An active machine learning technique for monitoring the voltage stability in transmission systems is presented. It has been shown that algorithms may be used to supplement traditional simulation approach, but they suffer from difficulties of online model update and offline training data preparation. We propose an solution enhance existing applications by actively interacting with prediction process. The identifies operating points where predictions based on power system measurements...

Artificial intelligence has found its use in various fields during the course of development, especially recent years with enormous increase available data. Its main task is to assist making better, faster and more reliable decisions. machine learning are increasingly finding their application medicine. This true for medical that utilize types biomedical images where diagnostic procedures rely on collecting processing a large number digital images. The helps consistency boosts accuracy...

This study evaluates the impact of web-based blended learning in physiology course at Faculty Medicine, University Montenegro. The two main goals were: to determine e-learning on student success mastering course, and assess user satisfaction after introduction e-learning. compared a group students who attended before, with Moodle platform was fully implemented as an educational tool. Formative summative assessment scores were between these groups. high vs. low use analyzed. among users...

The impact of digital transformation is becoming visible in every aspect our lives. Digital strategies mostly rely on disruptive technologies such as Internet Things, augmented reality, artificial intelligence, and blockchain. Blockchain provides decentralized, immutable, trustless database distributed across all participants it was initially proposed ledger for cryptocurrency. However, besides the finance sector, blockchain technology considered to be a major innovation enabler various...

This study goes through basic principles of environmental monitoring in order to propose a simple real-time based on the Internet Things technology. The proposed solution utilizes inexpensive and widely available hardware software components making it suitable for both personal commercial use. sensor node is an ESP32 microcontroller equipped with sensors monitoring. data collected integrated using Blynk's cloud-based web application as backbone developed system. Blynk cloud platform provide...

This research describes the use of high-performance computing (HPC) and deep learning to create prediction models that could be deployed on edge AI devices equipped with camera installed in poultry farms. The main idea is leverage an existing IoT farming platform HPC offline run train for object detection segmentation, where objects are chickens images taken farm. can ported from a new type computer vision kit enhance digital farm platform. Such sensors enable implementing functions such as...

As artificial neural network architectures grow increasingly more efficient in time-series prediction tasks, their use for day-ahead electricity price and demand prediction, a task with very specific rules highly volatile dataset values, grows attractive. Without standardized way to compare the efficiency of algorithms methods forecasting metrics, it is hard have good sense strengths weaknesses each approach. In this paper, we create models several predicting on HUPX market load Montenegro...

Blockchain offers decentralized, trustworthy and immutable data storage based on distributed ledger technology. technology is recognized as an innovation enabler in many areas, with the food supply chain being one of them. This paper provides a systematic literature review current state adoption blockchain agri-food sector, specifically focus wine chain. has potential to improve traceability authenticity provenance products, increase consumer trust, reduce fraud errors. With these goals...

This research explores the role of synthetic data in enhancing accuracy deep learning models for automated poultry farm management. A hybrid dataset was created by combining real images chickens with 400 FLUX.1 [dev] generated images, aiming to reduce reliance on extensive manual collection. The YOLOv9 model trained various compositions assess impact detection performance. Additionally, annotation techniques utilizing Grounding DINO and SAM2 streamlined labeling, significantly reducing...

A new iterative method for frequency estimation in the unbalanced three-phase power system is proposed. The proposed solution based on transformation of voltage signal into a complex exponential by using orthogonal constant modulus algorithm. In addition, transformed signal, simple adaptive algorithm Beside approach, any estimator designed single-tone can be used here, which also advantage method. Simulation results confirm that exhibits better performance compared to considered algorithms.

Fine frequency estimation of a single complex sinusoid is considered. We propose to refine the provided by state-of-the-art methods based on three discrete Fourier transform (DFT) samples around DFT maximum. To that end, parabolic interpolation periodogram peak used. With calculation only additional samples, Cramér-Rao lower bound met. Moreover, proposed method's performance does not depend displacement. Due its simplicity, efficiency and low computational burden, method suitable for...

In the recent years, rapid advances in technology, especially Artificial Intelligence (AI) and Machine Learning (ML), have impacted way economy functions, as well society at large. The integration of these technologies tourism plays a significant role cultural heritage preservation. This research explores intersection artificial intelligence, heritage, tourism, with focus on Montenegro's efforts to leverage digital transformation for development. Many monuments are not marked there is no...

Compared to traditional statistical models, Machine Learning (ML) algorithms provide the ability interpret, understand and summarize patterns regularities in observed data for making predictions an advanced more sophisticated way. The main reasons advantage of ML methods are a small number significant predictors which means limited informative capability, pseudo-correct regular patterns, used without previous understanding causality. Also, some methods, like Artificial Neural Networks, use...

One of the most interesting enabling technologies for digital transformation is computer vision. Object and character recognition has already become very popular it used in everyday life. This research focuses on use vision to read serial numbers from wine labels order enable applications based tracking tracing each individual bottle. After experimenting with several OCR tools, an open source software called Tesseract engine was selected pilot solution. The paper discusses implementation...

This paper presents a practical approach to evaluation of the operational health and reliability circuit breakers in substations. Motivated by recent failure surveys conducted CIGRE working groups, breaker condition data, obtained monitoring critical parameters such as timings, tank gas pressure temperature, mechanism traveling time speed, etc., is used evaluate proposed index. The measure would help identify transmission lines available for switching actions from view point. index also...

Counterfeit wine presents a significant issue for winemakers since it affects producer's reputation, profit and most importantly can be harmful the consumers. This paper describes brand protection anti-counterfeiting solution industry based on smart tags Cloud enabled technologies. The main idea behind is to utilize quick response codes functional inks supported by system two-way communication between winemaker end-user. proposed expected make counterfeiting hard unprofitable.

This paper describes an effort to utilize IoT, OCR, and blockchain technology create wine track trace system evaluated in a real-life environment. The research is focused on digital transformation traditional supply chain, using computer vision read the existing serial numbers labeled bottles, so as uniquely identifying individual bottles of item life-cycle. provides mobile app allow end consumers scan each bottle learn more about that particular product instance, its origin, authenticity,...