Vibration state monitoring method and device for variable speed rotating machinery

Abstract

The embodiment of the invention provides a vibration state monitoring method and device for a variable speed rotating machinery. The method comprises steps: during a process of increase to an upper limit rotating speed from a lower limit rotating speed in a normal state of the machinery, a vertical vibration speed, a horizontal vibration speed and a rotating speed of the machinery are acquired synchronously; the vertical vibration speed and the horizontal vibration speed are integrated, a vibration root mean square value of the integrated vibration speed is extracted, and mean and variance ofthe vibration root mean square value are determined; relation curves between the mean and the variance and the rotating speed are fit respectively to acquire corresponding continuous functions; the continuous functions are substituted to a preset decision function to acquire an alarm threshold model; and the vibration root mean square value and the rotating speed of the machinery currently are acquired in real time and are substituted to the alarm threshold model to acquire the current vibration state of the machinery. The embodiment of the invention can improve the variable speed rotating machinery vibration state alarm accuracy.

Description

technical field [0001] The present application relates to the technical field of state monitoring of variable-speed rotating mechanical equipment, in particular to a method and device for monitoring the vibration state of variable-speed rotating mechanical equipment. Background technique [0002] For rotating mechanical equipment (hereinafter referred to as equipment), most of the faults are closely related to mechanical movement or vibration. Therefore, equipment vibration status monitoring is of great significance. [0003] In order to determine the state of the equipment, it is often necessary to compare the collected state value with the alarm threshold. The traditional alarm threshold is usually a static threshold, and the common establishment methods mainly include statistical methods, regression analysis methods, and neural network methods. The most commonly used method in engineering practice is the 3σ method, that is, the mean M and standard deviation σ are calcul...

AI Technical Summary

Problems solved by technology

[0005] However, the inventors of the present application found that the above-mentioned dynamic models are obtained by continuously updating the training sample set and retraining, and the model does not consider the influence of speed changes when solving, and cannot effectively distinguish slow-change faults from speed changes, which affects the alarm. Accuracy

In addition, most of the above dynamic models are established based on a single channel signal. Due to factors such as rotor whirl characteristics, it is difficult to truly reflect the operating status of the equipment.

Images