The cases and the future of plant data analysis with AI technology

Abstract

In the fields of oil refining, chemicals, natural resource development, power generation, gas and LNG, etc., “System Invariant Analysis Technology” detects signs of anomalies through the real-time analysis of plant operation data. SIAT is an AI technology designed by NEC to identify the cause-and-effect relationship of large amounts of sensor data. This supports plant owners in preventing occurrences of operation trouble.

JGC and NEC jointly analyzed operation data of a number of plants and found anomalous signs at locations separate from the functional failure of each equipment. These examples show that process engineering knowledge combined with advanced AI such as SIAT works quite effective for reduction of plant downtime.