Quality prediction of chemical products by sensing the environment of Chemical Plant

Deep learning model successfully predicts quality of chemical products in the production process

NTT and Mitsui Chemicals have successfully created a technique that accurately and speedily predicts the quality of gas products and enables a safer and more stable production process.
Environmental data is analyzed using deep learning algorithms, an artificial intelligence (AI) methodology that automatically processes relevant factors in order to model and predict outcomes.
By estimating changes in concentration of the gas based on plant equipment, temperature and other environmental data, we succeeded in predicting gas quality 20 minutes into the future, keeping the discrepancies within a ±3% margin.
We are continuing this research into ways of improving the operational efficiency and safety of chemical plants.

Back to Zone