使用人工智能來預(yù)測地震

毛新紅

在科技迅猛發(fā)展的今天，人工智能技術(shù)被廣泛應(yīng)用在各個領(lǐng)域。越來越多的科學(xué)家表示，在人工智能技術(shù)的幫助下，他們分析地震數(shù)據(jù)的方式發(fā)生了變化，這可以幫助他們更好地了解地震并提供更快速、更準(zhǔn)確的預(yù)警。

An earthquake early warning system that uses artificial intelligence（AI） to predict how the ground will move during an earthquake can give several seconds advance notice that the shaking is coming.

A similar system that uses more traditional computing power is called ShakeAlert， and it works by detecting the first waves of earthquake motion and then calculating when the set of waves that cause most of the shaking will arrive.

The new system in development is called DeepShake， which is intended to provide a few seconds advance warning once an earthquake has started. However， DeepShake uses a deep neural network， a type of AI learning， to identify patterns from past earthquakes in order to predict how the shaking from a new quake will travel. This could lead to faster processing across different regions.

“When we set out on this project， our main goal was to beat the ground motion prediction equations that are currently used to program ShakeAlert system，” said Avoy Datta， who developed DeepShake. “They tend to be very slow because they need numerical solvers， running on supercomputers， and they can take minutes and even hours to process.”

Despite the fact that DeepShake is given no information about the earthquakes location or type， it is able to give warnings between 3 and 13 seconds before the earthquake happens. This is similar to the amount of advance notice with ShakeAlert. Researchers view this system as a competitor to make progress because DeepShake technology can be used to make up for ShakeAlert. The researchers hope to expand the testing to other faults.

Ground shaking at any given spot can be tricky to predict. For example， ShakeAlert once failed to send out warnings during the largest quakes in the Ridgecrest earthquake sequence in 2019 because the shaking was expected to not reach the programs threshold of “l(fā)ight shaking” in some areas that did indeed experience light shaking. The advantage of deeplearning networks， is that they can automatically collect information， because they are based on past experiences of shaking in that location. Unlike ShakeAlert， which uses more universal equations with assumptions built in， DeepShake will have to be retrained in each individual region where it is used. This training， however， will catch patterns that traditional equations may not.

1. How does ShakeAlert predict earthquakes according to the text？

A. By figuring out where most waves are.

B. By finding out most of the large waves.

C. By discovering when the waves will arrive.

D. By detecting the first waves of earthquakes.

2. Whats the aim of DeepShake？

A. To improve ShakeAlert.

B. To improve the predicting speed.

C. To prevent the terrible earthquakes.

D. To change the traditional methods.

3. What does the underlined word “tricky” in the last paragraph mean？

A. Interesting. B. Difficult. C. Impossible. D. Important.

4. Which section of the newspaper is the text probably taken from？

A. Science. B. Society. C. Environment. D. Geography.

Ⅰ. Difficult sentence in the text

Despite the fact that DeepShake is given no information about the earthquakes location or type， it is able to give warnings between 3 and 13 seconds before the earthquake happens.盡管DeepShake沒有得到地震位置或類型的信息，但它能夠在地震發(fā)生前3到13秒發(fā)出警告。

【點石成金】在本句中，DeepShake is given no information about the earthquakes location or type為that引導(dǎo)的同位語從句，解釋說明fact的內(nèi)容。

Ⅱ. Textcentered chunks

be intended to 打算……

identify patterns 識別模式

lead to 導(dǎo)致

set out 開始工作;展開任務(wù)

tend to 往往;趨向

be similar to 和……類似

view...as 把……視為

send out warnings 發(fā)出警告