Neural network for predicting lightning

• US 5,140,523 A
• Filed: 12/14/1990
• Issued: 08/18/1992
• Est. Priority Date: 09/05/1989
• Status: Expired due to Fees

• Current Assignee: KTAADN, Inc. (Rtx Corporation)
• Original Assignee: KTAADN, Inc. (Rtx Corporation)
• Inventors: Frankel, Donald S., Draper, James S.
• Primary Examiner(s): MacDonald, Allen R.
• Application Number:
  US07/628,695
• Time in Patent Office: 613 Days
• Field of Search: 364/513, 364/807, 364/420, 324/72, 73/170 R, 340/421, 340/601, 395/22, 395/931
• US Class Current: 702/4
• CPC Class Codes: G01K 2203/00 Application of thermometers in cryogenics
• CPC Class Codes: G01W 1/10 Devices for predicting weather conditions computers per se G06; display devices G09
• CPC Class Codes: Y10S 706/931 Weather

Abstract

A system and method are provided for the automated prediction of lightning strikes in a set of different spatial regions for different times in the future. In a preferred embodiment, the system utilizes measurements of many weather phenomena. The types of measurements that can be utilized in approximately the same geographical region as that for which the strike predictions are made. This embodiment utilizes a correlation network to relate these weather measurements to future lightning strikes.

Claims

1. A system for predicting the occurrence of lightning within a given area based on weather data collected and lightning occurrences detected during a plurality of past days and on weather data collected during the present day, the system comprising:

collection means for collecting weather data;

detection means for detecting the occurrence of lightning in the given area;

correlation means, in communication with the collection means and the detection means, for determining correlations between the weather data collected by the collection means during a plurality of past days and the occurrence of lightning detected by the detection means during the plurality of past days in the given area; and

prediction means, in communication with the collection means and the correlation means, for determining the probability of an occurrence of lightning in the given area at some future time, based on the weather data being collected during the present day and the correlations determined by the correlation means from the weather data collected on past days.

2. A system according to claim 1, wherein the collection means includes a plurality of anemometers for determining wind speed and direction, the anemometers being located at a plurality of locations.

3. A system according to claim 2, wherein the collection means further includes an electric field sensor for detecting the presence of a charged cloud.

4. A system according to claim 2, wherein the collection means further includes doppler radar for detecting precipitation.

5. A system according to claim 2, wherein the collection means includes an earth-orbiting satellite.

6. A system according to claim 2, wherein the collection means includes means for determining heights of nearby clouds and collecting data regarding the heights.

7. A system according to claim 1, wherein the detection means includes a plurality of direction-finding antennas for determining the location of lightning.

8. A system according to claim 1, wherein the collection means includes means for obtaining data regarding an overall pattern of weather.

9. A system according to claim 1, wherein the correlation means includes a neural network.

10. A method of operating a data processor to enable the data processor to predict the occurrence of lightning within a given area based on weather data collected and lightning occurrences detected during a plurality of past days and on weather data collected during the present day, the method comprising the steps of:

collecting weather data;

detecting the occurrence of lightning in the given area;

storing information regarding weather data and lightning occurrences in a storage area of the data processor;

determining and storing correlations between the weather data collected during a plurality of past days and the occurrence of lightning during the plurality of past days in the given area; and

determining the probability of an occurrence of lightning in the given area at some future time, based on the weather data collected during the present day and the stored correlations.

11. A method according to claim 10, wherein the step of collecting weather data includes the steps of providing a plurality of anemometers for determining wind sped and direction, and locating the anemometers at a plurality of locations.

12. A method according to claim 11, wherein the step of collecting weather data includes the step of providing an electric field sensor for detecting the presence of a charged cloud.

13. A method according to claim 11, wherein the step of collecting weather data includes providing doppler radar for detecting precipitation.

14. A method according to claim 11, wherein the step of collecting weather data includes providing an earth-orbiting satellite.

15. A method according to claim 11, wherein the step of collecting the weather data includes the steps of determining heights of nearby clouds and collecting data regarding the heights of nearby clouds.

16. A method according to claim 11, wherein the step of detecting lightning includes providing a plurality of direction-finding antennas for determining the location of lightning.

17. A method according to claim 10, wherein the step of collecting the weather data includes the step of collecting data regarding an overall pattern of weather.

18. A system for predicting the occurrence of lightning in a given area based on weather data collected and lightning occurrences detected during a plurality of past days and on weather data collected during present day, the system comprising:

collection means for collecting the weather data;

detection means for detecting the occurrence of lightning in the given area and for recording data regarding when the lightning occurred;

training means for receiving data from the collection means and from the detection means and for determining network coefficients, based on data regarding weather and lightning occurrences on past days received from the collection means and the detection means; and

prediction means for receiving data from the collection means and the network coefficients from the training means and for predicting the occurrence of lightning, based on data received from the collection means and the network coefficients received from the training means.

19. A system according to claim 18, wherein the collection means includes a plurality of anemometers for determining wind speed and direction, the anemometers being located at several locations.

20. A system according to claim 19, wherein the collection means further includes an electric field sensor for detecting the presence of a charged cloud.

21. A system according to claim 19, wherein the collection means further includes doppler radar for detecting precipitation.

22. A system according to claim 19, wherein the collection means includes an earth-orbiting satellite.

23. A system according to claim 19, wherein the collection means includes means for determining heights of nearby clouds and collecting data regarding the heights.

24. A system according to claim 19, wherein the detection means includes a plurality of direction-finding antennas for determining the location of lightning.

25. A system according to claim 18, wherein the collection means includes means for obtaining data regarding an overall pattern of weather.

26. A method of operating a data processor to enable the data processor to predict the occurrence of lightning in a given area during a day based on weather data collected and lightning occurrences detected during a plurality of days and on weather data collected during the day, the method comprising:

collecting weather data over a plurality of days and storing in a storage area of the data processor information regarding the weather data collected over the plurality of days;

detecting the occurrence of lightning in the given area during the plurality of days, generating data regarding when the lightning occurred and storing information regarding lightning occurrences in a storage area of the data processor;

determining and storing in a storage area of the data processor network coefficients, based on the stored data regarding weather and lightning occurrences on the plurality of days;

collecting weather data during the day and storing in a storage area of the data processor information regarding the weather data collected during the day;

determining the probability of lightning occurring in the given area, based on stored data collected during the day and the network coefficients.

27. A system for predicting the occurrence of lightning within a given area and within a given timeframe based on weather data collected and lightning occurrences detected during a plurality of past days and on weather data collected during the present day, the system comprising:

collection means for collecting weather data within and outside of the given area;

detection means for detecting the occurrence of lightning within the given area and for recording data regarding when the lightning occurred;

training means for receiving data from the collection means and from the detection means and for determining network coefficients, based on data regarding weather and lightning occurrences on past days received from the collection means and the detection means; and

prediction means for receiving data from the collection means and the network coefficients from the training means and for predicting the occurrence of lightning within the given area, based on data received from the collection means and the network coefficients received from the training means.

28. A system for predicting the occurrence of lightning in a plurality of given areas and within a plurality of given timeframes based on weather data collected and lightning occurrences detected during a plurality of past days and on weather data collected during the present day, the system comprising:

collection means for collecting weather data within and outside of a given area;

detection means for detecting the occurrence of lightning within the given area and for recording data regarding when the lightning occurred;

training means for receiving data from the collection means and from the detection means and for determining network coefficients, based on data regarding weather and lightning occurrences on past days received from the collection means and the detection means; and

prediction means for receiving data from the collection means and the network coefficients from the training means and for determining when lightning is likely to occur within the given area, based on data received from the collection means and the network coefficients received from the training means.