Machine tool cutter residual life prediction method based on LSTM + CNN
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A life prediction and tool technology, applied in prediction, neural learning method, measuring device and other directions, can solve the problems of tool wear, cost increase, tool waste, etc., to achieve the effect of fitting prevention, improving accuracy and improving accuracy
Pending Publication Date: 2019-11-19
SHANDONG INSPUR GENESOFT INFORMATION TECH CO LTD
View PDF9 Cites 13 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
[0013] The technical problem to be solved by the present invention is: in the process of using the cutting tool of the modern machine tool in the factory, the cutting tool itself will be worn out, and the parts processed by the worn cutting tool may not meet the specifications or even be scrapped, so the cutting tool needs to be replaced in time, and the timing of changing the cutting tool is very short. Important, if it is too early, it will cause waste of tools, increase the cost, and affect the profit. If it is too late, it will produce substandard parts, so it is necessary to predict the remaining life of the tool based on various sensor signals to determine the timing
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0047] Such as figure 1 As shown, the tool usage (historical data) of a milling machine is used to predict the remaining life of the tool according to the uploaded sensor signal and PLC signal;
[0048] If the frequency of the sensor signal and the PLC signal are different (generally different), the data needs to be merged;
[0049] There are many abnormal data in real data, and the data needs to be cleaned;
[0050] Then train LSTM first, then train CNN, judge whether the model is good or bad (error size), adjust parameters and then train to get a satisfactory model;
[0051] Then, according to the real-time data of the milling machine tool, the same steps of data cleaning are performed, and the model is input for prediction, and the tool is replaced according to the real-time prediction result.
Embodiment 2
[0053] Such as figure 2 As shown, between the X and h sequences in the figure is the simplified schematic part of the LSTM model. The LSTM+CNN model and the training and prediction process are as follows:
[0054] Using the training data, assign the data according to the cross-validation principle, and perform supervised learning on the LSTM;
[0055] Use the result data predicted by LSTM as a one-dimensional data variable, and the variables in the training LSTM data after feature selection and dimensionality reduction, such as feature selection for displacement in X, Y, and Z directions, and vibration signals in X, Y, and Z axes , select several variables that are most relevant to the remaining life of the tool, and the feature selection method is the Pearson correlation coefficient algorithm combined as the input of CNN to train it.
[0056] In the process of supervised learning, the accuracy of LSTM is higher by adjusting the LSTM network and training parameters.
[0057...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses a machine tool cutter residual life prediction method based on LSTM + CNN, and the method comprises the steps: carrying out the judgment of the signal features of uploaded training data, and distinguishing a continuous signal and a discrete signal; performing data merging on the real-time data of different frequencies sampled by the sensor; checking whether missing values or abnormal values exist in the training data and the real-time data or not; if the missing values or the abnormal values exist, using a moving average method to supplement the missing values or replacing the abnormal values, so as to enable the data to be complete and effective, and removing outliers; carrying out selection and dimension reduction on the training data and the real-time data according to data characteristics so as to facilitate model fitting and prevent an over-fitting phenomenon; and training and testing the LSTM + CNN model, and adjusting training parameters and model parameters according to the error, so as to reduce the error to a reasonable range. According to the method, the precision of the prediction result is improved by adopting a grouping mode and a dimension reduction mode, deterministic factors and uncertain factors are comprehensively considered, and the precision of the prediction result can be effectively improved.
Description
technical field [0001] The invention relates to the technical field of online monitoring, in particular to a method for predicting the remaining life of a machine tool based on LSTM+CNN. Background technique [0002] At this stage, the methods for tool life prediction in the factory are basically calculated through the Taylor model of the tool or other simple physical models. Wear predictions cannot cope with the level and complexity of today's metal cutting technology. Therefore, the mainstream thinking in academia is based on the method of data analysis, combining the signals of different sensors, performing big data preprocessing, selecting corresponding features, and finally using different complex models to fit the data to achieve the purpose of judging and predicting the remaining life of the tool . [0003] For example, someone at Huazhong University of Science and Technology proposed a real-time prediction method for the remaining life of CNC machine tools. The mai...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more