Radiant energy – Photocells; circuits and apparatus – Optical or pre-photocell system
Reexamination Certificate
2000-02-24
2002-04-23
Allen, Stephone (Department: 2878)
Radiant energy
Photocells; circuits and apparatus
Optical or pre-photocell system
C114S312000, C701S021000, C706S040000
Reexamination Certificate
active
06376831
ABSTRACT:
The present invention relates generally to obtaining data with respect to seaway conditions to which a submerged sea-going vessel is exposed under different operational conditions.
BACKGROUND OF THE INVENTION
The estimation of seaway conditions including seastate, wave direction relative to a submerged sea-going vessel, its heading and the seaway induced suction force on the vessel surfaces, which increases exponentially as the vessel approaches the water surface, is of interest in order to offset the suction force by corrective vessel control so as to maintain keel depth and avoid broach. Attempts have been made to deal with the foregoing problem by development of control methods based on seastate estimation algorithms. However, none of such algorithm estimation methods is sufficiently robust for satisfactory use under actual seastate operation.
The use of a neural network type of signal processing system is generally known in the art as disclosed in U.S. Pat. No. 5,180,911 to Grossman et al. Use of such neural network in the signal processing system as disclosed in the Grossman et al. patent is associated with a complex input signal generating arrangement including optical waveguide, light beam launcher and photodetectors.
It is therefore an important object of the present invention to provide a digital signal processing system having a neural network architecture for seastate estimations based on input signals derived from sea-going vessel measurements so as to predict vessel surface suction force with computational simplicity and operational robustness.
SUMMARY OF THE INVENTION
In accordance with present invention, depth, pitch, roll and forward speed measurements of a sea-going vessel during underwater travel are respectively processed into signal inputs supplied to four neural networks arranged in accordance with an algorithmic architecture involving supply of variable outputs from two of the neural networks to three detectors through which four outputs from two of the neural networks are converted into accurate estimates of stem, beam and bow directional headings of the vessel. The variable output from a third neural network is supplied to a fourth detector through which wave direction relative to the vessel is estimated. The fourth neural network provides one output varied as a linear variable function supplied to a fifth detector for conversion into an estimate of sea energy at the sea surface.
REFERENCES:
patent: 5180911 (1993-01-01), Grossman et al.
patent: 5845052 (1998-12-01), Baba et al.
patent: 6016763 (2000-01-01), Takahashi et al.
patent: 6308649 (2001-10-01), Gedeon
Allen Stephone
Shuster Jacob
The United States of America as represented by the Secretary of
LandOfFree
Neural network system for estimating conditions on submerged... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Neural network system for estimating conditions on submerged..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Neural network system for estimating conditions on submerged... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2847301