Factibilidad de la determinación de características del daño en rodamientos a partir de vibraciones.
Loading...
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
Los rodamientos son componentes primordiales dentro de las máquinas, ya que soportan y permiten el movimiento de elementos rotatorios u oscilantes. Las fallas en estos son una de las razones principales por las cuales varias máquinas rotativas tienden a deteriorarse, convirtiéndolos en objeto frecuente de mantención.
Muchos daños en rodamientos son provocados por desgaste o fatiga superficial, y a su vez cuando estos tipos de daños están presentes, se producen fuerzas del tipo impulsiva. Este tipo de fuerza presenta una corta duración, por lo que se pueden ver afectadas por interferencias provenientes de otras vibraciones, tales como el ruido o las vibraciones originadas por fuentes externas, generalmente de elementos de máquinas cercanos, por lo que es esencial preprocesar las señales correctamente para obtener resultados precisos.
En esta investigación se tiene por objetivo seleccionar e implementar técnicas para conocer las características de un daño en un rodamiento, particularmente el tamaño de un daño en la pista interior y exterior de un rodamiento, provocada por fatiga, debido a lo común que son estos daños. La metodología de este proyecto implica una revisión exhaustiva de la literatura para identificar técnicas prometedoras que permita estimar el tamaño de estos daños. A partir de estas técnicas, en esta investigación se propone el nuevo método DEW que combina la transformada de Wavelet con el análisis de la envolvente de las señales (Wavelet Denoising – Envolvente – WPD específicamente), para estimar el tamaño del daño puntual.
Para validar las técnicas empleadas, se utilizan señales simuladas, además de dos conjuntos de mediciones obtenidas de la bibliografía. Estos conjuntos de datos contienen señales correspondientes a rodamientos operando a diversas velocidades, cargas, tamaños de daño, entre otros.
Con los resultados obtenidos, fue posible estimar un intervalo de confianza para el tamaño del daño en las distintas condiciones planteadas. A pesar de algunas inconsistencias observadas, el método mostró resultados más precisos con cargas altas, tamaños de fallas mayores o velocidades bajas. En el caso de señales simuladas, se lograron buenos resultados incluso en escenarios con hasta 10 [dB] de relación señal-ruido. Sin embargo, para la evaluación con el conjunto de datos experimentales, esta estimación resultó insuficiente debido a la cantidad de ruido presente, que atenuó los componentes de interés, especialmente el pico de entrada.
Bearings are essential components within machines, due to the crucial role they play in supporting and enabling the motion of rotating or oscillating elements. Currently, despite their importance, there is no reliable strategy for predicting their failures. Damage to bearings is primarily caused by wear or surface fatigue, and when these types of damage are present, impact forces occur. This type of force has a short duration, so it can be affected by interferences from other vibrations, such as noise or vibrations originating from external sources, typically nearby machine components. Therefore, it is essential to preprocess the signals correctly to obtain accurate results. The objective of this research is to select and implement techniques to determine the characteristics of damage in a bearing, particularly the size of damage to the inner and outer raceways of a bearing, caused by fatigue, due to how common these types of damage are. The methodology of this project involves a comprehensive review of the literature to identify promising techniques for measuring the size of this damage. From these techniques, in this research, the new DEW method is proposed that combines the Wavelet transform with signal envelope analysis (specifically Wavelet Denoising – Envelope Analysis– WPD) to estimate the size of the localized damage. To validate the employed techniques, simulated signals are used, in addition to two sets of measurements obtained from the literature. These datasets contain signals corresponding to bearings operating at various speeds, loads, damage sizes, among others. With the results obtained, it was possible to estimate a confidence interval for the size of the damage under different conditions. Despite some observed inconsistencies, the method showed more accurate results with increased load, fault size, or decreased speed. In the case of simulated signals, good results were achieved even in scenarios with up to 10 [dB] of signal-to-noise ratio. However, for the evaluation with the experimental dataset, this estimation proved insufficient due to the amount of noise present, which attenuated the components of interest, especially the input peak.
Bearings are essential components within machines, due to the crucial role they play in supporting and enabling the motion of rotating or oscillating elements. Currently, despite their importance, there is no reliable strategy for predicting their failures. Damage to bearings is primarily caused by wear or surface fatigue, and when these types of damage are present, impact forces occur. This type of force has a short duration, so it can be affected by interferences from other vibrations, such as noise or vibrations originating from external sources, typically nearby machine components. Therefore, it is essential to preprocess the signals correctly to obtain accurate results. The objective of this research is to select and implement techniques to determine the characteristics of damage in a bearing, particularly the size of damage to the inner and outer raceways of a bearing, caused by fatigue, due to how common these types of damage are. The methodology of this project involves a comprehensive review of the literature to identify promising techniques for measuring the size of this damage. From these techniques, in this research, the new DEW method is proposed that combines the Wavelet transform with signal envelope analysis (specifically Wavelet Denoising – Envelope Analysis– WPD) to estimate the size of the localized damage. To validate the employed techniques, simulated signals are used, in addition to two sets of measurements obtained from the literature. These datasets contain signals corresponding to bearings operating at various speeds, loads, damage sizes, among others. With the results obtained, it was possible to estimate a confidence interval for the size of the damage under different conditions. Despite some observed inconsistencies, the method showed more accurate results with increased load, fault size, or decreased speed. In the case of simulated signals, good results were achieved even in scenarios with up to 10 [dB] of signal-to-noise ratio. However, for the evaluation with the experimental dataset, this estimation proved insufficient due to the amount of noise present, which attenuated the components of interest, especially the input peak.
Description
Tesis presentada para optar al grado de Magíster en Ciencias de la Ingeniería con Mención en Ingeniería Mecánica
Keywords
Rodamientos, Resistencia de materiales