Wind turbine condition monitoring method
View Patent ↗A wind turbine condition monitoring method includes the steps of generating in advance a reference curve for normal operation of a transmission chain of a wind turbine using position information, measuring position information and generating an actual condition monitoring curve; comparing the actual condition monitoring curve with the reference curve, and determining the current condition of the wind turbine from the comparison. With the proposed monitoring method, monitoring with a high signal-to-noise ratio can be achieved with minimal or no addition of hardware, such as an A/D module; moreover, computational task is greatly reduced, so that the monitoring system can be more easily integrated into the main control system. In addition, the position measurement system is more stable than a vibration measurement system, and the monitoring result is more reliable.
1. A method for monitoring a condition of a transmission chain of a wind turbine, comprising the steps of:
generating normal condition data with sensors located at a first position of the transmission chain and at a second position of the transmission chain of the wind turbine under a known normal operating condition when the transmission chain has no defects, by:
accumulating the generated normal condition data;
storing the accumulated data in a database;
with a data processing tool, performing probability analysis on the accumulated normal condition data, removing curves with relatively large deviations and abnormal points, and generating a long-term monitoring baseline of the transmission chain of the wind turbine after the curves are removed;
generating with the sensors located at the first and second position of the transmission chain an actual condition monitoring curve for the transmission chain of the wind turbine;
with the data processing tool, comparing the actual condition monitoring curve with the long-term monitoring baseline in a frequency domain or a time domain; and
determining an actual condition of the wind turbine from the comparison.
2. The method of claim 1 , wherein the actual condition monitoring curve is a transfer function waveform.
3. The method of claim 1 , wherein the actual condition monitoring curve is a backlash distribution waveform.
4. The method of claim 1 , wherein the transmission chain of the wind turbine includes at least one of a main transmission chain, a variable pitch propeller transmission chain and a yaw transmission chain.
5. The method of claim 1 , wherein the actual condition of the wind turbine includes at least one of a dynamic defect and a static defect.
6. The method of claim 1 , wherein comparing of the actual condition monitoring curve with the long-term monitoring baseline further comprises the steps of:
generating a residual curve after comparing;
monitoring the residual curve; and
sending out a warning if the residual curve exceeds a threshold value.
7. The method of claim 1 , wherein comparing of the actual condition monitoring curve with the long-term monitoring baseline further comprises the steps of:
performing frequency domain analysis on the actual condition monitoring curve;
monitoring a characteristic fault frequency band; and
sending out a warning if an amplitude of the fault frequency band exceeds a threshold value.