Scientists say they've developed a new system(系统). It can use "machine learning" to predict when and where lightning(闪电)will strike(闪击). The research was led by engineers from Switzerland.
European researchers have estimated(估算) between 6,000 and 24,000 people are killed by lightning worldwide each year. The strikes can also cause power cuts, destroy houses and start forest fires. For these reasons, some climate scientists have studied to develop methods to predict and control lightning. But no system has been created to effectively predict lightning.
The new system is being tested in the experiments(实验). It used a combination of data(数据资料) from weather stations and machine learning methods. The researchers developed a prediction model. It was trained to tell weather conditions that were possible to cause lightning.
The model was created with some data. The data was collected over a 12-year period from 12 Swiss weather stations in cities and mountain areas. The data related to(与……有关)four main surface conditions: air pressure, air temperature, relative humidity(相对湿度)and wind speed.
The atmospheric(大气的) data was put into a machine learning algorithm(计算程序), which compared it to records of lightning strikes. Researchers say the algorithm was then able to learn the conditions under which lightning happens.
Amirhossein Mostajabi is a PhD student who led the development of the new system. He said, "Common systems for collecting such data are slow and complex, and they require expensive collection equipment(设备)like radar or satellites (卫星)."
"Our system uses data that can be gotten from any weather station," Mostajabi said. "This will improve data collection in the areas which are not covered by radar and satellites or in places where communication systems have been cut," he added.