Researchers at the Indian Institute of Technology, Madras, have used artificial intelligence (AI) tools to study fuel production from biomass.
AI, computer simulation and modelling were used to reduce time and cost to understand the concept. It would provide quicker insights into building processes and plants for biomass, the researchers said.
The research has explored the use of biomass as a source of energy-dense fuel, using unconventional methods. They are finding methods to extract fuel from biomass like wood, grass and even waste organic matter.
According to them, such biomass-derived fuel is relevant to India, where currently about 750 million metric tonnes of biomass is available annually. The country can attain self-sufficiency by extracting fuel, they say.
Himanshu Goyal, assistant professor and Niket S. Kaisare, a professor in the Chemical Engineering department, led the team. Krishna Gopal Sharma, a fourth year B-Tech student and a Young Research Fellow of the Institute from the Department of Computer Science and Engineering, was also part of the team that published its research results in Reaction Chemistry and Engineering, a peer-reviewed journal of the Royal Society of Chemistry.
Across the globe, scientists are developing models to understand conversion of biomass into fuels and chemicals but most models take time to become operational. AI and machine learning could hasten the modelling process. It calls for training engineers on high-performance computing and machine learning skills to address the challenges of developing zero emission technologies, Mr. Goyal said.
Mr. Kaisare said the novelty of the machine learning approach was its ability to predict the composition of biofuel produced as a function of time the biomass spends in the reactor. “We used a statistical reactor for accurate data generation, which allows the model to be applied over a wide range of operating conditions,” he explained.
