Intelligent evaluation and identification method for early damage state of rolling bearing

Abstract

The invention discloses an intelligent evaluation and identification method for the early damage state of a rolling bearing, and the method comprises the following steps: firstly carrying out the processing of an early damage vibration feature data set of the rolling bearing through employing a minimum and maximum normalization method, and obtaining dimensionless feature data; then establishing asupport vector machine decision tree, and jointly optimizing kernel parameters and penalty parameters of the support vector machine through a particle swarm algorithm; and finally, carrying out automatic classification and intelligent decision making on the rolling bearing damage state data by utilizing a parameter-optimized support vector machine to obtain a state category corresponding to each state feature data, thereby realizing accurate evaluation and identification of the early damage state of the rolling bearing. Accurate classification of the early damage state of the rolling bearing can be achieved only through a small amount of effective data, the operation time is short, the calculation efficiency is high, the evaluation and recognition result of the early damage state of the rolling bearing can be automatically given, the intelligent degree is high, and the state recognition result is accurate and effective. And an important reference basis is provided for safe operation and real-time maintenance of the bearing and equipment.

Description

technical field [0001] The invention belongs to the technical field of mechanical equipment fault identification, and in particular relates to an intelligent evaluation and identification method for the early damage state of rolling bearings. Background technique [0002] If the rolling bearing is in a damaged state during use, it needs to be repaired and maintained in time to avoid equipment failure, especially the timely assessment and identification of the early damage state of the rolling bearing can effectively reduce the probability of equipment failure. At present, the evaluation and identification of rolling bearing damage status at home and abroad mainly use the vibration signal of the bearing to analyze, and then make identification and decision-making on the damage status and causes of rolling bearings. Among them, the fault diagnosis method based on machine learning has achieved better results. damage assessment and recognition. However, due to the strong nonlin...

AI Technical Summary

Problems solved by technology

However, due to the strong nonlinear characteristics of the rotor bearing system, the vibration signal of the bearing shows obvious non-stationary characteristics, and the operating environment of the bearing is complex and changeable, which makes it difficult to effectively identify the early damage state of the bearing until the damage degree of the rolling bearing is further aggravated. Abnormal bearing health status

That is to say, the early damage state of the bearing has not been evaluated and identified in time, which may lead to the complete failure of the rolling bearing or even the scrapping of the entire equipment

Images