Scientists say they have developed a system that uses machine learning to know when and where lightning will strike. Researchers report the system is able to tell that lightning strikes up to 30 minutes before they happen within a 30-kilometer area.
Lightning is a strong burst of electricity in the atmosphere. Since it carries an extremely powerful electrical charge(电荷), it can be destructive and deadly. European researchers have estimated that between 6, 000 and 24, 000 people are killed by lightning worldwide each year. For this reason, climate scientists have long sought to develop methods to predict lightning.
The system tested in the experiments uses a combination of data from weather stations and machine learning methods. The researchers developed a prediction model that was trained to recognize weather conditions that were likely to cause lightning.
The model was created with data collected over a 12-year period from 12 Swiss weather stations in cities and mountain areas. The data, related to four main surface conditions: air pressure, air temperature, relative humidity and wind speed, was placed into a unique machine learning algorithm(算法), which compared it to records of lightning strikes. Researchers say the algorithm was then able to learn the conditions under which lightning happens.
The researchers test-ran the system several times. They found that the system made predictions that proved correct almost 80 percent of the time. "It can now be used anywhere, " the Swiss Federal Institute of Technology said in a statement.
The researchers plan to keep developing the technology in partnership with a European effort that aims to create a lightning protection program. The effort is called the European Laser Lightning Rod project. Scientists working on the project are experimenting with a laser technology that could someday control lightning activity, taking lightning charges from clouds to the ground. They hope that such technology can one day be used as protection against lightning strikes. Possible uses could be at stations, airports or places where large crowds gather.
