PL
 |
EN
Preferencje help
Polish version English version Język
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 10

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  feedforward neural network
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Content available Forward and inverse kinematics solution of a robotic manipulator using a multilayer feedforward neural network
Sharkawy Abdel-Nasser
Journal of Mechanical and Energy Engineering
|
2022
|
Vol. 6, No 2
1--17
EN
In this paper, a multilayer feedforward neural network (MLFFNN) is proposed for solving the problem of the forward and inverse kinematics of a robotic manipulator. For the forward kinematics solution, two cases are presented. The first case is that one MLFFNN is designed and trained to find solely the position of the robot end-effector. In the second case, another MLFFNN is designed and trained to find both the position and the orientation of the robot end-effector. Both MLFFNNs are designed considering the joints’ positions as the inputs. For the inverse kinematics solution, a MLFFNN is designed and trained to find the joints’ positions considering the position and the orientation of the robot end-effector as the inputs. For training any of the proposed MLFFNNs, data is generated in MATLAB using two different cases. The first case is that data is generated assuming an incremental motion of the robot’s joints, whereas the second case is that data is obtained with a real robot considering a sinusoidal joint motion. The MLFFNN training is executed using the Levenberg-Marquardt algorithm. This method is designed to be used and generalized to any DOF manipulator, particularly more complex robots such as 6-DOF and 7-DOF robots. However, for simplicity, this is applied in this paper using a 2-DOF planar robot. The results show that the approximation error between the desired output and the estimated one by the MLFFNN is very low and it is approximately equal to zero. In other words, the MLFFNN is efficient enough to solve the problem of the forward and inverse kinematics, regardless of the joint motion type.
2
Content available remote Cancer prediction using cascade generalization and duo output neural network
Nilnumpetch Chatree, Amornsamankul Somkid, Kraipeerapun Pawalai
Annals of Computer Science and Information Systems
|
2021
|
Vol. 27
65--70
EN
This paper proposes the combination of cascade generalization and duo output neural network based on feedforward backpropagation neural networks for cancer prediction. Duo output neural network is a neural network that is created based on two opposite targets in order to predict two opposite results. Cascade generalization is a technique that consists of a set of machines that are sorted together in which the predicted output produced from the previous machine plus the original training input are used for the creation of each machine. In this study, cascade generalization is organized in two levels: the base level and the meta level. In this research, duo output neural network is trained in each level of cascade generalization. Two outputs produced from the base level which are truth output and non-falsity output are averaged. The average result plus the original input are used for training a machine in meta level. The proposed technique is tested using two cancer datasets from UCI machine learning repository and found that our technique provides the best overall results when compared with three individual techniques.
3
Content available remote Pathological Brain Detection via Wavelet Packet Tsallis Entropy and Real-Coded Biogeography-based Optimization
Wang S., Li P., Chen P., Phillips P., Liu G., Du S., Zhang Y.
Fundamenta Informaticae
|
2017
|
Vol. 151, nr 1/4
275--291
EN
(Aim) In order to detect pathological brains in a more efficient way, (Method) we proposed a novel system of pathological brain detection (PBD) that combined wavelet packet Tsallis entropy (WPTE), feedforward neural network (FNN), and real-coded biogeography-based optimization (RCBBO). (Results) The experiments showed the proposed WPTE + FNN + RCBBO approach yielded an average accuracy of 99.49% over a 255-image dataset. (Conclusions) The WPTE + FNN + RCBBO performed better than 10 state-of-the-art approaches.
4
Content available remote Abnormal Breast Detection in Mammogram Images by Feed-forward Neural Network Trained by Jaya Algorithm
Wang S., Rao R. V., Chen P., Zhang Y., Liu A., Wei L.
Fundamenta Informaticae
|
2017
|
Vol. 151, nr 1/4
191--211
EN
(Aim) Abnormal breast can be diagnosed using the digital mammography. Traditional manual interpretation method cannot yield high accuracy. (Method) In this study, we proposed a novel computer-aided diagnosis system for detecting abnormal breasts in mammogram images. First, we segmented the region-of-interest. Next, the weighted-type fractional Fourier transform (WFRFT) was employed to obtain the unified time-frequency spectrum. Third, principal component analysis (PCA) was introduced and used to reduce the spectrum to only 18 principal components. Fourth, feed-forward neural network (FNN) was utilized to generate the classifier. Finally, a novel algorithm-specific parameter free approach, Jaya, was employed to train the classifier. (Results) Our proposed WFRFT + PCA + Jaya-FNN achieved sensitivity of 92.26% ± 3.44%, specificity of 92.28% ± 3.58%, and accuracy of 92.27% ± 3.49%. (Conclusions) The proposed CAD system is effective in detecting abnormal breasts and performs better than 5 state-of-the-art systems. Besides, Jaya is more effective in training FNN than BP, MBP, GA, SA, and PSO.
5
Content available Repetitive neurocontroller with disturbance feedforward path active in the pass-to-pass direction for a VSI inverter with an output LC filter
Ufnalski B., Grzesiak L. M.
Bulletin of the Polish Academy of Sciences. Technical Sciences
|
2016
|
Vol. 64, nr 1
115--125
EN
An enhancement to the previously developed repetitive neurocontroller (RNC) is discussed and investigated in the paper. Originally, the time-base generator (TBG) has been used to produce the only input signal for the neural approximator. The resulting search space makes the dynamic optimization problem (DOP) of shaping the control signal solvable with the help of a function approximator such as the feed-forward neural network (FFNN). The plant under consideration, i.e. a constant-amplitude constant-frequency voltage-source inverter (CACF VSI) with an output LC filter, is assumed to be equipped with the disturbance load current sensor to enable implementation of the disturbance feed-forward (pDFF) path as a part of the non-repetitive subsystem acting in the along the pass p-direction. An investigation has been undertaken to explore potential benefits of using this signal also as an additional input for the RNC to augment the approximation space and potentially enhance the convergence rate of the real-time search process. It is numerically demonstrated in the paper that the disturbance feed-forward path active in the pass-to-pass k-direction (kDFF) improves the dynamics of the repetitive part as well indeed.
6
Content available remote Optimization of the test conditions for fault detection in nonlinear analog circuits using supply current
Kuczyński A., Ossowski M.
Przegląd Elektrotechniczny
|
2013
|
R. 89, nr 2a
267--269
EN
In this paper, selection of the optimum test conditions for catastrophic fault diagnosis of analog circuits containing MOS transistors is presented. The method of fault detection applies power supply current waveform IDD as an indicator of a device feature. The stimulate signal parameters and values of additional components are changed in optimization process to extend variation between the test signals for considered faults. An illustrative numerical example is presented.
PL
W pracy przedstawiono dobór warunków testu w metodzie wykrywania i lokalizacji uszkodzeń katastroficznych w układach analogowych zawierających tranzystory MOS. W zastosowanym algorytmie detekcji uszkodzeń informacje o właściwościach układu są zakodowane w przebiegu prądu źródła, zasilającego obwód w stanie nieustalonym. Parametry sygnału pobudzającego i wartości dodatkowych elementów są modyfikowane w procesie optymalizacyjnym tak, by powiększyc różnice między sygnałami testowymi odpowiadającymi rozważanym uszkodzeniom. Działanie algorytmu zilustrowano na praktycznym przykładzie.
7
Content available remote Parametric fault detection in analog circuits containing MOS transistors
Kuczyński A.
Przegląd Elektrotechniczny
|
2011
|
R. 87, nr 5
84-87
PL
W pracy przedstawiono metodę wykrywania uszkodzeń parametrycznych w układach analogowych zawierających tranzystory MOS. W zastosowanym algorytmie informacje o właściwościach układu są zakodowane w przebiegu prądu źródła, zasilającego obwód w stanie nieustalonym. Sygnał testowy jest filtrowany, by uwypuklić właściwości układu. Aby zachować istotę informacji o układzie, jako wektory uczące sieć neuronową zastosowano współczynniki wielomianów aproksymujących wybrany składnik sygnału testowego. Działanie algorytmu zilustrowano na praktycznym przykładzie.
EN
In this paper, algorithm for parametric fault diagnosis of nonlinear, analog circuits containing MOS is presented. This method applies power supply current waveform IDD as an indicator of a device feature. Test signal is filtered using a discrete wavelet transform filter bank to obtain signal sensitive to changes of device parameters. Coefficients of the polynomial approximating the component are calculated and used to formulate a learning vector of a feedforward neural network. Thus, it is possible to achieve data compression without the considerable loss of information about the tested device. An illustrative numerical example is presented.
8
Content available remote Comparative study of neural networks used in modeling and control of dynamic systems
Tiliouine H.
Przegląd Elektrotechniczny
|
2011
|
R. 87, nr 7
104-109
EN
In this paper, a diagonal recurrent neural network that contains two recurrent weights in the hidden layer is proposed for the designing of a synchronous generator control system. To demonstrate the superiority of the proposed neural network, a comparative study of performances, with two other neural networks, has been made. The investigated neural networks were: a feedforward neural network (FFNN), a first-order diagonal recurrent neural network (1_DRNN) and the proposed second-order diagonal recurrent neural network (2_DRNN). Moreover, to confirm the superiority of the proposed 2_DRNN, the results obtained with this network were also compared with those of the IEEE recommended conventional excitation control system (CECS).
PL
W artykule przedstawiono diagonalną rekurencyjną sieć neuronową do projektowania układu regulacji generatora synchronicznego. Proponowana sieć posiada dwie rekurencyjne wagi w warstwie ukrytej. Aby wykazać wyższość proponowanej sieci dokonano analizy porównawczej efektywności z dwoma innymi sieciami neuronowymi. Badanymi sieciami neuronowymi były: jednokierunkowa sieć neuronowa (FFNN), diagonalna rekurencyjna sieć neuronowa pierwszego rzędu (1_DRNN) oraz proponowana diagonalna rekurencyjna sieć neuronowa drugiego rzędu (2_DRNN). Ponadto, aby potwierdzić wyższość proponowanej sieci (2_DRNN), uzyskane wyniki dla tej sieci porównano z wynikami uzyskanymi dla konwencjonalnego układu regulacji (CECS) zalecanego przez IEEE.
9
Content available remote Random generalization by feedforward neural networks
Rataj A.
Theoretical and Applied Informatics
|
2008
|
Vol. 20, No. 2
117-124
EN
A generalization by a feedforward neural network is discussed, such that some samples may be generalized using very dierent, conflicting criteria. A training set is deliberately constructed to show that feedforward neural networks in such a case can generalize very spuriously and randomly. To illustrate the dierences between dierent learning machines, results given by a small subset of the support vector machines are also presented.
PL
W artykule dyskutowane jest uczenie jednokierunkowych sieci neuronowych w którym niektóre próbki mogą być uogólnianie przy użyciu bardzo odmiennych kryteriów. Zbiór uczący jest specjalnie skonstruowany w sposób pokazujący, że w przypadku istnienia bardzo odmiennych kryteriów uogólniania sieć neuronowa może generalizować w sposób przypadkowy, wynikły prawie całkowicie ze struktury wewnętrznej sieci, a nie z zawartości pliku uczącego. Sieci neuronowe są porównane też do SVM-ów, które przy odpowiednich parametrach nie wykazały takiej losowości, jednak z drugiej strony w testowanych przypadkach nie potrafiły uogólnić niektórych wzorców w pliku uczącym.
10
Content available remote Bayesian regression approaches on example of concrete fatigue failure prediction
Słoński M.
Computer Assisted Mechanics and Engineering Sciences
|
2006
|
Vol. 13, No. 4
655-668
EN
The focus of this paper is the application of two nonlinear regression models in the context of Bayesian inference to the problem of failure prediction of concrete specimen under repeated loads based on experimental data. These two models are compared with an empirical formulae. Results on testing data show that both models give better point predictions than empirical formulae. Moreover, Bayesian regression approach makes it possible to calculate prediction intervals (error bars) describing the reliability of the models predictions.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.