Method for managing failure of facility

Abstract

Provided is a method for managing the failure of a facility which has the sufficient-accuracy results of reasoning, can estimate an optimal repair time for a failure or other failures which possibly occur simultaneously with or subsequently to some failure, and can be adopted in large equipment used in a plant. The method for managing the failure of a facility is designed in such a manner that a monitoring means monitors the operational state of a facility, and an estimating means estimates the cause of the failure of the facility by using as a node the measurement data about the operational state of a facility obtained by the monitoring means when the failure occurs in the facility, wherein acoustic data closest to a sound produced from the facility in which the failure occurred is selected from a plurality of elements of acoustic data provided previously, form data closest to the form of the facility in which the failure occurred is selected from a plurality of elements of form data provided previously, the acoustic data and form data thus selected are added to the node, and then the cause of the failure of the facility is estimated by a first Bayesian network based on the node.

Description

technical field [0001] The present invention relates to a method for managing equipment failures, and more particularly to a management method for equipment failures that has sufficient accuracy in inference results and can be used for large-scale equipment such as gas turbines used in plants. Background technique [0002] In gas turbine combined cycle power generation (GCTT) equipment, land-use boiler equipment, wind power generation equipment, etc., when there is a malfunction such as abnormal sound or abnormal alarm of measured value in the equipment forming the plant, it is necessary to estimate its cause, and take countermeasures such as maintenance according to the presumed cause. [0003] Therefore, when a failure occurs in the equipment forming the above-mentioned plant, the operator of the plant deduces the cause of the failure based on past experience or the past performance of the plant, and takes an action that may solve the cause (Prior Art 1). [0004] In add...

AI Technical Summary

Problems solved by technology

[0008] However, in the technology disclosed in Patent Document 1, although the knowledge of an expert with sufficient knowledge and skills about the printer is collected in a Bayesian network, it is difficult to perform a bad situation in the Bayesian network. Quantification of the state of the printer may not be used for information such as sound and shape that are useful for inferring the cause of the failure, and the inference result may not have sufficient accuracy

[0009] In addition, in the case of large-scale equipment used in plant equipment, the cost of stopping the machine once is huge. Therefore, it is necessary to continue the operation for as long as possible. Should the machine be stopped immediately and the failure should be repaired? It is very important to operate on weekends other than weekdays, for example, and whether the operation after the limited operating conditions can be continued until regular maintenance is very important. Although patent document 1 discloses specific actions for adverse conditions, since it is related to small The equipment is the technology of the printer, therefore, information about the stop times important for the above-mentioned large machines is not proposed

[0010] In addition, although the specific action against the above-mentioned defect is disclosed, it does not describe the deterioration symptoms of some defects that are presumed to occur at the same time as the defect that has been taken or may occur in the future, and will be confirmed at the next regular inspection. Methods of confirming these deterioration states

[0011] From the above, it can be seen that it is very difficult to apply the technology disclosed in Patent Document 1, which is the technology of automatic diagnosis of printers, to large-scale equipment such as equipment used in plants.

Images