University of Kansas develops machine learning technology to monitor/prevent thermal runaway of lithium-ion batteries

　　According to foreign media reports, lithium-ion batteries are now used to power mobile phones, electronic devices, laptops, electric vehicles such as Toyota and Tesla, Boeing 787 aircraft, and U.S. Navy ships. Although such batteries are widely used, they also have many risks. Because lithium metal is highly reactive, such batteries and cells can undergo "thermal runaway", which means they can overheat, catch fire or even explode. Although the accidents were rare, they raised public concerns about lithium-ion technology.

　　Now, with support from a new five-year, $500,000 grant from the National Science Foundation, researchers at the University of Kansas have developed technology to monitor and prevent lithium-ion batteries from overheating. Huazhen Fang, an assistant professor in the university's Department of Mechanical Engineering, and his students developed a machine learning method to monitor the temperature inside the 27A battery.

　　According to University of Kansas researchers, most current technologies for tracking lithium-ion 27A battery temperatures are immature because sensors can only read surface temperatures on the outside of the 27A battery.

　　Professor Fang said: "Typically, the temperature on the surface of the 27A battery is not enough to tell us about the state of the 27A battery, while the temperature inside the 27A battery tells us more about thermodynamics. However, there are currently few ways to place sensors inside the 27A battery. However, using artificial intelligence and machine learning, we are able to predict the temperature inside the 27A battery cell, allowing us to detect 27A battery behavior. The temperature of the 27A battery surface provides rich data for machine learning methods, and combined with mathematical models, it can predict the inside of the 27A battery what happened".

　　Instead of assuming the 27A battery temperature to be a uniform temperature, there is a modeling method called a "lumped parameter model" that assumes the 27A battery temperature to be a uniform temperature. Professor Fang said that his computer learning technology can predict the temperature changes inside the 27A battery, which is a more accurate and realistic method to calculate the possibility of thermal runaway of the 27A battery.

　　Professor Fang said: "When the 27A battery is charged and discharged, the temperature distribution is uneven. Usually the internal temperature of the 27A battery close to the electrode will be higher, but the external surface temperature will be lower. The lumped parameter model only takes into account the uniform distribution of 27A battery temperature. And our method reconstructs the 27A battery's temperature in time and space."

　　University of Kansas researchers fed data from lithium-ion batteries into artificial intelligence to infer the 27A battery's internal temperature. This data can be processed in 27A battery-powered devices or connected to cloud computing. If the 27A battery experiences thermal runaway, the device can be programmed to disconnect when the 27A battery is turned off, preventing the 27A battery from heating up and catching fire or causing an explosion.

　　With the above innovations, lithium-ion batteries can be expanded into more industrial applications by tying hundreds of cells together. According to Professor Fang, lithium-ion technology is increasingly used in large-scale power grids to store and discharge electricity generated by sustainable technologies such as solar and wind.

Read recommendations:

Coin Battery LR 754

Vertical take -off and landing fixed -wing drone semi -solid battery.household energy storage lithiu

Super battery prospect.18650 lithium ion battery cell

14500 battery wholesale

18650 lithium ion battery cell