Adaptive real-time tool breakage detection method based on spindle current in numerical control machine tool

Abstract

The self-adaptive real-time tool breakage detection method based on the spindle current in the numerical control machine tool is suitable for an economical numerical control system without high-end sensor configuration. According to the method, a spindle current signal is obtained through a low-sampling-rate current sensor, and real-time online detection of a tool breaking event is achieved in combination with machining state recognition, historical template comparison and normalized correlation coefficient calculation. The system accurately recognizes the starting point of a machining section through a machining state alignment mechanism of'coarse alignment + fine correction ', and recognizes two typical cutter breaking behaviors of'load loss type' and'contact abnormal type 'in combination with current fluctuation characteristics. And meanwhile, a self-adaptive temperature drift calibration and historical template self-updating strategy is designed, so that the long-term stability and environment adaptability of the system are improved. The method has the advantages of being low in cost, high in robustness and short in response time, and is particularly suitable for tool breakage fault detection and intelligent upgrading of small and medium-sized machining workshops.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring and fault identification in the numerical control machining process, and specifically to an adaptive real-time tool breakage detection method based on spindle current in a numerically controlled machine tool. Background Art

[0002] Traditional tool breakage detection methods mostly rely on high-frequency vibration sensors, force sensors or vision systems. Although they have high precision, they are costly, complex to maintain, and difficult to popularize in machining scenarios with limited resources. And some tool breakage detection technologies relying on current signals have problems such as inaccurate recognition of machining states, fixed thresholds, and insufficient temperature drift compensation, resulting in frequent false positives and false negatives, and unable to meet the actual industrial needs. Therefore, there is an urgent need for a tool breakage detection method with high robustness, low cost, automatic adaptation to new working conditions, and the ability to adapt to fluctuations in machining conditions.

[0003] The comparison between the present application and the prior art is as follows:

[0004] Technical comparison with the application text CN105619178A "A real-time tool breakage detection method for a numerically controlled machine tool";

[0005] 1. The application text CN105619178A discloses a real-time tool breakage detection method based on the change law of the average current of the spindle motor, specifically as follows: Step 1, the numerical control system reads the normal machining current mean value and no-load current mean value saved in the tool process database; Step 2, set the current sampling timer; Step 3, take every L sampling points as an analysis segment, and calculate the spindle current mean value of each analysis segment respectively; Step 4, continuously compare the change of the current mean values of three adjacent analysis segments in a sliding window manner, and combine a fixed proportional coefficient to judge whether the tool breakage condition is met; Step 5, if the tool breakage condition is met, trigger an alarm instruction and abort the machining program. This method is integrated in the numerical control system and uses the time-domain current characteristics of the spindle for judgment. It is applicable to rough machining scenarios with large changes in spindle current and has the advantage of not requiring additional sensors. Its focus is on realizing the real-time detection of tool breakage events through the change trend of the average value.

[0006] 2. This application proposes an adaptive real-time self-breaking tool detection method based on spindle current. This method collects low-frequency spindle current signals through a series-connected low-cost current transformer, and combines modules such as spindle start-stop mutation recognition, historical template matching alignment, and fluctuation trend judgment to achieve high-precision time alignment of the current signal and historical data, identify the starting point of the machining section under multi-tool and multi-process conditions, and construct the aligned current sequence. Two types of judgment models are used to detect "load-loss type" and "abnormal contact type" tool breakage behaviors respectively; at the same time, a temperature drift calibration and historical template self-update mechanism is designed to adapt to changes in factors such as current drift and tool wear during the machining process, and to achieve long-term stable operation of the system. This method does not rely on high-frequency sampling or complex sensor systems, and has the characteristics of high robustness, self-updateability, and low-cost deployment. Its focus is on high-precision alignment at the machining section level and classification recognition of typical tool breakage types, and it improves the environmental adaptability under complex working conditions.

[0007] There are essential differences between the two in the tool breakage recognition mechanism, real-time data alignment granularity, and system design path.

[0008] Technical comparison with the application text CN114700802A "Tool Breakage Detection Method and Device";

[0009] 1. The application text CN114700802A discloses a tool breakage detection method and device, specifically: during the machining process of the machine tool, the current tool identification, machine tool current, and machining position are periodically obtained, and the sampling data of each tool is recorded in the corresponding file; at the end of the machining, for each tool file, the corresponding machining process is determined according to the machining position, and the load threshold corresponding to the process is calculated and recorded according to the machine tool current sampled in each process; tool breakage detection is carried out based on the load thresholds of each tool and each process recorded. This patent considers the automatic calculation problem of load thresholds in multi-tool and multi-process scenarios, aiming to improve the efficiency and accuracy of tool breakage detection. Its focus is on automatically dividing processes based on the machining position, and using historical sampling data (mainly the weighted average and maximum value of the current) to calculate static load thresholds for each process, and realizing tool breakage judgment for each tool and each process.

[0010] 2. This application proposes an adaptive real-time self-breaking tool detection method based on spindle current. This method collects low-frequency spindle current signals through a series-connected low-cost current transformer, and combines modules such as spindle start-stop mutation recognition, historical template matching alignment, and fluctuation trend judgment to achieve high-precision time alignment of the current signal and historical data, identify the starting point of the machining section under multi-tool and multi-process conditions, and construct the aligned current sequence. Two types of judgment models are used to detect "load-loss type" and "abnormal contact type" tool breakage behaviors respectively; at the same time, a temperature drift calibration and historical template self-update mechanism is designed to adapt to changes in factors such as current drift and tool wear during the machining process, and achieve long-term stable operation of the system. This method does not rely on high-frequency sampling or complex sensor systems, and has the characteristics of high robustness, self-updateability, and low-cost deployment. Its focus is on high-precision alignment at the machining section level and classification recognition of typical tool breakage types, and it improves the environmental adaptability under complex working conditions.

[0011] There are essential differences between the two in the tool breakage recognition mechanism, the principle of process division, and the environmental adaptation mechanism.

[0012] Technical comparison with the application text CN115781411A "Online Tool Breakage Detection Method, Device and System"

[0013] 1. The application text CN115781411A discloses an online tool breakage detection method, specifically: detecting the current no-load voltage value corresponding to the spindle no-load current and the real-time machining voltage value corresponding to the real-time machining current under the current machining conditions respectively; calculating the difference between the two to obtain the real-time voltage difference; obtaining a pre-set reference voltage threshold (determined based on the difference between the normal machining voltage and the no-load voltage); judging tool breakage by comparing the real-time voltage difference with the reference voltage threshold. This patent focuses on the voltage difference comparison mechanism, and uses the dynamic threshold of the voltage difference between the no-load and machining states to judge tool breakage. Its core technology lies in ensuring the reliability of the threshold through high-precision voltage sampling calibration, and optimizing the detection stability for multi-spindle and multi-board scenarios in PCB processing. The focus is on solving the problem of the acquisition accuracy of voltage signals, and making a binary judgment (whether there is tool breakage or not) through a pre-set fixed reference voltage threshold.

[0014] 2. This application proposes an adaptive real-time self-breaking tool detection method based on spindle current. This method collects low-frequency spindle current signals through a series-connected low-cost current transformer, and combines modules such as spindle start-stop mutation recognition, historical template matching alignment, and fluctuation trend judgment to achieve high-precision time alignment between the current signal and historical data, identify the starting point of the machining section under multi-tool and multi-process conditions, and construct the aligned current sequence. Two types of judgment models are used to detect "load-loss type" and "contact abnormal type" tool breakage behaviors respectively; at the same time, a temperature drift calibration and historical template self-update mechanism is designed to adapt to changes in factors such as current drift and tool wear during the machining process, and achieve the long-term stable operation of the system. This method does not rely on high-frequency sampling or complex sensor systems, and has the characteristics of high robustness, self-updateability, and low-cost deployment. Its focus is on high-precision alignment at the machining section level and classification recognition of typical tool breakage types, and improves the environmental adaptability under complex working conditions.

[0015] There are essential differences between the two in the core detection principle, real-time data alignment granularity, and environmental adaptation mechanism. Summary of the Invention

[0016] Aiming at the problems of high deployment cost, low machining state recognition accuracy, poor anti-environmental interference ability, and lack of adaptability in the existing numerical control machining tool breakage detection methods, the present invention proposes an adaptive real-time tool breakage detection method based on spindle current in a numerical control machine tool, aiming to achieve efficient and reliable recognition of tool breakage status under low-cost and low-sampling-rate conditions. By integrating the current change characteristics during the spindle start-stop process, the template matching alignment mechanism, the current fluctuation analysis and tool breakage abnormal mode recognition, and the self-update mechanism of the historical current template, the system can quickly and accurately detect two typical tool breakage types ("no-load tool breakage" and "contact abnormal"), and improve the intelligent monitoring ability of the numerical control system in a complex machining environment.

[0017] To achieve the above object, the technical solution adopted by the present invention is:

[0018] An adaptive real-time self-breaking tool detection method based on spindle current in a numerical control machine tool includes the following specific steps:

[0019] 1) Obtain the current signal of the spindle motor and transmit it to the industrial control system in real time;

[0020] 2) Obtain the temperature drift correction amount in the template library and calculate the calibrated current;

[0021] 3) Based on the current mutation during the spindle start-stop process, identify the rough machining start time ;

[0022] 4) Extract the machining state recognition template sequence of the corresponding tool in the historical machining data and the historical machining current ;

[0023] 5) Perform a sliding match search near , calculate the normalized cross - correlation coefficient NCC, obtain the optimal alignment time , and construct a calibrated and aligned machining current sequence ;

[0024] 6) Use the current fluctuation trend to identify the tool - break state, including the no - load type and the abnormal contact type;

[0025] 7) Automatically update the historical template library during the non - abnormal period to improve the model's adaptability.

[0026] As a further improvement of the present invention, the calculation formula of the normalized cross - correlation coefficient NCC in step 5) is:

[0027]

[0028] where is the template sequence, is the sliding window sequence, and are the and th elements of the sequences respectively, and are their means respectively, N is the length of the T and S sequences. When the maximum normalized cross - correlation coefficient NCC is lower than the threshold , the system records that the starting point recognition fails and skips the detection of this cycle.

[0029] As a further improvement of the present invention, the tool - break detection in the process of using the current fluctuation trend to identify the tool - break state in step 6) includes:

[0030] No - load type recognition: If the absolute value of the difference of the calibrated real - time current is continuously less than the first threshold, and the absolute value of the difference between the calibrated real - time current and the historical no - load value is continuously less than the second threshold, it is determined as a no - load type tool - break;

[0031] Abnormal contact type recognition: If the difference between the calibrated real - time current and the historical machining current continuously exceeds the third threshold, it is determined as an abnormal contact type tool - break.

[0032] As a further improvement of the present invention, the self - update mechanism steps of the template library in step 7) to update the historical template library are as follows:

[0033] During the initial machining under the new working conditions, record the current of one no-load machining and one normal machining, and obtain the no-load and load machining intervals of the historical current through difference comparison and incorporate them into the historical template library; when there is no tool breakage anomaly in the current machining cycle, the system automatically incorporates the current data of the current machining segment into the historical template library for reference in the next machining cycle.

[0034] The present invention includes the following key steps and modules:

[0035] 1. Current acquisition and preprocessing;

[0036] Install a low-cost current transformer in series in the spindle input circuit of the CNC machine tool. The acquisition frequency is about 30 Hz. The analog signal is transmitted to the industrial control host through the ADC module to construct a real-time current data stream. Use a circular buffer structure to achieve high-efficiency data reading and storage, avoiding data loss or processing delay.

[0037] 2. Machining segment starting point identification and alignment mechanism;

[0038] Combined with the sudden change characteristics of the spindle start-stop current, the system uses the "current sudden rise point" as the initial rough identification basis to obtain the machining start time . Construct a sliding window sequence near this time point and calculate the normalized correlation coefficient (NCC) with the machining state recognition template sequence . Select the maximum correlation point as the final accurate machining starting point . This mechanism effectively eliminates misalignment problems caused by spindle response delay, unstable machining rhythm, etc. The alignment accuracy can be controlled within one sampling period.

[0039] 3. Dynamic historical template modeling and updating mechanism;

[0040] The system initially collects the no-load and normal machining data of each tool for the first time to construct a machining current template library, and incorporates the load interval and the no-load interval , the reference current mean value and the machining waveform into it. To adapt to the current drift caused by the temperature drift of the machining environment and tool wear, a historical template dynamic update mechanism is introduced: if there is no anomaly in the current cycle, the current data of this machining segment is used to update the template, improving the robustness and accuracy of the system during long-term operation.

[0041] 4. Tool breakage state recognition model;

[0042] Based on the difference between the real-time current of the machining segment and the historical reference value, the system respectively identifies the following two types of tool breakage behaviors:

[0043] (1) Type Ⅰ: Loss-of-load type (the tool flies off after breaking), the differential fluctuation of the current decreases significantly, and the overall value is close to the no-load reference current, determining that the tool has broken and lost the load.

[0044] (2) Type Ⅱ: Abnormal contact type (the broken part of the tool remains in the workpiece), the real-time current is significantly higher than the historical machining value, and the fluctuation is intense, reflecting that the remaining parts of the broken tool are still in contact with the material and generating additional cutting resistance, which belongs to a hidden high-risk fault.

[0045] 5. Current correction parameter calibration mechanism;

[0046] During the long continuous machining cycle of the system, the collected current value is greatly affected by temperature. A temperature drift correction amount is introduced to calibrate the current value: when the machining cycle of a certain tool is completed normally, the average no-load section current of the current machining cycle is calculated according to the historical no-load interval. and The difference is used as the new temperature drift calibration correction amount to achieve continuous dynamic calibration of the current reference.

[0047] 6. Alarm and machine tool anti-control interface linkage;

[0048] After the system detects the tool breakage state, it can send an alarm signal to the CNC control unit to trigger the shutdown operation, realizing equipment-level linkage and avoiding further machining errors or equipment damage.

[0049] Beneficial effects: Compared with the prior art, the present invention has the following advantages and practical values:

[0050] (1) Low-cost deployment: There is no need for high-frequency current sensors, vibration sensors or vision systems. Only relying on low-speed current transformers can realize tool breakage detection, and the hardware cost is greatly reduced.

[0051] (2) High alignment accuracy and response speed: Combining event-driven and sliding template matching, the time alignment error of the machining section is as low as one sampling period, ensuring the accuracy of subsequent judgment logic.

[0052] (3) Strong adaptive anti-interference ability: By introducing a dynamic historical template library and a self-update mechanism, the system can adapt to the influences of temperature drift, tool wear and load changes, etc., and has the stability for long-term operation.

[0053] (4) Support for classification and recognition of typical tool breakage situations: It can accurately identify two types of tool breakage behaviors, namely loss-of-load type and contact load type, with a wide coverage rate, and both the misjudgment rate and the missed judgment rate are lower than 1%.

[0054] (5) Strong adaptability to the industrial field: The system has been successfully deployed on a conventional three-axis CNC platform, has good adaptability and real-time response performance, and is suitable for various small-batch high-frequency tool change scenarios. Brief Description of the Drawings

[0055] Figure 1 is the flowchart of the method of the present invention;

[0056] Figure 2 is the flowchart of the state recognition module of the present invention;

[0057] Figure 3 is the flowchart of the broken tool detection module of the present invention;

[0058] Figure 4 is the current curve when detecting a broken tool in the engineering application of the present invention; Among them, Figure (a) is the current curve in the case of a load-loss type broken tool in the abnormal state, and Figure (b) is the current curve in the case of a contact-abnormal type broken tool in the abnormal state;

[0059] Figure 5 is the flowchart of the template library update module of the present invention. Detailed Description of the Invention

[0060] The present invention will be further described in detail below in conjunction with the drawings and the specific embodiments:

[0061] Embodiment 1: System Composition and Deployment Method;

[0062] As shown in the attached Figure 1 , the broken tool detection system constructed by the present invention is applicable to the standard CNC machine tool platform, and the overall system includes the following six core function modules:

[0063] (1) Current acquisition module: Install a current transformer non-invasively on the input side of the spindle of the CNC machine tool, and set the sampling frequency to 30 Hz. The collected current signal is transmitted to the industrial control host in real time through an analog-to-digital converter (ADC) to achieve high-efficiency and low-latency data acquisition. The data structure adopts a circular buffer mechanism to ensure the continuity and real-time nature during the acquisition process.

[0064] (2) State recognition module: By detecting the change trend of the spindle start-stop current, accurately identify the start and end times of the machining section of the tool from "tool change → machining → tool change". Among them, the spindle current will mutate instantaneously at the start of machining, which is used as the key feature for identifying the start point of machining.

[0065] (3) Data matching and alignment module: Combine the dual strategies of start-stop mutation point recognition and sliding template matching to construct a machining start time recognition mechanism of "coarse alignment + fine correction". By calculating the normalized cross-correlation coefficient (NCC), select the optimal matching point , and achieve high-precision time alignment of the spindle machining current.

[0066] (4) Broken tool detection module: Based on the real-time current fluctuation characteristics during the machining process and the deviation from the historical template, it discriminates the tool breakage behavior. It respectively identifies "load-loss type" and "abnormal contact type" broken tools to improve the detection sensitivity and risk response ability.

[0067] (5) Dynamic template library update module: At the beginning of machining, the system needs to continuously collect two groups of reference data: no-load rotation current and current during normal loaded machining, and extract the load interval and no-load interval , and establish a standard current library including machining state recognition templates , no-load reference and historical machining current . On the premise that no abnormal tool breakage event is detected in the current machining cycle, the system automatically uses the machining current data for historical template update. Through the current correction parameter calibration mechanism, slow-changing factors such as temperature drift and machine tool aging are compensated to maintain the stability and accuracy of the model during long-term use.

[0068] The above modules can be integrated into the existing industrial control system, with low deployment cost and small reconstruction workload, and have good engineering feasibility and on-site promotion value.

[0069] Example 2: Machining section starting point identification and time alignment method;

[0070] The present invention proposes a fusion type spindle current alignment mechanism, and its steps are as follows:

[0071] Coarse alignment stage: As shown in the appendix Figure 2 , during the machining process, detect the sudden rise point of the spindle current. If the current of a number of consecutive sampling points exceeds the "state recognition threshold" for the first time, it is identified as the rough machining starting point .

[0072] Fine alignment stage:

[0073] (1) Extract a fixed-length template sequence from historical data (for example, 30 sampling points);

[0074] (2) For the current current sequence , starting from , construct a sliding window sequence , and calculate the normalized correlation coefficient (NCC) with within the range of 10 sampling points in turn:

[0075]

[0076] where is the template sequence, is the sliding window sequence, and are the and th elements of the sequences respectively, and and are their mean values respectively. N is the length of sequences T and S. When the maximum normalized cross-correlation coefficient NCC is lower than the threshold , the system records that the starting point recognition fails and skips the detection of this cycle.

[0077] (3) Find the matching position with the maximum correlation coefficient as the final machining starting point , and the current value at this point is used as the first value of the machining current sequence after alignment of the current tool ;

[0078] (4) If the maximum NCC is less than the preset threshold (such as 0.75), it is recorded as alignment failure, and this cycle detection can be skipped or manual intervention can be triggered.

[0079] This mechanism controls the alignment error of the machining starting point within 1 sampling cycle, which is significantly better than the traditional machining section recognition method based on CNC status codes. The latter is limited by communication delay and system refresh cycle, and the error is often up to hundreds of milliseconds.

[0080] Embodiment 3: Broken tool state recognition method;

[0081] As shown in the appendix Figure 3 , after the machining section recognition is completed, when the cutting machining enters the load interval, that is belonging to , the system compares the real-time current with the historical template , and recognizes the broken tool behavior according to the current change trend and deviation characteristics, which is divided into the following two typical modes:

[0082] (1) Type I: Unloaded broken tool (the broken part of the tool flies off);

[0083] It is manifested as a sudden drop in the spindle load, and the current quickly drops and tends to be stable. The judgment criteria are as follows:

[0084] ① The current difference is continuously lower than the first threshold for m consecutive times;

[0085] ② And the difference between and is lower than the second threshold

[0086] If the above conditions are met, it is determined that the tool has broken and is unloaded.

[0087] (2) Type II: abnormal contact and tool breakage (the broken part of the tool remains in the workpiece);

[0088] The manifestation is that the residual tool after the breakage is still in contact with the workpiece surface, causing the current to increase abnormally. The judgment logic is:

[0089] like - continuous Above the third threshold ; It is judged that there is abnormal contact load, which is a high-risk hidden tool breakage.

[0090] like Figure 4 , where Figure (a) is the current curve when the abnormal state is the load-loss type tool breakage, and Figure (b) is the current curve when the abnormal state is the contact abnormal type tool breakage. According to the attached figures, the curve comparison between the real-time current and the reference current of the two types of tool breakage can be seen; once the above two types of states are detected, the system will trigger the tool breakage alarm signal and link the CNC system to execute the shutdown operation.

[0091] Embodiment 4: Template update and temperature drift calibration mechanism;

[0092] Since the performance of electronic components such as ADC and current transformer is affected by temperature, the measured current value also shows a trend of changing with temperature. In order to improve the long-term stability of the system and avoid finding the complex nonlinear relationship between multiple components, temperature and current value offset to calibrate the current value, the present invention establishes the following adaptive update strategy, as shown in the attached Figure 5 :

[0093] (1) At the beginning of processing, the system needs to collect two sets of reference data: no-load rotation current and normal load processing current, and complete it within 2 hours. Extract the processing section and no-load section data respectively, and establish a processing state recognition template , no-load reference , Load range , No-load section , Historical cutting temperature drift correction And historical processing current The standard current library.

[0094] (2) If the machining cycle of a single tool is within 1 hour, it is considered that the ambient temperature change is extremely small and the current offset value can be ignored. The current value calibration within the current cycle can use the same temperature drift calibration correction value. ;

[0095] (3) If the current processing cycle is completed normally, the current data of the current processing section will be automatically included in the template library to replace the old value; the average current of the no-load section of the current processing cycle will be calculated and The difference is used as a new temperature drift calibration correction amount .

[0096] This mechanism can continuously adapt to current drift problems caused by room temperature changes, tool wear, workpiece hardness changes, etc., and reduce the false alarm rate.

[0097] The above are only the preferred embodiments of the present invention, and are not any other form of limitation to the present invention. Any modification or equivalent change made according to the technical essence of the present invention still belongs to the scope claimed by the present invention.

Claims

1. An adaptive real-time tool breakage detection method based on spindle current in a numerically controlled machine tool, characterized in that: It includes the following specific steps: 1) Obtain the current signal of the spindle motor and transmit it to the industrial control system in real time; 2) Obtain the temperature drift correction amount in the template library and calculate the calibrated current; 3) Identify the rough machining start time based on the current mutation during the spindle start-stop process ; 4) Extract the machining state recognition template sequence of the corresponding cutting tool from the historical machining data and the historical machining current ; 5) Perform a sliding matching search near to calculate the normalized cross-correlation coefficient NCC, obtain the optimal alignment time , and construct a calibrated and aligned machining current sequence ; 6) Use the current fluctuation trend to identify the tool breakage state, including the no-load type and the abnormal contact type; 7) Automatically update the historical template library during the period without abnormalities to improve the model's adaptability.

2. The adaptive real-time tool breakage detection method based on spindle current in the numerically controlled machine tool according to claim 1, wherein: The calculation formula of the normalization correlation coefficient NCC in step 5) is: ； wherein is the template sequence, is the sliding window sequence, and are the and th elements of the sequences respectively, and are their respective means, N is the length of the T and S sequences, and when the maximum normalized cross - correlation coefficient NCC is lower than the threshold the system records that the starting point recognition fails and skips the cycle detection.

3. The adaptive real-time tool breakage detection method based on spindle current in the numerically controlled machine tool according to claim 1, wherein: In the process of using the current fluctuation trend to identify the tool breakage state in step 6), the tool breakage detection includes: No-load type identification: If the absolute value of the difference of the calibrated real-time current is continuously less than the first threshold, and the absolute value of the difference between the calibrated real-time current and the historical no-load value is continuously less than the second threshold, it is determined as a no-load type tool breakage; Abnormal contact type identification: If the difference between the calibrated real-time current and the historical machining current continuously exceeds the third threshold, it is determined as an abnormal contact type tool breakage.

4. The adaptive real-time tool breakage detection method based on spindle current in the numerically controlled machine tool according to claim 1, wherein: The self-update mechanism steps of the template library in step 7) to update the historical template library are as follows: During the first machining under the new working condition, record the current of one no-load machining and one normal machining, and obtain the no-load and load machining intervals of the historical current through difference comparison and incorporate them into the historical template library; when there is no tool breakage abnormality in the current machining cycle, the mechanism automatically incorporates the current data of the current machining section into the historical template library for reference in the next machining cycle.

Images