This summer, Matthew Russell, Evan King, Chad Parrish, and electrical and computer engineering assistant professor with a joint appointment in mechanical engineering Peng Wang were recognized by the North American Manufacturing Research Institution and SME Scientific Committee with the Outstanding Paper Award in the Manufacturing Systems Track at the 49th North American Manufacturing Research Conference (NAMRC 49). Their paper was titled, “Stochastic modeling for tracking and prediction of gradual and transient battery performance degradation.” The honor was awarded to the top paper related to manufacturing systems presented during the virtual conference hosted by the University of Cincinnati from June 21-25 earlier this year. In addition, the paper was one of three submissions fast-tracked for publication in the Journal of Manufacturing Systems, a leading venue for operations and manufacturing research.

Wang leads the Augmented Intelligence for Smart Manufacturing (AISM) lab at UK, a rapidly growing research group that develops AI techniques for improving data-driven industrial manufacturing. Russell is a current Ph.D. student in the lab, and in collaboration with King and Parrish, recent master’s program graduates in ECE, the group proposed a novel technique for modeling gradual and transient capacity degradation in lithium-ion batteries, a phenomenon that causes the cells to store less energy as they get older. Simple tracking is insufficient for making realistic predictions of battery degradation because Li-ion cells often experience capacity regeneration events after prolonged rest periods, creating discontinuous jumps in the historical trends. The team from UK introduced both a new model of gradual degradation which outperformed existing approaches and investigated a method of modeling the regeneration events using a mathematical technique known as a Compound Poisson Process (CPP). The models were fit to historical data collected by the Prognostics Center of Excellence at NASA Ames Research Center by using particle filtering and Markov-chain Monte Carlo (MCMC), two methods for estimating parameters of a stochastic, or random, process. Accurate tracking and prediction of Li-ion battery capacity degradation are important for maintaining the health of batteries used in electric vehicle fleets used by distributors and automated guided vehicles (AGV) that can improve logistics on the factory floor. Up-to-date information about battery health informs predictive maintenance activities that can reduce downtime-related costs and improve efficiency.

First held in 1973, NAMRC remains an influential conference within the field of manufacturing research after nearly fifty years of existence, and Wang and the AISM lab look forward to future opportunities to showcase similar applications of machine learning for smart manufacturing at NAMRC in the coming years. 

This summer, Matthew Russell, Evan King, Chad Parrish, and electrical and computer engineering assistant professor with a joint appointment in mechanical engineering Peng Wang were recognized by the North American Manufacturing Research Institution and SME Scientific Committee with the Outstanding Paper Award in the Manufacturing Systems Track at the 49th North American Manufacturing Research Conference (NAMRC 49). Their paper was titled, “Stochastic modeling for tracking and prediction of gradual and transient battery performance degradation.” The honor was awarded to the top paper related to manufacturing systems presented during the virtual conference hosted by the University of Cincinnati from June 21-25 earlier this year. In addition, the paper was one of three submissions fast-tracked for publication in the Journal of Manufacturing Systems, a leading venue for operations and manufacturing research.

Wang leads the Augmented Intelligence for Smart Manufacturing (AISM) lab at UK, a rapidly growing research group that develops AI techniques for improving data-driven industrial manufacturing. Russell is a current Ph.D. student in the lab, and in collaboration with King and Parrish, recent master’s program graduates in ECE, the group proposed a novel technique for modeling gradual and transient capacity degradation in lithium-ion batteries, a phenomenon that causes the cells to store less energy as they get older. Simple tracking is insufficient for making realistic predictions of battery degradation because Li-ion cells often experience capacity regeneration events after prolonged rest periods, creating discontinuous jumps in the historical trends. The team from UK introduced both a new model of gradual degradation which outperformed existing approaches and investigated a method of modeling the regeneration events using a mathematical technique known as a Compound Poisson Process (CPP). The models were fit to historical data collected by the Prognostics Center of Excellence at NASA Ames Research Center by using particle filtering and Markov-chain Monte Carlo (MCMC), two methods for estimating parameters of a stochastic, or random, process. Accurate tracking and prediction of Li-ion battery capacity degradation are important for maintaining the health of batteries used in electric vehicle fleets used by distributors and automated guided vehicles (AGV) that can improve logistics on the factory floor. Up-to-date information about battery health informs predictive maintenance activities that can reduce downtime-related costs and improve efficiency.

First held in 1973, NAMRC remains an influential conference within the field of manufacturing research after nearly fifty years of existence, and Wang and the AISM lab look forward to future opportunities to showcase similar applications of machine learning for smart manufacturing at NAMRC in the coming years.