Wind turbine generator fault early warning method based on graph neural network

Abstract

The invention belongs to the field of wind turbine generator state monitoring and fault early warning, and relates to a wind turbine generator fault early warning method based on a graph neural network, and the method comprises the steps: S1, obtaining a multivariable time sequence, and carrying out the data preprocessing; s2, the influence of the working condition change on the temperature variable is decoupled, and decoupled temperature sensor data is obtained; s3, inputting the decoupled health data into a space-time diagram network, and advancing space-time correlation features; s4, setting a threshold value according to the verification set; s5, inputting online data into the model, calculating an abnormal score, and judging whether fault early warning occurs or not according to a threshold value; according to the method, the influence of the working condition change on the temperature state parameters is decoupled, and the dynamic spatio-temporal characteristics among different temperature sensor parameters are effectively extracted by using the graph neural network, so that the fault early warning accuracy and reliability are improved.

Description

technical field [0001] The invention belongs to the field of state monitoring and fault early warning of wind turbines, and relates to a graph neural network-based fault early warning method for wind turbines. Background technique [0002] In recent years, with the increase in the number of wind turbine installations, due to its complex operating conditions and harsh working environment, wind turbines and their key components are extremely prone to failure, and even cause the unit to shut down in severe cases. This often results in huge economic losses and bad social impact. Therefore, it is of great significance to realize early warning of unit failure. Early detection of potential faults based on fault development trends can formulate optimal maintenance strategies, reduce fault rates, reduce operation and maintenance costs, and prevent major faults through fault warnings, thereby avoiding major property losses and ensuring personal and equipment safety. [0003] At pres...

AI Technical Summary

Problems solved by technology

However, the wind turbine is a strongly coupled system with multiple subsystems working together. The information content of a single monitoring parameter is low, and it is difficult to fully reflect the abnormal state of the system.

Moreover, the existing technology cannot effectively model and mine the dynamic space-time relationship between different SCADA sensor parameters, resulting in low accuracy and reliability of fault warning

Images