A Real-time Statistical Method of Machine Tool Workpieces Based on Current Signal Segmentation

Abstract

The invention proposes a real-time statistics method for machine tool workpieces based on current signal segmentation, which includes two stages: a segmentation model training stage and an online segmentation and statistics stage for workpieces. For data collection and processing in the segmentation model training stage, first collect the historical current signal and historical work order information of the machine tool, and after cleaning and normalizing the data, obtain the training samples; for model training, use the GaussianHMM model to train the samples and save them. The trained model is used as a predictor; the workpiece online segmentation statistics stage: the GaussianHMM model is used to segment the current signal, and a single workpiece processing signal segment is extracted to realize the online real-time statistics of the workpiece. The model-based workpiece segmentation algorithm can realize accurate and real-time online statistics of workpieces without pre-establishing a workpiece processing signal template library.

Description

【Technical field】 [0001] The invention relates to the technical field of industrial machine tool processing, in particular to a real-time statistical method for machine tool workpieces based on current signal segmentation. 【Background technique】 [0002] The traditional method of statistics of machine tool workpiece processing progress information is mainly through manual statistics, especially when small processing plants often change the type and quantity of processed workpieces, manual statistics are inefficient and error-prone. The artificial statistics method has the following problems: first, post-event statistics, poor real-time performance, and cannot dynamically grasp the real-time progress of the processing site; second, manual statistics are prone to errors. [0003] Some new methods collect the signal of the equipment to make statistics, mainly to solve the statistics of the batch production of the same kind of workpiece or the progress of large workpieces. In t...

AI Technical Summary

Problems solved by technology

[0002] The traditional method of statistics of machine tool workpiece processing progress information is mainly through manual statistics, especially when small processing plants often change the type and quantity of processed workpieces, manual statistics are inefficient and error-prone

The artificial statistics method has the following problems: First, post-event statistics, poor real-time performance, and cannot dynamically grasp the real-time progress of the processing site; second, manual statistics are prone to errors

However, in many industrial enterprises, the motor of ordinary machine tools will always be in the power-on state, and it is impossible to divide a workpiece by simply starting and stopping; and for CNC machine tools, there will also be a short-time start and stop state during the processing of a workpiece. , using the above method will divide a workpiece into multiple parts, resulting in statistical errors

The second problem is that in terms of data acquisition and processing, the above method requires an acquisition frequency of 20Hz and an accuracy of 12 bits; it requires high acquisition equipment and processing capabilities, and the acquisition cost is high, so it is difficult to promote in general industrial enterprises

Images