Heating furnace abnormality cause identification method, heating furnace abnormality cause identification device, machine learning method, and heating furnace abnormality cause identification model

摘要

To provide a furnace abnormality reason identification method, a furnace abnormality reason identification device, a machine learning method and a furnace abnormality reason identification model capable of identifying a reason for abnormality including one which has not caused the abnormality in the past.SOLUTION: A furnace abnormality reason identification method which identifies a reason for abnormality occurring in a furnace comprises: an acquisition step to acquire operation data of the furnace; and an identification step to identify the reason for the abnormality occurring in the furnace by inputting the operation data acquired in the acquisition step into an abnormality reason identification model which is a model having operation data of the furnace as an input variable and the reason for the abnormality of the furnace as an output variable and has undergone machine learning with mock operation data created on the basis of the actual operation data in normal time for identifying the reason for abnormality where actual operation data corresponding thereto has not acquired or the number of pieces of actual operation data corresponding thereto has not been equal to or more than a predetermined number.SELECTED DRAWING: Figure 4