Прегледај по Аутор "Cosovic, Marijana"

Сада се приказује 1 - 3 од 3
  • Сличица
    Ставка
    BGP Anomaly Prediction Using Ensemble Learning
    (2019) Cosovic, Marijana; Junuz, Emina
    This paper investigates anomalies such as worms, power outages, and routing table leak (RTL) events occurring in Border Gateway Protocol (BGP) that can cause connectivity and data loss issues. Ensemble learning is a machine learning model employing multiple classifiers in order to reliably identify network anomalies. We use bagging, boosting, and random forests ensemble models trained on network anomaly datasets for classification improvement. Models were compared with respect to the following performance metrics: F-measure, Matthews correlation coefficient (MCC), Receiver operating characteristic (ROC) curve, precision-recall (PR) curves and model execution time. We observed improvement in performance measures when ensemble classifiers realized in Python were used in comparison to our previously reported results on single classifiers. Further improvement in most performance measures was observed by using sampling techniques (oversampling and undersampling) on anomalous datasets. This approach increases model execution time which is not favorable for real-time anomaly detection models.
  • Сличица
    Ставка
    Determination of Normative Value Power Losses in Distribution power grids with Renewable Energy Sources using Criterion Method
    (Faculty of Electrical Engineering, University of East Sarajevo, 2020) Gundebommu, Sree Lakshmi; Cosovic, Marijana; Rubanenko, Olena
    The paper presents the possibility of using criterion programming and neuro-fuzzy modeling in determining the value of planning technical power losses. Proposed is an optimal control in normal mode power grids which considers the value of planned technical power losses. Improved method for determining normative values of technical energy losses using the criteria programming and neuro-fuzzy modeling is presented.
  • Сличица
    Ставка
    Predicting the Power Generation from Renewable Energy Sources by using ANN
    (Faculty of Electrical Engineering, University of East Sarajevo, 2021) Rubanenko, Olena; Gundebommu, Sree Lakshmi; Cosovic, Marijana; Lesko, Vladislav
    This paper proposes power generation forecasting for photovoltaic power plants by using Adaptive Neuro-Fuzzy Inference Systems library in MATLAB and considering meteorological factors. Renewable energy sources (RES) introduce compensation instability problems in the grid hence forecasting methods are considered. Especially important for grid operators is a day ahead forecasting as it can reduce negative imbalance price. Means of ensuring the balance reliability of the power system in terms of RES integration are presented. The installation of charging stations for electric vehicles or use of hydrogen technologies and modern storage systems can provide grid balance. In addition, decreasing the deviation of the current (real) value from the predicted value of power generation is a way to compensate for power unbalance.