Flood Prediction Based on Recurrent Neural Network Time Series Classification Boosted by Modified Metaheuristic Optimization DOI
Igor Markovic, Jovana Krzanovic, Luka Jovanovic

и другие.

Lecture notes in networks and systems, Год журнала: 2024, Номер unknown, С. 289 - 303

Опубликована: Янв. 1, 2024

Язык: Английский

Recognition of Longitudinal Cracks on Slab Surfaces Based on Particle Swarm Optimization and eXtreme Gradient Boosting Model DOI Open Access
Yu Liu, Jiang Lai, Jing Shi

и другие.

Processes, Год журнала: 2024, Номер 12(6), С. 1087 - 1087

Опубликована: Май 25, 2024

Longitudinal cracks are a common defect on the surface of continuous casting slabs, and cause increases in additional processing costs or long-time interruptions. The accurate identification longitudinal is helpful to ensure process adjusted time, which significantly improves quality slabs. In this research, typical temperature characteristics thermocouples at position their adjacent locations were extracted. principal component analysis (PCA) method was used reduce dimensions these remove redundant information. particle swarm optimization (PSO) introduced optimize parameter. On basis, recognition model established, based optimization–eXtreme gradient boosting (XGBOOST) model. Finally, trained tested using crack non-longitudinal samples compared with decision tree, tree (GBDT) XGBOOST models. test results showed that PSO-XGBOOST had best performance all evaluation indexes. accuracy, F1 score alarm rate 95.8%, 92.3% 100%, respectively, false as low 5.5%. research provide theoretical basis reliable for identification.

Язык: Английский

Процитировано

3

An AI-Based Framework for Characterizing the Atmospheric Fate of Air Pollutants Within Diverse Environmental Settings DOI Creative Commons

Nataša Radić,

Mirjana Perišić, Gordana Jovanović

и другие.

Atmosphere, Год журнала: 2025, Номер 16(2), С. 231 - 231

Опубликована: Фев. 18, 2025

This study introduces a novel artificial intelligence (AI) modeling framework that combines machine learning algorithms optimized through metaheuristics with explainable AI to capture complex interactions among pollutant concentrations, meteorological data, and socio-economic indicators. Applied COVID-19-related dataset comprising 404 variables, benzene concentrations as the target—measured using proton transfer reaction–mass spectrometry in Belgrade, Serbia—the demonstrated exceptional sensitivity assessing impact of environmental societal changes during pandemic. Explainable techniques, such SHAP SAGE, were employed reveal influence each predictor, while clustering values identified distinct settings influenced behavior. Three regarding levels onset state emergency. The first, involving local petroleum-related activities, biomass burning, chemical manufacturing, traffic, led 15.7% reduction levels. second, characterized by non-combustion processes, nocturnal chemistry, specific context, resulted 51.9% increase. third, driven industrial contributed modest 2.33% reduction. underscored critical role shaping air behavior, emphasizing importance integrating broader contexts into models gain more comprehensive understanding pollutants their dynamics.

Язык: Английский

Процитировано

0

The eXtreme Gradient Boosting Method Optimized by Hybridized Sine Cosine Metaheuristics for Ship Vessel Classification DOI
Milos Bukumira, Miodrag Živković, Miloš Antonijević

и другие.

Lecture notes in networks and systems, Год журнала: 2024, Номер unknown, С. 255 - 270

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

2

Speeding Classification by a Deep Learning Audio Analysis System Optimized by the Reptile Search Algorithm DOI

Tea Dogandžić,

Aleksandar Petrović, Luka Jovanovic

и другие.

Algorithms for intelligent systems, Год журнала: 2024, Номер unknown, С. 73 - 88

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

2

Availability optimization of power generating units used in sewage treatment plants using metaheuristic techniques DOI Creative Commons
Monika Saini, Ashish Kumar, Dinesh Kumar Saini

и другие.

PLoS ONE, Год журнала: 2023, Номер 18(5), С. e0284848 - e0284848

Опубликована: Май 4, 2023

Metaheuristic techniques have been utilized extensively to predict industrial systems' optimum availability. This prediction phenomenon is known as the NP-hard problem. Though, most of existing methods fail attain optimal solution due several limitations like slow rate convergence, weak computational speed, stuck in local optima, etc. Consequently, present study, an effort has made develop a novel mathematical model for power generating units assembled sewage treatment plants. Markov birth-death process adopted development and generation Chapman-Kolmogorov differential-difference equations. The global discovered using metaheuristic techniques, namely genetic algorithm particle swarm optimization. All time-dependent random variables associated with failure rates are considered exponentially distributed, while repair follow arbitrary distribution. switch devices perfect independent. numerical results system availability derived different values crossover, mutation, generations, damping ratio, population size value. were also shared plant personnel. Statistical investigation justifies that optimization outdoes predicting power-generating systems. In study proposed optimized performance evaluation plant. developed one can be useful designers establishing new plants purposing maintenance policies. same procedure other industries too.

Язык: Английский

Процитировано

5

Environmental Monitoring and Analysis of Persistent Organic Pollutants DOI Creative Commons
Vlasta Drevenkar, Gordana Mendaš

Toxics, Год журнала: 2023, Номер 11(6), С. 535 - 535

Опубликована: Июнь 15, 2023

Persistent organic pollutants (POPs) are a group of 28 toxic compounds different chemical classes listed in the Stockholm Convention on POPs, which aims to protect environment and human health [...].

Язык: Английский

Процитировано

4

Audio analysis speeding detection techniques based on metaheuristic-optimized machine learning models DOI
Luka Jovanovic, Nebojša Bačanin, Vladimir Šimić

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 133, С. 108463 - 108463

Опубликована: Апрель 30, 2024

Язык: Английский

Процитировано

1

Marine Vessel Trajectory Forecasting Using Long Short-Term Memory Neural Networks Optimized via Modified Metaheuristic Algorithm DOI

Ana Toskovic,

Aleksandar Petrović, Luka Jovanovic

и другие.

Algorithms for intelligent systems, Год журнала: 2024, Номер unknown, С. 51 - 66

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0

Flood Prediction Based on Recurrent Neural Network Time Series Classification Boosted by Modified Metaheuristic Optimization DOI
Igor Markovic, Jovana Krzanovic, Luka Jovanovic

и другие.

Lecture notes in networks and systems, Год журнала: 2024, Номер unknown, С. 289 - 303

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0