当前位置: X-MOL 学术Robot. Comput.-Integr. Manuf. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
An efficient machine tool control instruction compression method for networked numerical control systems
Robotics and Computer-Integrated Manufacturing ( IF 9.1 ) Pub Date : 2020-07-14 , DOI: 10.1016/j.rcim.2020.102027
Miao Liu , Yingxue Yao , Jianjun Du

With the development of computer and information technology, function modules of computer numerical control (CNC) systems are moving up to network environment to simplify local machine tool controller architecture and increase function reusability. In this context, network-based CNC system generates machine control instructions (MCI) in network environment and send MCI to local machines. Compared with building a remote real-time communication channel between networked CNC and local machines, applying offline interpolator is more feasible under state-of-the-art hardware and software. However, the size of MCI generated by offline interpolator is large. Transmitting MCI without compression will be inefficient and increase lead time. Currently there has been few researches on the features and compression of offline MCI. In this paper, a simple and effective compression method for MCI named multichannel differential run-length encoding (MDRL) is proposed. Utilizing the limited variation and high repetition features of MCI, MDRL algorithm first calculates the differences of MCI, then encodes the amplitudes and repetition of difference values using fixed length codes. Compression performance discrepancy among different channels is controlled using compressed data chunks. The decompression process of MDRL only needs a few hundred of bytes of space and can process in real-time. Compression performance of MDRL is tested using three common manufacturing tasks under different system configurations. The results show that the mean compression ratio of MDRL is 3.41%-15.86%. The advantages of MDRL are further validated by comparing with three other data compression algorithms. By using MDRL, MCI transmission efficiency can be significantly increased for networked CNC systems to increase function reusability and shorten lead time.



中文翻译:

网络数控系统的高效机床控制指令压缩方法

随着计算机和信息技术的发展,计算机数控(CNC)系统的功能模块正在向网络环境发展,以简化本地机床控制器的体系结构并提高功能的可重用性。在这种情况下,基于网络的CNC系统在网络环境中生成机器控制指令(MCI),并将MCI发送到本地机器。与在联网的CNC与本地计算机之间建立远程实时通信通道相比,在最新的硬件和软件下应用离线插值器更为可行。但是,离线内插器生成的MCI的大小很大。在不压缩的情况下发送MCI将会效率低下,并增加前置时间。当前,关于脱机MCI的功能和压缩的研究很少。在本文中,提出了一种简单有效的MCI压缩方法,称为多通道差分游程编码(MDRL)。利用MCI的有限变化和高重复特征,MDRL算法首先计算MCI的差,然后使用固定长度代码对差值的幅度和重复进行编码。使用压缩数据块控制不同通道之间的压缩性能差异。MDRL的解压缩过程仅需要几百个字节的空间,并且可以实时处理。在不同的系统配置下,使用三个常见的制造任务来测试MDRL的压缩性能。结果表明,MDRL的平均压缩率为3.41%-15.86%。通过与其他三种数据压缩算法进行比较,进一步验证了MDRL的优势。

更新日期:2020-07-14
down
wechat
bug