Artificial Intelligence system can predict boiling crisis

The easiest cooking technique boiling can do a lot more than just heating vegetables in the pan. Conversion of liquid into gas prevents hot surfaces from overheating. Hot  surfaces can encounter problems due to boiling crises. Boiling crisis occurs when a vapour film is formed between the heater and the liquid. 

 

During a boiling crisis overheated bubbles cover the liquid surface. The bubbles strongly combine to form a vapour layer. Overheated bubbles increase the chances of disastrous events. It is important to predict the boiling crisis, in advance to decrease the chances of a catastrophic event. 

 

(Must Check: How is Transfer Learning done in Neural Networks and Convolutional Neural Networks?)

 

Researchers at MIT have developed a system that employs artificial intelligence for predicting  boiling crises. Previously the same team of researchers developed a technique that employed machine learning for image processing. In the past researchers were required to manually measure the bubble count and its characteristics. 

 

Advancing the research , Bucci trained a neural network to perform the bubble measurement task. This training cut down the rigorous  three week process to mere five seconds.

 

Objective behind the research was to estimate the boiling crisis. The system considered 17 factors that were given by artificial intelligence nucleation density, infrared radiation are few factors that enabled better radiation understanding. AI tackled the most difficult analysis challenge. 

 

As reported by MIT News, Rishi Raj, an associate professor at the Indian Institute of Technology said, “Boiling has such complicated physics.” It involves at least two phases of matter, and many factors contributing to a chaotic system. “It’s been almost impossible, despite at least 50 years of extensive research on this topic, to develop a predictive model.”