Wyniki wyszukiwania - BazTech

1

Improving road safety: supervised machine learning analysis of factors influencing crash severity

Li Khanh Giang

Zeszyty Naukowe. Transport / Politechnika Śląska

|

2025

|

z. 127

129--153

EN

Road traffic crash severity is shaped by a complex interplay of human, vehicular, environmental, and infrastructural factors. While machine learning (ML) has shown promise in analyzing crash data, gaps remain in model interpretability and region-specific insights, particularly for the UK context. This study addresses these gaps by evaluating supervised ML models – Decision Tree, Support Vector Machine (SVM), and LightGBM – to predict crash severity using 2022 UK accident data. The research emphasizes interpretability through SHapley Additive exPlanations (SHAP) to identify critical factors influencing severity outcomes. Results demonstrate that LightGBM outperforms other models in predictive performance, with police officer attendance at the scene, speed limits, and the number of vehicles involved emerging as pivotal determinants of severity. The analysis reveals that higher speed limits and single-vehicle collisions correlate with severe outcomes, while police presence may mitigate accident severity. However, the study acknowledges limitations, including dataset constraints. By integrating ML with post-hoc interpretability techniques, this work advances actionable insights for policymakers to prioritize road safety interventions, such as optimizing enforcement strategies and revising speed regulations. The findings underscore the potential of interpretable ML frameworks to enhance understanding of crash dynamics and inform targeted safety measures, contributing to global efforts to reduce traffic-related fatalities and injuries.

2

A comparative study of traditional excel forecasting methods and machine learning techniques for freight volume prediction in Uzbekistan

Turaev Komil, Budzynski Artur, Sładkowski Aleksander

Transport Problems

|

2025

|

T. 20, z. 2

19--31

EN

Accurate forecasting of freight volumes is essential for effective transportation planning and infrastructure development. Previous research on Uzbekistan’s railway sector primarily relied on single-method approaches, either using traditional statistical tools or machine learning techniques. This study adopts an innovative dual-method framework, combining Excel-based models—such as regression equation, exponential smoothing, and moving average—with advanced machine learning techniques, including decision tree, random forest, gradient boosting, and extreme gradient boosting. Freight shipment data and socio-economic variables, such as gros domestic product and operational railway length. Model performance was evaluated using root mean square error and mean absolute percentage error. The regression equation model demonstrated exceptional precision with a mean absolute percentage error of 0.001%, though its simplicity raised concerns about overfitting and limited scalability. Meanwhile, machine learning models showcased superior robustness and generalization capabilities, achieving low and balanced error rates, making them more suitable for capturing complex, non-linear relationships in freight dynamics. According to the compound annual growth rate projection, freight volumes are expected to increase significantly, reaching 106 million tons by 2030. This underscores the growing importance of strategic infrastructure investment, modernization, and policy interventions to accommodate future demand. The findings provide valuable insights for policymakers and transportation planners, offering a practical and comprehensive framework for sustainable development in Uzbekistan’s railway sector. This study aims to lay a foundation for informed decision-making and long-term growth planning by leveraging a mix of traditional and modern forecasting approaches.

3

Optimizing home energy balance with Bayesian models: predicting solar panel efficiency in variable weather conditions

Poniatowski Adrian, Baranowski Jerzy

Przegląd Elektrotechniczny

|

2025

|

R. 101, nr 4

62--70

EN

Photovoltaic (PV) technology is revolutionizing renewable energy by providing a sustainable and inexhaustible source of electricity. This paper explores Bayesian models for predicting home energy balance when using solar panels, with a focus on optimizing energy production under varying weather conditions in Poland. The research emphasizes the importance of personalized energy management solutions due to the high variability in household consumption patterns. Two distinct Bayesian models are developed to predict energy production and consumption, leveraging factors such as temperature, insulation, cloudiness, and day length. Data collected over five months from a specific region in Poland was used to validate these models. The results demonstrate the models’ effectiveness in capturing the variability in energy production and provide insights into optimizing the efficiency of PV systems. This study offers valuable guidelines for individuals considering investments in photovoltaic panels, highlighting potential profitability and efficiency improvements.

PL

Technologia fotowoltaiczna (PV) rewolucjonizuje odnawialne źródła energii, zapewniając zrównoważone i niewyczerpane źródło energii elektrycznej. W niniejszym artykule zbadano modele bayesowskie służące do przewidywania bilansu energetycznego w domu podczas korzystania z paneli słonecznych, ze szczególnym uwzględnieniem optymalizacji produkcji energii w zmiennych warunkach pogodowych w Polsce. Badania podkreślają znaczenie spersonalizowanych rozwiązań w zakresie zarządzania energią ze względu na dużą zmienność wzorców zużycia energii w gospodarstwach domowych. Opracowano dwa odrębne modele bayesowskie w celu przewidywania produkcji i zużycia energii, wykorzystując takie czynniki jak temperatura, izolacja, zachmurzenie i długość dnia. Dane zebrane w ciągu pięciu miesięcy z określonego regionu w Polsce zostały wykorzystane do walidacji tych modeli. Wyniki pokazują skuteczność modeli w uchwyceniu zmienności produkcji energii i dostarczają spostrzeżeń na temat optymalizacji wydajności systemów PV. Niniejsze badanie oferuje cenne wskazówki dla osób rozważających inwestycje w panele fotowoltaiczne, podkreślając potencjalną rentowność i poprawę wydajności.

4

Assessment of the impact of soil characteristics on crack formation and their prediction using machine learning

Wyborski Piotr

Materiały Budowlane

|

2025

|

nr 9

160--167

EN

The aim of the article was to develop a model identifying soil characteristics and environmental factors that determine its susceptibility to cracking. Additionally, a predictive algorithm was developed to forecast the intensity of soil cracking. The random forest method was used, based on 261 cases. The most significant variables influencing crack intensity were found to be soil moisture, specific density, and soil sample size. The developed algorithm demonstrated high predictive accuracy.

PL

W artykule zaprezentowano model identyfikujący cechy gruntu oraz właściwości środowiskowe decydujące o jego podatności na pękanie. Dodatkowo opracowano algorytm predykcyjny służący do prognozowania intensywności pęknięć gruntu. Wykorzystano metodę lasu losowego na bazie 261 przypadków. Najistotniejszymi zmiennymi wpływającymi na intensywność pęknięć okazały się: wilgotność gruntu, gęstość właściwa i wymiary próbki. Opracowany algorytm charakteryzuje się bardzo dużą dokładnością.

5

Optimizing an Artificial Intelligence Model to Predict The Tunnel Face Area after Blasting

Do Hoang Hiep, Bui Manh Tung, Nguyen Chi Thanh

Inżynieria Mineralna

|

2025

|

Vol. 1, no. 2

291--303

EN

In underground tunnel construction for mining, the drilling and blasting method is widely used due to its advantages, such as low cost, simple calculation and implementation, and applicability in various geological and hydrogeological conditions. The drilling and blasting method is also suitable for tunnels with different cross-sectional shapes. One parameter that significantly influences the effectiveness of the drilling and blasting method is the area of the tunnel face after blasting. In this study, 136 datasets of influencing parameters and the tunnel face area after blasting from the DeoCa tunnel construction project were used to develop an artificial neural network (ANN) model capable of predicting the tunnel face area after blasting. The paper developed an ANN model and proposed a hybrid model based on the ANN model combined with a genetic algorithm (GA) to predict the area of the tunnel face after blasting. The input variables for the models included the designed tunnel face area (Sd), the specific charge (SC), the average borehole length (L), and the rock mass rating (RMR) of the rock mass on the tunnel face. This paper demonstrates that the hybrid GA-ANN model provides more accurate calculations and predictions for the tunnel face area after blasting than the ANN model alone.

6

Designing efficient Model Predictive Control algorithm using easy-to-obtain fuzzy models

Marusak Piotr M.

International Journal of Electronics and Telecommunications

|

2025

|

Vol. 71, No. 4

art. no. 33, s. 1-8

EN

The paper describes the design process of an efficient model predictive control (MPC) algorithm based on fuzzy models. An interesting feature of the proposed approach is that it uses easy–to–obtain fuzzy Takagi–Sugeno (TS) models composed of a few step responses employed as local models; one of these models is used to derive the dynamic matrix, and the second one, being a skillful modification of the first one, to generate the free response. The designed MPC algorithm uses formulation as an efficient quadratic optimization task. Still, it offers control quality compared with the MPC algorithm formulated as a nonlinear optimization task, thanks to the skillful generation of the free response. The efficiency of the proposed approach is tested and demonstrated in the simulated control system of the nonlinear and non–minimum phase process of the chemical reactor with the van de Vusse reaction.

7

Prediction of machine state for non-Gaussian degradation model using Hidden Markov Model approach

Janczura Joanna, Żuławiński Wojciech, Shiri Hamid, Barszcz Tomasz, Zimroz Radosław, Wyłomańska Agnieszka

Eksploatacja i Niezawodność

|

2025

|

Vol. 27, no. 1

art. no. 193898

EN

Machinery health management becomes an essential issue in many sectors. The ultimate goal is to predict machinery degradation and accordingly plan maintenance actions. However, prediction becomes much harder if data is noisy. We propose a procedure for on-line prediction of the forthcoming machine state. This procedure is dedicated to the non-Gaussian (impulsive) health index (HI) data. It is based on a simplified degradation model with three machine states, i.e. healthy, warning and alarm, described in terms of a Hidden Markov Model (HMM). Using simulated trajectories we demonstrate that the α-stable HMM dedicated to time series with impulsive behaviour outperforms the classical Gaussian approach and can be an efficient alternative in such a case. In particular, the percentage errors of the predicted alarm state transition points decrease from 20%-45% to 1%-6%, if the α-stable HMM is used instead of the Gaussian one. We illustrate the proposed methodology for two datasets acquired during experiment on the VIBstand test rig and for a benchmark FEMTO dataset.

8

Prediction of dynamics change in biowaste quantity collected in functionally different regions. A case study of Poland

Tałałaj Izabela Anna, Hajduk Sławomira, Wiszniewska Anna

Economics and Environment

|

2025

|

No. 2

art. no. 920

EN

The aim of the paper is to determine the dynamics of change in biowaste quantity as well as to forecast the amount of biowaste generated in 4 functionally different regions of Poland. The analysis was made for a period of 16 years (2007-2022), and a prognosis was made for the next 4 years (2023-2026). Based on the obtained data, the following calculations were made: share of biowaste from households in the quantity of total municipal biowaste, accumulation rate of biowaste from households, medium-term change rate in the amount of biowaste from households, and prediction of changes in the biowaste accumulation index until 2026. In all the analysed regions, an increasing trend in the collected biowaste mass index has been observed. The agricultural and recreational regions were characterised by the highest dynamics of changes in collected biowaste quantity (T=0.21 and 0.25, respectively) and by the lowest values of their accumulation indicator (48.9 and 44.7 kg/ca per year, respectively). The highest quantity of biowaste is predicted to be generated in urbanised and industrialised regions (62.1 and 53.2 kg/ca per year, respectively).

PL

Celem artykułu jest określenie dynamiki zmian ilości bioodpadów oraz prognoza ilości generowanych bioodpadów w czterech funkcjonalnie różnych regionach Polski. Analizy zostały przeprowadzone dla okresu 16 lat (2007–2022), a prognozy obejmują kolejne 4 lata (2023–2026). Na podstawie uzyskanych danych obliczono: udział bioodpadów pochodzących z gospodarstw domowych w całkowitej ilości odpadów komunalnych, wskaźnik akumulacji bioodpadów z gospodarstw domowych, średniookresową dynamikę zmian ilości bioodpadów zebranych z gospodarstw domowych oraz prognozę zmian wskaźnika akumulacji bioodpadów do 2026 roku. We wszystkich analizowanych regionach zaobserwowano rosnący trend wskaźnika nagromadzenia zebranych bioodpadów. Obszary rolnicze i rekreacyjne charakteryzowały się najwyższą dynamiką zmian ilości bioodpadów (T = 0,21 i 0,25, odpowiednio) oraz najniższymi wartościami wskaźnika ich nagromadzenia (odpowiednio 48,9 i 44,7 kg/os. rocznie). Największe ilości zbieranych bioodpadów prognozuje się dla regionów zurbanizowanych i uprzemysłowionych (odpowiednio 62,1 i 53,2 kg/os. rocznie).

9

Enhancing urban flood hazard assessment: A comparative analysis of frequency ratio and xgboost models for precision risk mapping

El Haou Mohamed, Ourribane Malika, Ismaili Maryem, Krimissa Samira, Namous Mustapha

Ecological Engineering & Environmental Technology

|

2025

|

Vol. 26, iss. 3

265--285

EN

Urban flood hazard prediction should effectively balance accuracy and interpretability. This paper compares the performances of the Frequency Ratio method, a simple statistical technique, and XGBoost, a state-of-the-art machine learning algorithm for flood Hazard mapping in Beni Mellal (Morocco). The dataset was derived from preprocessed and standardized Sentinel-2 and Landsat 8 images, a Digital Elevation Model, and geological and soil maps. A flood inventory map was produced, it was then divided into training and testing subsets in the ratio of 70:30 for model calibration and validation, respectively. The FR method highlights key geographical variables such as slope, proximity to rivers, and vegetation indices to deliver rapid, interpretable flood risk assessments. In contrast, XGBoost captures complex, nonlinear relationships by integrating natural and anthropogenic factors for precise risk mapping. The results indicate that while FR is efficient for preliminary assessments in data-scarce environments, XGBoost significantly outperforms it in accuracy, reliability, and detailed hazard differentiation. XGBoost achieved an area under the curve (AUC) of 90.71% in testing datasets compared to 86.1% for FR. Flood distribution analysis showed that FR identified 21.3% of the study area as low-risk and 11.3% as very high-risk, suitable for broad evaluations. XGBoost, however, mapped 73.0% as very low-risk and 12.0% as very high-risk, making it valuable for resource-efficient interventions. This study highlights the complementary strengths of both approaches and advocates for integrating FR’s rapid insights with XGBoost’s precision. Together, they provide a robust framework for comprehensive flood hazard management in semi-arid regions, balancing strategic planning with localized interventions.

10

Application of continuous radon gas concentration telemonitoring for predictive seismic hazard assessment in Manado, Indonesia

Pratama Thomas Oka, Wijatna Agus Budhie, Sunarno, Sari Hermin Kartika, Yanti Ressy Jaya, Wijaya Rony, Waruwu Memory Motivanisman

Ecological Engineering & Environmental Technology

|

2025

|

Vol. 26, iss. 2

122--130

EN

Abnormal increases in radon gas (222Rn) concentrations in soil, groundwater, and atmosphere have been consistently observed as precursors of seismic activity, especially near active faults. In this study, we focus on earthquake prediction using IoT-based radon monitoring near the active fault in Manado, North Sulawesi, Indonesia, where seismic activity is high due to interactions between the Eurasian, Pacific, and Philippine plates. Radon gas concentration telemonitoring collected in real-time every minute between October 2023 and August 2024 was analyzed along with seismic data above M4.5 to predict earthquakes with magnitude 4.5 and above. This telemonitoring system enables continuous data storage every minute, with data accessible on the dataalamdiy web server, despite radon concentration readings on the detector updating every 10 minutes to filter out emissions from Thoron and Actanium sources.The results showed that earthquake date prediction sensitivity was 84%, accuracy was 75%, and the average prediction time was 2.65 days before the earthquake. The prediction was based on statistical algorithms derived from the daily average of radon gas concentration fluctuations, which resulted in an effective early warning system. One of the largest earthquakes M6.7 on January 9, 2024, was predicted 2 days ago. These findings highlight the possibility of integrating radon gas concentration anomaly analysis into disaster prevention strategies and provide an important lead time for preparedness efforts in seismically active areas. This research will significantly contribute to earthquake prediction methodology in Indonesia, especially in less-studied areas such as North Sulawesi, improving regional disaster preparedness and resilience.

11

Wind power prediction in Poland using temporal fusion transformers and numerical weather prediction

Jachuła Weronika, Wydra Michał

|

Vol. 73, nr 5

art. no. e155038

EN

Predicting wind power generation is essential to ensure the stability and efficiency of power systems. Accurate predictions enable better planning and management of energy reserves, minimizing operational costs and helping grid operators mitigate the adverse effects of wind generation fluctuations. The primary objective of the presented study is to develop an accurate wind power prediction method and apply it to Poland’s conditions. Among many emerging methods, the temporal fusion transformers (TFT) method is particularly well-suited for wind power generation forecasting, as it models complex, nonlinear dependencies in time series data. The TFT method combines self-attention mechanisms and recurrent networks, capturing long-term dependencies and short-term changes in input data. Additionally, TFT enables the effective use of contextual information, improving forecast accuracy. The numerical weather data was collected, and the feature extraction was performed. The features, such as time series data, have been used to train and test the different TFT networks. After the training and testing stage, an error analysis was performed. The final results showed similar or improved accuracy in wind generation forecasts compared to other methods in increased variability of weather conditions.

12

The method of predicting tightening torque and preload for bolts embedded in softwood using steel washers

Kawecki Bartosz

Archives of Civil Engineering

|

2025

|

Vol. 71, nr 1

595--613

EN

The paper presents an investigation on tightening torque and preload prediction for bolts embedded in softwood using steel washers. A basis for the research was a lack of any information on the tightening torque value for bolted connections in the timber structures’ design codes. For this reason, two experimental tests, theoretical analysis and Finite Element modelling, were performed in the paper. The first experiment based on finding the tightening torque to relative displacement relationship. The next one enabled the author to check the maximal compressing force determined by theoretical approach. In this test, dependencies between plastic modulus including material’s compaction and modulus of elasticity were found too and then applied to the numerical model. Tightening torque was calculated according to agreed formulas elaborated for steel structures based on the obtained preload force value. The high correlation between results from the prepared numerical models and experimental tests was observed. The research presented in the paper has multiple applications, as estimating a proper tightening torque value that should clamp a bolted connection, predicting stresses in connection’s components and clamping pressure when connecting several elements due to tightening torque and bolt preload force introduction or predicting the structural response of multiple bolts connections in the first phase of the loading.

PL

W artykule przedstawiono badania dotyczące przewidywania momentu dokręcenia i siły sprężającej dla śrub osadzonych w miękkim drewnie iglastym przy użyciu podkładek stalowych. Podstawą badań był brak informacji o wartości momentu dokręcenia śrub w połączeniach śrubowych w normach projektowych konstrukcji drewnianych. Z tego powodu w artykule przeprowadzono dwa badania eksperymentalne, analizę teoretyczną oraz modelowanie Metodą Elementów Skończonych. Pierwsze doświadczenie polegało na wyznaczeniu zależności momentu dokręcenia od przemieszczenia względnego. Kolejne umożliwiło autorowi sprawdzenie maksymalnej siły ściskającej określonej teoretycznie. W teście tym znaleziono także zależności pomiędzy modułem plastycznym uwzględniającym zagęszczeniem materiału i modułem sprężystości, które następnie zastosowano do modelu numerycznego. Moment dokręcenia obliczono na podstawie ustalonych wzorów opracowanych dla konstrukcji stalowych bazując na uzyskanej wartości siły sprężającej. Zaobserwowano wysoką korelację pomiędzy wynikami opracowanych modeli numerycznych i wynikami badań eksperymentalnych. Badania przedstawione w artykule mają wiele zastosowań, jak oszacowanie właściwej wartości momentu dokręcenia, jaki powinien zostać wprowadzony do połączenia śrubowego, przewidywanie naprężeń w elementach połączenia i docisku podczas łączenia kilku elementów na skutek wprowadzenia momentu dokręcenia i siły sprężającej śruby, czy przewidywanie reakcji konstrukcji połączeń na wiele śrub w pierwszej fazie obciążenia.

13

Prediction of tool wear based cutting forces during end milling of Inconel 718 using artificial neural networks

Felusiak-Czyryca Agata, Twardowski Paweł

Advances in Science and Technology. Research Journal

|

2025

|

Vol. 19, no. 7

394--405

EN

During the research, correlation between the input parameters (cutting parameters and cutting forces measure like peak to peak, root mean square and root mean square of ripple) and the variables were searched for, and the sensitivity of the network to input parameters was determined. In this paper artificial neural networks (ANNs) to prediction of tool wear based on cutting forces were used. Multilayer perceptron (MLP) networks with backward error propagation were used. The research shows that for the tested material and in the tested range, the cutting parameters are not diagnostically significant for the prediction of VBC (band width of the corner wear). The authors of this article focus on simplifying the model and analyzing the influence of variables on the prediction error. Neural networks show a correlation of about 95% for test sets.

14

Deep learning long short-term memory methods for instantaneous fuel consumption prediction: Experimental study and comparison of modeling strategies

Grabowski Łukasz, Sochaczewski Rafał, Biały Michał

Advances in Science and Technology. Research Journal

|

2025

|

Vol. 19, no. 12

197--209

EN

This paper presents an empirical study on the prediction of the instantaneous fuel consumption of public transport buses using LSTM type recurrent neural networks. The analyses were conducted on selected repetitive Sort 2 driving cycles. This allowed for stable test conditions and control of data variability. For the analyses, valid measurements including vehicle speed, accelerator pedal position (APP) and instantaneous fuel consumption (l/h) were used. Five LSTM modelling strategies were developed and compared: a baseline model, an in-depth model with dropout, an advanced model with callbacks, a model with a special weighted loss function for idling periods and a FuelNet model for fuel consumption prediction . The results indicate high prediction performance (MAE, RMSE, R²) and the potential for practical implementation of the model in fleet management systems.

15

Using GMM-HMM (Gaussian Mixture Model) to predict the state of the cryptocurrency market based on historical data

Błaszczyk Norbert, Michalik Adrian

Sustainable Production, Instrumentation and Engineering Sciences

|

2024

|

vol. 3, nr 1

1--8

EN

In recent years, the popularity of the cryptocurrency market has in-creased; two examples of well-known cryptocurrencies are Bitcoin and Ethereum. The Aim of this paper is to examine the (GMM-HMM) Gaussian Mixture Model-Hidden Markov Model ’s potential for forecasting the bitcoin market. Effective forecasting is essential for making well-informed investing decisions given the volatility and uncertainty of the market. The GMM-HMM model can be used to forecast the bitcoin market because it can take into account stochastic elements and ambiguity as well as exam-ine several possible outcomes. The effectiveness of the GMM-HMM model will be assessed along with its use in predicting future bitcoin price values and market trends. In order to learn more about prices, sales volume, mar-ket capitalization, and other potential market influences for bitcoin, his-torical market data will be examined. Statistical and correlation analysis will then be performed to look for any relevant relationships between these factors. The Prophet library’s forecasts were more accurate with a smaller variation from the actual exchange rate than the GMM-HMM model’s, which had predictions that were not quite as accurate. By incor-porating other variables into the model, such as news sentiment analysis, or by experimenting with different time series forecasting methods, the existing strategy could still be made better. This study attempted to pre-dict bitcoin exchange rates using a GMM-HMM technique, however it was unsuccessful. The study also emphasized the difficulties in predicting the cryptocurrency market and offered ideas for potential improvements, in-cluding as the addition of more factors and the investigation of other fore-casting methods.

16

Short-term prediction of UT1-UTC and LOD via Dynamic Mode Decomposition and combination of least-squares and vector autoregressive model

Michalczak Maciej, Ligas Marcin

Reports on Geodesy and Geoinformatics

|

2024

|

Vol. 117

45--54

EN

This study presents a short-term forecast of UT1-UTC and LOD using two methods, i.e. Dynamic Mode Decomposition (DMD) and combination of Least-Squares and Vector Autoregression (LS+VAR). The prediction experiments were performed separately for yearly time spans, 2018-2022. The prediction procedure started on January 1 and ended on December 31, with 7-day shifts between subsequent 30-day forecasts. Atmospheric Angular Momentum data (AAM) were used as an auxiliary time series to potentially improve the prediction accuracy of UT1-UTC and LOD in LS+VAR procedure. An experiment was also conducted with and without elimination of effect of zonal tides from UT1-UTC and LOD time series. Two approaches to using the best steering parameters for the methods were applied:. First, an adaptive approach, which observes the rule that before every single forecast, a preliminary one must be performed on the pre-selected sets of parameters, and the one with the smallest prediction error is then used for the final prediction; and second, an averaged approach, whereby several forecasts are made with different sets of parameters (the same parameters as in adaptive approach) and the final values are calculated as the averages of these predictions. Depending on the method and data combination mean absolute prediction errors (MAPE) for UT1-UTC vary from 0.63 ms to 1.43 ms for the 10th day and from 3.07 ms to 8.05 ms for the 30th day of the forecast. Corresponding values for LOD vary from 0.110 ms to 0.245 ms for the 10th day and from 0.148 ms to 0.325 ms for the 30th day.

17

Wpływ czynników zewnętrznych na predykcję zużycia energii elektrycznej w budynku biurowym z panelami fotowoltaicznymi

Kawa Kazimierz, Mularczyk Rafał, Bauer Waldemar, Kucharska Edyta

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 2

221--226

PL

Planowanie zużycia energii elektrycznej ze względu na pojawiające się niespodziewanie przerwy w dostawie staje istotnym aspektem zarządzania utrzymania budynków. Analiza szeregów czasowych pozwala na predykowanie zużycia energii elektrycznej w kolejnych latach na podstawie danych historycznych. Celem badania jest weryfikacja wpływu czynników zewnętrznych na predykcję ilości zużycia energii elektrycznej. W badaniach zostały wykorzystane metody analizy szeregów czasowych: model naiwny z sezonowością, regresji liniowej oraz Facebook Prophet. Wyniki pokazują, że zaproponowane modele w zadawalający sposób są w stanie prognozować zapotrzebowanie na energię.

EN

Planning for electricity consumption due to power outages occurring unexpectedly is becoming an important aspect of building maintenance management. Time series analysis makes it possible to predict electricity consumption in future years based on historical data. The purpose of the study is to verify the influence of external factors on the prediction of the amount of electricity consumption. The study used time series analysis methods: naive model with seasonality, linear regression and Facebook Prophet. The results show that the proposed models are able to predict energy demand satisfactorily.

18

Prediction passenger numbers in light rail transit using seasonal autoregressive integrated moving average (SARIMA)

Saputra Aldi, Gustriansyah Rendra, Sanmorino Ahmad, Mair Zaid Romegar, Sartika Dewi, Puspasari Shinta

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 10

43--45

EN

Light Rail Transit (LRT) plays a role in supporting the mobility of the people of a city. However, the increase in LRT use presents challenges, requiring effective solutions to anticipate changes in the number of passengers. This research aims to design and implement a prediction model using the Seasonal Autoregressive Integrated Method Moving Average to anticipate and predict the number of LRT passengers. The prediction results using the parameter model (0,1,1)(0,1,0) obtained a MAPE value of 16.69%, thus, the accuracy level obtained was 83.31%.

PL

Tranzyt koleją lekką (LRT) odgrywa rolę we wspieraniu mobilności mieszkańców miasta. Jednakże wzrost wykorzystania LRT stwarza wyzwania wymagające skutecznych rozwiązań umożliwiających przewidywanie zmian w liczbie pasażerów. Celem badania jest zaprojektowanie i wdrożenie modelu predykcyjnego wykorzystującego sezonową, zintegrowaną metodę autoregresyjną, średnią ruchomą do przewidywania i przewidywania liczby pasażerów LRT. Wyniki predykcji z wykorzystaniem modelu parametrycznego (0,1,1)(0,1,0) uzyskały wartość MAPE na poziomie 16,69%, a zatem uzyskany poziom dokładności wyniósł 83,31%.

19

Prediction of the service life of LED lamps based on the extrapolation of the luminous flux conservation factor

Baghirov Sabir, Basova Yulia, Guba Liudmyla, Kozhushko Hryhorii

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 2

190--192

EN

The paper presents the results of research into the luminous flux conservation factors of commercial samples of light-emitting diode lamps for general lighting during tests of up to 6.000 and 10.000 hours, as well as the results of their service life assessment based on the extrapolation of the values of the luminous flux conservation factor until the moment when this coefficient in 50% of the lamps will decrease to 70% of the initial value. The measurement of the luminous flux of the lamps was carried out every 1000 hours operation of the lamps in the mode with switching cycles: 150 minutes on at full power, after which the lamps are turned off for 30 minutes. Selection of the empirical curve of the luminous flux conservation factor should be performed by finding the initial constant and rate of change of the luminous flux using the method of least squares. The forecast of the service life, according to the recommendations of the IEC 62612 standard, was carried out for four times the test time according to the test results up to 6 thousand hours and 10 thousand hours. It is shown that the service life of lamp samples which can be declared based on test results, is up to 6.000 hours is at least 24 thousand hours (calculated 27.2 thousand hours). Forecast based on tests up to 10 thousand hours is about 36.500 hours, which is 3.500 hours less than declared by the manufacturer.

PL

W artykule przedstawiono wyniki badań współczynników zachowania strumienia świetlnego komercyjnych próbek lamp diodowych elektroluminescencyjnych do oświetlenia ogólnego podczas testów do 6000 i 10000 godzin, a także wyniki oceny ich żywotności na podstawie ekstrapolacji wartości współczynnika zachowania strumienia świetlnego do momentu, gdy współczynnik ten w 50% lamp spadnie do 70% wartości początkowej. Pomiar strumienia świetlnego lamp wykonywano co 1000 godzin pracy lamp w trybie z cyklami załączania: 150 minut pracy z pełną mocą, po czym lampy są wyłączane na 30 minut. Doboru empirycznej krzywej współczynnika zachowania strumienia świetlnego należy dokonać poprzez znalezienie stałej początkowej i szybkości zmian strumienia świetlnego metodą najmniejszych kwadratów. Prognozę żywotności, zgodnie z zaleceniami normy IEC 62612, przeprowadzono dla czterokrotności czasu badania według wyników badań do 6 tys. godzin i 10 tys. godzin. Wykazano, że żywotność próbek lamp, którą można zadeklarować na podstawie wyników badań, wynosi do 6000 godzin, czyli co najmniej 24 tysiące godzin (obliczono 27,2 tysiąca godzin). Prognoza na podstawie testów do 10 tys. godzin to około 36.500 godzin, czyli o 3.500 godzin mniej niż deklaruje producent.

20

Macrogenetic program of distributed windings of electromechanical converters

Shynkarenko Vasyl, Shymanska Anna, Kotliarova Viktoriia, Krasovskyi Pavlo

Przegląd Elektrotechniczny

|

2024

|

R. 100, nr 9

114--118

EN

The paper considers the definition of macrogenetic programs of distributed windings. It is shown that the windings perform the role of an energy and genetic core which determines the principles of genetic structure formation and evolution of an arbitrary complex electromechanic system. The systematic basis for the organization of further structural and systemic studies of structural and functional classes of windings (primarily, twin-type, hybrid and spatially adaptive) was determined.

PL

W artykule rozważono definicję programów makrogenetycznych uzwojeń rozproszonych. Pokazano, że uzwojenia pełnią rolę rdzenia energetycznego i genetycznego, który wyznacza zasady tworzenia struktury genetycznej i ewolucji dowolnego złożonego układu systemu elektromechanicznego. Określono systematyczne podstawy organizacji dalszych badań strukturalno-systemowych klasów strukturalnych i systemowych uzwojeń (przede wszystkim typu bliźniaczego, hybrydowego i adaptacyjnego przestrzennie).