Extranet Systems Inc.
● With "correlation extraction method" as the core technology, we upgraded defect detection in the production line. Furthermore, by applying "causal analysis", we improved estimation accuracy of defects or abnormal factors, and started supporting creation of a mechanism that integrates operation and maintenance. Lubrication failure and misalignment of rotating machine are monitored by vibration data at existing plant., and "proactive maintenance (cause removal type)" is carried out easily.
■ We received the TPM Excellence Product Award (Development Award) from the Japan Institute of Plant Maintenance in 2014 !
■ Received Chairman's Award of the Japan Society for the Advancement of Inventions at the 45th Invention Award in 2019 (March 6, 2020)
● Main results of this service Total number of units as of the end of December 2023: 1470 units
●Developed and started sales of automatic estimation app for deterioration factors of rotating machine (from November 2023)
● Completed verification of fuel conversion plant monitoring function (detects signs of abnormality earlier than operators)
■“Condition monitoring technology based on correlation extraction method”; Peer-reviewed and published in the February 2024 issue of the Journal of the Japan Society of Facility Management
●Selected as a NEDO Challenge Candidate for Manufacturing DX Prize
●Selected as a Member of the Ministry of Land, Infrastructure, Transport and Tourism’s Infrastructure Facility Management AI Council
Major feature and effectiveness is to realize proactive maintenance using the vibration data by the degradation diagnosis service based on "correlation extraction method".
On the basis of the state-based maintenance,proactive maintenance comprises two elements, cause eleminating maintenance and predictive maintenance. Especially, by monitoring insufficient lubrication and axis misalignment, it becomes possible to early detect deteriorationof bearing parts,etc. and to take measures/repair at an appropriate time, which is to realize "Life extension of a machine with no trouble" so as to complete its original designed life. Furthermore, we can constructa causal model for an extracted factor group and estimate the cause of occurrence probability in combination with Bayesian inference.
5 patents registered .
(1) Method and apparatus for evaluating object of diagnosis : Patent No. 4049331
(2) Method and apparatus for totally diagnosing plant device : Patent No. 5753301
(3) Cause searching method for state change of object : Patent No. 6154523
(4) Integrated monitoring method of quality and equipment in production line : Patent application No. 6680430
(5) Method for building a circular/low-carbon production system in manufacturing industry: Patent No. 7054487