ID
× COMMENTARYCOVER STORYIN THE NEWSNEWSFRONTSCHEMENTATOR + Show More BUSINESS NEWSTECHNICAL & PRACTICALFACTS AT YOUR FINGERTIPSTECHNOLOGY PROFILESOLIDS PROCESSINGENGINEERING PRACTICEEQUIPMENT & SERVICESFOCUSNEW PRODUCTS + Show More

Comment Automation & Control

‘Deep-learning-based’ AI quickly predicts furnace products

By Tetsuo Satoh |

NTT Communications Corp. (NTT Com; www.ntt.com) and Mitsui Chemicals Inc. (both Tokyo, Japan; www.mitsuichem.com) have demonstrated the ability to predict, with a high precision, the concentration of gaseous products generated in a furnace. The predictions are produced in just 20 minutes after sampling process data, by modeling the relationships between process data and raw material, and furnace conditions, using deep-learning-based artificial intelligence (AI) developed by NTT Com. Over 50 process parameters are used in the modeling, including temperatures and flowrates of raw materials and furnace-reactor parameters. The two companies have been working together since 2015. Mitsui Chemicals plans to enhance the reliability and efficiency of its production facilities by studying next-generation production technologies using the internet of things (IoT), big data and AI.

Meanwhile, NTT Com plans to use AI-based technology to enhance the efficiency of chemical plant operations, by quickly clarifying the causes of quality abnormalities. In a project supported by the New Energy and Industrial Technology Development Organization (NEDO) under the authority of the Ministry of Economy, Trade and Industry (METI), NTT Com, Yokogawa Electric Corp. and Yokogawa Solution Service Corp. have started demonstration testing of a high-level Energy Management System (EMS), which aims to optimize production facilities in the chemical and pulp-and-paper industries. NTT Com will develop the analysis support system that handles process data using an IoT/AI deep-learning system. Yokogawa will develop ways to optimize the inter-production process, using plant big data. Yokogawa Solutions will develop highly sophisticated production-control technology for mass-production factories.

Related Content
A New Mentality in Process Modeling
Predictive process models enable more efficient operations and higher-value products When they hear the words “process modeling,” some may envision…
Going digital
In these times of fast-changing automation, more and more companies are talking about “going digital.” Converting to digital data is…

Chemical Engineering publishes FREE eletters that bring our original content to our readers in an easily accessible email format about once a week.
Subscribe Now
Purdue University Saves $400,000 Annually with Local Vacuum Networks
Bag filter Housings/Vessels
Innovative Backwashable Media Filter
Automated Vertical Tower Filter Press
Visit VACUUBRAND at ACS in San Francisco!

View More

Live chat by BoldChat