Machine tool cutting intelligent monitoring system based on Internet of Things

Abstract

The invention relates to the technical field of machine tool cutting monitoring, in particular to an intelligent machine tool cutting monitoring system based on the Internet of Things. The system comprises the following steps: determining a transmission path between a starting point and an ending point of a to-be-cut object, and conveying the to-be-cut object to a machine tool cutting position according to the transmission path; obtaining an image of a to-be-cut object, and performing cutting track recognition on the image of the to-be-cut object through a cutting track recognition algorithm to obtain cutting track points; spline curve fitting is carried out based on the cutting track points, and a cutting track curve after cloud fitting is obtained; and performing comparative analysis based on the standard cutting track curve and the cutting track curve, and judging whether the cutting track curve meets the cutting requirement or not. The shortest transmission path of the object to be cut can be planned in real time, the cutting efficiency is improved, the abnormal cutting track can be recognized in time, early warning of the abnormal cutting track is achieved, and the cutting monitoring accuracy and efficiency of a machine tool are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of machine tool cutting monitoring, in particular to an intelligent machine tool cutting monitoring system based on the Internet of Things. Background Art

[0002] In the machine tool cutting process, the optimization and control of cutting parameters are of great significance for improving machining accuracy, reducing tool wear and extending the service life of machine tools. However, traditional cutting process monitoring methods often rely on manual inspections and regular testing, which have problems such as long monitoring cycles, slow response speeds and low data accuracy.

[0003] In recent years, the rapid development of Internet of Things technology has provided a new solution for real-time monitoring of machine tool cutting processes. Internet of Things technology converts various physical quantities in the machine tool cutting process into digital signals through sensors, wireless communication technology, cloud computing and other means, and transmits and processes them in real time. This technology can not only realize real-time monitoring of cutting parameters, but also discover potential faults and problems through data analysis, providing strong support for production decisions.

[0004] However, the existing machine tool cutting monitoring systems based on the Internet of Things still have some shortcomings. For example, some systems can only achieve simple data collection and transmission, and lack the functions of transmission path planning for cutting objects and abnormal cutting trajectory warning.

[0005] Therefore, there is an urgent need for an intelligent machine tool cutting monitoring system based on the Internet of Things to solve the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent monitoring system for machine tool cutting based on the Internet of Things: to solve the technical problems that the existing solutions for traditional cutting process monitoring methods often rely on manual inspections and regular inspections, and have long monitoring cycles, slow response speeds, and low data accuracy.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] An intelligent monitoring system for machine tool cutting based on the Internet of Things, the system includes a transmission path planning module, a cutting trajectory recognition module, a cutting trajectory point cloud fitting module and a cutting trajectory comparison module;

[0009] The transmission path planning module is used to obtain the current coordinate position of the object to be cut, and based on the current position as the starting point and the machine tool cutting position as the end point, traverse and calculate to obtain the node closest to the current node, and at the same time eliminate the nodes that have been visited. When the end point is visited, the calculation is stopped, the transmission path between the starting point and the end point is determined, and the object to be cut is transmitted to the machine tool cutting position according to the transmission path;

[0010] The cutting trajectory recognition module is used to obtain the image of the object to be cut, and perform cutting trajectory recognition on the image of the object to be cut through the cutting trajectory recognition algorithm to obtain cutting trajectory points;

[0011] The cutting trajectory point cloud fitting module is used to perform spline curve fitting based on the cutting trajectory points to obtain the cutting trajectory curve after cloud fitting;

[0012] The cutting trajectory comparison module is used to perform comparative analysis based on the standard cutting trajectory curve and the cutting trajectory curve to determine whether the cutting trajectory curve meets the cutting requirements.

[0013] Furthermore, traverse and calculate the node closest to the current node, and at the same time eliminate the nodes that have been visited. Stop the calculation when reaching the end point. Determining the transmission path between the start point and the end point specifically includes the following process:

[0014] Step 1: Use the current coordinate position of the object to be cut as the start point S, and the machine tool cutting position as the end point E;

[0015] Step 2: Create an array D[S] to store the distances from all nodes to S, and perform initialization settings D[S]=0. For other nodes V between the start point S and the end point E, the distance between node V and the current coordinate position of the object to be cut is D[V];

[0016] Step 3: Create a data heap Q to store the information of the node closest to the start point S and the distance, and initialize it to S, 0;

[0017] Step 4: Create a list P of visited nodes s , and initialize the algorithm to P s null value;

[0018] Step 5: Traverse the adjacent nodes of the current node in turn. Each time during the traversal calculation, take out the information (V, D[V]) in the data heap Q, and query whether the node is in the list P s ; Among them, when calculating each adjacent node n of node V, judge whether D[V]+W[V,n]<D[n] holds. If so, let D[V]+W[V,n]=D[n], and in the data heap Q (n, D[n]). If not, no update processing is performed; W[V,n] is the distance between node V and adjacent node n, and D[n] is the distance between node n and the current coordinate position of the object to be cut;

[0019] Step 6: If node V is in P a , it means that the node has been visited, return to Step 5, and continue to traverse the next adjacent node; if node V is not in P s , then add the node to the list P s ;

[0020] Step 7: The traversal is completed, the algorithm ends, and the shortest distance from node S to node E and the nodes passed through are obtained. The transmission path between the starting point and the end point is generated based on the shortest distance and the nodes passed through.

[0021] Furthermore, obtaining an image of the object to be cut, and performing cutting trajectory recognition on the image of the object to be cut by a cutting trajectory recognition algorithm to obtain cutting trajectory points specifically includes the following process:

[0022] The image of the object to be cut is converted into a grayscale image, and the pixels of the grayscale image are defined as I(u, v);

[0023] The pixel I(u, v) gradient is calculated according to the different directions of the pixel plane coordinate axis:

[0024] G X (u, v) = S X *I(u,v);

[0025] G Y (u, v) = S Y *I(u,v);

[0026] Among them, G X (u, v) is the pixel gradient in the X-axis direction of the pixel plane, G Y (u, v) is the pixel gradient in the Y-axis direction of the pixel plane, S X , S Y They are the Sobel operators of the pixel points on the X-axis and Y-axis of the pixel plane coordinate system, respectively, and * is the convolution operation;

[0027] Based on G X (u, v) and G Y (u, v) calculates the gradient value g at each pixel XY (u, v):

[0028]

[0029] For each pixel, the gradient direction θ of the pixel is obtained based on the partial derivatives of the pixel with respect to the X-axis and Y-axis:

[0030] θ=arctan(gx(u,v) / gY(u,v));

[0031] Among them, g X (u, v) is the partial derivative of the pixel with respect to the X axis, g Y (u, v) is the partial derivative of the pixel with respect to the Y axis;

[0032] Assume that the gradient direction of pixel A is along the horizontal direction, and the pixels adjacent to pixel A in the upward direction are B and C. Compare the gradient values ​​of pixels A, B, and C. If the gradient value of pixel A is the largest, pixel A is determined to be an edge point, and the edge point is discarded, and the remaining pixels are retained as cutting trajectory points.

[0033] Furthermore, spline curve fitting is performed based on the cutting trajectory points to obtain a cutting trajectory curve after cloud fitting, which specifically includes the following process:

[0034] The cutting trajectory points are defined as P = (P1, P2, ..., P m ), where P i represents the cutting trajectory point, i = 1, 2, ..., m;

[0035] Calculate the distance fe between the spline curve and the cutting trajectory point:

[0036]

[0037] Among them, d(C(t), P i )=min||C(t)-P i ||, represents the minimum Euclidean distance from the cutting trajectory point Pi to the control point of the spline curve C(t);

[0038] Solve f using matrix method e :

[0039] Represent the control points of the spline curve C(t) as a column vector C s (t), the cutting trajectory point P is represented by a column vector P s (t):

[0040] C s (t)=[C(1), C(2),..., C(N)] T ;

[0041] Where N is the number of control points of the spline curve C(t);

[0042] P s (t) = [P1, P2, ..., P m ] T ;

[0043] Then the distance f between the spline curve and the cutting trajectory point is e It is expressed as:

[0044]

[0045] f e Find the first-order derivative and set it to 0. The calculated spline curve C(j) is the cutting trajectory curve after cloud fitting.

[0046] Furthermore, based on the comparative analysis between the standard cutting trajectory curve and the cutting trajectory curve, judging whether the cutting trajectory curve meets the cutting requirements specifically includes the following process:

[0047] Calculate the error E between the standard cutting trajectory curve and the cutting trajectory curve r :

[0048]

[0049] Among them, H is the number of trajectory points between the standard cutting trajectory curve and the cutting trajectory curve, Q l is the coordinate of the trajectory point corresponding to the standard cutting trajectory curve on the projection plane, P h is the coordinate of the trajectory point corresponding to the cutting trajectory curve on the projection plane;

[0050] Based on the error E r Determine whether the cutting trajectory curve meets the cutting requirements.

[0051] Furthermore, based on the error E r Judging whether the cutting trajectory curve meets the cutting requirements specifically includes the following processes:

[0052] loading an error threshold, wherein the error threshold is stored in the system;

[0053] Judgment error E r Whether it exceeds the error threshold, if so, it is determined that the cutting trajectory curve does not meet the cutting requirements, if not, it is determined that the cutting trajectory curve meets the cutting requirements.

[0054] Furthermore, the system also includes the following process:

[0055] Based on the cutting trajectory curve that does not meet the cutting requirements, the running time, cutting times and tool replacement frequency of multiple tools of the CNC machine tool are recorded, the use frequency of multiple tools is analyzed, and tool use frequency data is generated;

[0056] Using the tool usage frequency data, the average wear rate of the tool is estimated by calculating the proportion of the tool's operating time under various loads and speeds, and the wear rate evaluation data is generated;

[0057] According to the wear rate evaluation data, by analyzing the usage data of multiple tools of CNC machine tools, the remaining service life of the tools is predicted and the tool life prediction results are generated.

[0058] Furthermore, by using the tool usage frequency data, the average wear rate of the tool is estimated by calculating the operating time ratio of the tool under various loads and speeds. The generation of wear rate evaluation data specifically includes the following process:

[0059] According to the established wear model, the tool wear rate under each load and speed is calculated;

[0060] The wear rate is calculated by dividing the amount of wear by the time of use;

[0061] Calculate the average wear rate:

[0062] The wear rate is weighted averaged according to the proportion of operating time at each load and speed;

[0063] The average wear rate is the product of the wear rate and the operating time ratio;

[0064] The calculated average wear speed is recorded in a wear speed evaluation data table; the wear speed evaluation data table includes tool type, use conditions, operating time ratio, and average wear speed;

[0065] Analyze the differences in wear rates under different tool types and usage conditions;

[0066] Identify the key factors affecting wear rate and provide a basis for optimizing tool use and maintenance strategies.

[0067] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0068] The present invention determines the transmission path between the starting point and the end point of the object to be cut, and transmits the object to be cut to the cutting position of the machine tool according to the transmission path; obtains the image of the object to be cut, and identifies the cutting trajectory of the image of the object to be cut by a cutting trajectory recognition algorithm to obtain cutting trajectory points; performs spline curve fitting based on the cutting trajectory points to obtain a cutting trajectory curve after cloud fitting; compares and analyzes the standard cutting trajectory curve with the cutting trajectory curve to determine whether the cutting trajectory curve meets the cutting requirements. The shortest transmission path of the object to be cut can be planned in real time to improve cutting efficiency, and abnormal cutting trajectories can be identified in a timely manner to achieve early warning of abnormal cutting trajectories, thereby improving the accuracy and efficiency of machine tool cutting monitoring;

[0069] Furthermore, the system can realize real-time monitoring, intelligent analysis and abnormal warning of key parameters in the machine tool cutting process, improve production efficiency and product quality, and reduce production costs and safety risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0071] Figure 1It is a system block diagram of a machine tool cutting intelligent monitoring system based on the Internet of Things according to an embodiment of the present invention;

[0072] Figure 2 This is a workflow diagram of a machine tool cutting intelligent monitoring system based on the Internet of Things according to an embodiment of the present invention;

[0073] Figure 3 This is another workflow diagram of an IoT-based machine tool cutting intelligent monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0076] This embodiment provides an intelligent monitoring system for machine tool cutting based on the Internet of Things. Figure 1 is a system block diagram of a machine tool cutting intelligent monitoring system based on the Internet of Things according to an embodiment of the present invention, such as Figure 1 As shown, the system includes a transmission path planning module, a cutting trajectory recognition module, a cutting trajectory point cloud fitting module and a cutting trajectory comparison module;

[0077] The transmission path planning module is used to obtain the current coordinate position of the object to be cut, and based on the current position as the starting point and the machine tool cutting position as the end point, traverse and calculate to obtain the node closest to the current node, and at the same time eliminate the nodes that have been visited. When the end point is visited, the calculation is stopped, the transmission path between the starting point and the end point is determined, and the object to be cut is transmitted to the machine tool cutting position according to the transmission path;

[0078] A cutting trajectory recognition module is used to obtain an image of the object to be cut, and to perform cutting trajectory recognition on the image of the object to be cut by a cutting trajectory recognition algorithm to obtain cutting trajectory points;

[0079] A cutting trajectory point cloud fitting module is used to perform spline curve fitting based on cutting trajectory points to obtain a cutting trajectory curve after cloud fitting;

[0080] The cutting trajectory comparison module is used to compare and analyze the standard cutting trajectory curve with the cutting trajectory curve to determine whether the cutting trajectory curve meets the cutting requirements.

[0081] In summary, the present invention determines the transmission path between the starting point and the end point of the object to be cut, and transmits the object to be cut to the cutting position of the machine tool according to the transmission path; obtains the image of the object to be cut, and identifies the cutting trajectory of the image of the object to be cut by the cutting trajectory recognition algorithm to obtain the cutting trajectory points; performs spline curve fitting based on the cutting trajectory points to obtain the cutting trajectory curve after cloud fitting; compares and analyzes the standard cutting trajectory curve with the cutting trajectory curve to determine whether the cutting trajectory curve meets the cutting requirements. The shortest transmission path of the object to be cut can be planned in real time to improve cutting efficiency, and abnormal cutting trajectories can be identified in a timely manner to achieve early warning of abnormal cutting trajectories, thereby improving the accuracy and efficiency of machine tool cutting monitoring;

[0082] Furthermore, the system can realize real-time monitoring, intelligent analysis and abnormal warning of key parameters in the machine tool cutting process, improve production efficiency and product quality, and reduce production costs and safety risks.

[0083] In some embodiments, the traversal calculation obtains the node closest to the current node, and the nodes that have been visited are eliminated. When the end point is visited, the calculation is stopped. Determining the transmission path between the starting point and the end point specifically includes the following process:

[0084] Step 1: Take the current coordinate position of the object to be cut as the starting point S and the cutting position of the machine tool as the end point E;

[0085] Step 2: Create an array D[S] to store the distances from all nodes to S, and initialize it to set D[S] = 0. For other nodes V between the starting point S and the end point E, the distance between the node V and the current coordinate position of the object to be cut is D[V];

[0086] Step 3: Create a data heap Q to store the node and distance closest to the starting point S in the current state, and initialize it to (S, 0);

[0087] Step 4: Create a list of visited nodes P s , the algorithm is initialized to P a Null value;

[0088] Step 5: Traverse the adjacent nodes of the current node in turn. Each time the traversal is calculated, the information (V, D[V]) in the data pile Q is taken out, and the node is checked whether it is in the list P. sIn the process, when calculating each adjacent node n of node V, it is determined whether D[V]+W[V,n]<D[n] holds. If it holds, let D[V]+W[V,n]=D[n], and add (n,D[n]) to the data heap Q. If not, no update process is performed; W[V,n] is the distance between node V and adjacent node n, and D[n] is the distance between node n and the current coordinate position of the object to be cut.

[0089] Step Six: If node V is in P s it means that the node has been visited, return to Step Five, and continue to traverse the next adjacent node; if node V is not in P s then add this node to the list P s ;

[0090] Step Seven: After the traversal is completed, the algorithm ends, and the shortest distance from node S to node E and the nodes passed through are obtained. A transmission path between the starting point and the ending point is generated based on the shortest distance and the nodes passed through.

[0091] In some embodiments, obtaining an image of the object to be cut, and performing cutting trajectory recognition on the image of the object to be cut through a cutting trajectory recognition algorithm to obtain cutting trajectory points specifically includes the following process:

[0092] Convert the image of the object to be cut into a grayscale image, and define the pixels of the grayscale image as I(u,v);

[0093] Calculate the gradient of pixel I(u,v) respectively according to the different directions of the pixel plane coordinate axes:

[0094] G X (u,v)=S X *I(u,v);

[0095] G Y (u,v)=S Y *I(u,v);

[0096] where G X (u,v) is the pixel gradient in the X-axis direction of the pixel plane, G Y (u,v) is the pixel gradient in the Y-axis direction of the pixel plane, S X and S Y are the Sobel operators of the pixel point on the X-axis and Y-axis of the pixel plane coordinate system respectively, and * is the convolution operation;

[0097] Based on G X (u,v) and G Y (u,v), calculate the gradient value magnitude g XY (u,v) at each pixel point:

[0098]

[0099] For each pixel, the gradient direction θ of the pixel is obtained based on the partial derivatives of the pixel with respect to the X-axis and Y-axis:

[0100] θ=arctan(gX(u,v) / gY(u,v));

[0101] Among them, g X (u, v) is the partial derivative of the pixel with respect to the X axis, g Y (u, v) is the partial derivative of the pixel with respect to the Y axis;

[0102] Assume that the gradient direction of pixel A is along the horizontal direction, and the pixels adjacent to pixel A in the upward direction are B and C. Compare the gradient values ​​of pixels A, B, and C. If the gradient value of pixel A is the largest, pixel A is determined to be an edge point, and the edge point is discarded, and the remaining pixels are retained as cutting trajectory points.

[0103] In some embodiments, performing spline curve fitting based on cutting trajectory points to obtain a cutting trajectory curve after cloud fitting specifically includes the following process:

[0104] The cutting trajectory points are defined as P = (P1, P2, ..., P m ), where P i represents the cutting trajectory point, i = 1, 2, ..., m;

[0105] Calculate the distance fe between the spline curve and the cutting trajectory point:

[0106]

[0107] Among them, d(C(t), P i )=min||C(t)-P i ||, represents the cutting trajectory point P i Minimum Euclidean distance to the control point of the spline curve C(t);

[0108] Solve f using matrix method e :

[0109] Represent the control points of the spline curve C(t) as a column vector C s (t), the cutting trajectory point P is represented by a column vector P s (t):

[0110] C s (t)=[C(1), C(2),..., C(N)] T ;

[0111] Where N is the number of control points of the spline curve C(t);

[0112] P s (t) = [P1, P2, ..., P m ] T ;

[0113] Then the distance f between the spline curve and the cutting trajectory point is e It is expressed as:

[0114]

[0115] f e Find the first-order derivative and set it to 0. The calculated spline curve C(j) is the cutting trajectory curve after cloud fitting.

[0116] In some embodiments, based on the comparison and analysis between the standard cutting trajectory curve and the cutting trajectory curve, determining whether the cutting trajectory curve meets the cutting requirements specifically includes the following process:

[0117] Calculate the error E between the standard cutting trajectory curve and the cutting trajectory curve r :

[0118]

[0119] Among them, H is the number of trajectory points between the standard cutting trajectory curve and the cutting trajectory curve, Q l is the coordinate of the trajectory point corresponding to the standard cutting trajectory curve on the projection plane, P h is the coordinate of the trajectory point corresponding to the cutting trajectory curve on the projection plane;

[0120] Based on the error E r Determine whether the cutting trajectory curve meets the cutting requirements.

[0121] Furthermore, based on the error E r Judging whether the cutting trajectory curve meets the cutting requirements specifically includes the following processes:

[0122] loading an error threshold, wherein the error threshold is stored in the system;

[0123] Judgment error E r Whether it exceeds the error threshold, if so, it is determined that the cutting trajectory curve does not meet the cutting requirements, if not, it is determined that the cutting trajectory curve meets the cutting requirements.

[0124] In some embodiments, Figure 2 This is a workflow diagram of a machine tool cutting intelligent monitoring system based on the Internet of Things according to an embodiment of the present invention. Figure 2 As shown, the system also includes the following steps:

[0125] Step S201: recording the operation time, cutting times and tool replacement frequency of multiple tools of a CNC machine tool based on a cutting trajectory curve that does not meet the cutting requirements, analyzing the usage frequency of multiple tools, and generating tool usage frequency data;

[0126] Step S202: using the tool usage frequency data, by calculating the operating time ratio of the tool under various loads and speeds, estimating the average wear speed of the tool, and generating wear speed evaluation data;

[0127] Step S203: Based on the wear rate evaluation data, by analyzing the usage data of multiple tools of the CNC machine tool, the remaining service life of the tool is predicted, and a tool life prediction result is generated.

[0128] In some embodiments, in step S202, Figure 3 This is another workflow diagram of the machine tool cutting intelligent monitoring system based on the Internet of Things according to an embodiment of the present invention. Figure 3 As shown in the figure, using the tool usage frequency data, by calculating the proportion of the tool's operating time under various loads and speeds, the average wear rate of the tool is estimated, and the wear rate evaluation data is generated, which specifically includes the following steps:

[0129] Step S301: Calculate the tool wear rate under each load and speed according to the established wear model;

[0130] Among them, the wear rate is calculated by dividing the wear amount by the use time;

[0131] Step S302: calculating the average wear rate;

[0132] Specifically, the wear rate is weighted averaged according to the proportion of operating time at each load and speed;

[0133] The average wear rate is the product of the wear rate and the operating time ratio;

[0134] The calculated average wear speed is recorded in a wear speed evaluation data table; the wear speed evaluation data table includes tool type, use conditions, operating time ratio, and average wear speed;

[0135] Step S303: Analyze the differences in wear rates under different tool types and usage conditions, identify key factors affecting the wear rate, and provide a basis for optimizing tool usage and maintenance strategies.

[0136] In some embodiments, according to the wear rate evaluation data, by analyzing the usage data of multiple tools of the CNC machine tool, predicting the remaining service life of the tool, generating the tool life prediction result includes the following process:

[0137] First, it is necessary to collect the usage data of multiple tools in CNC machine tools. This data includes but is not limited to:

[0138] Cutting parameters: such as cutting speed, feed rate, cutting depth, etc. These data directly affect the wear rate of the tool.

[0139] Processing material type: Different materials have different properties such as hardness and toughness that will have different effects on tool wear.

[0140] Environmental conditions: such as temperature, humidity, etc., although their impact on tool wear is relatively small, they also need to be considered in some specific cases.

[0141] Tool wear status: Regularly check and record the tool wear status, such as the amount of flank wear, etc.

[0142] The collected raw data often has problems such as noise and missing values, and needs to be preprocessed to improve data quality. The preprocessing steps include:

[0143] Data cleaning: Remove invalid, erroneous or duplicate data.

[0144] Data normalization: Convert data of different dimensions into the same dimension to facilitate subsequent analysis.

[0145] Data partitioning: Divide the dataset into training and testing sets for training prediction models and evaluating model performance.

[0146] Select features that can effectively predict tool life from the preprocessed data. These features should be able to reflect the main influencing factors of tool wear, such as cutting parameters, processing material type, etc.

[0147] Select a suitable prediction model for training. In tool life prediction, commonly used models include grey model, long short-term memory network (LSTM), etc.

[0148] Grey model: It is suitable for situations with less data and can make full use of the information in a small amount of data for prediction. After the tool enters the normal wear stage, a small amount of measurement point data can be extracted for prediction.

[0149] LSTM model: It can learn long-term dependencies in data sequences and is suitable for processing tool wear data that changes over time. By collecting various data on tool use, cleaning, normalizing, and feature selection, the LSTM network is used to train the data to predict the time point of tool failure.

[0150] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0151] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0152] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0153] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0154] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0155] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent monitoring system for machine tool cutting based on the Internet of Things, characterized in that: The system includes a transmission path planning module, a cutting trajectory recognition module, a cutting trajectory point cloud fitting module and a cutting trajectory comparison module; The transmission path planning module is used to obtain the current coordinate position of the object to be cut, and based on the current position as the starting point and the machine tool cutting position as the end point, traverse and calculate to obtain the node closest to the current node, and at the same time eliminate the nodes that have been visited. When the end point is visited, the calculation is stopped, the transmission path between the starting point and the end point is determined, and the object to be cut is transmitted to the machine tool cutting position according to the transmission path; A cutting trajectory recognition module is used to obtain an image of the object to be cut, and to perform cutting trajectory recognition on the image of the object to be cut by a cutting trajectory recognition algorithm to obtain cutting trajectory points; A cutting trajectory point cloud fitting module is used to perform spline curve fitting based on cutting trajectory points to obtain a cutting trajectory curve after cloud fitting; The cutting trajectory comparison module is used to compare and analyze the standard cutting trajectory curve with the cutting trajectory curve to determine whether the cutting trajectory curve meets the cutting requirements.

2. According to the Internet of Things-based machine tool cutting intelligent monitoring system according to claim 1, it is characterized in that: The traversal calculation obtains the node closest to the current node, and the nodes that have been visited are eliminated. The calculation stops when the end point is visited. The specific process of determining the transmission path between the starting point and the end point includes the following: Step 1: Take the current coordinate position of the object to be cut as the starting point S and the cutting position of the machine tool as the end point E; Step 2: Create an array D[S] to store the distances from all nodes to S, and initialize it to set D[S] = 0. For other nodes V between the starting point S and the end point E, the distance between the node V and the current coordinate position of the object to be cut is D[V]; Step 3: Create a data heap Q to store the node and distance closest to the starting point S in the current state, and initialize it to (S, 0); Step 4: Create a list of visited nodes P s , the algorithm is initialized to P s Null value; Step Five: Traverse the adjacent nodes of the current node in sequence. Each time during the traversal calculation, retrieve the information (V, D[V]) from the data heap Q, and query whether the node is in the list P s ; where, when calculating each adjacent node n of node V, determine whether D[V] + W[V,n] < D[n] holds. If so, let D[V] + W[V,n] = D[n], and (n, D[n]) in the data heap Q. If not, no update processing is performed; W[V,n] is the distance between node V and adjacent node n, and D[n] is the distance between node n and the current coordinate position of the object to be cut Step 6: If node V is in P s If the node V is not in P, it means that the node has been visited, and we return to step 5 to continue traversing the next adjacent node. s In the list P s Add this node in Step 7: The traversal is completed, the algorithm ends, and the shortest distance from node S to node E and the nodes passed through are obtained. The transmission path between the starting point and the end point is generated based on the shortest distance and the nodes passed through.

3. The machine tool cutting intelligent monitoring system based on the Internet of Things according to claim 1 is characterized in that: Acquire an image of the object to be cut, and use a cutting trajectory recognition algorithm to perform cutting trajectory recognition on the image of the object to be cut, and obtain cutting trajectory points, which specifically includes the following processes: The image of the object to be cut is converted into a grayscale image, and the pixels of the grayscale image are defined as I(u, v); The pixel I(u, v) gradient is calculated according to the different directions of the pixel plane coordinate axis: G X (u,v)=S X *I(u,v); G Y (u,v)=S Y *I(u,v); Among them, G X (u, v) is the pixel gradient in the X-axis direction of the pixel plane, G Y (u, v) is the pixel gradient in the Y-axis direction of the pixel plane, S X , S Y They are the Sobel operators of the pixel points on the X-axis and Y-axis of the pixel plane coordinate system, respectively, and * is the convolution operation; Based on G X (u, v) and G Y (u, v) calculates the gradient value g at each pixel XY (u, v): For each pixel, the gradient direction θ of the pixel is obtained based on the partial derivatives of the pixel with respect to the X-axis and Y-axis: θ=arctan(g X (u,v) / g Y (u,v)); Among them, g X (u, v) is the partial derivative of the pixel with respect to the X axis, g Y (u, v) is the partial derivative of the pixel with respect to the Y axis; Assume that the gradient direction of pixel A is along the horizontal direction, and the pixels adjacent to pixel A in the upward direction are B and C. Compare the gradient values ​​of pixels A, B, and C. If the gradient value of pixel A is the largest, pixel A is determined to be an edge point, and the edge point is discarded, and the remaining pixels are retained as cutting trajectory points.

4. The machine tool cutting intelligent monitoring system based on the Internet of Things according to claim 1 is characterized in that: Based on the cutting trajectory points, spline curve fitting is performed to obtain the cutting trajectory curve after cloud fitting, which specifically includes the following processes: The cutting trajectory points are defined as P = (P1, P2, ..., P m ), where P i represents the cutting trajectory point, i = 1, 2, ..., m; Calculate the distance f between the spline curve and the cutting trajectory point e : Among them, d(C(t), P i )=min||C(t)-P i ||, represents the cutting trajectory point P i Minimum Euclidean distance to the control point of the spline curve C(t); Solve f using matrix method e : The spline curve C(t ) The control point is represented as a column vector C s (t), the cutting trajectory point P is represented by a column vector P s (t): C s (t)=[C(1),C(2),...,C(N)] T ; Where N is the number of control points of the spline curve C(t); P s (t)=[P1,P2,...,P m ] T ; Then the distance f between the spline curve and the cutting trajectory point is e It is expressed as: f e Find the first-order derivative and set it to 0. The calculated spline curve C(j) is the cutting trajectory curve after cloud fitting.

5. The machine tool cutting intelligent monitoring system based on the Internet of Things according to claim 1 is characterized in that: Based on the comparative analysis between the standard cutting trajectory curve and the cutting trajectory curve, judging whether the cutting trajectory curve meets the cutting requirements specifically includes the following processes: Calculate the error E between the standard cutting trajectory curve and the cutting trajectory curve r : Among them, H is the number of trajectory points between the standard cutting trajectory curve and the cutting trajectory curve, Q l is the coordinate of the trajectory point corresponding to the standard cutting trajectory curve on the projection plane, P h is the coordinate of the trajectory point corresponding to the cutting trajectory curve on the projection plane; Based on the error E r Determine whether the cutting trajectory curve meets the cutting requirements.

6. The machine tool cutting intelligent monitoring system based on the Internet of Things according to claim 5 is characterized in that: Based on the error E r Judging whether the cutting trajectory curve meets the cutting requirements specifically includes the following processes: loading an error threshold, wherein the error threshold is stored in the system; Judgment error E r Whether it exceeds the error threshold, if so, it is determined that the cutting trajectory curve does not meet the cutting requirements, if not, it is determined that the cutting trajectory curve meets the cutting requirements.

7. The machine tool cutting intelligent monitoring system based on the Internet of Things according to claim 6 is characterized in that: The system also The process includes: Based on the cutting trajectory curve that does not meet the cutting requirements, the running time, cutting times and tool replacement frequency of multiple tools of the CNC machine tool are recorded, the use frequency of multiple tools is analyzed, and tool use frequency data is generated; Using the tool usage frequency data, the average wear rate of the tool is estimated by calculating the proportion of the tool's operating time under various loads and speeds, and the wear rate evaluation data is generated; According to the wear rate evaluation data, by analyzing the usage data of multiple tools of CNC machine tools, the remaining service life of the tools is predicted and the tool life prediction results are generated.

8. The machine tool cutting intelligent monitoring system based on the Internet of Things according to claim 7 is characterized in that: Using the tool usage frequency data, the average wear rate of the tool is estimated by calculating the proportion of the tool's operating time under various loads and speeds. The wear rate evaluation data is generated by the following process: According to the established wear model, the tool wear rate under each load and speed is calculated; The wear rate is calculated by dividing the amount of wear by the time of use; Calculate the average wear rate: The wear rate is weighted averaged according to the proportion of operating time at each load and speed; The average wear rate is the product of the wear rate and the operating time ratio; Record the calculated average wear rate in the wear rate evaluation data table; The wear rate evaluation data table includes tool type, usage conditions, operating time ratio, and average wear rate; Analyze the differences in wear rates under different tool types and usage conditions; Identify the key factors affecting wear rate and provide a basis for optimizing tool use and maintenance strategies.

Images