Data mining based wind turbine generator system fault diagnosis method

Abstract

The invention provides a data mining based wind turbine generator system fault diagnosis method. The method comprises the following steps of: S1, removing data noises and processing missing values of wind turbine generator system parameters acquired in real time; S2, carrying out amplitude domain analysis; S3, carrying out correlation analysis; S4, building a fault decision tree for data samples with different fault types; S5, adopting a post-pruning method for the built fault decision tree to remove the noises in the data and training abnormity caused by outliers; S6, extracting the knowledge shown by the post-pruned fault decision tree to give the fault decision rules in the form of IF-THEN; and S7, repeating the steps S1, S2 and S3, carrying out fault diagnosis on the processed data by utilizing the fault decision rules extracted in the step S6 and displaying the diagnosis results. The method has the following beneficial effects: the fault parts of the wind turbine generator system can be timely and effectively determined; the fault types and reasons can be determined; and major accidents caused by fault deterioration can be prevented.

Description

technical field [0001] The invention relates to a fault diagnosis method of a wind power generator set, in particular to a data mining-based fault diagnosis method of a wind power generator set. Background technique [0002] The increasingly severe world energy crisis has made the use of renewable energy, represented by wind power generation, greatly developed. While the total installed capacity of wind power in the world is increasing rapidly, the faults of wind turbines are also emerging one after another. The faults of wind turbines mainly include wind rotor faults, transmission system faults, generator faults, and pitch system faults. Wind rotor faults include wind rotor mass imbalance faults, aerodynamic imbalance faults, blade icing faults, blade crack damage, etc. Transmission system faults include main shaft misalignment faults, main bearing damage faults, coupling damage faults, and gear tooth damage , gear bearing damage, etc. Generator faults include generator b...

AI Technical Summary

Problems solved by technology

There are many types of wind turbine faults, and due to its complex nonlinearity, when a fault occurs, it is difficult to determine the location of the fault and determine the cause of the fault

At present, the fault handling of wind turbines mostly focuses on the control method of wind turbines when the power grid fails. [1] , Wind turbine blade fault diagnosis [2] , Wind turbine inverter fault diagnosis [3] Fault diagnosis of specific components such as wind turbines, unable to realize fault location and fault identification of the whole wind turbine

Images