A prediction method of machine tool spindle spare parts based on running covariance analysis

Abstract

The invention provides a machine tool spindle spare part prediction method based on operation covariance analysis, which relates to the technical field of numerical control machine tools. Firstly, thehistorical fault data of an NC machine tool spindle and the related factors of running covariates which affect the life of parts are extracted, and the reliability model of spare parts for equipmentparts is established by using a proportional risk model. Then the covariates are screened and the regression coefficients of the final covariates are determined based on the data analysis software SPSS. Finally, the update process model is used to calculate the number of faults in a specific period of time for the independently distributed parts with irreparable fault data, and then the forecast number of spare parts required for the parts is obtained. The machine tool spindle spare parts prediction method based on the operation covariance analysis provided by the invention improves the accuracy of the required spare parts prediction data, reduces the storage amount of the spare parts of the equipment, and greatly reduces the operation cost of the enterprise.

Description

technical field [0001] The invention relates to the technical field of numerical control machine tools, in particular to a method for predicting machine tool spindle spare parts based on analysis of running covariates. Background technique [0002] The timely supply of spare parts is the basic guarantee for the immediate maintenance of the machine system. The timely supply of spare parts can improve the effectiveness of machinery and equipment production and reduce downtime losses; however, excessive spare parts reserves will cause inventory backlogs, generate inventory costs, and occupy a large amount of working capital. Therefore, accurate spare parts forecasting is not only related to production, but also involves cost accounting. [0003] The empirical method is the most basic spare parts forecasting method. Compared with this kind of method, there are time series-based methods with slightly higher prediction accuracy, mainly including exponential smoothing method (SES)...

AI Technical Summary

Problems solved by technology

However, the Bootstrap method assumes that there is an autocorrelation theory in the demand time series, which is difficult to guarantee, and can only obtain the previous demand distribution.

[0004] The failure law of the equipment is determined by its inherent reliability, but the factors affecting the external working environment will advance or delay the occurrence cycle of the failure point to a certain extent, changing the reliability of the equipment within a certain period of time.

Ghodrati neglected the effect of covariates when calculating the number of spare parts required for hydraulic jacks (lift cylinders) for mine unloaders in Kiruna mines, which may have resulted in a variance of about 20%.

Images