Nonlinear identification of eddy current sensors based on artificial neural networks

2008 ◽  
Author(s):  
Hairong Zou
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Eddy Current ◽  
Nonlinear Identification ◽  
Artificial Neural ◽  
Current Sensors

Identification of Conductivity Distribution Using Eddy Current Tomography System and Artificial Neural Networks

2010 ◽  
Vol 670 ◽  
pp. 336-344
Author(s):  
Tomasz Chady ◽  
Ireneusz Spychalski ◽  
Takashi Todaka
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Data Base ◽  
Eddy Current ◽  
Current System ◽  
Identification Problem ◽  
Conductivity Distribution ◽  
Tomography System ◽  
Dimensional Distribution ◽  
Artificial Neural

In certain applications (security, biomedical, food and wood testing etc.) it is necessary to detect and identify position of small metal particles with high precision. This paper presents an eddy current system designated for evaluation of conductivity distribution. The system was modeled using the finite element method as well as it was constructed and the measurements were carried out. Using these results a data base of the signals achieved for various configurations of the test objects were created. The data base was utilized to solve the identification problem. Artificial neural networks were utilized as the inverse models in order to reconstruct two-dimensional distribution of conductivity. Selected results achieved for simulated signals were presented.

scholarly journals Evaluation of Loose Assemblies Using a Multi-frequency Eddy Current Method and Artificial Neural Networks 

2021 ◽  
Vol 31 (3) ◽  
pp. 78-82
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Eddy Current ◽  
Current Method ◽  
Eddy Current Method ◽  
The Neural Network ◽  
Artificial Neural ◽  
Aluminum Plates ◽  
The Relationship ◽  
Non Destructive

The paper deals with the non-destructive evaluation of the airgap existing between parts in loose metallic assemblies, using the eddy current (EC) method. In this study, the relationship between the variations of the impedance of a ferrite-cored coil sensor and an assembly featuring two aluminum plates is analyzed. Then artificial neural networks, based on statistical learning of the relationship between a sensor and an assembly are proposed and developed using both simulated and measured multi-frequency EC data, so as to estimate the distance between the assembly parts in a range from 0 µm to 500 µm. For the neural network built on experiment data, the inaccuracy of obtained results is smaller than 1.06%.

Intelligence Monitor and Diagnosis to High Speed Brushes Aeration Mechanics Based on Turbulent Flow Displacement Sensing Theory

2010 ◽  
Vol 426-427 ◽  
pp. 191-196
Author(s):  
H.I. Liu ◽  
X.P. Li ◽  
Yan Nian Rui ◽  
Ying Ping He
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Eddy Current ◽  
High Speed ◽  
Current Effect ◽  
Vibration Signals ◽  
Flow Displacement ◽  
Eddy Current Effect ◽  
Artificial Neural ◽  
Main Spindle

High Speed Brushes Aeration Mechanics are the effective aeration equipments which are widely used in the environmental protection. Because of the big span of main spindle and its high speed when it is working, the breakdown sometimes occurs. It is very importance to monitor its condition and diagnose its breakdowns. Turbulent Flow Displacement Sensors are the non-contact types which are based on eddy current effect. It has many advantages, such as good linearity, wide frequency response scope, convenience installment and so on. So it is very suitable for the main spindle’s vibration signals of a high speed brushes aeration mechanic are monitored. With the development of Artificial Neural Networks technology, the equipment breakdown diagnosis has realized intellectualization. The recognition of equipment failure types is one of the most important studying domains of Artificial Neural Networks at present. Based on the research of eddy current effect and Artificial Neural Networks, we build up a test system which can monitor condition and diagnose breakdown to a GSB-12 high speed brushes aeration mechanic. With the help of it, the vibration signals of the measurement points on the main spindle are measured at two mutually vertical positions. The signals’ base frequency and multiplicative frequency are taken as characteristic value. Six common breakdowns are selected and to be taken as the standard sample and there are 3 lays in the neural network. Using FBP algorithm, we get a satisfied effect. The experiment has confirmed that this method is advanced, reliable and practical. It provides a new method about intelligent monitor and breakdown diagnosis to high speed brushes aeration mechanics’ condition.

Sign in / Sign up

Export Citation Format

Share Document