Disclosure of Invention
In order to solve the technical problems, the invention provides an intelligent monitoring system and method for a numerical control machine tool.
The technical problems solved by the invention can be realized by adopting the following technical scheme:
a first aspect of the present invention provides an intelligent monitoring system for a numerically-controlled machine tool, including:
the acquisition end of the image acquisition unit covers a cutter of the numerical control machine tool and is used for acquiring image data of the cutter;
the analysis unit is connected with the image acquisition unit and is used for processing the image data and determining the current cutter abrasion degree;
the device comprises a first cutter life prediction unit, a second cutter life prediction unit and a third cutter life prediction unit, wherein a pre-trained cutter abrasion model is arranged in the first cutter life prediction unit and used for representing the relation between the cutter abrasion degree and the cutter life so as to determine and obtain the residual life of a first cutter according to the current cutter abrasion degree;
the second cutter life prediction unit is pre-established with a mathematical model between the cutter life and the use condition and is used for determining and obtaining the residual life of a second cutter according to the current use condition and the mathematical model;
the monitoring unit is respectively connected with the first cutter life prediction unit and the second cutter life prediction unit and is used for determining the service life of the cutter according to the residual life of the first cutter and the residual life of the second cutter.
Preferably, the image acquisition unit comprises a camera;
the cameras comprise two cameras, and the two cameras are oppositely arranged on two sides of the cutter.
Preferably, the analysis unit comprises:
the image preprocessing module is used for preprocessing the image data to obtain preprocessed image data;
the contour extraction unit is connected with the image preprocessing module and used for extracting a cutter contour in the preprocessed image data;
the feature extraction unit is connected with the contour extraction unit and is used for carrying out feature extraction according to the cutter contour to obtain feature information used for representing the abrasion degree of the cutter;
and the wear degree evaluation unit is connected with the characteristic extraction unit and is used for inputting the characteristic information into a pre-trained wear degree evaluation model so as to output and obtain the current cutter wear degree.
Preferably, the method further comprises:
a first data collection unit for collecting a training data set, wherein the training data set comprises the use data of a cutter in the numerical control machine tool, and the use data of the cutter comprise the use time, cutting parameters, the abrasion degree and the service life of the cutter;
the model training unit is respectively connected with the first data collection unit and the first cutter life prediction unit, and is used for carrying out model training, testing and verification according to the training data set to obtain the cutter abrasion model, and transmitting the cutter abrasion model to the first cutter life prediction unit.
Preferably, the tool wear model adopts at least one of a linear regression model, a support vector machine model and a decision tree model.
Preferably, the method further comprises:
the second data collection unit is used for collecting full-period use data of the cutter in the numerical control machine tool;
and the statistics unit is respectively connected with the second data collection unit and the second cutter life prediction unit, and is used for carrying out statistical analysis on the full-period use data, establishing and obtaining the mathematical model between the cutter life and the use condition, and transmitting the mathematical model to the second cutter life prediction unit.
Preferably, the service life of the tool is a weighted average of the remaining life of the first tool and the remaining life of the second tool.
Preferably, the weight of the remaining life of the first tool is greater than the weight of the remaining life of the second tool.
The second aspect of the present invention provides an intelligent monitoring method for a numerically-controlled machine tool, which is applied to the intelligent monitoring system for a numerically-controlled machine tool, and comprises the following steps:
collecting image data of a cutter of a numerical control machine tool;
processing the image data to determine the current cutter wear degree;
inputting the current tool wear degree into a pre-trained tool wear model, wherein the tool wear model is used for representing the relation between the tool wear degree and the tool life so as to determine and obtain a first tool residual life according to the current tool wear degree;
determining and obtaining the residual life of a second cutter according to the current use condition, the pre-established cutter life and a mathematical model between the use conditions;
and determining the service life of the cutter according to the residual life of the first cutter and the residual life of the second cutter.
The technical scheme of the invention has the advantages that:
according to the invention, the corresponding mathematical model and physical model are established in the tool life prediction of the numerical control machine tool, and the prediction results of the mathematical model and the physical model are comprehensively considered, so that the problem that the tool life prediction deviates from the actual state and the normal work of the numerical control machine tool is influenced due to model errors is avoided.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
The invention is further described below with reference to the drawings and specific examples, which are not intended to be limiting.
Referring to fig. 1, in a preferred embodiment of the present invention, based on the above-mentioned problems existing in the prior art, there is now provided an intelligent monitoring system for a numerically-controlled machine tool, comprising:
the image acquisition unit 1 covers a cutter of the numerical control machine tool at an acquisition end of the image acquisition unit 1 and is used for acquiring image data of the cutter;
the analysis unit 2 is connected with the image acquisition unit 1 and is used for processing the image data and determining the current cutter abrasion degree;
the first tool life prediction unit 3 is connected with the analysis unit 2, and a pre-trained tool wear model is arranged in the first tool life prediction unit 3 and used for representing the relation between the tool wear degree and the tool life so as to determine and obtain a first tool residual life according to the current tool wear degree;
a second tool life prediction unit 4, in which a mathematical model between the tool life and the usage condition is pre-established in the second tool life prediction unit 4, for determining a second tool remaining life according to the current usage condition and the mathematical model;
and the monitoring unit 5 is respectively connected with the first cutter life prediction unit 3 and the second cutter life prediction unit 4 and is used for determining the service life of the cutter according to the residual life of the first cutter and the residual life of the second cutter.
Specifically, in this embodiment, the service life of the tool is determined by comprehensively considering the remaining life of the first tool based on the wear degree of the tool and the remaining life of the second tool based on the service condition of the tool, the service life of the tool is monitored in real time, and the prediction results of the mathematical model and the physical model are comprehensively considered, so that the problem that the normal operation of the numerical control machine tool is affected due to the fact that the prediction of the service life of the tool deviates from the actual state due to the model error is avoided.
Meanwhile, the intelligent monitoring system of the numerical control machine tool has the functions of accurately monitoring the abrasion degree of the cutter, predicting the residual life of the cutter and monitoring the service life of the cutter in real time; meanwhile, the service life of the cutter is monitored in real time, and corresponding measures such as replacing the cutter or adjusting processing parameters are taken in time so as to ensure the processing quality and the production efficiency.
As a preferred embodiment, wherein the image acquisition unit 1 comprises a camera;
the camera includes two, and two camera opposite directions set up in the both sides of cutter.
Specifically, the image acquisition unit 1 is a device for acquiring image data, and its core component is a camera. In this embodiment, the image acquisition unit 1 includes two cameras, and two cameras are set up in the both sides of cutter in opposite directions, provide more comprehensive visual angle to realize the omnidirectional image acquisition, thereby can acquire complete image information, supply follow-up image processing and analysis very important, can improve accuracy and efficiency in the cutter use.
Further, due to possible vibration, impact and other factors during the use of the cutter, a single camera may be disturbed, resulting in degradation of image quality. By using two cameras, the two cameras can be mutually supplemented and corrected, and the stability and reliability of image acquisition are improved.
As a preferred embodiment, wherein, as shown in fig. 2, the analysis unit 2 includes:
an image preprocessing module 21, configured to preprocess image data to obtain preprocessed image data;
a contour extraction unit 22 connected to the image preprocessing module 21 for extracting a cutter contour in the preprocessed image data;
a feature extraction unit 23 connected to the contour extraction unit 22, and configured to perform feature extraction according to the contour of the tool, so as to obtain feature information for characterizing the wear degree of the tool;
the wear level evaluation unit 24 is connected to the feature extraction unit 23, and is configured to input the feature information into a pre-trained wear level evaluation model, so as to output and obtain the current tool wear level.
Specifically, in the present embodiment, in the process of determining the current tool wear degree, first, two tool images in image data acquired by two cameras are preprocessed respectively. The pretreatment comprises denoising, graying, binarization and the like. Denoising may use a filter (e.g., a gaussian filter) to remove noise from the image; then, converting the image into a gray scale image for subsequent processing; finally, the gray scale image is converted to a binary image using a suitable thresholding method to better extract the contours of the tool.
Then, the extraction of the cutter contour is performed, and the edge in the image can be detected by using an image processing algorithm (such as a Canny edge detection algorithm) to realize the extraction of the cutter contour so as to better identify the shape of the cutter.
Next, features are extracted to describe the extent of wear of the tool from its profile. For example, the area, circumference, form factor, etc. characteristics of the tool may be calculated to represent the wear of the tool. Further, a wear level assessment model can be built based on the extracted features. The model may be trained, for example, using a machine learning algorithm (e.g., support vector machine, random forest, etc.) to predict the degree of wear of the tool based on the characteristic information. In training the model, a cutter image of known wear level may be used as a training sample.
Further, the wear degree of the tool can be displayed according to the result of the evaluation model, for example, different colors or marks are used for representing different degrees of wear, so that a user can more intuitively know the condition of the tool.
As a preferred embodiment, as shown in fig. 3, the method further includes:
a first data collection unit 6 for collecting a training data set including usage data of a tool in the numerical control machine tool, the usage data of the tool including a tool usage time, a cutting parameter, a tool wear degree, and a life;
the model training unit 7 is respectively connected with the first data collecting unit 6 and the first tool life predicting unit 3, and is used for performing model training, testing and verifying according to the training data set to obtain a tool wear model, and transmitting the tool wear model to the first tool life predicting unit 3.
Specifically, in the present embodiment, during the training process of the tool wear model, first, the usage data of the tool in the numerically-controlled machine tool is collected, including parameters such as the usage time, cutting force, cutting speed, cutting depth, etc. of the tool, and the wear degree and life of the tool. The usage data of the tool may be full cycle usage data.
Further, characteristic data related to the degree of wear and the life of the tool can be extracted from the collected data. Features may be selected and extracted using, for example, statistical methods, machine learning methods, or domain knowledge. The collected data is then preprocessed, including data cleansing, missing value processing, outlier processing, etc., to ensure data quality and integrity in the training dataset.
Then, the collected training data set is divided into a training set, a testing set and a verification set so as to train, test and verify the cutter abrasion model. In the training process, according to a loss function and an optimization algorithm of the model, parameters of the model are adjusted, so that the model can be better fitted with training data; then, evaluating the trained model by using a test set, calculating indexes such as prediction accuracy, precision, recall rate and the like of the model, and evaluating the performance and generalization capability of the model; then, according to the evaluation result, the model is optimized, for example, the method of adjusting the super parameters of the model, adding more features, using more complex model structures and the like is adopted, so that the performance of the model is improved. Finally, the optimized model is verified by using a verification set, and the model after verification can be applied to actual tool wear prediction so as to predict the wear degree and service life of the tool according to new tool use data.
As a preferred embodiment, the tool wear model uses at least one of a linear regression model, a support vector machine model, and a decision tree model.
As a preferred embodiment, as shown in fig. 3, the method further includes:
a second data collection unit 8 for collecting full cycle usage data of the tool in the numerical control machine;
the statistics unit 9 is connected with the second data collection unit 8 and the second tool life prediction unit 4 respectively, and is used for performing statistical analysis on the full-cycle usage data, establishing a mathematical model between the tool life and the usage conditions, and transmitting the mathematical model to the second tool life prediction unit 4.
Specifically, in this embodiment, by collecting full-cycle usage data of a tool in a numerical control machine tool, processing the collected full-cycle usage data by a statistical analysis method, such as regression analysis, correlation analysis, etc., determining a relationship between a tool life and a usage condition, and establishing a mathematical model.
When the mathematical model is applied to actual production, the service life of the cutter can be predicted by monitoring and recording the cutter use data and inputting the cutter use data into the mathematical model. The cutter use data can be provided by a numerical control machine tool, no additional monitoring equipment is needed, and the monitoring cost is reduced.
In a preferred embodiment, the service life of the tool is a weighted average of the remaining life of the first tool and the remaining life of the second tool.
Specifically, in this embodiment, by respectively assigning corresponding weights to the first remaining life of the tool based on the degree of wear of the tool and the second remaining life of the tool based on the service condition of the tool, the weighted average value thereof is recorded as the service life of the tool obtained by final arbitration, and the influence of the degree of wear of the tool and the service condition on the service life of the tool is comprehensively considered, thereby obtaining a more accurate prediction result of the service life of the tool.
It should be noted that different tool types and machining materials may be selected with different weight distribution methods according to actual situations, so as to obtain more accurate prediction results.
As a preferred embodiment, wherein the weight of the remaining life of the first tool is greater than the weight of the remaining life of the second tool.
Referring to fig. 4, a second aspect of the present invention provides an intelligent monitoring method for a numerically-controlled machine tool, which is applied to the intelligent monitoring system for a numerically-controlled machine tool, including:
s1, collecting image data of a cutter of a numerical control machine tool;
s2, processing the image data and determining the abrasion degree of the current cutter;
s3, inputting the current tool wear degree into a pre-trained tool wear model, wherein the tool wear model is used for representing the relation between the tool wear degree and the tool life so as to determine and obtain the residual life of a first tool according to the current tool wear degree;
s4, determining and obtaining the residual life of a second cutter according to the current use condition, the pre-established cutter life and a mathematical model between the use conditions;
s5, determining the service life of the cutter according to the residual life of the first cutter and the residual life of the second cutter.
Specifically, in this embodiment, firstly, image data of a tool is collected and processed, so that the wear degree of the current tool can be accurately determined, so that the wear condition of the tool can be known in time, and the problem of tool failure or processing quality degradation caused by excessive wear is avoided.
Secondly, based on a cutter abrasion model, the residual service life of the first cutter can be determined according to the abrasion degree of the current cutter; meanwhile, based on a mathematical model used for representing the service life of the cutter and the service condition, the residual service life of the second cutter can be determined according to the current service condition and the mathematical model.
Finally, the service life of the cutter is determined by comprehensively considering the residual life of the first cutter based on the abrasion degree of the cutter and the residual life of the second cutter based on the service condition of the cutter, the service life of the cutter is monitored in real time, and the prediction results of the mathematical model and the physical model are comprehensively considered, so that the problem that the normal work of the numerical control machine tool is influenced due to the fact that the prediction of the service life of the cutter deviates from the actual state caused by model errors is avoided.
The technical scheme has the advantages that: according to the invention, the corresponding mathematical model and physical model are established in the tool life prediction of the numerical control machine tool, and the prediction results of the mathematical model and the physical model are comprehensively considered, so that the problem that the tool life prediction deviates from the actual state and the normal work of the numerical control machine tool is influenced due to model errors is avoided.
The foregoing is merely illustrative of the preferred embodiments of the present invention and is not intended to limit the embodiments and scope of the present invention, and it should be appreciated by those skilled in the art that equivalent substitutions and obvious variations may be made using the description and illustrations herein, which should be included in the scope of the present invention.