基于多维时间序列的数控加工中心运行状态预测方法研究

he prediction of the operation status of CNC machining centers in modern enterprises is the key to equipment failure warning, pre-maintenance and efficient operation of equipment. At present, there are many researches on equipment operation state prediction and fault diagnosis, but there are few researches on the numerical control equipment operation state prediction method. The process data is rarely used to analyze and research the numerical control equipment, resulting in waste of data. This paper proposes a multi-dimensional time series prediction model to predict the running state of numerical control equipment. Through the similarity search of the multi-dimensional time series model, the similarity set of the equipment running state is obtained, and the clustering analysis and the difference degree analysis of the similar set are obtained. The optimal similar sequence of the equipment running state is obtained, and the average absolute percentage of the result is predicted by the process data. The error is 5.42-9.56%, indicating that the model has strong practicability.