Ograniczanie wyników
Czasopisma help
Autorzy help
Lata help
Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 108

Liczba wyników na stronie
first rewind previous Strona / 6 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  ANN
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 6 next fast forward last
EN
Knowing the material properties is of a crucial importance when planning to manufacture some structure. That is true for the steel structures, as well. Thus, for the proper planning of a certain steel part or a structure production, one must be aware of the properties of the material, to be able to make a qualified decision, which material should be used. Considering that the manufacturing of steel products is constantly growing in various branches of industry and engineering, the problem of predicting the material properties, needed to satisfy the requirements for the certain part efficient and reliable functioning, becomes an imperative in the design process. A method of predicting four material properties of the two stainless steels, by use of the artificial neural network (ANN) is presented in this article. Those properties were predicted based on the particular steels’ known chemical compositions and the corresponding material properties available in the Cambridge Educational System EDU PACK 2010 software, using neural network module of MathWorks Matlab. The method was verified by com-paring the values of the material properties predicted by this method to known values of properties for the two stainless steels, X5CrNi18-10 (AISI 304), X5CrNiMo17-12-2 (AISI 316). The difference be-tween the two sets of values was below 5% and, in some cases, even negligible.
EN
This work aims to create an ANN-based system for a musical improviser. An artificial improviser of "hearing" music will create a melody. The data supplied to the improviser is MIDItype musical data. This is the harmonic-rhythmic course, the background for improvisation, and the previously made melody notes. The harmonic run is fed into the system as the currently ongoing chord and the time to the next chord, while the supplied few dozen notes performed earlier will indirectly carry information about the entire run and the musical context and style. Improvisation training is carried out to check ANN as a correctlooking musical improvisation device. The improviser generates several hundred notes to be substituted for a looped rhythmicharmonic waveform and examined for quality.
EN
There is a demand for worm gearboxes in diversified industrial fields that include machinery such as escalators, ribbon blenders, pulverisers, bowl mills, etc. because of their peculiar characteristics like torque and quick retardation. The most commonly occurring defects in a worm gear box are scratches that develop in the worm gear and in bearings. Early defect categorization is required to prevent a sudden breakdown that would decrease production. The defect is depicted in different cases, which include defects in the gear tooth and the outer and inner races of the bearing. In another case, the defect is considered in the gear tooth as well as the bearing. The severity is designated using the ANN. The experiments were performed under these conditions with a good worm gearbox to capture vibration response signatures. Using these values as an input to the ANN, the model is trained. Experimental results show that vibration amplitude increases with fault progression in the worm gearbox, and the trained ANN model effectively categorizes worm gearbox faults with an accuracy of 97.12%.
EN
The main objective of this work is to select the most reliable machine learning model to predict the generated solid flow in the Tafna basin (North-West of Algeria). It is about the artificial neural networks (ANN) and long short-term memory (LSTM). The sediment load is recorded through three hydrometric stations. The efficiency and performance of the two models is verified using the correlation coefficient (R2), the Nash-Sutcliffe coefficient (NSC) and the root mean square error (RMSE). The obtained simulated solids load shows a very good correlation in terms of precision although the ANN model gave relatively better results compared to the LSTM model where low RMSE values were recorded, which confirms that the artificial intelligence models remain also effective for the treatment and the prediction of hydrological phenomena such as the estimation of the solid load in a such watershed.
EN
The concept of equivalence of the realistic, initial reference model and the simplified, reduced model is proposed. In reduced models, the action of the soil on the structure is replaced by the action of a layer with prescribed properties, defined by a set of parameters. The main difficulty here is to find the parameter values required by the simplified theory. The subject of this work is to find the dependence of the parameters of the reduced model on the parameters of the full model, including realistic soil behavior, in order to ensure the equivalence of both models. We show the potential of the method by presenting two examples: Winkler and Pasternak's model of a plate on the ground. We assume that both models are equivalent if they give identical results (displacements) at a finite number of observation points. An artificial neural network (ANN) is built in order to approximate and record the dependence of the parameters of the reduced model (at the network output) from the parameters of the full model (given at the network input). The complex network acts as a formula that assigns the parameters of the reduced model to a realistic description of the soil structure that is used for finite element method (FEM) modeling. The formalism we propose is quite general and can be applied to many engineering problems. The presented procedure is entirely numerical; it allows to calculate the parameters of the reduced model without resorting to symbolic calculations or additional theoretical considerations.
EN
In this study, a novel method is proposed to optimize the reinforced parameters influencing the bearing capacity of a shallow square foundation resting on sandy soil reinforced with geosynthetic. The parameters to be optimized are reinforcement length (L), the number of reinforcement layers (N), the depth of the topmost layer of geosynthetic (U), and the vertical distance between two reinforcement layers (X). To achieve this objective, 25 laboratory small-scale model tests were conducted on reinforced sand. This laboratory-scale model has used two geosynthetics as reinforcement materials and one sandy soil. Firstly, the effect of reinforcement parameters on the bearing load was investigated using the analysis of variance (ANOVA). Both response surface methodology (RSM) and artificial neural networks (ANN) tools were applied and compared to model bearing capacity. Finally, the multiobjective genetic algorithm (MOGA) coupled with RSM and ANN models was used to solve multi objective optimization problems. The design of bearing capacity is considered a multi-objective optimization problem. In this regard, the two conflicting objectives are the need to maximize bearing capacity and minimize the cost. According to the obtained results, an informed decision regarding the design of the bearing capacity of reinforced sand is reached.
EN
This study develops a non-invasive method to predict blood glucose through image processing. For investigation, several invasive images and glucose levels were taken. Types of samples based on age classification, 20-60 years. For accuracy and simple analysis, 37 images of participants as volunteers, samples were evaluated and investigated under the gray level co-occurrence matrix (GLCM). In this study, an artificial neural network (ANN) was used for all training and hand texture testing to detect glucose levels. The performance of this model is evaluated using Root Mean Square Error (RMSE) and correlation coefficient (r). Clarke Error Grid Analysis (EGA) variance was used in this investigation to determine the accuracy of the method. The results showed that the RMSE was close to the standard value, the regression coefficient was 0.95, and the Clarke EGA analysis: 81.08% was in the A .% zone. So that the blood glucose prediction model using the GLCM-ANN method is feasible to apply.
PL
Niniejsze badanie rozwija nieinwazyjną metodę przewidywania stężenia glukozy we krwi poprzez przetwarzanie obrazu. W celu zbadania wykonano kilka inwazyjnych obrazów i poziomów glukozy. Rodzaje próbek na podstawie klasyfikacji wiekowej, 20-60 lat. Dla dokładności i prostej analizy, 37 obrazów uczestników jako ochotników, próbki zostały ocenione i zbadane w ramach macierzy współwystępowania poziomu szarości (GLCM). W tym badaniu sztuczna sieć neuronowa (ANN) została wykorzystana do wszystkich testów treningu i tekstury dłoni w celu wykrycia poziomu glukozy. Wydajność tego modelu ocenia się za pomocą błędu średniokwadratowego (RMSE) i współczynnika korelacji (r). W tym badaniu zastosowano analizę wariancji siatki błędów Clarke'a (EGA) w celu określenia dokładności metody. Wyniki pokazały, że RMSE była zbliżona do wartości standardowej, współczynnik regresji wyniósł 0,95, a analiza Clarke EGA: 81,08% znajdowała się w strefie A.%. Aby model przewidywania stężenia glukozy we krwi przy użyciu metody GLCM-ANN był możliwy do zastosowania.
EN
River discharge is affected by many factors, such as water level, rainfall, and precipitation. This study proposes a new hybrid framework named LAES (LASSO-ANN-EMD-SVM) to model the relationship of daily river discharge with meteorological variables. This hybrid framework is a composite of the least absolute shrinkage and selection operator (LASSO), an artificial neural network (ANN), and an error correction method. In the first stage, LASSO identifies meteorological variables that have a significant influence on the generation of river discharge. Next, the ANN model is used to predict river discharge using meteorological variables selected by LASSO, and the error series is determined. The error series is decomposed into intrinsic mode functions and residuals using empirical mode decomposition (EMD). The EMD components are modeled using the support vector machine (SVM) model, and the error predictions are aggregated. In the last stage, the LASSO-ANN predictions and the predicted error series are aggregated as the final discharge prediction. The proposed hybrid framework is illustrated on the Kabul River of Pakistan. The performance of the proposed hybrid framework is compared with six models using various performance measures and the Diebold-Mariano test. These models include multiple linear regression (MLR), SVM, ANN, LASSO-MLR, LASSO-SVM, and LASSO-ANN models. The findings reveal that the proposed hybrid model outperforms all other models considered in the study. In the testing phase, the root mean square error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE) of the proposed LAES hybrid model are 337.143 m3/s, 32.354%, and 218.353 m3/s which are smaller than all other models compared in the study. Our proposed hybrid system is an efficient model for river discharge prediction that will be helpful in water management and protection against floods. Long-term prediction can help to identify the major effects of climate change and to make evidence-based environmental policies.
PL
Zastosowanie Sztucznych Sieci Neuronowych (SSN) do sterowania procesem przemiału cementu jest w pełni uzasadnione ze względu na złożoność procesu mielenia oraz nieliniowość charakteryzujących go parametrów. Stabilna praca młyna uzyskana przy wsparciu samouczących się SSN może przełożyć się na minimalizację jednostkowego zużycia energii przy utrzymaniu właściwego stopnia rozdrobnienia. Jako dane wejściowe zasilające algorytm SSN wykorzystano wybrane parametry technologiczne monitorowane podczas pracy młyna kulowego pracującego w warunkach przemysłowych. Eksperymenty wykazały, że mały błąd predykcji dają modele uwzględniające małą liczbę parametrów, biorące pod uwagę dane wejściowe z krótszego okna czasowego i 30-minutowym oknem wygładzania danych wejściowych. Najlepsze konfiguracje sieci neuronowej pozwalają na predykcję parametrów pracy młyna ze średnim bezwzględnym błędem procentowym poniżej 5% dla horyzontu czasowego 10 min oraz poniżej 7% dla horyzontu czasowego 15 min.
EN
The use of Artificial Neural Networks (ANNs) to control the cement grinding process is fully justified, taking into account the complexity of the grinding process and the non-linearity of its parameters. Stable operation of the mill, obtained with the support of self-learning ANNs, may translate into minimization of unit energy consumption while maintaining the desired degree of fragmentation. As input data powering the ANN algorithm, selected technological parameters monitored during the operation of the ball mill in an industrial setting were used. Experiments have shown that models with a smaller number of parameters, taking into account input data from a shorter time window and a 30-minute input smoothing window, yield a smaller prediction error. The best configurations of the neural network allow for the prediction of the mill operation parameters with an average absolute percentage error of less than 5% for the time horizon of 10 minutes and less than 7% for the time horizon of 15 minutes.
EN
This work focuses on optimizing process parameters in turning AISI 4340 alloy steel. A hybridization of Machine Learning (ML) algorithms and a Non-Dominated Sorting Genetic Algorithm (NSGA-II) is applied to find the Pareto solution. The objective functions are a simultaneous minimum of average surface roughness (Ra) and cutting force under the cutting parameter constraints of cutting speed, feed rate, depth of cut, and tool nose radius in a range of 50–375 m/min, 0.02–0.25 mm/rev, 0.1–1.5 mm, and 0.4–0.8 mm, respectively. The present study uses five ML models – namely SVR, CAT, RFR, GBR, and ANN – to predict Ra and cutting force. Results indicate that ANN offers the best predictive performance in respect of all accuracy metrics: root-mean-squared-error (RMSE), mean-absolute-error (MAE), and coefficient of determination (R2). In addition, a hybridization of NSGA-II and ANN is implemented to find the optimal solutions for machining parameters, which lie on the Pareto front. The results of this multi-objective optimization indicate that Ra lies in a range between 1.032 and 1.048 μm, and cutting force was found to range between 7.981 and 8.277 kgf for the five selected Pareto solutions. In the set of non-dominated keys, none of the individual solutions is superior to any of the others, so it is the manufacturer's decision which dataset to select. Results summarize the value range in the Pareto solutions generated by NSGA-II: cutting speeds between 72.92 and 75.11 m/min, a feed rate of 0.02 mm/rev, a depth of cut between 0.62 and 0.79 mm, and a tool nose radius of 0.4 mm, are recommended. Following that, experimental validations were finally conducted to verify the optimization procedure.
11
EN
In this paper, two methods to predict and calculate the area of the tunnel face after the blasting were used. The first one is an artificial intelligence method using an artificial neural network system (ANN) model, and the second one – the support vector regression (SVR). After building predictive models for the area of the tunnel face after blasting by both methods, on the basis of comparing the results obtained in both methods, the performance of these models was assessed through the root mean square error RMSE and the coefficient of determination R2. RMSE and R2 values of the artificial neural network system (ANN) model were obtained as 0.1473 and 0.903 in training datasets, respectively. These values are 0.1497 and 0.9107 in testing datasets. In the SRV model, RMSE and R2 were equaled to 0.1228 and 0.9331 in training datasets, respectively. These values are 0.1708 and 0.9055, respectively in testing datasets. It can be concluded that artificial intelligence using ANN and SVM models can be used to predict the area of the tunnel face after blasting with high accuracy.
EN
Mining-induced road subsidence is a significant concern in areas with extensive underground mining activities. Therefore, the prediction of road subsidence is crucial for effective land management and infrastructure planning. This paper applies an artificial neural network (ANN) to predict road subsidence caused by underground mining activities in Vietnam. The ANN model proposed in this study is adopted relying on the recursive multistep prediction process, in which the predicted value in the previous step is appended to the time series to predict the next value. The entire dataset of 12 measured epochs covering 12 months with a 1-month repeat time is divided into the training set by the first 9 measured epochs and the test set by the last 3 measured epochs. K-fold cross validation is first applied to the training set to determine the best model’s hyperparameters, which are then adopted to predict land subsidence of the test set. Absolute errors of the predicted road subsidence depend on the separated time between the last measured epoch and the predicted epoch. Those errors at the 10th month of the three tested points are 3.0%, 0.1 %, and 0.1%, which increase to 4.8%, 3.3%, and 1.5% at the 11th month, and 7.2%, 2.5% and 1.3% at the 12th month. The absolute errors are found to be small, which were all ranged with 0.5 mm and demonstrates that the proposed method utilizing ANN in this study can produce good prediction for road subsidence time series at mining areas.
EN
Customer churn prediction is used to retain customers at the highest risk of churn by proactively engaging with them. Many machine learning-based data mining approaches have been previously used to predict client churn. Although, single model classifiers increase the scattering of prediction with a low model performance which degrades reliability of the model. Hence, Bag of learners based Classification is used in which learners with high performance are selected to estimate wrongly and correctly classified instances thereby increasing the robustness of model performance. Furthermore, loss of interpretability in the model during prediction leads to insufficient prediction accuracy. Hence, an Associative classifier with Apriori Algorithm is introduced as a booster that integrates classification and association rule mining to build a strong classification model in which frequent items are obtained using Apriori Algorithm. Also, accurate prediction is provided by testing wrongly classified instances from the bagging phase using generated rules in an associative classifier. The proposed models are then simulated in Python platform and the results achieved high accuracy, ROC score, precision, specificity, F-measure, and recall.
EN
Groundwater is a valuable resource whose purity is necessary for human survival. It serves as a significant source of water for household, industrial, and agricultural purposes. Traditional groundwater pollution remediation technologies include pump & treat, phase extraction, aeration gas of groundwater, bioremediation, and chemical oxidation. Permeable reactive barrier (PRB) is one of the most key technology being developed as alternatives to the pump and manage method for the remedying contaminated groundwater. An overview on the groundwater significant as important sources for water, sources of groundwater contamination, transport of contaminants, and groundwater remediation technologies have been discussed in this paper. In addition to reactive media, the design and installation of PRBs of funnel-gate configurations and their application as a remediation technique have been covered in this review. Finally reaction mechanisms in groundwater, contaminant transport governing equation, isotherms sorption models, kinetic sorption models, breakthrough curves modeling have been presented in this review. PRB technique provides financial benefits while also encouraging waste material reuse, so contributing to environmental sustainability. Funnel and gate PRB can offer one or more dense treatment areas for maximizing groundwater pollution plume capture. Funnel-gate PRB is characterized by smaller reaction area, ease in replacement and removal during the blocking of the reactive barrier by fine soil particles and reactive sediments.
EN
The wind energy conversion systems (WECS) suffer from an intermittent nature of source (wind) and the resulting disparity between power generation and electricity demand. Thus, WECS are required to be operated at maximum power point (MPP). This research paper addresses a sophisticated MPP tracking (MPPT) strategy to ensure optimum (maximum) power out of the WECS despite environmental (wind) variations. This study considers a WECS (fixed pitch, 3KW, variable speed) coupled with a permanent magnet synchronous generator (PMSG) and proposes three sliding mode control (SMC) based MPPT schemes, a conventional first order SMC (FOSMC), an integral back-stepping-based SMC (IBSMC) and a super-twisting reachability-based SMC, for maximizing the power output. However, the efficacy of MPPT/control schemes rely on availability of system parameters especially, uncertain/nonlinear dynamics and aerodynamic terms, which are not commonly accessible in practice. As a remedy, an off-line artificial function-fitting neural network (ANN) based on Levenberg-Marquardt algorithm is employed to enhance the performance and robustness of MPPT/control scheme by effectively imitating the uncertain/nonlinear drift terms in the control input pathways. Furthermore, the speed and missing derivative of a generator shaft are determined using a high-gain observer (HGO). Finally, a comparison is made among the stated strategies subjected to stochastic and deterministic wind speed profiles. Extensive MATLAB/Simulink simulations assess the effectiveness of the suggested approaches.
EN
The quality of machine components surfaces plays an important impact on their functional performance. Product performance may be restricted by changes to surface integrity, which includes changes to roughness, hardness, and microstructure. In this research, the impact of cutting variables in CNC turning under the conventional cooling condition on surface hardness of Duplex Stainless Steel. Cutting variables under conventional cooling, including cutting speed, feed, and depth of cut, have been optimized utilizing Taguchi’s L9 orthogonal array designed with three stages of turning variables. The optimal variable stages and the degree of significance of the cutting variables, respectively, were determined utilizing the analysis of means (ANOM) and analysis of variance (ANOVA). Effectiveness tests with optimum stages of variables were done to prove the viability of optimization by utilizing Taguchi. It has been found that the maximum surface hardness is most strongly affected by the feed 71.29%, followed by the depth of cut 12.1%, and finally the cutting speed 11.61%.
EN
Missing data cause problems in meteorological, hydrological, and climate analysis. The observation data should be complete and cover long periods to make the research more accurate and reliable. Artificial intelligence techniques have attracted interest for completing incomplete meteorological data in recent years. In this study the abilities of machine learning models, artificial neural networks, the nonlinear autoregressive with exogenous input (NARX) model, support vector regression, Gaussian processes regression, boosted tree, bagged tree (BAT), and linear regression to fill in missing precipitation data were investigated. In developing the machine learning model, 70% of the dataset was used for training, 15% for testing, and 15% for validation. The Bayburt, Tercan, and Zara precipitation stations, which are closest to the Erzincan station and have the highest correlation coefficients, were used to fill the data gaps. The accuracy of the constructed models was tested using various statistical criteria, such as root-mean-square error (RMSE), mean absolute error (MAE), Nash–Sutcliffe model efficiency coefficient (NSE), and determination coefficient (R2) and graphical approaches such as scattering, box plots, violin plots, and Taylor diagrams. Based on the comparison of model results, it was concluded that the BAT model with R2: 0.79 and NSE: 0.79 and error (RMSE: 11.42, and MAE: 7.93) was the most successful in the completion of missing monthly precipitation data. The contribution of this research is assist in the choice of the best and most accurate method for estimating precipitation data in semi-arid regions like Erzincan.
EN
In this paper it has been assumed that the use of artificial intelligence algorithms to predict the level of air quality gives good results. Our goal was to perform a comparative analysis of machine learning algorithms based on an air pollution prediction model. By repeatedly performing tests on a number of models, it was possible to establish both the positive and negative influence of the parameters on the result generated by the ANN model. The research was based on some selected both current and historical data of the air pollution concentration altitude and weather data. The research was carried out with the help of the Python 3 programming language, along with the necessary libraries such as TensorFlow and Jupyter Notebook. The analysis of the results showed that the optimal solution was to use the Long Stort Term Memory LSTM algorithm in smog prediction. It is a recursive model of an artificial neural network that is ideally suited for prediction tasks. Further research on the models may develop in various directions, ranging from increasing the number of trials which would be linked to more reliable data, ending with increasing the number of types of algorithms studied. Developing the models by testing other types of activation and optimization functions would also be able to improve the understanding of how they affect the data presented. A very interesting developmental task may be to focus on a self-learning artificial intelligence algorithm, so that the algorithm can learn on a regular basis, not only on historical data. These studies would contribute significantly to the amount of data collected, its analysis and prediction quality in the future.
EN
This paper aims to propose a useful modeling diagnostic method for solar plants. The study was performed on the basis of the localization of the failing panel obtained by an effective comparison of measured output voltages and estimator voltages. The comparison is done with the ideal solar plant using learning approach based on artificial neuronal network (ANN). The partial shading failure was detected by the given equation d²ΔV/dI²>0. The obtained results using MATLAB/Simulink environment show a satisfactory performance in terms of rapidity and precision under variable shading conditions.
PL
Celem artykułu jest zaproponowanie użytecznej metody diagnostycznej modelowania dla elektrowni słonecznych. Badania przeprowadzono na podstawie lokalizacji uszkodzonego panelu uzyskanej poprzez efektywne porównanie zmierzonych napięć wyjściowych i napięć estymatorów. Porównanie jest dokonywane z idealną elektrownią słoneczną przy użyciu podejścia uczenia opartego na sztucznej sieci neuronowej (ANN). Częściowe zacienienie zostało wykryte za pomocą podanego równania d²ΔV/dI²>0. Uzyskane wyniki w środowisku MATLAB/Simulink wykazują zadowalające działanie pod względem szybkości i precyzji w zmiennych warunkach zacienienia.
EN
There are many research on electric vehicles to reduce environmental pollution due to vehicles that use fossil fuels. The advantages of using a BLDC motor are high efficiency, high torque, reduced noise, long lifetime, and easy maintenance. Using of BLDC motors in electric vehicles is sometimes not optimal due to varying set points and presence of loads. Then a speed motor is needed to be controlled so the motor can work properly. In this research using the Artificial Neural Network (ANN) method. The ANN on this speed controller is practical as a 3-phase inverter input voltage control so the speed of BLDC motor can match the set point. In the simulation in this research, controlled based ANN is applied to electric buses with large torque, from the simulation it can be seen that Controlled based ANN can work well.
PL
Istnieje wiele badań dotyczących pojazdów elektrycznych mających na celu zmniejszenie zanieczyszczenia środowiska przez pojazdy wykorzystujące paliwa kopalne. Zalety stosowania silnika BLDC to wysoka sprawność, wysoki moment obrotowy, obniżony poziom hałasu, długa żywotność i łatwa konserwacja. Stosowanie silników BLDC w pojazdach elektrycznych czasami nie jest optymalne ze względu na różne nastawy i obecność obciążeń. Następnie konieczne jest sterowanie prędkością silnika, aby silnik mógł działać prawidłowo. W badaniach wykorzystano metodę Sztucznej Sieci Neuronowej (ANN). SSN na tym regulatorze prędkości jest praktycznym sterowaniem napięcia wejściowego falownika 3-fazowego, dzięki czemu prędkość silnika BLDC może być zgodna z wartością zadaną. W symulacji w niniejszych badaniach, kontrolowany SSN jest stosowany do autobusów elektrycznych o dużym momencie obrotowym, z symulacji widać, że SSN w oparciu o sterowanie może dobrze działać.
first rewind previous Strona / 6 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.