Wyniki wyszukiwania - Biblioteka Nauki

1

A comparative study of CNN, LSTM, BiLSTM, AND GRU architectures for tool wear prediction in milling processes

100%

Zegarra F. C. , Vargas-Machuca J. , Coronado A. M.

Journal of Machine Engineering

|

2023

|

tom Vol. 23, No. 4

122--136

EN

Accurately predicting machine tool wear requires models capable of capturing complex, nonlinear interactions in multivariate time series inputs. Recurrent neural networks (RNNs) are well-suited to this task, owing to their memory mechanisms and capacity to construct highly complex models. In particular, LSTM, BiLSTM, and GRU architectures have shown promise in wear prediction. This study demonstrates that RNNs can automatically extract relevant information from time series data, resulting in highly precise wear models with minimal feature engineering. Notably, this approach avoids the need for excessively large window sizes of data points during model training, which would increase model complexity and processing time. Instead, this study proposes a procedure that achieves low prediction errors with window sizes as small as 100 data points. By employing Bayesian hyperparameter optimization and two preprocessing techniques (detrend and offset), RMSE errors consistently fall below 10. A key difference in this study is the use of boxplots to provide a better representation of result variability, as opposed to solely reporting the best values. The proposed approach matches more complex state of-the-art. methods and offers a powerful tool for wear prediction in engineering applications.

2

A new cast-resin transformer thermal model based on recurrent neural networks

100%

Azizian D. , Bigdeli M.

Archives of Electrical Engineering

|

2017

|

tom Vol. 66, nr 1

17--28

EN

Thermal modeling in the transient condition is very important for cast-resin dry-type transformers. In the present research, two novel dynamic thermal models have been introduced for the cast-resin dry-type transformer. These models are based on two artificial neural networks: the Elman recurrent networks (ELRN) and the nonlinear autoregressive model process with exogenous input (NARX). Using the experimental data, the introduced neural network thermal models have been trained. By selecting a typical transformer, the trained thermal models are validated using additional experimental results and the traditional thermal models. It is shown that the introduced neural network based thermal models have a good performance in temperature prediction of the winding and the cooling air in the cast-resin dry-type transformer. The introduced thermal models are more accurate for the temperature analysis of this transformer and they will be trained easily. Finally, the trained and validated thermal models are employed to evaluate the life-time and the reliability of a typical cast-resin dry-type transformer.

3

Detecting objects using Rolling Convolution and Recurrent Neural Network

100%

Huang W. , Huang M. , Wang Y.

|

tom Vol. 65, No. 2

293--301

EN

At present, most of the existing target detection algorithms use the method of region proposal to search for the target in the image. The most effective regional proposal method usually requires thousands of target prediction areas to achieve high recall rate.This lowers the detection efficiency. Even though recent region proposal network approach have yielded good results by using hundreds of proposals, it still faces the challenge when applied to small objects and precise locations. This is mainly because these approaches use coarse feature. Therefore, we propose a new method for extracting more efficient global features and multi-scale features to provide target detection performance. Given that feature maps under continuous convolution lose the resolution required to detect small objects when obtaining deeper semantic information; hence, we use rolling convolution (RC) to maintain the high resolution of low-level feature maps to explore objects in greater detail, even if there is no structure dedicated to combining the features of multiple convolutional layers. Furthermore, we use a recurrent neural network of multiple gated recurrent units (GRUs) at the top of the convolutional layer to highlight useful global context locations for assisting in the detection of objects. Through experiments in the benchmark data set, our proposed method achieved 78.2% mAP in PASCAL VOC 2007 and 72.3% mAP in PASCAL VOC 2012 dataset. It has been verified through many experiments that this method has reached a more advanced level of detection.

4

Adaptive ANC System with Nonlinear Secondary Path Based on Recurrent Neural Network

100%

Falkiewicz J. , Okoński M. A. M. A. , Kustra G. G.

Mechanika / Akademia Górniczo-Hutnicza im. St. Staszica

|

2003

|

tom T. 22, z. 3

263-270

EN

In this paper, the active noise control system based on an adaptively trained recurrent neural network (RNN) is proposed. In our research, we assume that secondary path is modelled by nonlinear transfer function, i.e. linear filter followed by a saturation nonlinearity. To compare commonly used filtered-X LMS (FXLMS) and filtered-X RLS (FXRLS) algorithms with the proposed RNN-based scheme, computer simulations are carried out. Results of experiments have shown that the adaptively trained recurrent neural network provides a significant improvement in an operation of the adaptive ANC systems in the steady-state in comparison with classically used FXLMS and FXRLS algorithms.

5

An arma type pi-sigma artificial neural network for nonlinear time series forecasting

100%

Akdeniz E. , Egrioglu E. , Bas E. , Yolcu U.

Journal of Artificial Intelligence and Soft Computing Research

|

2018

|

tom Vol. 8, No. 2

121--132

EN

Real-life time series have complex and non-linear structures. Artificial Neural Networks have been frequently used in the literature to analyze non-linear time series. High order artificial neural networks, in view of other artificial neural network types, are more adaptable to the data because of their expandable model order. In this paper, a new recurrent architecture for Pi-Sigma artificial neural networks is proposed. A learning algorithm based on particle swarm optimization is also used as a tool for the training of the proposed neural network. The proposed new high order artificial neural network is applied to three real life time series data and also a simulation study is performed for Istanbul Stock Exchange data set.

6

Neural Based Autonomous Navigation of Wheeled Mobile Robots

100%

Al-Sagban M. , Dhaouadi R.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2016

|

tom Vol. 10, No. 2

64--72

EN

This paper presents a novel reactive navigation algorithm for wheeled mobile robots under non-holonomic constraints and in unknown environments. Two techniques are proposed: a geometrical based technique and a neural network based technique. The mobile robot travels to a pre-defined goal position safely and efficiently without any prior map of the environment by modulating its steering angle and turning radius. The dimensions and shape of the robot are incorporated to determine the set of all possible collision-free steering angles. The algorithm then selects the best steering angle candidate. In the geometrical navigation technique, a safe turning radius is computed based on an equation derived from the geometry of the problem. On the other hand, the neural-based technique aims to generate an optimized trajectory by using a user-defined objective function which minimizes the traveled distance to the goal position while avoiding obstacles. The experimental results demonstrate that the algorithms are capable of driving the robot safely across a variety of indoor environments.

7

A Neuro-Fuzzy System Combined with Particle Swarm Optimization for Handwritten Character Recognition

88%

Chang B.-M.

Fundamenta Informaticae

|

2014

|

tom Vol. 133, nr 4

345--366

EN

A novel character recognition method, called a Neuro-Fuzzy system combined with Particle swarm optimization for Handwritten Character Recognition (NFPHCR), is proposed in this paper. The NFPHCR method integrates Recurrent Neural Network (RNN), Fuzzy Inference System (FIS), and Particle Swarm Optimization (PSO) algorithm to recognize handwritten characters. It employs the RNN to effectively extract oriented features of handwritten characters, and then, these features are applied to create the FIS. Finally, the FIS combined with the PSO algorithm can powerfully estimate similarity ratings between the recognized character and sampling characters in the character database. Experimental results demonstrate that the NFPHCR method achieves a satisfying recognition performance and outperforms other existing methods under considerations.

8

Character-based recurrent neural networks for morphological relational reasoning

88%

Mogren O. , Johansson R.

Journal of Language Modelling

|

2019

|

tom Vol. 7, No. 1

139--170

EN

We present a model for predicting inflected word forms based on morphological analogies. Previous work includes rule-based algorithms that determine and copy affixes from one word to another, with limited support for varying inflectional patterns. In related tasks such as morphological reinflection, the algorithm is provided with an explicit enumeration of morphological features which may not be available in all cases. In contrast, our model is feature-free: instead of explicitly representing morphological features, the model is given a demo pair that implicitly specifies a morphological relation (such as write: writes specifying infinitive:present). Given this demo relation and a query word (e.g. watch), the model predicts the target word (e.g. watches). To address this task, we devise a character-based recurrent neural network architecture using three separate encoders and one decoder. Our experimental evaluation on five different languages shows that the exact form can be predicted with high accuracy, consistently beating the baseline methods. Particularly, for English the prediction accuracy is 94.85%. The solution is not limited to copying affixes from the demo relation, but generalizes to words with varying inflectional patterns, and can abstract away from the orthographic level to the level of morphological forms.

9

Monitoring regenerative heat exchanger in steam power plant by making use of the recurrent neural network

88%

Niksa-Rynkiewicz T. , Szewczuk-Krypa N. , Witkowska A. , Cpałka K. , Zalasiński M. , Cader A.

|

tom Vol. 11, No. 2

143--155

EN

Artificial Intelligence algorithms are being increasingly used in industrial applications. Their important function is to support operation of diagnostic systems. This paper presents a new approach to the monitoring of a regenerative heat exchanger in a steam power plant, which is based on a specific use of the Recurrent Neural Network (RNN). The proposed approach was tested using real data. This approach can be easily adapted to similar monitoring applications of other industrial dynamic objects.

10

Basic concepts of dynamic recurrent neural networks development

88%

Boyko N. , Pobereyko P.

ECONTECHMOD : An International Quarterly Journal on Economics of Technology and Modelling Processes

|

2016

|

tom Vol. 5, No 2

63--68

EN

In this work formulated relevance, set out an analytical review of existing approaches to the research recurrent neural networks (RNN) and defined precondition appearance a new direction in the field neuroinformatics – reservoir computing. Shows generalized classification neural network (NN) and briefly described main types dynamics and modes RNN. Described topology, structure and features of the model NN with different nonlinear functions and with possible areas of progress. Characterized and systematized wellknown learning methods RNN and conducted their classification by categories. Determined the place RNN with unsteady dynamics of other classes RNN. Deals with the main parameters and terminology, which used to describe models RNN. Briefly described practical implementation recurrent neural networks in different areas natural sciences and humanities, and outlines and systematized main deficiencies and the advantages of using different RNN. The systematization of known recurrent neural networks and methods of their study is performed and on this basis the generalized classification of neural networks was proposed.

11

Single-ended quality measurement of a music content via convolutional recurrent neural networks

88%

Organiściak K. , Borkowski J.

Metrology and Measurement Systems

|

2020

|

tom Vol. 27, nr 4

721--733

EN

The paper examines the usage of Convolutional Bidirectional Recurrent Neural Network (CBRNN) for a problem of quality measurement in a music content. The key contribution in this approach, compared to the existing research, is that the examined model is evaluated in terms of detecting acoustic anomalies without the requirement to provide a reference (clean) signal. Since real music content may include some modes of instrumental sounds, speech and singing voice or different audio effects, it is more complex to analyze than clean speech or artificial signals, especially without a comparison to the known reference content. The presented results might be treated as a proof of concept, since some specific types of artefacts are covered in this paper (examples of quantization defect, missing sound, distortion of gain characteristics, extra noise sound). However, the described model can be easily expanded to detect other impairments or used as a pre-trained model for other transfer learning processes. To examine the model efficiency several experiments have been performed and reported in the paper. The raw audio samples were transformed into Mel-scaled spectrograms and transferred as input to the model, first independently, then along with additional features (Zero Crossing Rate, Spectral Contrast). According to the obtained results, there is a significant increase in overall accuracy (by 10.1%), if Spectral Contrast information is provided together with Mel-scaled spectrograms. The paper examines also the influence of recursive layers on effectiveness of the artefact classification task.

12

Detecting objects using Rolling Convolution and Recurrent Neural Network

88%

Huang W. , Huang M. , Wang Y.

|

nr 2

13

Gray-level co-occurrence matrix of Smooth Pseudo Wigner-Ville distribution for cognitive workload estimation

75%

Mirzaeian R. , Ghaderyan P.

Biocybernetics and Biomedical Engineering

|

2023

|

tom Vol. 43, no. 1

261--278

EN

Automatic, cost-effective, and reliable cognitive workload estimation (CWE) is one of the important issues in diagnosis and treatment of neurocognitive diseases, cognitive performance improvement and error preventive strategies. To address this issue, this paper has proposed a novel and robust CWE method by detecting the time–frequency (TF) changes of electrodermal activities (EDA). Firstly, the local and global properties of the time-variant characteristics of EDA have been presented using Smooth Pseudo WignerVille distribution with enhanced TF resolution. Then, the transient changes in TF images of EDA signals have been quantified using a set of textural features based on Gray Level Co-occurrence Matrix descriptor (GLCM). Several static and dynamic classifiers, such as support vector machine, K- k-nearest neighbor, cascade forward neural network, and recurrent neural network have been explored. A real EDA data experiment recorded during arithmetic task with different workload levels have been used to evaluate the performance of the proposed method. The obtained results have confirmed that it can achieve a high estimation performance of 97.71% using contrast feature for discrimination of three workload levels. Further analysis has also suggested that the model is robust to GLCM parameters and classifiers and can provide a better tradeoff between computational complexity and high performance using minimum number of textural features in comparison with previous studies.

14

Combining homomorphic filtering and recurrent neural network in gait signal analysis for neurodegenerative diseases detection

75%

Saljuqi M. , Ghaderyan P.

Biocybernetics and Biomedical Engineering

|

2023

|

tom Vol. 43, no. 2

476--493

EN

Automatic, cost-effective, and reliable detection of neurodegenerative diseases (NDs) is one of the important issues in clinical practice. The main idea of the proposed method in this study is to utilize the advantages of gait time series that can provide low-cost and noninvasive measures, homomorphic filtering that can effectively separate muscle activity from body dynamic and recurrent neural network or cascade forward neural network that can learn sequential time-varying data. Experimental results on gait time series of 16 healthy control subjects, 13 patients with amyotrophic lateral sclerosis, 15 patients with Parkinson’s disease and 20 patients with Huntington’s disease have demonstrated high detection performance with an accuracy rate of 100 % using K-fold cross validation for all three types of NDs outperforming other existing methods. The results have also indicated that the use of real cepstral coefficients, oscillation components, or basic statistics feature set has improved the detection performance.

15

Heuristic modeling using recurrent neural networks: simulated and real-data experiments

75%

Przystałka P.

Computer Assisted Mechanics and Engineering Sciences

|

2007

|

tom Vol. 14, No. 4

715-727

EN

The focus of this paper is on the problems of system identification, process modeling and time series forecasting which can be met during the use of locally recurrent neural networks in heuristic modeling technique. However, the main interest of this paper is to survey the properties of the dynamic neural processor which is developed by the author. Moreover, a comparative study of selected recurrent neural architectures in modeling tasks is given. The results of experiments showed that some processes tend to be chaotic and in some cases it is reasonable to use soft computing models for fault diagnosis and control.

16

Application of Recurrent Neural Networks for User Verification based on Keystroke Dynamics

75%

Kobojek P. , Saeed K.

|

tom nr 3

80--90

EN

Keystroke dynamics is one of the biometrics techniques that can be used for the verification of a human being. This work briefly introduces the history of biometrics and the state of the art in keystroke dynamics. Moreover, it presents an algorithm for human verification based on these data. In order to achieve that, authors’ training and test sets were prepared and a reference dataset was used. The described algorithm is a classifier based on recurrent neural networks (LSTMand GRU). High accuracy without false positive errors as well as high scalability in terms of user count were chosen as goals. Some attempts were made to mitigate natural problems of the algorithm (e.g. generating artificial data). Experiments were performed with different network architectures. Authors assumed that keystroke dynamics data have sequence nature, which influenced their choice of classifier. They have achieved satisfying results, especially when it comes to false positive free setting.

17

Convolutional and Recurrent Neural Networks for Face Image Analysis

75%

Yüksel K. , Skarbek W.

|

2019

|

tom Vol. 44, No. 3

331--347

EN

In the presented research two Deep Neural Network (DNN) models for face image analysis were developed. The first one detects eyes, nose and mouth and it is based on a moderate size Convolutional Neural Network (CNN) while the second one identifies 68 landmarks resulting in a novel Face Alignment Network composed of a CNN and a recurrent neural network. The Face Parts Detector inputs face image and outputs the pixel coordinates of bounding boxes for detected facial parts. The Face Alignment Network extracts deep features in CNN module while in the recurrent module it generates 68 facial landmarks using not only this deep features, but also the geometry of facial parts. Both methods are robust to varying head poses and changing light conditions.

18

Identyfikacja rekurencyjnych modeli neuronowych typu RBF

75%

Ławryńczuk M.

|

tom R. 12, nr 2

558--570

PL

W pracy omówiono algorytm identyfikacji (uczenia) rekurencyjnych sieci neuronowych typu RBF (ang. Radial Basis Function), które mogą być zastosowanie w modelowaniu nieliniowych procesów dynamicznych. W porównaniu z powszechnie stosowanym klasycznym algorytmem wstecznej propagacji błędów, który prowadzi do wyznaczenia modeli służących do predykcji jednokrokowej, proponowany algorytm umożliwia identyfikację predyktorów wielokrokowych. Przedstawiony algorytm wykorzystano do modelowania reaktora chemicznego.

EN

This paper details an identification (training) algorithm of RBF (Radial Basis Function) recurrent neural networks which can be used for modelling nonlinear dynamic processes. In comparison with the widely used classical backpropagation algorithm, which leads to one-step ahead predictors, the described one results in many-steps ahead predictors. The algorithm is used for modelling a chemical reactor.

19

Future Graduate Salaries Prediction Model Based On Recurrent Neural Network

75%

Siłka J. , Wieczorek M. , Woźniak M.

Annals of Computer Science and Information Systems

|

2020

|

tom Vol. 21

427--430

EN

Prediction models are widely applied in several fields. In this study we present a discussion on using Recurrent Neural Network as predictor for salaries of future graduates. The model is based on feature analysis which leads to input values of the predictor. We have analyzed several compositions and ideas. As a result we have selected Recurrent Neural Network to be the most accurate. Presented results confirm this selection and show high precision.

20

Recurrent neural networks for non-linear system identification

63%

Krejsa J. , Dolezal R.

Zeszyty Naukowe Katedry Mechaniki Stosowanej / Politechnika Śląska

|

1999

|

tom z. 11

37-40

EN

The contribution shows Elman neural network used for non-linear system identification. A simple example of non-linear dynamic system is used to test the performance of networks with different number of hidden units. Results shows that higher number of hidden neurons surprisingly degrades the performance of the network both in training and generalisation abilities.

PL

W pracy przedstawiono zastosowanie neuronowej sieci Elmana do identyfikacji układu nieliniowego. Na przykładzie prostego nieliniowego układu dynamicznego zbadano osiągi sieci z różną liczbą ukrytych neuronów. Wyniki wskazują, że większe liczb)' ukrytych neuronów zmniejszają zdolności treningowe i uogólniające sieci.