Direct Inverse Control of Sensors by Neural Networks for Static/Low Frequency Applications
piecewise linear input-output transfer function;
hysteresis occurring in the input-output transfer function; with the aim of establishing whether some relationship exists between the severity of different nonlinearities and the complexity of the network required to control such nonlinearities in static/low-frequency sensor applications.
The compensation is performed using an artificial neural networks approach. The networks chosen were a static MLP and a tap-delayed line MLP, both trained by an improved BKP method which included a form of dynamic learning management.
KeywordsNetwork Configuration Neural Network Architecture Artificial Neural Network Approach Network Approximation Seismic Mass
Unable to display preview. Download preview PDF.
- 1.Kraft, M.: “Closed-loop accelerometer employing oversampling conversion”, PhD Thesis, Coventry University, 1997.Google Scholar
- 2.Gaura E., Burian A., “A dedicated medium for the synthesis of BKP networks”, Romanian J. of Biophysics, Vol. 5, No. 15, 1995, Bucharest, Romania.Google Scholar
- 3.Irwin, G.W., Warwick, K., Hunt, K.J., “Neural networks applications in control”, IEE Control Engineering Series 53, Short Run Press Ltd., UK, 1995Google Scholar
- 4.Godjevac, J., Steele, N. “Fuzzy systems and neural networks”, Autosoft J. Intelligent Automation and Soft Computing, 1995Google Scholar
- 5.Poopalasindam, S., “Neural network based digital compensation schemes for industrial pressure sensors”, Ph.D. Thesis, Coventry University, Sep 1995Google Scholar
- 6.Trifa, R. Munteanu, E. Gaura, “Neural Network based modelling and simulation of PM-Hybrid Stepping Motor Drives”, Proc. International Aegean Conference on Electrical Machines and Power Electronics, Vol. 2/2, pp. 460–464, Kusadasi, Turkey.Google Scholar