A framework for robust neural network-based control of nonlinear servomechanisms

Lee, T.H.; Wang, Q.G.; Tan, W.K.

doi:10.1109/icit.1994.467183

Cited by 2 publications

(1 citation statement)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Surface form data have been obtained using a laser-based system [2], and it is intended that surface data be obtained for system control using this method, which will subsequently be fed into an ANN used in the control loop in order to provide the performance necessary for real time control of the system. ANNs are highly suited to the control of hydraulic servo mechanisms since they feature rapid response and the ability to cope with non-linearities [21] and time variant properties inherent to hydraulic servos [22]. For example, oil column stiffness will drop over a period of several hours of run time owing to increased temperature and ingress of air into the system.…”

Section: Surface Finish Simulationmentioning

confidence: 99%

Improving wood surface form by modification of the rotary machining process—a mechatronic approach

Brown

Parkin

1999

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

Rotary machining is extensively used for planing and moulding operations within the woodworking industry. Although the surface form produced by this machining method is acceptable, the rotary machining action produces cutter marks on the wood surface so that further finishing operations, such as sanding, are often required to generate a product of acceptable standard. It has been theorized that the surface finish of planed and moulded timber products may be improved by oscillation of the cutter block in either a vertical or horizontal plane. This paper describes the use of a rapid surface simulation algorithm to predict surface finish and the use of computer simulation to model cutterblock oscillation. The result is a tool for effective design and optimization of a hydraulic oscillation system in order to improve surface form. NOTATIONhydraulic ram area (m 2 ) A artificial neural network ANN bulk modulus (N/m 2 ) B capacitance (F) C C cv coefficient of viscous friction d strain developed/applied field (C/N) dielectric displacement (C/m 2 ) D field (V/m) E f d feed rate of workpiece (m/s) f k feed per knife (m) resonant frequency (Hz) f m f n antiresonant frequency (Hz) f r feed per revolution (m) F force (N) strain developed/applied charge density g (m 2 /C) depth of knife marking (m) h H original horizontal sample array I(t) x instantaneous cutter tip locus position (x axis) I(t) y instantaneous cutter tip locus position (y axis) k A drive amplifier constant (mA/V) k LE leakage coefficient (cm 3 /N m 2 ) spool displacement gain (mm/mA) k s M total ram, slideway and cutterblock mass (kg) n cutterhead rotational speed (rad/s) number of knives in the cutterhead produc-N ing a wave on the workpiece p hydraulic fluid pressure (N/m 2 ) P pitch of knife marking (mm) P f pressure required to overcome viscous friction hydraulic fluid flow, single acting ram input q (cm 3 /s) Q general hydraulic fluid flow (cm 3 /s) energy supplied/dissipated per cycle Q m radius of cutting circle (m) r ct s Laplace operator S d scallop depth: depth of knife marking (mm) mechanical compliance (m 2 /N) s E S strain surface simulation algorithm SSA t time (s) T stress (N/m 2 ) valve actuator time constant 6 m V volume of oil column (m 3 ) V v valve voltage (V) x hydraulic ram displacement (m) valve spool displacement (m) X s The MS was recei6ed on

show abstract

Section: Surface Finish Simulationmentioning

confidence: 99%