PI: Carlos Romero
Co-PI(s): Zheng Yao
University: Lehigh University
Industry partner: N/A
This project will develop artificial intelligence (AI) algorithms for system data processing from real-time laser induced breakdown spectroscopy (LIBS) and Raman spectroscopy. Municipal solid waste (MSW) is a very heterogeneous material, with large variability in its physical, chemical, and biological characteristics. This poses significant challenges in utilizing it as a feedstock for producing biofuels and bioproducts. Conventional MSW laboratory analyses lack the time resolution required to effect efficient feedstock preprocessing or biofuel production process control. This project is for the development of artificial intelligence algorithms for system data processing. These algorithms will be used in the future for rapid detection and analysis of MSW streams. The rapid detection and analysis system will be based on laser induced breakdown spectroscopy and Raman spectroscopy, operating simultaneously, to provide real-time, in-situ spectra for further analysis by AI. AI algorithms will be developed for processing of the combined LIBS/Raman spectra, and validated with standardized methods. The uniqueness of the project relies in that machine learning and unsupervised AI algorithms will be used for concurrent optical data processing and correlation to produce statistically accurate monitoring of elemental, molecular and chemical content of elements/species of interest in MSW feedstock, as well as higher order variables, such as those included in MSW proximate and ultimate analyses. This will represent a leap in developing on-line measurement technology for application in MSW conversion processes, and an improvement in MSW characterization throughput over baseline off-line methods