Wyniki wyszukiwania - BazTech

1

An unscented Kalman filter in designing dynamic GMDH neural networks for robust fault detection

Mrugalski M.

International Journal of Applied Mathematics and Computer Science

|

2013

|

Vol. 23, no. 1

157--169

EN

This paper presents an identification method of dynamic systems based on a group method of data handling approach. In particular, a new structure of the dynamic multi-input multi-output neuron in a state-space representation is proposed. Moreover, a new training algorithm of the neural network based on the unscented Kalman filter is presented. The final part of the work contains an illustrative example regarding the application of the proposed approach to robust fault detection of a tunnel furnace.

2

Modelowanie rekrystalizacji dynamicznej za pomocą dynamicznych sieci neuronowych

Stanisławczyk A., Kusiak J.

Hutnik, Wiadomości Hutnicze

|

2009

|

Vol. 76, nr 8

659-661

PL

W pracy zaprezentowano neuronowy model naprężenia uplastyczniającego dla mikrostopowej stali z dodatkiem niobu. W modelowaniu wykorzystano dynamiczną sieć neuronową typu LRN (Layer-Recurrent Network). Model uwzględnia dwa kluczowe w procesie rekrystalizacji dynamicznej parametry: temperaturę i prędkość odkształcenia. Opracowany model cechuje się wysoką dokładnością przewidywania wartości naprężenia uplastyczniającego (błąd RMSE równy 3,1 MPa), oraz bardzo dużą szybkością działania. Przedstawione wyniki potwierdzają przydatność dynamicznych sieci neuronowych w modelowaniu zjawisk i procesów dynamicznych.

EN

The paper presents the neural network model of the flow stress for nobium microalloyed steel. The dynamic neural network of the type LRN (Layer-Recurrent Network) was used in modelling. The model takes into account two key parameters in the dynamic recrystallization: temperature and strain rate. The model characterizes high accuracy of flow stress prediction (RMSE equals 3.1 MPa), and the very high speed. Introduced model confirms the usefulness of dynamic neural networks in the dynamic processes modelling.

3

Towards robustness in neural network based fault diagnosis

Patan K., Witczak M., Korbicz J.

International Journal of Applied Mathematics and Computer Science

|

2008

|

Vol. 18, no 4

443-454

EN

Challenging design problems arise regularly in modern fault diagnosis systems. Unfortunately, classical analytical techniques often cannot provide acceptable solutions to such difficult tasks. This explains why soft computing techniques such as neural networks become more and more popular in industrial applications of fault diagnosis. Taking into account the two crucial aspects, i.e., the nonlinear behaviour of the system being diagnosed as well as the robustness of a fault diagnosis scheme with respect to modelling uncertainty, two different neural network based schemes are described and carefully discussed. The final part of the paper presents an illustrative example regarding the modelling and fault diagnosis of a DC motor, which shows the performance of the proposed strategy.

4

Neural network control with fuzzy logic compensation

Li X. Z., Bajić V. B., Sha D. H., Wang H. Y.

Journal of Applied Computer Science

|

2000

|

Vol. 8, nr 2

31-44

EN

This paper presents a new AI based control strategy. A dynamic neural network is used to identify the plant on-line. The control signal is then calculated iteratively according to the responses of a reference model and the identified neural model of the process. A fuzzy logic block with four very simple rules is added to the loop to improve the overall loop properties. This synergetic AI control paradigm is tested via simulation. The results demonstrate that the proposed control strategy provides better disturbance rejection and tracking properties of the control loop than those achieved by an optimally tuned PID controller.

5

Dynamic Neural Networks for Process Modelling in Fault Detection and Isolation Systems

Korbicz J., Patan K., Obuchowicz A.

International Journal of Applied Mathematics and Computer Science

|

1999

|

Vol. 9, no 3

519-546

EN

A fault diagnosis scheme for unknown nonlinear dynamic systems with modules of residual generation and residual evaluation is considered. Main emphasis is placed upon designing a bank of neural networks with dynamic neurons that model a system diagnosed at normal and faulty operating points.To improve the quality of neural modelling, two optimization problems are included in the construction of such dynamic networks: searching for an optimal network architecture and the network training algorithm. To find a good solution, the effective well-known cascade-correlation algorithm is adapted here. The residuals generated by a bank of neural models are then evaluated by means of pattern classification. To illustrate the effectiveness of our approach, two applications are presented: a neural model of Narendra's system and a fault detection and identification system for the two-tank process.

6

Development of Dynamic Neural Networks With Application to Observer-Based Fault Detection and Izolation

Marcu T., Mirea L., Frank P. M.

International Journal of Applied Mathematics and Computer Science

|

1999

|

Vol. 9, no 3

547-570

EN

The paper suggests a neural-network approach to the design of robust fault diagnosis systems. The main emphasis is placed upon the development of neural observer schemes. They are built based on dynamic neural networks, i.e. dynamic multi-layer perceptrons with mixed structure. The goal is to achieve an adequate approximation of process outputs for known classes of the process behaviour. The obtained symptoms are then classified by means of static artificial nets. Appropriate decision mechanisms are designed for each type of observer schemes. An application to a laboratory process is included. It refers to component and instrument fault detection and isolation in a three-tank system.

7

Robust identification by dynamic neural networks using sliding mode learning

Poznyak A. S., Yu W., Sanchez E. N., Sira-Ramirez H.

Applied Mathematics and Computer Science

|

1998

|

Vol. 8, no 1

135-144

EN

The problem of identification of continuous, uncertain nonlinear systems in the presence of bounded disturbances is implemented using dynamic neural networks. The proposed neural identifier guarantees a bound for the state estimation error. This bound turns out to be a linear combination of internal and external uncertainty levels. The neural net weights are updated on-line by a learning algorithm based on the sliding mode technique. To the best of the authors' knowledge, such a learning scheme is proposed for dynamic neural networks for the first time. Numerical simulations illustrate its effectiveness, even for highly nonlinear systems in the presence of important disturbances.