Fault diagnosis method and device for blower

Abstract

The invention relates to a fault diagnosis method and device for a blower, which are used for real-time fault diagnosis of the blower. The method includes the following steps: collecting blower operating data under normal conditions, using the median average filtering method to process the existing data and forming a matrix, constructing an initial model according to the recursive dynamic principal component analysis method, and obtaining statistical control limits. Collect the real-time operation data of the blower for monitoring, judge whether the blower is faulty according to whether the obtained data statistics exceed the statistical control limit, and judge the cause of the fault according to the contribution rate of each statistical quantity. It solves the problem that the blower operation interference noise and the large amount of data in the traditional method lead to the inaccurate diagnosis of the blower system.

Description

technical field [0001] The invention relates to the technical field of fault detection and diagnosis, and in particular designs a fault diagnosis method and device for a blower, based on sensor data information under normal conditions, and judging whether the blower is faulty or not based on the real-time sensor data of the blower. Background technique [0002] The blower operates eccentrically by the offset rotor in the cylinder, and makes the volume change between the blades in the rotor slot suck, compress and spit out the air. It is widely used in sewage treatment, smelting blast furnace, coal washing plant, mine flotation, chemical manufacturing Gas, vacuum and other fields. During the operation of the equipment, the failure of the blower will cause serious consequences such as gas transmission errors, blower burnout, etc. Therefore, the blower needs a good operating state and accurate fault location performance when a fault occurs. Existing fault diagnosis methods nee...

AI Technical Summary

Problems solved by technology

[0007] The technology of the present invention solves the problem: Aiming at the problems that a large amount of noise is generated during the operation of the existing blower system, and the amount of information and data collected by the traditional fault diagnosis method is relatively large, a fault diagnosis method and device for the blower are provided. The blower sensor data is filtered to eliminate the influence of measurement noise in the actual operation process, which can effectively avoid false alarms caused by high-frequency noise, and solve the problem of blower operation interference noise and the large amount of data caused by the traditional method. Accurately Diagnosed Problems

Images