart grids and technologies lead to the evolution of nanogrids, which overcome the innate flaws of conventional power distribution systems. DC nanogrids provide suitable provisions for the incorporation of sustainable energy sources and PEV charging stations. In this paper, an optimal PEV charge scheduling algorithm is presented for charging PEV in a DC nanogrid with solar photovoltaic (PV) system. The charge scheduling algorithm is based on the Time of Use Price (ToUP) tariff scheme. The algorithm schedules the charging of PEVs in slots with minimum cost and interrupts the charging during peak hours. The main aim of the charge scheduling algorithm is to minimize the charging cost of PEVs. The charging of different PEVs using the scheduling algorithm under different cases were carried out to analyze the reliability and efficacy of the scheduling algorithm. The algorithm is developed using MATLAB Thu, 30 Jun 2022 00:00:00 GMT http://210.212.227.212:8080/xmlui/handle/123456789/376 2022-06-30T00:00:00Z http://210.212.227.212:8080/xmlui/handle/123456789/375 SOLAR RADIATION FORECASTING USING MACHINE LEARNING TECHNIQUES Sarath, Sasidharan; Baiju R, Naina A well-known statistical modelling method named ARIMA has been used to forecast the total daily solar radiation generated by a solar panel located in a research facility. The beauty of the ARIMA model lies in its simplicity and it can only be applied to stationary time series. So, our time series data, which is seasonal and non-stationary, is transformed into a stationary one for applying the ARIMA model. The model is developed using sophisticated statistical techniques. The optimum model is chosen and validated using Akaike information criterion (AIC) and residual sum of squares (SSE). Another method used for solar radiation prediction is LSTM. Long short-term memory (LSTM) models based on specialized deep neural network based architecture have emerged as an important model for forecasting time-series. existing models are not good at learning long-term historical dependencies, lead to compromise in modeling non- linear solar irradiance patterns. In this paper, a novel prediction model Long Short Term Memory (LSTM) from deep neural network family is used to predict hourly solar irradiance with enhanced prediction accuracy by considering long-term historical data dependencies. The proposed model is compared with Random forest and Extreme Gradient Boost (XGBoost). Thu, 30 Jun 2022 00:00:00 GMT http://210.212.227.212:8080/xmlui/handle/123456789/375 2022-06-30T00:00:00Z http://210.212.227.212:8080/xmlui/handle/123456789/374 SMART GRID STABILITY PREDICTION AND EVALUATION USING MODERN MACHINE LEARNING ALGORITHMS Safna, S The global demand for energy is rapidly rising. As a result, energy systems must evolve and be upgraded in order to become more efficient, adaptable, and sustainable. A smart grid reduces workforce while providing consumers with safe, reliable, high-quality, and long-lasting electricity. Smart grids use digital communication technology to enable two-way flow of electricity and data. This vast amount of data needs to be processed for making better decisions for maintaining the grid stability. ML and AI approaches are utilised to acquire, store, and manage this data. This study compares the performance of modern machine learning methods for predicting smart grid stability. The dataset that was chosen contains findings from a smart grid simulation. XGBoost, Adaboost, Gradient Boosting Method (GBM), HistGBM, LightGBM and CatBoost algorithms have been implemented to forecast smart grid stability. Performance of the ML model has been evaluated based on Classification evaluation metrics. The following evaluation metrics such as accuracy, precision, recall, F1-score, MCC, specificity, training time, predicting time, AUC-ROC curve and AUC-PR CURVE are used for classification evaluation.An efficient Stacking Ensemble Classifier (SEC) model developed by using the above mentioned machine learning algorithms and compared the evaluated results of individual machine learning models with the SEC model. Thu, 30 Jun 2022 00:00:00 GMT http://210.212.227.212:8080/xmlui/handle/123456789/374 2022-06-30T00:00:00Z http://210.212.227.212:8080/xmlui/handle/123456789/373 MULTI-CLASS CLASSIFICATION OF TRANSFORMER FAULT FROM DISSOLVED GAS ANALYSIS Nowfiya, B S; Dr. Sabeena, Beevi K Power transformers are a critical part of the power system. The early-stage fault detection of Power transformer is essential for the protection and prevention of further technical and financial losses. Dissolved Gas Analysis (DGA) is a commonly used diagnosis tool for keeping track of transformer status,but the existing DGA methods are based on expertise and personal experience,so their reliability can never be guaranteed,which can lead to unreliable diagnosis.Nowadays,artificial intelligence-based techniques are widely used to enhance DGA fault detection accuracy.Here,the machine learning model tries to overcome the deficiency of conventional DGA by converting DGA into a pattern recognition problem by establishing a connection between gas concentration and incipient faults. In this thesis, a new deep learning based Bidirectional Long short-term Memory(BLSTM) is introduced for multi-class classification of transformer fault and the model’s performance is compared with that of other deep learning models. Thu, 30 Jun 2022 00:00:00 GMT http://210.212.227.212:8080/xmlui/handle/123456789/373 2022-06-30T00:00:00Z