Fault diagnosis and online monitoring method of wind turbine main control system based on data-driven

Abstract

The invention discloses a fault diagnosis and online monitoring method for a draught fan master control system based on data driving, and belongs to the technical filed of fault diagnosis. The method comprises the steps of calculating control limits of normal data of the draught fan master control system; calculating statistical magnitude of fault data of the draught fan master control system; determining fault occurrence time T according to the relation between the statistical magnitude and the control limits and determining the variety contribution rate and the fault variety of each master control system in the collected fault data when determining the time T; calculating the signal to noise ratio, the number of principal elements and the statistical magnitude which correspond to the fault variety in real-time data of the draught fan master control system sequentially, wherein if the statistic magnitude is within the control limits, the draught fan master control system is normal in operating process, otherwise, a sensor which corresponds to the fault variety is adopted in a fault sensor. According to the fault diagnosis and online monitoring method for the draught fan master control system based on the data driving, the use of complicated modeling by mechanism and signal analysis of the draught fan is avoided; multiple faults in the same time frame can be monitored, and the fault capable of being found is multidimensional; the biggest signal to noise ratio is adopted to determine the number of the principal elements, and thus the monitoring sensitivity of the faults is improved.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis, and in particular relates to a data-driven fault diagnosis and online monitoring method for a main control system of a fan. Background technique [0002] As a new energy source, wind power has been paid more and more attention by the government. In the cost of wind power enterprises, maintenance costs and repair costs account for a large proportion. Reasonable and effective monitoring of the operation process of the fan and accurate fault diagnosis of the fan are of great significance to the maintenance and operation of the fan. [0003] In recent years, research on wind turbine monitoring and fault diagnosis has emerged in an endless stream: In "Research on the Condition Monitoring and Fault Diagnosis System of Double-fed Asynchronous Wind Turbines [D], Song Lei, 2015", the faults of wind turbines under variable working conditions have been solved Feature extraction, but a large amount...

AI Technical Summary

Problems solved by technology

Most of the existing studies use pattern recognition or analytical model methods, which are cumbersome and complicated to implement.

Images