Intelligent monitoring system and method for numerical control machine tool

Abstract

The invention discloses an intelligent monitoring system and method for a numerical control machine tool, and relates to the field of machine tool control, and the system comprises a material attribute similarity obtaining module which is used for obtaining the material attribute similarity information of a target machining part and a historical machining part; the processing code similarity acquisition module is used for acquiring code similarity information of the target processing part and the historical processing part; the historical abnormal information acquisition module is used for acquiring abnormal pause information of historical processing parts; the early warning prompt information generation module is used for generating early warning prompt information according to the material attribute similarity information, the code similarity information and the abnormal pause information; and the auxiliary function control module is used for controlling the working state of the auxiliary function module according to the early warning prompt information.

Description

Technical Field

[0001] This specification relates to the field of machine tool control. More specifically, the present application relates to an intelligent monitoring system and method for a numerically controlled machine tool. Background Art

[0002] With the wide application of numerically controlled machine tools in the manufacturing industry, their processing efficiency and product quality directly affect the production cost and market competitiveness of enterprises. However, due to various complex factors involved in the processing process (such as material properties, processing technology, machine tool status, and human operation, etc.), various abnormal situations may occur during the actual operation of numerically controlled machine tools. Once faults such as tool breakage, spindle overload, or abnormal feed occur, it often leads to machine tool shutdown, deterioration of processing quality, and even equipment damage, causing economic losses and production delays to enterprises.

[0003] Currently, for the intelligent monitoring and early warning of the processing process of numerically controlled machine tools, most methods only focus on the monitoring of single dimensions such as the processing program or tool wear, or use manual experience for post-event investigation, lacking the fusion and real-time prediction of multi-source data.

[0004] Therefore, it is necessary to propose an intelligent monitoring system and method for a numerically controlled machine tool to solve at least some of the above problems. Summary of the Invention

[0005] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further elaborated in the Detailed Description section. The Summary of the Invention section of the present application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the protection scope of the claimed technical solution.

[0006] In a first aspect, the present application proposes an intelligent monitoring system for a numerically controlled machine tool, including:

[0007] A material attribute similarity acquisition module, configured to acquire the material attribute similarity information between a target processed part and a historical processed part;

[0008] A processing code similarity acquisition module, configured to acquire the code similarity information between the above-mentioned target processed part and the above-mentioned historical processed part;

[0009] A historical anomaly information acquisition module, configured to acquire the abnormal pause information of historical processed parts;

[0010] An early warning prompt information generation module, configured to generate early warning prompt information according to the above-mentioned material attribute similarity information, the above-mentioned code similarity information, and the above-mentioned abnormal pause information;

[0011] An auxiliary function control module, configured to control the working state of the auxiliary function module according to the above-mentioned early warning prompt information.

[0012] Second aspect, the present application proposes an intelligent monitoring method for a numerically controlled machine tool, which is used for the intelligent monitoring system for a numerically controlled machine tool in the first aspect, and includes:

[0013] Obtain the material attribute similarity information between the target machined part and the historical machined part;

[0014] Obtain the code similarity information between the above-mentioned target machined part and the above-mentioned historical machined part;

[0015] Obtain the abnormal pause information of the historical machined part;

[0016] Generate a warning prompt message according to the above-mentioned material attribute similarity information, the above-mentioned code similarity information and the above-mentioned abnormal pause information;

[0017] Control the working state of the auxiliary function module according to the above-mentioned warning prompt message.

[0018] In a feasible implementation manner, the obtaining of the material attribute similarity information between the target machined part and the historical machined part includes:

[0019] Determine the material static similarity information according to the hardness information, density information and thermal conductivity of the above-mentioned target machined part and the above-mentioned historical machined part;

[0020] Determine the material dynamic similarity information according to the cutting speed, feed rate, tool wear amount and surface roughness of the above-mentioned target machined part and the above-mentioned historical machined part;

[0021] Determine the above-mentioned material attribute similarity information according to the above-mentioned static similarity information and the above-mentioned dynamic similarity information.

[0022] In a feasible implementation manner, the obtaining of the code similarity information between the above-mentioned target machined part and the above-mentioned historical machined part includes:

[0023] Extract the key instructions in the G code to identify the process type;

[0024] Judge the machining stage according to the M code and the tool information;

[0025] Determine the static weight coefficient according to the above-mentioned process type;

[0026] Determine the stage sensitivity factor according to the above-mentioned machining stage;

[0027] Determine the dynamic weight coefficient according to the above-mentioned static weight coefficient and the above-mentioned stage sensitivity factor;

[0028] Extract the key machining features according to different above-mentioned process types;

[0029] Calculate the above code similarity information based on the above dynamic weight coefficients of the target processed part and the above historical processed part and the above key processing features.

[0030] In a feasible implementation manner, the above abnormal pause information includes equipment abnormal pause information and manual emergency pause information.

[0031] Generate a warning prompt message based on the above material attribute similarity information, the above code similarity information, and the above abnormal pause information, including:

[0032] Classify the above abnormal pause information to obtain an equipment abnormal pause information set and a manual emergency pause information set.

[0033] Generate equipment abnormal warning information based on the above material attribute similarity information and the above code similarity information in the above equipment abnormal pause information set.

[0034] Generate manual pause prompt information based on the above material attribute similarity information and the above code similarity information in the above manual emergency pause information set.

[0035] In a feasible implementation manner, the above generating equipment abnormal warning information based on the above material attribute similarity information and the above code similarity information in the above equipment abnormal pause information set includes:

[0036] Calculate the first processing condition similarity information based on the above material attribute similarity information and the above code similarity information in the above equipment abnormal pause information set.

[0037] In the case that there is equipment possible abnormal information with the first processing condition similarity information greater than the first preset threshold in the above equipment abnormal pause information set, obtain the first abnormal statistical information of all the above equipment possible abnormal information.

[0038] Generate equipment abnormal warning information based on the above first abnormal statistical information.

[0039] In a feasible implementation manner, the above abnormal statistical information includes abnormal occurrence frequency, abnormal severity, and abnormal category statistical information.

[0040] The above generating equipment abnormal warning information based on the above abnormal statistical information includes:

[0041] Calculate the abnormal risk score based on the above abnormal occurrence frequency, abnormal severity, and the above abnormal category statistical information.

[0042] Generate equipment abnormal warning information based on the above abnormal risk score.

[0043] In a feasible implementation manner, controlling the working state of the auxiliary function module according to the above warning prompt information includes:

[0044] Obtaining the equipment abnormal risk level information of the above equipment abnormal warning information;

[0045] When the above abnormal risk level information is of low risk, adjusting the sampling frequency of the data monitoring module; and / or,

[0046] When the above abnormal risk level is of medium risk, determining the high-probability abnormal category based on the above abnormal category statistical information;

[0047] Adjusting the processing parameters based on the above high-probability abnormal category and the above abnormal risk level information; and / or,

[0048] When the above abnormal risk level is of high risk, starting the machine tool emergency stop program.

[0049] In a feasible implementation manner, the above artificial pause prompt information in the above artificial emergency pause information set according to the above material attribute similarity information and the above code similarity information includes:

[0050] Calculating the second processing condition similarity information in the above artificial emergency pause information set according to the above material attribute similarity information and the above code similarity information;

[0051] When there is an artificial pause prompt information in the above equipment abnormal pause information set where the second processing condition similarity information is greater than the second preset threshold, obtaining the second abnormal statistical information of all the above artificial pause prompt information;

[0052] Generating artificial pause prompt information according to the above second abnormal statistical information.

[0053] In a feasible implementation manner, controlling the working state of the auxiliary function module according to the above warning prompt information includes:

[0054] Controlling the coolant flow information and the light prompt information according to the above artificial pause prompt information.

[0055] In summary, this application takes into account both the similarity of material properties and the similarity of processing codes, and combines historical abnormal pause information to conduct a comprehensive comparison of the target processing parts and historical processing parts in different dimensions. Compared with the traditional monitoring method based only on single factors such as process codes or tool wear, it can more accurately capture potential anomalies in complex processing environments, identify high-risk parts and failure modes, and thus significantly improve the accuracy of abnormal warnings. Through the mining and analysis of high-similarity historical abnormal cases, the system can predict the possible abnormalities of the target parts at the beginning or during processing and output "early warning prompt information". Compared with traditional manual experience or post-event investigation mode, the method of this application can detect potential fault signs earlier, reduce economic losses caused by unexpected machine tool shutdowns and tool scrapping, and improve the start-up rate and production efficiency of CNC machine tools. After generating the "early warning prompt information", the method proposed in this application will link the auxiliary function module to control and adjust the cooling flow, feed speed, alarm signal, etc. according to the early warning information. Compared with the traditional method that relies on manual operation, it can automatically adjust process parameters during machine tool operation and prompt operators to check key links, greatly reducing the reliance on manual experience and enhancing the safety and automation level of machine tools. Therefore, this application constructs a more complete abnormal warning and auxiliary function linkage mechanism through a multi-dimensional comprehensive analysis of material property similarity, processing code similarity, and historical abnormal information, which can improve the accuracy and timeliness of prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0057] Figure 1 A structural schematic diagram of an intelligent monitoring system for a CNC machine tool provided in an embodiment of the present application;

[0058] Figure 2 A flowchart of an intelligent monitoring method for a CNC machine tool provided in an intelligent monitoring system for a CNC machine tool according to an embodiment of the present application. DETAILED DESCRIPTION

[0059] In the description and claims of this application and the above-mentioned drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices. The technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments.

[0060] Please refer to Figure 1 , which is a structural schematic diagram of the intelligent monitoring system 10 for a numerically controlled machine tool provided by an embodiment of this application. Specifically, it may include:

[0061] A material property similarity acquisition module 101, configured to acquire material property similarity information between a target machined part and a historical machined part;

[0062] A machining code similarity acquisition module 102, configured to acquire code similarity information between the above-mentioned target machined part and the above-mentioned historical machined part;

[0063] A historical anomaly information acquisition module 103, configured to acquire abnormal pause information of historical machined parts;

[0064] An early warning prompt information generation module 104, configured to generate early warning prompt information according to the above-mentioned material property similarity information, the above-mentioned code similarity information, and the above-mentioned abnormal pause information;

[0065] An auxiliary function control module 105, configured to control the working state of the auxiliary function module according to the above-mentioned early warning prompt information.

[0066] Exemplarily, the material property similarity acquisition module 101 acquires the respective material characteristic parameters (such as hardness, density, thermal conductivity, etc.) of the target machined part and the historical machined part, and performs a comparison operation, thereby obtaining "material property similarity information" reflecting the similarity degree between the two in terms of materials. When performing anomaly prediction or process optimization, material characteristics often directly affect key factors such as tool wear and cutting parameter selection.

[0067] The machining code similarity acquisition module 102 parses the numerical control programs (such as G codes, M codes, tool calls, etc.) of the target machined part and the historical machined part respectively, extracts the corresponding process feature information (such as process type, machining stage, feed rate, tool path, etc.), and then calculates the similarity of this information to obtain the "code similarity information". This can help the system evaluate the similarity degree of the target part and the historical part in terms of machining strategy and machining sequence.

[0068] The historical anomaly information acquisition module 103 collects and collates the "abnormal pause information" from the historical machining data, including the shutdown triggered by the automatic alarm of the equipment (such as spindle overload, tool breakage alarm, etc.) and the emergency stop triggered manually. Through these anomaly records, the faults or abnormal situations that occurred under similar working conditions can be understood.

[0069] The early warning prompt information generation module 104 comprehensively evaluates the possible anomalies or risks of the target machined part under the current working conditions based on the "material property similarity information", "code similarity information", and "abnormal pause information", and then outputs the corresponding "early warning prompt information". If the system determines that the target part is highly similar to a certain historical part in terms of material and process and the historical part has frequently experienced shutdowns, a high-risk prompt will be issued to remind of preventive measures.

[0070] The auxiliary function control module 105 controls the auxiliary functions of the CNC machine tool (such as the cooling and lubrication system, feed adjustment, alarm device, etc.) according to the early warning prompt information. The system may actively increase the coolant flow rate, reduce the feed rate, or turn on the prompt lights and display information to reduce tool or machine tool failures and ensure machining safety and stability.

[0071] In summary, this application takes into account both the similarity of material properties and the similarity of processing codes, and combines historical abnormal pause information to conduct a comprehensive comparison of the target processing parts and historical processing parts in different dimensions. Compared with the traditional monitoring method based only on single factors such as process codes or tool wear, it can more accurately capture potential anomalies in complex processing environments, identify high-risk parts and failure modes, and thus significantly improve the accuracy of abnormal warnings. Through the mining and analysis of high-similarity historical abnormal cases, the system can predict the possible abnormalities of the target parts at the beginning or during processing and output "early warning prompt information". Compared with traditional manual experience or post-event investigation mode, the method of this application can detect potential fault signs earlier, reduce economic losses caused by unexpected machine tool shutdowns and tool scrapping, and improve the start-up rate and production efficiency of CNC machine tools. After generating the "early warning prompt information", the method proposed in this application will link the auxiliary function module to control and adjust the cooling flow, feed speed, alarm signal, etc. according to the early warning information. Compared with the traditional method that relies on manual operation, it can automatically adjust process parameters during machine tool operation and prompt operators to check key links, greatly reducing the reliance on manual experience and enhancing the safety and automation level of machine tools. Therefore, this application constructs a more complete abnormal warning and auxiliary function linkage mechanism through a multi-dimensional comprehensive analysis of material property similarity, processing code similarity, and historical abnormal information, which can improve the accuracy and timeliness of prediction.

[0072] In a second aspect, the present application proposes an intelligent monitoring method for a CNC machine tool, which is used in the intelligent monitoring system 10 for a CNC machine tool of the first aspect, comprising:

[0073] S110, obtaining material property similarity information of the target processed part and the historical processed part;

[0074] Exemplarily, the system mainly collects the physical characteristics of the target parts and historical parts (such as hardness, density, thermal conductivity, etc.), and compares these material properties to obtain material property similarity information. This similarity is used to determine the degree of matching between the target parts and historical parts in terms of material properties, providing a basis for subsequent judgment of fault risks.

[0075] S120, obtaining code similarity information between the target processed part and the historical processed part;

[0076] For example, by analyzing the NC program (G code, M code, etc.), the target part and the historical part are compared in terms of processing technology, process path, processing stage, etc., and the code similarity information is calculated. This information reflects their proximity in processing strategy and can provide a reference for fault prediction or process optimization.

[0077] S130, obtaining abnormal pause information of historically processed parts;

[0078] Exemplarily, the system retrieves the records related to "abnormal suspension" from the historical database, including equipment abnormalities (the machine tool automatically alarms and shuts down) and manual emergency suspensions (the operator manually shuts down the machine), etc. Through sorting and classification, this data is used to locate possible faults or problems under similar working conditions.

[0079] S140. Generate a warning prompt message based on the above material attribute similarity information, the above code similarity information, and the above abnormal suspension information;

[0080] Exemplarily, the system comprehensively evaluates the risks that may occur to the target machined part by considering the material similarity, code similarity, and abnormal suspension records in historical machining, and outputs the corresponding "warning prompt message". For example, if the target part is highly similar to a certain historical part in terms of both material and process, and this historical part has repeatedly experienced abnormalities such as tool breakage and spindle overload, the system will prompt a relatively high potential risk.

[0081] S150. Control the working state of the auxiliary function module according to the above warning prompt message.

[0082] Exemplarily, if the system determines that there is an abnormal risk, it will control the auxiliary function modules of the CNC machine tool (such as the cooling system, tool management module, lighting warning, etc.) and take corresponding measures to prevent or mitigate possible faults. For example, actively increasing the coolant flow rate, reducing the feed rate, prompting the operator to check the tool condition, etc., thereby reducing the possibility of downtime and damage.

[0083] In summary, this application takes into account both the similarity of material properties and the similarity of processing codes, and combines historical abnormal pause information to conduct a comprehensive comparison of the target processing parts and historical processing parts in different dimensions. Compared with the traditional monitoring method based only on single factors such as process codes or tool wear, it can more accurately capture potential anomalies in complex processing environments, identify high-risk parts and failure modes, and thus significantly improve the accuracy of abnormal warnings. Through the mining and analysis of high-similarity historical abnormal cases, the system can predict the possible abnormalities of the target parts at the beginning or during processing and output "early warning prompt information". Compared with traditional manual experience or post-event investigation mode, the method of this application can detect potential fault signs earlier, reduce economic losses caused by unexpected machine tool shutdowns and tool scrapping, and improve the start-up rate and production efficiency of CNC machine tools. After generating the "early warning prompt information", the method proposed in this application will link the auxiliary function module to control and adjust the cooling flow, feed speed, alarm signal, etc. according to the early warning information. Compared with the traditional method that relies on manual operation, it can automatically adjust process parameters during machine tool operation and prompt operators to check key links, greatly reducing the reliance on manual experience and enhancing the safety and automation level of machine tools. Therefore, this application constructs a more complete abnormal warning and auxiliary function linkage mechanism through a multi-dimensional comprehensive analysis of material property similarity, processing code similarity, and historical abnormal information, which can improve the accuracy and timeliness of prediction.

[0084] In a feasible implementation manner, the above-mentioned obtaining of the material property similarity information of the target processed part and the historical processed part includes:

[0085] Determine material static similarity information based on the hardness information, density information and thermal conductivity of the target processed part and the historical processed part;

[0086] Determining material dynamic similarity information based on the cutting speed, feed rate, tool wear and surface roughness of the target machined part and the historical machined part;

[0087] The material property similarity information is determined according to the static similarity information and the dynamic similarity information.

[0088] Exemplarily, material static similarity measures the closeness between the target part and the historical part in physical properties, mainly including hardness (H), density (D) and thermal conductivity (K).

[0089] The similarity of individual material properties is calculated using normalized Euclidean distance or weighted cosine similarity:

[0090]

[0091] Where: H target ,Dtarget , K target is the hardness, density, and thermal conductivity of the target part, H history , D history , K history are the corresponding property values of the historical part, H max , D max , K max are the maximum values in this material category respectively.

[0092] Calculate the comprehensive static similarity S static :

[0093] S static = α H × S H + α D × S D + α K × S K

[0094] Where: α H , α D , α K are the weight coefficients, and α H + α D + α K = 1

[0095] The weights can be optimized according to process experience or data training. For example: If the material hardness has a greater impact on machining, then set α H = 0.5, α D = 0.3, α K = 0.2.

[0096] The material dynamic similarity measures the similarity between the target part and the historical part during the machining process, mainly considering the cutting speed (V), feed rate (F), tool wear (W), and surface roughness (R a )

[0097] Calculate the dynamic similarity using weighted cosine similarity or Euclidean distance:

[0098]

[0099] Where: V target , F target , W target , R a,target are the cutting speed, feed rate, tool wear, and surface roughness of the target part, V history , F history , W history , R a,history are the corresponding cutting speed, feed rate, tool wear, and surface roughness of the historical part, V max , F max,W max ,R a,max is the normalization coefficient.

[0100] Calculate the comprehensive dynamic similarity S dynamic :

[0101] S dynamic = β V × S V + β F × S F + β W × S W + β R × S R

[0102] Where: β V , β F , β W , β R are weight coefficients, and β V + β F + β W + β R = 1. For example, if tool wear has a greater impact on anomalies, β W can be set to 0.4, β V to 0.3, β F to 0.2, and β R to 0.1

[0103] The final material property similarity S mat is comprehensively calculated from the static similarity and the dynamic similarity:

[0104] S mat = γ static × S static + γ dynamic× S dynamic

[0105] Where: γ static and γ dynamic are fusion weights, and γ static + γ dynamic = 1. If the material physical properties are more important than the processing parameters, then γ static = 0.6, γ dynamic = 0.4. If the processing parameters have a greater impact on the performance, then γ static = 0.4, γ dynamic = 0.6.

[0106] In summary, this embodiment takes into account both the characteristics of the material itself and the key parameters in the processing process, ensuring that the similarity calculation is closer to the actual processing situation. The influence weights of physical properties and processing parameters can be adjusted according to process requirements to improve adaptability. Combining historical abnormal data, it is possible to predict equipment failures or processing problems that may occur in the target part, thereby improving production stability.

[0107] In a feasible implementation manner, obtaining the code similarity information between the target machining part and the historical machining part includes:

[0108] Extract the key instructions in the G code to identify the operation type;

[0109] Judge the processing stage according to the M code and tool information;

[0110] Determine the static weight coefficient according to the above operation type;

[0111] Determine the stage sensitivity factor according to the above processing stage;

[0112] Determine the dynamic weight coefficient according to the above static weight coefficient and the above stage sensitivity factor;

[0113] Extract the key processing features according to different above operation types;

[0114] Calculate the above code similarity information according to the above dynamic weight coefficient and the above key processing features between the target machining part and the historical machining part.

[0115] Exemplarily, a numerical control machining program contains multiple G codes, which determine the specific operations of machining. For example: G01: Linear interpolation (commonly used in milling and turning); G02 / G03: Circular interpolation (suitable for curve machining); G81 - G89: Drilling, tapping and other cyclic machining; G71 / G72: Rough turning and finish turning cycles.

[0116] Scan the G code line by line, extract the key instructions, and identify the operation type. For example: G01 represents linear milling, G02 / GO3 represents circular machining, G81 represents drilling, and G84 represents tapping.

[0117] Represent the machining process as an operation vector:

[0118] V op =[x1,x2,…,x n

[0119] where x i represents the i-th operation (such as linear milling, drilling, etc.).

[0120] ​M codes are used to control the machine tool status (such as spindle start / stop, coolant control), which can help determine the machining stage. For example: M03 (spindle forward rotation), M08 (coolant on) → rough machining; M05 (spindle stop), M09 (coolant off) → non-machining stage; M06 (tool change) → tool change information; T01, T02. (tool number) → identify the tool used.

[0121] For example: Rough machining has a large feed rate and high cutting depth, and the code features are: M03 + G71 / G72 (rough turning), G01 (rapid feed). Finish machining has a small feed rate and high precision, and the code features are G01 low feed speed and M08 coolant on.

[0122] Construct a machining stage vector, and set the machining stage vector V stage :

[0123] V stage = [s1, s2, …, s n

[0124] where s i represents each stage (such as rough machining, finish machining).

[0125] Different machining processes have different impacts on the overall similarity. Therefore, set the static weight coefficient W static : Drilling W drill = 0.4, Milling W mill = 0.3, Turning W turn = 0.2, Tapping W tap = 0.1.

[0126] Comprehensively calculate the static weight coefficient W static :

[0127]

[0128] where: is the proportion of this process in the machining program, is the static weight of this process

[0129] Define the stage sensitivity factor S stage :

[0130]

[0131] where: is the proportion of this stage, is the sensitivity of this stage (rough machining = 1.5, finish machining = 1.2)

[0132] Determine the dynamic weight coefficient W based on the above static weight coefficient and the above stage sensitivity factor dynamic :

[0133] W dynamic = W static × S stage

[0134] The key processing features include: path complexity (coordinate point density, tool path change rate), cutting parameters (feed rate, spindle speed, etc.), and processing time (the proportion of the total time occupied by different processes).

[0135] Construct the feature vector:

[0136] V feature = [f1, f2, …, f n

[0137] Where: f1 is the path complexity, f2 is the cutting parameter, and f3 is the processing time.

[0138] Calculate the code similarity S between the target part and the historical part based on the cosine similarity code :

[0139]

[0140] Among them, V feature,i is the key feature vector of the target part, is the corresponding feature vector of the historical part, and W dynamic,i is the calculated dynamic weight.

[0141] In summary, the method proposed in this embodiment combines G code, M code, and tool information, can accurately judge the processing stage, and improve the accuracy of similarity calculation. By adjusting the influence of different processing steps through the process type and processing stage sensitivity factors, the matching degree of complex processing programs is improved. Combining process parameters such as tool path, feed rate, and spindle speed makes the code similarity calculation more in line with the actual processing situation. It can improve the intelligence of machine tool monitoring, detect possible processing abnormalities in advance, and improve the safety and production efficiency of the machine tool.

[0142] In a feasible implementation manner, the above abnormal pause information includes equipment abnormal pause information and manual emergency pause information.

[0143] Generating warning prompt information based on the above material attribute similarity information, the above code similarity information, and the above abnormal pause information includes:

[0144] Classify the above abnormal pause information to obtain an equipment abnormal pause information set and a manual emergency pause information set;

[0145] Generate equipment abnormal warning information based on the above material attribute similarity information and the above code similarity information in the above equipment abnormal pause information set;

[0146] ​Manually pause the prompt message in the above-mentioned manual emergency pause information set according to the above-mentioned material attribute similarity information and the above-mentioned code similarity information.

[0147] Exemplarily, multiple abnormal pause records are stored in the historical processing database, and the records include:

[0148] 1. Equipment abnormal pause (R eqp ): Abnormalities automatically detected by the system, such as spindle overload, tool breakage, etc.

[0149] 2. Manual emergency pause (R man ): Emergency shutdown triggered manually, such as when the operator discovers abnormal tool or vibration.

[0150] In order to determine whether historical abnormalities may occur in the target part, it is first necessary to calculate the processing condition similarity S cond (Both the first processing condition similarity information and the second processing condition similarity information are determined by this formula):

[0151] S cond = αS mat + (1 - α)S code

[0152] Where: S mat is the material attribute similarity, which measures the closeness between the target part and the historical part in terms of material hardness, density, thermal conductivity, etc. S code is the code similarity, which measures the similarity between the target part and the historical part in terms of processing technology, process, cutting parameters, etc.

[0153] In a feasible implementation manner, generating the equipment abnormal warning information according to the above-mentioned material attribute similarity information and the above-mentioned code similarity information in the above-mentioned equipment abnormal pause information set includes:

[0154] Calculating the first processing condition similarity information according to the above-mentioned material attribute similarity information and the above-mentioned code similarity information in the above-mentioned equipment abnormal pause information set;

[0155] When there is equipment possible abnormal information with the first processing condition similarity information greater than the first preset threshold in the above-mentioned equipment abnormal pause information set, obtaining the first abnormal statistical information of all the above-mentioned equipment possible abnormal information;

[0156] Generating equipment abnormal warning information according to the above-mentioned first abnormal statistical information.

[0157] Exemplarily, when the processing condition similarity S cond is greater than the first preset threshold τ1, it means that the processing condition of the target part is highly similar to the historical processing record, and it is necessary to check whether there are abnormal pause records in the historical data for this condition.

[0158] If S cond > τ1, it enters the first abnormal statistical information, where τ1 is the first preset threshold for the similarity of processing conditions (such as 0.7).

[0159] For the selected historical abnormal processing cases, count their abnormal data, including:

[0160] 1. Abnormal occurrence frequency R eqp :

[0161]

[0162] Where M is the number of cases with equipment abnormalities in the historical data of similar working conditions, and N is the total number of processing times under similar working conditions.

[0163] 2. Abnormal severity R sev :

[0164]

[0165] For example, slight tool wear (weight = 1), tool breakage (weight = 3), spindle overload (weight = 5).

[0166] 3. Abnormal category statistics

[0167]

[0168] Based on the comprehensive abnormal statistical information, calculate the equipment abnormal risk score R risk :

[0169] In a feasible implementation manner, the above abnormal statistical information includes abnormal occurrence frequency, abnormal severity, and abnormal category statistical information.

[0170] The above-mentioned generation of equipment abnormal warning information based on the above abnormal statistical information includes:

[0171] Calculate the abnormal risk score according to the above abnormal occurrence frequency, abnormal severity, and the above abnormal category statistical information;

[0172] Generate equipment abnormal warning information according to the above abnormal risk score.

[0173] Exemplarily, calculate the abnormal risk score R risk :

[0174]

[0175] Where: λ1, λ2, λ3 are adjustable weight parameters (such as λ1 = 0.5, λ2 = 0.3, λ3 = 0.2). Tool breakage W 刀具 = 3.0, spindle overload W主轴 = 5.0, servo alarm W 伺服 = 4.0.

[0176] In a feasible implementation manner, controlling the working state of the auxiliary function module according to the above warning prompt information includes:

[0177] Obtaining the equipment abnormal risk level information of the above equipment abnormal warning information;

[0178] When the above abnormal risk level information is of low risk, adjusting the sampling frequency of the data monitoring module; and / or,

[0179] When the above abnormal risk level is of medium risk, determining the high-probability abnormal category based on the above abnormal category statistical information;

[0180] Adjusting the processing parameters based on the above high-probability abnormal category and the above abnormal risk level information; and / or,

[0181] When the above abnormal risk level is of high risk, starting the machine tool emergency stop program.

[0182] Exemplarily, setting the threshold τ of different risk levels. When the abnormal risk level information is of low risk (R risk <1.0), giving a mild warning, adjusting the sampling frequency of the data monitoring module, and closely monitoring the working state of the machine tool. When the above abnormal risk level is of medium risk (1.0 ≤ R risk <2.0), determining the high-probability abnormal category based on the above abnormal category statistical information, and determining the high-probability abnormal category based on the above abnormal category statistical information. When the above abnormal risk level is of high risk (R risk ≥ 2.0), starting the machine tool emergency stop program.

[0183] Specifically, 1. When the high-probability abnormal category is tool breakage warning and the spindle load is too high, reducing the feed rate F:

[0184] F ′ = F × (1 - ΔF)

[0185] Let ΔF be the adjustment amplitude, usually taking 5% - 15%.

[0186] Reducing the spindle speed N:

[0187] N ′ = N × (1 - ΔN)

[0188] If the warning indicates spindle overload, the spindle speed can be reduced to reduce the load.

[0189] 2. When the high-probability abnormal category is to optimize the cooling / lubrication system, increasing the coolant flow rate:

[0190] Q c ′ oolant = Q coolant ×(1 + ΔQ)

[0191] where ΔQ is the adjustment range.

[0192] The nozzle direction can also be adjusted to improve the cooling efficiency. If the system is equipped with a controllable nozzle, the coolant flow direction can be dynamically optimized. Increase the cutting oil concentration

[0193] In a feasible implementation manner, the artificial pause prompt information in the above artificial emergency pause information set according to the above material attribute similarity information and the above code similarity information includes:

[0194] Calculate the second processing condition similarity information in the above artificial emergency pause information set according to the above material attribute similarity information and the above code similarity information;

[0195] When there is artificial pause prompt information in the above equipment abnormal pause information set where the second processing condition similarity information is greater than the second preset threshold, obtain the second abnormal statistical information of all the above artificial pause prompt information;

[0196] Generate artificial pause prompt information according to the above second abnormal statistical information.

[0197] Exemplarily, when the processing condition similarity S cond is greater than the second preset threshold τ2, it means that the processing condition of the target part is highly similar to the historical processing record, and it is necessary to check whether there is artificial pause prompt information for this condition in the historical data.

[0198] If S cond > τ2, then enter the second abnormal statistical information, and τ2 is the second preset threshold of the processing condition similarity (such as 0.9).

[0199] The second abnormal statistical information includes: the number R of artificial pause prompt information man and the time distribution information of the artificial pause prompt information and the parameter modification information after the artificial pause.

[0200] Statistical time distribution information of artificial pause prompt information, such as: at the beginning stage of processing (such as 0% - 10% of the processing time): may be caused by incorrect initial settings. In the middle stage of processing (such as 30% - 70% of the processing time): may be due to excessive feed rate, rapid tool wear, etc. In the later stage of processing (such as more than 80%): may be due to the decline in cutting quality or the approaching failure of the tool.

[0201] The parameter modification information ΔP after the artificial pause = Pafter - Pbefore, where Pafter and P before are the processing parameters adjusted before and after the pause, respectively.

[0202] In a feasible implementation manner, controlling the working state of the auxiliary function module according to the above warning prompt information includes:

[0203] Controlling the coolant flow information and the light prompt information according to the above manual pause prompt information.

[0204] Exemplarily, the system combines the number R of manual pause prompt messages man , the pause time distribution information, and the parameter adjustment ΔP, and adopts the hierarchical control strategy shown in Table 1:

[0205]

[0206] Table 1

[0207] In this embodiment, through data-driven coolant adjustment and light prompt, abnormal shutdowns are reduced, and the stability of the CNC machine tool is improved. The coolant adjustment strategy is optimized to reduce the processing temperature, improve the tool life and the processing accuracy. Yellow / red lights are used for warning to enhance the operator's abnormal response speed and reduce the risk of misadjustment. The processing parameters are automatically adjusted to improve the automation level of the CNC machine tool, which is applicable to large-scale production scenarios.

[0208] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. An intelligent monitoring system for a CNC machine tool, characterized in that: include: A material property similarity acquisition module is used to obtain the material property similarity information between the target processed part and the historical processed part; A processing code similarity acquisition module, used to acquire code similarity information between the target processing part and the historical processing part; A historical abnormal information acquisition module is used to obtain abnormal pause information of historical processed parts; A warning prompt information generating module, used for generating warning prompt information according to the material property similarity information, the code similarity information and the abnormal suspension information; The auxiliary function control module is used to control the working state of the auxiliary function module according to the early warning prompt information.

2. An intelligent monitoring method for a CNC machine tool, used in the intelligent monitoring system for a CNC machine tool according to claim 1, characterized in that: include: Obtaining the material property similarity information between the target processed parts and the historical processed parts; Acquire code similarity information between the target processed part and the historical processed part; Get abnormal pause information of historically processed parts; generating early warning prompt information according to the material property similarity information, the code similarity information and the abnormal suspension information; The working state of the auxiliary function module is controlled according to the early warning prompt information.

3. The intelligent monitoring method for CNC machine tools according to claim 2, characterized in that: The obtaining of the material property similarity information between the target processed part and the historical processed part includes: Determining material static similarity information according to hardness information, density information and thermal conductivity of the target processed part and the historical processed part; Determining material dynamic similarity information according to the cutting speed, feed rate, tool wear and surface roughness of the target processed part and the historical processed part; The material property similarity information is determined according to the static similarity information and the dynamic similarity information.

4. The intelligent monitoring method for a CNC machine tool according to claim 2, characterized in that: The obtaining of code similarity information between the target processed part and the historical processed part includes: Extract key instructions from G-code to identify the process type; Determine the processing stage based on the M code and tool information; Determining a static weight coefficient according to the process type; determining a stage sensitivity factor according to the processing stage; Determine a dynamic weight coefficient according to the static weight coefficient and the stage sensitivity factor; Extract key processing features according to different types of the process; The code similarity information is calculated according to the dynamic weight coefficient and the key processing feature of the target processing part and the historical processing part.

5. The intelligent monitoring method for a CNC machine tool according to claim 2, characterized in that: The abnormal suspension information includes equipment abnormal suspension information and manual emergency suspension information. The generating of the warning prompt information according to the material property similarity information, the code similarity information and the abnormal suspension information includes: Classifying the abnormal suspension information to obtain a device abnormal suspension information set and a manual emergency suspension information set; generating equipment abnormality warning information according to the material property similarity information and the code similarity information in the equipment abnormality suspension information set; In the manual emergency pause information set, manual pause prompt information is provided according to the material attribute similarity information and the code similarity information.

6. The intelligent monitoring method for a CNC machine tool according to claim 5, characterized in that: The generating of equipment abnormality warning information according to the material property similarity information and the code similarity information in the equipment abnormality suspension information set includes: Calculating first processing condition similarity information in the equipment abnormal suspension information set according to the material property similarity information and the code similarity information; In the case where there is possible abnormal information of equipment whose first processing condition similarity information is greater than a first preset threshold in the equipment abnormal suspension information set, obtaining first abnormal statistical information of all possible abnormal information of the equipment; Generate equipment abnormality warning information according to the first abnormality statistical information.

7. The intelligent monitoring method for a CNC machine tool according to claim 6, characterized in that: The abnormal statistical information includes abnormal occurrence frequency, abnormal severity and abnormal category statistical information, The generating device abnormality warning information according to the abnormal statistical information includes: Calculate an abnormality risk score according to the abnormality occurrence frequency, abnormality severity and the abnormality category statistical information; Generate equipment abnormality warning information according to the abnormality risk score.

8. The intelligent monitoring method for a CNC machine tool according to claim 7, characterized in that: The controlling the working state of the auxiliary function module according to the early warning prompt information includes: Obtaining device abnormality risk level information of the device abnormality warning information; When the abnormal risk level information is low risk, adjusting the sampling frequency of the data monitoring module; and / or, When the abnormal risk level is medium risk, determining a high probability abnormal category based on the abnormal category statistical information; Adjusting processing parameters based on the high probability abnormality category and the abnormality risk level information; and / or, When the abnormal risk level is high risk, the machine tool emergency stop program is started.

9. The intelligent monitoring method for a CNC machine tool according to claim 5, characterized in that: The manual emergency pause prompt information according to the material property similarity information and the code similarity information in the manual emergency pause information set includes: Calculating the second processing condition similarity information in the manual emergency pause information set according to the material property similarity information and the code similarity information; In the case where there is manual pause prompt information whose second processing condition similarity information is greater than a second preset threshold in the equipment abnormal pause information set, obtaining second abnormal statistical information of all the manual pause prompt information; Generate manual pause prompt information according to the second abnormal statistical information.

10. The intelligent monitoring method for a CNC machine tool according to claim 9, characterized in that: The controlling the working state of the auxiliary function module according to the early warning prompt information includes: The coolant flow information and the light prompt information are controlled according to the manual pause prompt information.

Images