Boots – shoes – and leggings
Patent
1996-05-24
1998-03-03
Voeltz, Emanuel T.
Boots, shoes, and leggings
324521, G01R 3100
Patent
active
057242470
DESCRIPTION:
BRIEF SUMMARY
This application is a 371 of International application PCT/DE94/01151 filed Sep. 20, 1994.
FIELD OF THE INVENTION
The present invention relates to a method for generating a signal indicating the direction of a short-circuit current in relation to a control point on an electric transmission line to be monitored.
BACKGROUND OF THE INVENTION
A method for generating a signal indicating the direction of a short-circuit current is described in the publication "An Artificial Neural Network Based Directional Discriminator for Protecting Transmission Lines" in "Proc. of the Canadian Conference on Electrical and Computer Engineering," 1993, Vol. 1, pp. 205-208. In the process described therein, sample values, formed from current and voltage signals derived from currents and voltages of a power transmission line to be monitored, are supplied to the input of a neural network. For the neural network to be able to recognize the direction of a short-circuit current, the neural network is previously taught with suitable sample values. A neural network thus taught is capable of determining a direction immediately after the occurrence of a short-circuit with approximately ten sample values.
SUMMARY OF THE INVENTION
The present invention provides a method for generating a signal indicating the direction of a short-circuit current in relation to a control point on an electric transmission line to be monitored in which current and voltage signals derived from phase currents and phase voltages of the power transmission line are used for generating the direction signal, wherein each derived current and voltage signal is sampled and normalized separately for each phase to obtain different series of normalized sample values, and in which signals are supplied to a neural network consisting of an input layer with input neurons, a hidden layer, and an output layer with one output neuron, and which is taught by coaching through simulation of different short-circuits at different points of the power transmission line to be monitored, so that the signal of the output neuron assumes one predefined value for a short-circuit in one direction and another predefined value for a short-circuit in the other direction.
The present invention further provides for forming a triggering reference value, obtained from the sample value immediately previous to triggering and, in the event of triggering, from consecutively generated triggering reference values for each current sample value, and, in the event of triggering, the triggering reference value corresponding to the same time of this sample value is subtracted from the current sample values of the different series and thus a series of difference values is formed for each phase; the difference values of all series are simultaneously supplied to different input neurons of the neural network, which has a second hidden layer and which is taught by coaching through simulation of different short-circuits at different points of the power transmission line to be monitored, so that the signal of the output neuron exceeds a predefined upper threshold in the event of a short-circuit in one direction and remains below a predefined lower threshold in the event of a short-circuit in the other direction; the signal of the output neuron is examined for its magnitude and, when a threshold value is reached, the direction signal is generated.
One advantage of the method according to the present invention is that, after triggering, the aforementioned difference values are supplied to the input neurons of the neural network, the difference values having been ultimately produced only with triggering reference values obtained from consecutively formed triggering reference values.
An advantageous embodiment of the method according to the present invention provides for the use of a neural network for detecting a short-circuit between the power transmission line to be monitored and another power transmission line running parallel to it; the said neural network is additionally taught through simulation of short-ci
REFERENCES:
patent: 4591992 (1986-05-01), Yamaura
patent: 5159660 (1992-10-01), Lu et al.
patent: 5576618 (1996-11-01), Calero
Proceedings of Canadian Conference On Electrical and Computer Engineering, Bd. 1, 17, Sep. 1993, Vancouver CA, pp. 205-208, Sidhu "An Artificial Neural Network Based Directional Discriminator For Protecting Transmission Lines".
IEEE Transactions On Power Systems, Bd. 7, Nr. 1, May 1992, New York, US, pp. 812-819, Kandil "Fault Identification In An AC-DC Transmission System Using Neural Networks".
Proceedings of the 27th Intersociety Energy Conversion Engineering Conference, Bd. 1, Aug. 1992, San Diego, US, pp. 1191-1196, Fernando "High Impedance Fault Detection Using Artificial Neural Network Techniques".
Muller et al., Selekivschutz Elektrischer Anlagen, Aug. 1990, pp. 73-76.
Peeso Thomas
Siemens Aktiengesellschaft
Voeltz Emanuel T.
LandOfFree
Method of generating a signal indicating the direction of a shor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of generating a signal indicating the direction of a shor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of generating a signal indicating the direction of a shor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2254740