A Method of Equipment Fault Prediction Based on Recognition of Correlation Between Factors and Events

Abstract

The invention discloses an equipment fault prediction method based on factor-event correlation recognition. The method includes: acquiring normal running historical information and fault historical information of equipment, and storing running information and ambient information related to an equipment fault moment in a data sheet to form an equipment fault data set; forming a normal equipment running data set through the running information and ambient information when the equipment normally runs; statistically analyzing equipment fault parameters online, recognizing an equipment running abnormality indicator to quickly recognize a grid equipment fault hazard, and giving an early warning to provide scheduling and shipping inspectors with decision basis. The method has the advantages that fault criterions can be provided, fault factors and a fault indicator range can be selected, a clear warning indicator is provided, and a scientific method of warning is provided.

Description

technical field [0001] The invention relates to the electrical industry, in particular to an equipment failure prediction method based on identification of correlations between factors and events. Background technique [0002] With the rapid development of power grid construction, my country is at the forefront of the world in terms of power transmission capacity, equipment and technical level, and has put forward higher requirements for the performance and operation reliability of power transmission and transformation equipment, so fault warning is required Fundamentally avoid failures. However, the existing equipment failure early warning technology is based on knowledge base expert system modeling and artificial neural network modeling. Although the former has inherited the existing operating experience, for some complex systems, especially the source of knowledge is not enough to express and reflect the accident. Although the latter has better self-learning ability, it i...

AI Technical Summary

Problems solved by technology

However, the existing equipment failure early warning technology is based on knowledge base expert system modeling and artificial neural network modeling. Although the former has inherited the existing operating experience, for some complex systems, especially the source of knowledge is not enough to express and reflect the accident. characteristics; although the latter has better self-learning ability, model maintenance is very difficult, modeling requires a time-consuming learning process, and the selection of learning samples is also lacking in basis

At the same time, the traditional method may cause misjudgment or missed judgment due to errors in the model itself or errors in modeling, or the model itself is not general.

Therefore, the traditional early warning method can no longer solve the early warning problems faced by the current equipment. A new method is needed to identify the abnormal signs of the equipment and give early warning to help the power grid to take measures in advance to avoid the occurrence of failures. Planned Downtime is Planned Downtime

Images