Scientists at Derzhavin University have developed a method for predicting sudden failure using weak acoustic and electromagnetic signals emitted by an aluminum alloy. The results can be used in developing vehicle safety systems made from high-tech alloys.
According to the head of the research team, Doctor of Physical and Mathematical Sciences, Professor Alexander Shibkov, aluminum alloys with magnesium, lithium, copper, or zinc are valued for their high specific strength and corrosion resistance. They help reduce the weight of aircraft, rockets, and ships, saving fuel. However, during processing and use, so-called localized deformation bands spontaneously appear in such materials. These bands accumulate damage unnoticed and can lead to sudden failure of the structure under load. Detecting them in operating equipment using traditional methods is extremely difficult. "We've developed a unique set of sensitive methods for detecting localized deformation bands directly during mechanical testing," says Alexander Shibkov. "The system includes recording the metal's own noise—acoustic, electromagnetic, and electrochemical—generated by active localized deformation bands, which are recorded synchronously with surface video recording at up to 100,000 frames per second. This allowed us to literally 'see' the moment danger zones emerge."
Derzhavin scientists note that video data processed using an original computer program makes it possible to detect these bands even under challenging conditions, such as in an aggressive chemical environment or on a rough surface where visual inspection is difficult.
"We've established that before macroscopic failure, emission signals exhibit critical, power-law statistics, which can serve as a harbinger of catastrophic failure of the structural material," emphasized Professor Shibkov. "This is a kind of metal 'cry for help,' indicating that the load has become critical, and failure is inevitable unless intervention is made. We now know which patterns in the alloy's 'noise' to look for in order to predict the risk level and time of material failure.
In the future, the 'emission reconnaissance' method for damage could be integrated into onboard diagnostic systems for automobiles, aircraft, and underwater vehicles, including unmanned ones, where higher overloads are permissible, enabling a transition from scheduled repairs to predictive maintenance based on the actual condition of the material."
