Research On Process Data Compression Technology In Visual Monitoring Of NC Machine

Under the background of intelligent manufacturing,and with the enterprises' increasing demand on the intellectualization visual monitoring system of NC machine,massive monitoring process data are collected in real time.They have the characteristics of large amount,rapid growth and great value.How to compress and store these massive data has become an urgent problem in the process of enterprise informatization.Therefore,in this paper,the monitoring process data of NC machine are studied in depth.On this basis,the compression strategy of monitoring process data is put forward,and they are applied to the development of visual monitoring system of NC machine.The main contents of this paper are as follows:(1)Through the study of existing compression algorithms,the limitations of applying these data compression algorithms to process data of NC machine monitoring system are analyzed.In this paper,the structure,characteristics and rules of data in NC machine monitoring process are deeply studied.Aiming at the switching data,three-dimensional trajectory data and analog data in monitoring process,the corresponding compression strategies are proposed based on the different changing rules of data.(2)Aiming at the three-dimensional trajectory data of NC machine monitoring,a data compression method based on NC program analysis is proposed in this paper.By studying the corresponding relationship between three-dimensional trajectory data and NC code,the NC code matching method of real-time data and the recognition method of abnormal data are proposed.The position error between the real-time coordinates and the theoretical trajectory based on NC code is used as the judgment basis of compression process,and the error limit is adjusting dynamically to obtain a better compression effect.(3)An improved SDT data compression algorithm based on similarity matching is proposed for the analog data of NC machine monitoring.A real-time state model and a historical state matrix for analogue data of monitoring process are constructed.Aiming at the inaccurate similarity calculation caused by the time delay between the state model and the subsequence of the historical state matrix,a similarity measurement method of state model based on window translation is proposed.According to different compression intervals,the real-time state model is matched with the subsequence of the historical state matrix by this similarity measure method,so that the threshold value of SDT algorithm can be adjusted dynamically,so that the algorithm can adapt to the changing trend of analog data in monitoring process.(4)The compression method proposed in this paper is applied to the development of visual monitoring system for NC machine to verify the reliability and practicability of the algorithm.Unity3 D and Eclipse platform are used to complete the development of monitoring system,and real-time data compression and decompression are used to provide data support for the monitoring system,so as to realize three-dimensional visual monitoring,historical data query,data statistics and other functions of the system.