Health Condition Assessment Of Air Compressor Method Based On Mahalanobis Distance

In recent years,mechanical and electrical equipment as an important part of industrial production,its healthy operation is related to the safety and stability of the system,attracting widespread attention.Due to the complexity of the internal structure of the device,it is very difficult to find and eliminate the fault in a timely manner by means of human beings.Therefore,it is very important to evaluate the health degradation condition of such complex devices quickly and accurately.In this paper,the related concepts of equipment failure prediction,health management and health assessment are expounded firstly,and the current research status of equipment condition assessment technology in China and abroad is analyzed.Then some condition assessment methods presented in this paper are introduced and applied to Oxygen compressor health assessment.The main contents of this article are as follows:(1)This paper studies the improved Mahalanobis distance evaluation method.In order to overcome the shortcomings of the Mahalanobis distance,which is too sensitive to the small parameters,to reduce the accuracy of the estimation,an improved Mahalanobis distance algorithm optimization is proposed,in which the standard parameters of the Mahalanobis distance are introduced to replace the sample mean.By studying the time series of Mahalanobis distance,the health condition of equipment was evaluated.The two algorithms are applied to the assessment of the health condition of air compressor,which verifies that the improved Mahalanobis distance evaluation method is better in real time.(2)A state evaluation method for constructing degradation index was studied.Based on the improvement of the Mahalanobis distance,the histogram and the sliding time window are introduced to build a new metric of degradation index,which is applied to the air compressor evaluation example and the operating environment of the oxygen system where the air compressor is located when the abnormal observation point occurs is monitored.The method is proved to be able to remove environmental disturbance and instantaneous fault well and diagnose the early degradation of air compressor earlier,which has better stability and real-time performance.