August 13 news, with the continuous development of technology, in recent years, the concept of “intelligent fire”. “Intelligent fire” aims to use the Internet of Things, artificial intelligence, virtual reality, mobile Internet + and other latest technologies to minimize the risk and impact of fire.
Recently, the Hong Kong Polytechnic University and foreign research institutions published a new study on the use of artificial intelligence to accomplish firefighting tasks. The study describes that the most frightening thing in a fire is not a roaring blaze, but a flash fire that is difficult to detect in advance.
Flash fire is a deadly fire phenomenon in which almost anything combustible in the environment suddenly catches fire, making it impossible for firefighters on fire-fighting missions to escape. Although firefighters try to identify flash fires through black smoke and high temperatures, flash fires are still difficult to predict.
For this reason, the research group has developed a predictive neural model called FlashNet. In over 41,000 digital fire simulations, the model predicted flash fires 30 seconds in advance and with 92.1% accuracy. Thus, with this artificial intelligence technology, firefighters will be able to predict flash fires in advance. This will effectively improve the safety of firefighters when fighting fires, and may be able to save firefighters’ lives at critical times.
However, the researchers say there is still a long way to go before this neural model can be put into practical application. The next step is for the researchers to conduct real-world tests and put it to practical use as soon as possible.
The study was published in the journal “Engineering Applications of Artificial Intelligence” and can be viewed here.
