A blade condition detection method and system

Abstract

This invention discloses a blade condition detection method and system, relating to the field of detection technology. It is applied to a main control unit connected to a sensing device, which includes a transmitter and a receiver arranged opposite each other. The receiver has multiple independent photosensitive channels, each outputting an electrical signal and selected via a channel selection switch. The method selects each channel in a time-division multiplexing manner and reads the electrical signals. Based on the electrical signals, it determines the channel position corresponding to the junction of the obscured and unobscured areas as the current junction position, compares it with historical junction positions, and configures the selection state of the channel selection switch based on the comparison result to ensure the initial light transmittance matches a preset benchmark. Then, it detects blade damage and / or wear based on the electrical signals. A long strip-shaped detection light curtain covers the effective exposed area of ​​the blade, eliminating the need for machine downtime for adjustment; automatic calibration is completed after blade replacement or wear, improving detection efficiency and accuracy; and multiple channels suppress random noise in the spatial dimension, ensuring detection stability and reliability.

Description

Technical Field

[0001] This invention relates to the field of detection technology, and in particular to a method and system for detecting the condition of a cutting blade. Background Technology

[0002] In semiconductor wafer dicing (dicing) processes, blade breakage detection is a critical quality control step. Traditional fiber-optic spot detection systems have inherent limitations: their effective detection area (diameter ≤ 1 mm) is smaller than the effective exposure of the blade (> 1 mm). During actual dicing, blade wear causes its relative position to constantly change with the sensor, requiring frequent machine shutdowns and manual sensor adjustments, severely impacting the continuous dicing efficiency, detection efficiency, and accuracy of the equipment.

[0003] To address the aforementioned pain points in engineering applications, the industry has explored the use of elongated, strip-shaped light spot sensors with a larger coverage area. However, with the light source power remaining constant, the increased spot area leads to a sharp decrease in optical power density per unit area. This low power density results in a significant drop in the signal-to-noise ratio of the detection signal. Under actual operating conditions such as cutting water, the effective signal is easily overwhelmed by noise, and the detection algorithm fails, affecting the continuity and reliability of detection.

[0004] Therefore, improving the continuity and reliability of blade inspection without frequent downtime for adjustments is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a blade condition detection method and system to solve the technical problems of traditional spot light systems requiring frequent shutdowns for manual adjustment of sensor positions, and existing long strip light spot sensors being easily overwhelmed by noise under actual working conditions such as cutting water, leading to detection algorithm failure and affecting the continuity and reliability of detection.

[0006] To address the aforementioned technical problems, this invention provides a blade state detection method, applied to a main control unit connected to a sensing device for blade state monitoring. The sensing device includes a transmitter and a receiver arranged opposite to each other, with a working distance between the transmitter and receiver for accommodating the blade. The receiver includes a photosensitive element and a signal receiving circuit board. The photosensitive element has multiple independent photosensitive channels arranged along the length of a long strip-shaped detection light curtain. Each independent photosensitive channel receives light at a corresponding position in the light curtain and outputs an electrical signal individually. The signal receiving circuit board is equipped with a channel selection switch, and the output terminal of each independent photosensitive channel is electrically connected to the corresponding input terminal of the channel selection switch. The method includes: Each independent photosensitive channel is selected in a time-division manner by the channel selection switch, and the electrical signal output by each independent photosensitive channel is read. Based on the comparison results between the electrical signals of each independent photosensitive channel and the preset threshold, the photosensitive channel number or position corresponding to the junction point between the occluded area and the unoccluded area is determined as the current junction point position; The current handover point position is compared with the historical handover point position, and the selection state of the channel selection switch is configured according to the comparison result so that the initial light transmittance of the sensing device is consistent with the preset reference. The independent photosensitive channel is selected in a time-division manner by the channel selection switch, the electrical signal of the independent photosensitive channel is read, and the blade is subjected to damage detection and / or wear detection based on the electrical signal.

[0007] Preferably, configuring the selection state of the channel selection switch according to the comparison result includes: The change in the position of the blade is determined based on the positional deviation between the current handover point position and the historical handover point position. Based on the change in position, determine the independent photosensitive channel that needs to be opened, and control the channel selection switch to open the determined independent photosensitive channel.

[0008] Preferably, the wear detection of the blade based on the electrical signal includes: During the rotation of the blade with the spindle, electrical signals from multiple sampling points within one revolution of the blade are collected. The average electrical signal of the current cycle is obtained by averaging the electrical signals of the sampling points within a cycle. If the difference between the average electrical signal of the current loop and a preset value is greater than a preset difference, it is determined that the blade has abnormal wear.

[0009] Preferably, determining the preset value includes: The average electrical signal of each circle in multiple target circles is continuously acquired; wherein, the multiple target circles are multiple circles preceding the current circle; The average value is calculated for the average electrical signals corresponding to the plurality of target circles to obtain the preset value; The step of determining that the blade has abnormal wear when the difference between the average electrical signal of the current loop and a preset value is greater than a first preset difference includes: If the difference between the average electrical signal of the current cycle and the preset value is greater than the first preset difference, record the current number of consecutive exceedances. If the number of consecutive exceedances exceeds the threshold, it is determined that the blade has abnormal wear. If the current consecutive number of exceedances is less than or equal to the exceedance threshold, then return to the step of collecting electrical signals from multiple sampling points within one revolution of the blade.

[0010] Preferably, the step of detecting the wear amount of the blade based on the electrical signal includes: The change in the first electrical signal of the target independent photosensitive channel within a preset time period is obtained; wherein, the target independent photosensitive channel is the independent photosensitive channel corresponding to the maximum change in the second electrical signal; the change in the second electrical signal is the change in the electrical signal of each independent photosensitive channel determined by comparing the current electrical signal obtained at the detection time with the reference electrical signal corresponding to the independent photosensitive channel for each independent photosensitive channel. Based on the pre-calibrated correspondence between the electrical signal change and the light-shielding width change of the target independent photosensitive channel, the current light-shielding width change corresponding to the first electrical signal change is determined. The change in the current light-shielding width is determined as the current wear amount of the blade.

[0011] Preferably, the damage detection of the blade based on the electrical signal includes: During the rotation of the blade with the spindle, an electrical signal is collected for at least two rotations of the blade. If the difference between the electrical signals at the same sampling position in at least two consecutive cycles is greater than a second preset difference, it is determined that the blade is damaged.

[0012] Preferably, after determining that the blade is damaged, the method further includes: Record the current number of consecutive breaks; If the number of consecutive breaks is greater than the breakage number threshold, then the blade is determined to be damaged. If the number of consecutive breaks is less than or equal to the breakage number threshold, then return to the step of collecting the electrical signal of the blade rotating at least two revolutions.

[0013] Preferably, after obtaining the average electrical signal of the current rotation, before detecting the relationship between the difference between the average electrical signal of the current rotation and a preset value and a preset difference, and / or, after acquiring the electrical signal of the blade rotating at least two rotations, before detecting the relationship between the difference between the electrical signals at the same sampling position in at least two consecutive rotations and a second preset difference, the method further includes: The average electrical signal and / or the electrical signal of at least two turns are subjected to frequency domain filtering, and a bandpass filter is used to retain the frequency band component where the blade characteristic signal is located and to filter out noise interference. And / or, perform nonlinear filtering on the average electrical signal and / or the electrical signal of at least two turns to remove impulse noise; The electrical signal after the frequency domain filtering and / or the nonlinear filtering is used as the noise-reduced electrical signal.

[0014] Preferably, after the step of time-division multiplexing the independent photosensitive channel via the channel selection switch and reading the electrical signal of the independent photosensitive channel, and before the step of performing damage detection and / or wear detection on the blade based on the electrical signal, the method further includes: If a continuous abnormality in the electrical signal of a single independent photosensitive channel is detected, it is determined that the independent photosensitive channel is being interfered with, and the interfered independent photosensitive channel is turned off by the channel selection switch. The step of performing damage detection and / or wear detection on the blade based on the electrical signal includes: The blade is subjected to damage detection and / or wear detection based on the electrical signals from other independent photosensitive channels that are not turned off.

[0015] To address the aforementioned technical problems, the present invention also provides a blade condition detection system, comprising a sensing device and a main control unit for blade condition monitoring. The sensing device includes a transmitter and a receiver arranged opposite to each other, with a working distance between the transmitter and the receiver for accommodating the blade. The receiver includes a photosensitive element and a signal receiving circuit board. The photosensitive element has multiple independent photosensitive channels arranged along the length of an elongated detection light curtain. Each independent photosensitive channel is used to receive light at a corresponding position in the light curtain and outputs an electrical signal individually. The signal receiving circuit board is provided with a channel selection switch, and the output terminal of each independent photosensitive channel is electrically connected to the corresponding input terminal of the channel selection switch. The main control unit is used to select each of the independent photosensitive channels in a time-division manner through the channel selection switch, and read the electrical signals output by each of the independent photosensitive channels; based on the comparison result of the electrical signals of each of the independent photosensitive channels with a preset threshold, determine the photosensitive channel number or position corresponding to the junction point between the occluded area and the unoccluded area, and use it as the current junction point position; compare the current junction point position with the historical junction point position, and configure the selection state of the channel selection switch according to the comparison result, so that the initial light transmittance of the sensing device is consistent with the preset benchmark; select the independent photosensitive channels in a time-division manner through the channel selection switch, read the electrical signals of the independent photosensitive channels, and perform damage detection and / or wear detection on the blade based on the electrical signals.

[0016] The blade condition detection method provided by this invention is applied to a main control unit connected to a sensing device for blade condition monitoring. The sensing device includes a transmitter and a receiver arranged opposite each other. The receiver has multiple independent photosensitive channels arranged along the length of a long strip-shaped detection light curtain. Each channel outputs an electrical signal independently. A channel selection switch is provided on the signal receiving circuit board, and the output terminal of each channel is electrically connected to the corresponding input terminal of the channel selection switch. This method uses the channel selection switch to select each channel in a time-division manner and reads the electrical signals. Based on the comparison result of each channel's electrical signal with a preset threshold, the channel number or position corresponding to the intersection point of the obstructed area and the unobstructed area is determined as the current intersection point position. The current intersection point position is compared with historical intersection point positions. Based on the comparison result, the selection state of the channel selection switch is configured to ensure that the initial light transmittance of the sensing device is consistent with a preset reference. Then, the channel selection switch is used again to select each channel in a time-division manner and read the electrical signals. The blade is then subjected to damage detection and / or wear detection based on the electrical signals. Because the elongated detection light curtain covers the effective exposure area of ​​the blade, and in conjunction with multiple independent photosensitive channels, any change in the blade's position can be detected by the corresponding channel, eliminating the need for manual adjustment during machine downtime. After blade replacement or wear, the main control unit automatically determines the current handover point position and compares it with historical handover point positions. Based on the comparison results, it automatically configures the channel selection state to ensure that the initial light transmittance is consistent with the preset benchmark. The entire process is completed automatically without manual intervention, improving cutting and detection efficiency, eliminating the uncertainty of manual adjustment, and enhancing detection accuracy. Each channel outputs an electrical signal independently, and the handover point position is determined by comparing the electrical signals of each channel with preset thresholds. The handover point position information can be obtained through multiple channels, and even if a single channel is affected by noise, it can still be obtained through the electrical signals of other channels. This spatially suppresses random noise and ensures the stability and reliability of the detection.

[0017] In addition, the present invention also provides a blade condition detection system, which has the same or corresponding technical features as the blade condition detection method mentioned above, and has the same effect. Attached Figure Description

[0018] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of a sensing device for blade condition monitoring provided in an embodiment of the present invention; Figure 2 A flowchart of a blade state detection method provided in an embodiment of the present invention; Figure 3A flowchart of an adaptive calibration method provided in an embodiment of the present invention; Figure 4 A flowchart of a method for detecting abnormal blade wear provided in an embodiment of the present invention; Figure 5 A flowchart of a method for detecting blade wear provided in an embodiment of the present invention; Figure 6 This is a flowchart of a blade breakage detection method provided in an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present invention.

[0021] The core of this invention is to provide a blade status detection method and system to solve the technical problems of traditional spot light systems requiring frequent shutdowns for manual adjustment of sensor positions, and existing long strip light spot sensors being easily overwhelmed by noise under actual working conditions such as cutting water, leading to detection algorithm failure and affecting the continuity and reliability of detection.

[0022] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The present invention provides a blade condition detection method, applied to a main control unit connected to a sensing device for blade condition monitoring. Figure 1 This is a schematic diagram of a sensing device for blade condition monitoring provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the sensing device includes a transmitter and a receiver arranged opposite each other, with a working gap between them to accommodate the blade. The receiver includes a photosensitive element 201 and a signal receiving circuit board 202. The photosensitive element 201 has multiple independent photosensitive channels arranged along the length of the elongated detection light curtain. Each independent photosensitive channel is used to receive light at a corresponding position in the light curtain and outputs an electrical signal individually. The signal receiving circuit board 202 is equipped with a channel selection switch, and the output terminal of each independent photosensitive channel is electrically connected to the corresponding input terminal of the channel selection switch. The output terminal of the channel selection switch is electrically connected to a signal line 203, which is used to electrically connect to an external main control unit.

[0023] There are no restrictions on the structure of the transmitter, as long as it can form a long strip of detection light curtain. Figure 1 The diagram shows a transmitter that forms a long, strip-shaped detection light curtain. (For example...) Figure 1As shown, the emitting unit includes a light source 101, a first beam shaping element 102 (using a Powell prism as an example), a collimating lens 103, a second beam shaping element 104 (using a Powell prism as an example), and a light path deflection element 105 (using a collimating prism as an example), arranged sequentially along the optical path. The first beam shaping element 102 converts the point light emitted from the light source 101 into a line light. The second beam shaping element 104 expands and homogenizes the energy of the line light collimated by the collimating lens 103, forming a long strip-shaped detection light curtain with uniform light intensity distribution along its length within the working spacing. The light path deflection element 105 deflects the beam emitted from the second beam shaping element 104 towards the receiving unit, aligning the long strip-shaped detection light curtain with the photosensitive area of ​​the receiving unit. Furthermore, the knob 3 in the sensing device adjusts the installation position of the blade, ensuring that the blade is within the effective detection area of ​​the light curtain.

[0024] Figure 2 A flowchart of a blade state detection method provided in an embodiment of the present invention is shown below. Figure 2 As shown, the method includes: S10: Select each independent photosensitive channel in a time-division manner through the channel selection switch, and read the electrical signal output by each independent photosensitive channel; S11: Based on the comparison results of the electrical signals of each independent photosensitive channel with the preset threshold, determine the photosensitive channel number or position corresponding to the junction point between the occluded area and the unoccluded area, and use it as the current junction point position; S12: Compare the current handover point position with the historical handover point position, and configure the selection state of the channel selection switch according to the comparison result so that the initial light transmittance of the sensing device is consistent with the preset reference. S13: Select an independent photosensitive channel in a time-division manner by using a channel selection switch, read the electrical signal of the independent photosensitive channel, and perform damage detection and / or wear detection on the blade based on the electrical signal.

[0025] Specifically, the main control unit uses a channel selection switch to time-division select each independent photosensitive channel and reads the electrical signals output by each independent photosensitive channel. In a specific implementation, the photosensitive element has multiple (e.g., 16) independent photosensitive channels arranged along the length of the elongated light spot, and the position and physical size of each channel are known. The main control unit sequentially switches channels using a multiplexer and collects the voltage value of each channel. Then, based on the comparison between the electrical signals of each independent photosensitive channel and a preset threshold, the main control unit determines the photosensitive channel number or position corresponding to the junction of the blocked and unblocked areas, which is taken as the current junction point position. This junction point accurately reflects the real-time position of the blade edge in the detection light curtain. The preset threshold is not limited and is determined according to the actual situation. For example, since the blade will block part of the light path, the output voltage of the blocked channel is close to 0, while the output voltage of the unblocked channel is relatively large.

[0026] To accommodate changes in blade position due to wear or replacement, the main control unit compares the current handover point with historical handover points. These historical handover points can be the positions recorded during initial blade installation or the last calibration. Based on the comparison results, the main control unit automatically configures the channel selection switch's activation state. For example, when a blade position shift is detected, the system adjusts the required channel combination to ensure the initial light transmittance of the sensor remains consistent with a preset reference. The preset reference refers to the initial light transmittance target value set by the system when the blade was initially installed and calibrated.

[0027] This process is designed for multi-channel low signal-to-noise ratio data. It updates the internal detection benchmark in real time by analyzing the collective offset of the channels. It is fully automated and requires no manual intervention, achieving sub-millimeter-level intelligent calibration and blade breakage detection capabilities without the need for physical probe movement.

[0028] After completing the adaptive calibration, the main control unit again selects the independent photosensitive channel in a time-division manner via the channel selection switch to read the electrical signal. The main control unit then performs damage detection and / or wear detection on the blade based on the electrical signal.

[0029] The blade condition detection method provided in this embodiment is applied to a main control unit connected to a sensing device for blade condition monitoring. The sensing device includes a transmitter and a receiver arranged opposite each other. The receiver has multiple independent photosensitive channels arranged along the length of a long strip-shaped detection light curtain. Each channel outputs an electrical signal independently. A channel selection switch is provided on the signal receiving circuit board, and the output terminal of each channel is electrically connected to the corresponding input terminal of the channel selection switch. This method uses the channel selection switch to select each channel in a time-division manner and reads the electrical signals. Based on the comparison result of each channel's electrical signal with a preset threshold, the channel number or position corresponding to the intersection point of the obstructed area and the unobstructed area is determined as the current intersection point position. The current intersection point position is compared with historical intersection point positions. Based on the comparison result, the selection state of the channel selection switch is configured to ensure that the initial light transmittance of the sensing device is consistent with a preset reference. Then, the channel selection switch is used again to select each channel in a time-division manner and read the electrical signals. The blade is then subjected to damage detection and / or wear detection based on the electrical signals. Because the elongated detection light curtain covers the effective exposure area of ​​the blade, and in conjunction with multiple independent photosensitive channels, any change in the blade's position can be detected by the corresponding channel, eliminating the need for manual adjustment during machine downtime. After blade replacement or wear, the main control unit automatically determines the current handover point position and compares it with historical handover point positions. Based on the comparison results, it automatically configures the channel selection state to ensure that the initial light transmittance is consistent with the preset benchmark. The entire process is completed automatically without manual intervention, improving cutting and detection efficiency, eliminating the uncertainty of manual adjustment, and enhancing detection accuracy. Each channel outputs an electrical signal independently, and the handover point position is determined by comparing the electrical signals of each channel with preset thresholds. The handover point position information can be obtained through multiple channels, and even if a single channel is affected by noise, it can still be obtained through the electrical signals of other channels. This spatially suppresses random noise and ensures the stability and reliability of the detection.

[0030] After automatically identifying the blade position, in order to further accurately select the working channel based on the position offset and achieve intelligent channel switching, in one possible implementation, the selection state of the channel selection switch is configured according to the comparison result, including: The change in the blade's position is determined based on the positional deviation between the current handover point and the historical handover point. The independent photosensitive channel that needs to be opened is determined based on the amount of position change, and the channel selection switch is used to open the determined independent photosensitive channel.

[0031] Figure 3 A flowchart of an adaptive calibration method provided in an embodiment of the present invention is shown below. Figure 3 As shown, the method includes: S14: Turn on the LED light; S15: Time-division opening of the receiving channel; S16: Read the voltage value of each channel; S17: Determine the positional relationship between the blade and the sensor; S18: Automatically calibrate the initial value at this position.

[0032] In implementation, the main control unit first calculates the positional deviation between the current junction point and the historical junction point. This deviation directly reflects the displacement of the blade edge within the light curtain, i.e., the amount of blade wear or the change in installation position. Subsequently, based on this positional change, the main control unit determines the independent photosensitive channels that need to be activated under the current operating conditions. For example, if the junction point shifts two channels to the left, the system correspondingly shifts the selection window to the left to ensure that the channel combination used for detection always surrounds the optimal photosensitive area of ​​the blade edge. Finally, the main control unit controls the channel selection switch to precisely activate the determined independent photosensitive channels and deactivate other irrelevant channels. This process achieves dynamic configuration of the detection channels, ensuring that regardless of blade wear, the system can automatically select the channel with the best signal quality for data acquisition, thereby maintaining the consistency of initial light transmittance and providing a stable and optimized signal source for subsequent high-precision damage and wear detection.

[0033] In order to accurately detect abnormal wear of the cutting tool, in one possible implementation, blade wear detection based on an electrical signal includes: During the rotation of the blade with the spindle, electrical signals from multiple sampling points within one revolution of the blade are collected. The average electrical signal of the current cycle is obtained by averaging the electrical signals at each sampling point within a cycle. If the difference between the average electrical signal of the current loop and the preset value is greater than the preset difference, it is determined that there is abnormal wear on the blade.

[0034] To eliminate random interference and prevent false alarms caused by single signal fluctuations, thereby further improving the reliability of wear detection, the following preset values ​​are specifically determined: The average electrical signal of each circle in multiple target circles is continuously acquired; where multiple target circles are the circles preceding the current circle. The average value is calculated for the average electrical signals corresponding to multiple target circles to obtain the preset value.

[0035] Correspondingly, if the difference between the average electrical signal of the current loop and a preset value is greater than a first preset difference, it is determined that the blade has abnormal wear, including: If the difference between the average electrical signal of the current cycle and the preset value is greater than the first preset difference, record the current number of consecutive exceedances. If the number of consecutive exceedances exceeds the threshold, it is determined that the blade has abnormal wear. If the current consecutive number of exceedances is less than or equal to the exceedance threshold, then return to the step of collecting electrical signals from multiple sampling points within one revolution of the blade.

[0036] There is no limit on the number of times the limit is exceeded; it is determined based on the actual situation. Figure 4 A flowchart of a method for detecting abnormal blade wear provided in an embodiment of the present invention is shown below. Figure 4 As shown, the method includes: S19: Controls spindle rotation; S20: Close the blade guard; S21: Acquire the electrical signal collected by the sensor; S22: Continuously record the average value of the electrical signal for each of the multiple revolutions, and obtain the average electrical signal for the multiple revolutions; S23: Calculate the difference between the average electrical signal value of the current cycle and the average electrical signal value of multiple cycles; S24: If the detected difference is greater than the preset difference, record one instance of exceeding the limit; S25: Determine whether the number of consecutive over-limits exceeds the over-limit threshold; if yes, proceed to step S26; if no, return to step S21. S26: Output a prompt message to indicate abnormal wear of the cutting tool.

[0037] There is no limit to the frequency of acquiring the electrical signals from the sensor, such as once per second. When continuously recording the average value of the electrical signal for each of multiple revolutions, a sliding list method can be used. The specific value of the preset difference can be set through the interface. The threshold for the number of times the limit is exceeded can be set through the configuration file.

[0038] Specifically, the blade is first driven to rotate at a constant speed with the spindle. During one revolution of the blade, the main control unit collects electrical signals from multiple sampling points at equal intervals through a configured channel selection switch and stores these signals as an array Vol_buf[n]. Then, the main control unit (program) calculates the average value of the electrical signals from each sampling point within one revolution to obtain the average electrical signal of the current revolution. This value represents the overall light transmittance or occlusion of the blade at the current time. During continuous cutting, the system continuously calculates the average electrical signal for each revolution. When the difference between the average electrical signal of the current revolution and a preset value (e.g., a baseline value determined based on the wear trend of the previous multiple revolutions) is greater than the preset difference, it indicates that a sudden change has occurred in the light transmittance of the blade. This sudden change is usually caused by exposure of the blade substrate, particle shedding, or abnormal wear. Therefore, the main control unit determines that the blade has abnormal wear. This method effectively suppresses random noise and instantaneous vibration interference by averaging the data of the entire revolution, achieving reliable detection of abnormal blade wear.

[0039] To improve the reliability of wear detection, the system continuously acquires the average electrical signal of each of the previous 100 target cycles (e.g., the first 100 cycles) and calculates the average value of these average electrical signals for each target cycle, using this as a preset value, i.e., the current dynamic baseline. During judgment, if the difference between the average electrical signal of the current cycle and the preset value is greater than a first preset difference, the system does not immediately issue an alarm but records the current number of consecutive exceedances. The program then determines whether this number of consecutive exceedances exceeds a preset threshold (e.g., 3 times). Only when exceedances are detected multiple times consecutively (e.g., 3 consecutive cycles) does the system finally determine that the blade has abnormal wear; if the current number of consecutive exceedances is less than or equal to the threshold, it is considered occasional noise or disturbance, and the system returns to continue acquiring the electrical signal of the next cycle for cyclical judgment. This continuous multiple judgment greatly enhances the robustness of the detection system and effectively avoids false alarms caused by non-wear factors such as vibration or momentary water mist obstruction.

[0040] In practice, in addition to detecting the wear condition, it is also necessary to detect the amount of wear. Figure 5 A flowchart of a method for detecting blade wear provided in an embodiment of the present invention is shown below. Figure 5 As shown, the wear of the cutting blade is detected based on an electrical signal, including: S27: Obtain the first electrical signal change of the target independent photosensitive channel within a preset time period; wherein, the target independent photosensitive channel is the independent photosensitive channel corresponding to the maximum second electrical signal change; the second electrical signal change is the electrical signal change of each independent photosensitive channel determined by comparing the current electrical signal obtained at the detection time with the reference electrical signal corresponding to the independent photosensitive channel for each independent photosensitive channel. S28: Based on the correspondence between the electrical signal change and the light-shielding width change of the pre-calibrated target independent photosensitive channel, determine the current light-shielding width change corresponding to the first electrical signal change. S29: Determine the current change in the light-shielding width as the current wear amount of the blade.

[0041] In implementation, the main control unit uses a channel selection switch to time-division multiplex each independent photosensitive channel. At the current detection moment, it reads the current electrical signal output by each channel and retrieves the reference electrical signal stored during the initial blade installation. For each independent photosensitive channel, the current electrical signal acquired at the current detection moment is compared with the corresponding reference electrical signal, and the difference is calculated to obtain the change in the second electrical signal for each channel. The larger the change in the second electrical signal, the greater the increase in light transmission in that channel, meaning a greater reduction in the light-blocking width of the blade at the corresponding position, and more severe wear. The main control unit selects the channel with the largest change in the second electrical signal as the target independent photosensitive channel, representing the position of the most severe blade wear.

[0042] After selecting a target independent photosensitive channel, the main control unit acquires the electrical signal value of the target independent photosensitive channel at the start of the preset time and the electrical signal value of the target independent photosensitive channel at the end of the preset time. The electrical signal value at the end time is subtracted from the electrical signal value at the start time to obtain the electrical signal change of the target independent photosensitive channel within the preset time, which is recorded as the first electrical signal change.

[0043] The main control unit pre-stores the calibration relationship for the target's independent photosensitive channel, which is a correspondence between changes in electrical signal and changes in light-blocking width. Substituting the first electrical signal change into this calibration relationship, the current change in light-blocking width corresponding to that first electrical signal change can be determined by looking up a table. For example, if the electrical signal of the target's independent photosensitive channel is 2.0V at the start of a preset duration and 2.5V at the end, the first electrical signal change is 0.5V. Looking up the calibration table, the corresponding light-blocking width decreases by 0.15mm. The main control unit determines this current change in light-blocking width as the current wear amount of the blade and outputs it, enabling simultaneous cutting and detection.

[0044] In addition to the process of determining wear as described above, in practice, wear can also be determined in the following ways: While controlling the blade to rotate with the spindle, it descends along the Z-axis at a set speed until the occlusion of the set independent photosensitive channel reaches the set value, then stops descending and records the current Z-axis coordinate value. After the preset cutting time, the control blade returns to rotate with the spindle and descends along the Z-axis at a set speed until the occlusion of the set independent photosensitive channel reaches the set value, then stops descending and records the current Z-axis coordinate value. Obtain the change in the Z-axis coordinate value, and determine the wear amount of the cutting tool based on the change in the Z-axis coordinate value.

[0045] The preset cutting time is not limited and is determined based on actual conditions. Upon detecting abnormal wear, the system enters a wear quantification process. First, the main control unit controls the blade to rotate with the spindle and simultaneously controls the blade to descend slowly along the Z-axis at a set speed. During descent, the system monitors the occlusion of a designated independent photosensitive channel (e.g., a specific channel near the junction point) in real time. When the occlusion of this designated independent photosensitive channel reaches a preset value, the main control unit controls the Z-axis to stop moving and records the current Z-axis coordinate value z1. After the equipment continues cutting for a period of time (i.e., when actual wear of the blade may occur), the system automatically repeats the above steps, i.e., descending the Z-axis again until the same designated channel reaches the same occlusion amount, and records a new Z-axis coordinate value z2. Since the two positioning references the same physical location, the difference between z1 and z2 is calculated to obtain the accurate value of the actual wear of the blade during that time period. This method utilizes the spatial resolution characteristics of multi-channel technology to convert blade wear into a high-precision Z-axis displacement, achieving closed-loop accurate measurement of wear and providing reliable data support for process control.

[0046] Besides the aforementioned wear and tear on the blade, blade breakage may also occur in practice. To detect blade breakage, in one possible implementation, detection based on an electrical signal includes: During the rotation of the blade with the spindle, an electrical signal is collected for at least two rotations of the blade. If the difference between the electrical signals at the same sampling position in at least two consecutive cycles is greater than a second preset difference, it is determined that the blade is damaged.

[0047] To avoid false alarms due to single signal jitter or accidental foreign objects (such as water droplets) and to improve the accuracy of damage alarms, after determining that the blade is damaged, the following steps are also included: Record the current number of consecutive breaks; If the number of consecutive breaks detected exceeds the breakage threshold, it is determined that the blade is damaged. If the number of consecutive breaks detected is less than or equal to the breakage number threshold, the process returns to the step of collecting electrical signals from at least two rotations of the blade. The second preset difference and the breakage number threshold are not limited and are determined based on the actual situation.

[0048] To enable those skilled in the art to better understand the process of blade breakage detection, the following description, in conjunction with the accompanying drawings and specific embodiments, will continue. Figure 6 A flowchart of a blade breakage detection method provided in an embodiment of the present invention is shown below. Figure 6 As shown, the method includes: S30: Spindle rotation; S31: Read the blade breakage status; S32: Valid for obtaining blade breakage status; S33: Record the number of times a single breakage occurs; S34: Determine whether the number of consecutive breaks is greater than the breakage number threshold; if yes, proceed to step S35; if no, return to step S31. S35: Outputs a warning message indicating blade breakage.

[0049] There is no limit to the frequency of reading the blade breakage status, such as once per second. When the blade breakage status value is read as 1, it indicates that the blade breakage status is valid. The breakage count threshold can be set through the configuration file.

[0050] In practice, during the normal rotation of the blade with the spindle, the main control unit collects the electrical signals of the blade during at least two consecutive rotations via a channel selection switch. The core of the damage detection algorithm lies in comparing the signal differences at the same angular position across different rotations. Specifically, the system stores the electrical signal values ​​of each sampling point in the previous rotation of the blade. When data for the current rotation is collected, the program compares the electrical signal at a specific sampling position in the current rotation with the electrical signal at the same sampling position in the previous rotation (or previous few rotations). Since the edge of a normal blade is continuous, the amount of light transmitted at the same position across different rotations should be basically consistent. If a tear appears on the blade edge, when the tear rotates to the position of the light curtain, additional light will pass through, causing a sudden change in the electrical signal at that position. Therefore, when the system detects that the difference between the electrical signals at the same sampling position in at least two consecutive rotations is greater than a second preset difference, it indicates that the phenomenon of abnormally increased light flux has repeatedly occurred at that position, which is sufficient to determine that the blade is damaged at that position. This method utilizes the periodic characteristics of blade rotation and, through cross-period correlation comparison, can effectively suppress the slow drift interference caused by ambient light and water mist, thus achieving sensitive identification of small-sized breaks.

[0051] To improve the accuracy of damage alarms, when the main control unit detects consecutive electrical signal differences exceeding the limit at the same sampling location, the system does not immediately output a final damage signal. Instead, it records the current consecutive damage count. The system then determines whether this consecutive damage count exceeds a preset damage count threshold (e.g., 3 times). Only when the system detects damage features at the same location more than this threshold is it finally determined that the blade is indeed damaged, and an alarm is output. If the detected consecutive damage count is less than or equal to the damage count threshold, it is considered an occasional interference, and the system returns to continuously collect signals for the next cycle for monitoring. By comparing features from the same sampling point across multiple cycles, spike pulse interference in single-cycle signals is effectively avoided, ensuring high reliability of the damage determination results and preventing accidental shutdowns during normal cutting.

[0052] To effectively extract weak break feature signals submerged in noise under complex cutting conditions (especially when cutting water is activated) and significantly improve the system's signal-to-noise ratio, this invention adds a signal preprocessing step before threshold comparison. In one possible implementation, after obtaining the average electrical signal of the current rotation, before detecting the relationship between the difference between the average electrical signal of the current rotation and a preset value, and / or, after acquiring electrical signals from at least two rotations of the blade, before detecting the relationship between the difference between the electrical signals at the same sampling position in at least two consecutive rotations and a second preset difference, the method further includes: The average electrical signal and / or at least two turns of the electrical signal are subjected to frequency domain filtering. A bandpass filter is used to retain the frequency band component where the blade characteristic signal is located and to filter out noise interference. And / or, perform nonlinear filtering on the average electrical signal and / or at least two turns of the electrical signal to remove impulse noise; The electrical signal after frequency domain filtering and / or nonlinear filtering is used as the noise-reduced electrical signal.

[0053] In practical implementation, this method adds a signal preprocessing step after obtaining the average electrical signal of the current cycle (for wear detection) and / or after acquiring electrical signals from at least two cycles (for damage detection), before performing the corresponding threshold comparison judgment. First, the main control unit performs frequency domain filtering on the acquired raw electrical signal. Specifically, a bandpass filter is used, with its passband specifically set to pass through the frequency band components of the blade's characteristic signal, effectively filtering out low-frequency noise generated by the cutting water flow, high-frequency vibration interference from the spindle, and power frequency noise. In addition, the system also uses nonlinear filtering algorithms (such as median filtering) to process the signal to effectively remove impulse noise (salt-and-pepper noise). The electrical signal obtained after the above frequency domain filtering and / or nonlinear filtering is used as a clean, denoised signal for subsequent threshold comparison and damage / wear judgment. Through this signal processing flow, even if the signal-to-noise ratio of the original signal is so low that the effective signal is almost completely submerged, the algorithm can extract and enhance the characteristic components, providing a reliable data foundation for the detection algorithm.

[0054] To fully utilize the hardware redundancy of the multi-channel receiving system and ensure normal operation even when some channels are partially contaminated or blocked, this invention also provides a redundancy fault-tolerant mechanism. In one possible implementation, after the step of time-division multiplexing the independent photosensitive channels via channel selection switches and reading the electrical signals of the independent photosensitive channels, and before the step of performing damage detection and / or wear detection on the blade based on the electrical signals, the following method is also included: If a continuous abnormality in the electrical signal of a single independent photosensitive channel is detected, it is determined that the independent photosensitive channel is being interfered with, and the interfered independent photosensitive channel is shut down via a channel selection switch.

[0055] Blade breakage and / or wear detection based on electrical signals includes: The blade is subjected to damage detection and / or wear detection based on the electrical signals from other independent photosensitive channels that are not turned off.

[0056] In practice, after reading the electrical signals of each independent photosensitive channel via the channel selection switch, the main control unit continuously monitors the signal quality of each channel. When a persistent abnormality is detected in the electrical signal of a single (or part of) independent photosensitive channel—for example, if the signal remains saturated for an extended period or is zero for a long time and does not change with blade rotation—the system determines that the independent photosensitive channel is affected by dirt, obstruction, or hardware failure. To prevent the data from this abnormal channel from contaminating the overall detection results, the main control unit shuts down the interfered independent photosensitive channel via the channel selection switch, isolating it from the detection system. In subsequent damage detection and / or wear detection steps, the main control unit makes judgments based solely on the electrical signals of the other independent photosensitive channels that have not been shut down. Because multiple channels are arrayed along the length of the light curtain, the failure of a single channel does not affect the continued effective operation of other channels, achieving the fault tolerance capability brought by hardware redundancy and greatly improving the long-term operational stability of the system in harsh environments.

[0057] The blade state detection method provided by this invention, while retaining the advantage of complete blade coverage in the detection area, offers a more stable and high-quality signal transmission physical basis. Users do not need to stop the machine to manually adjust the sensor due to blade position changes throughout the cutting process, achieving true self-adaptation, thereby maximizing equipment cutting efficiency and significantly improving user experience. Multi-channel receiving technology is introduced. Through parallel signal acquisition and redundant fusion processing at the hardware level, the system can effectively suppress random noise in the spatial dimension, significantly improving the overall signal-to-noise ratio. This allows even under complex conditions such as water cooling, weak blade breakage features can be clearly extracted, fundamentally solving the core problem of poor signal quality from a single long-strip light spot sensor and ensuring high detection reliability. Specifically designed for multi-channel hardware characteristics and optimized signal features, it can intelligently compensate for signal reference drift caused by minor changes in blade position or environmental interference, achieving sub-millimeter-level adaptive calibration accuracy. This makes the detection system more adaptable to environmental changes (such as blade holder vibration and cutting water mist), with stable output signals and robust detection logic. In addition, the hardware design of multi-channel reception provides redundancy and fault tolerance. Even if a single or a few reception channels are subject to local contamination, obstruction, or sudden interference, the system can still maintain effective detection through other normal channels, which greatly improves the operational stability of the system.

[0058] The above describes a blade condition detection method. This embodiment also provides a blade condition detection system, including a sensing device and a main control unit for blade condition monitoring. The sensing device includes a transmitter and a receiver arranged opposite to each other, with a working distance between the transmitter and receiver for accommodating the blade. The receiver includes a photosensitive element and a signal receiving circuit board. The photosensitive element has multiple independent photosensitive channels arranged along the length of the elongated detection light curtain. Each independent photosensitive channel is used to receive light at a corresponding position in the light curtain and output an electrical signal separately. The signal receiving circuit board is provided with a channel selection switch, and the output terminal of each independent photosensitive channel is electrically connected to the corresponding input terminal of the channel selection switch. The main control unit is used to select and activate each independent photosensitive channel in a time-division manner via a channel selection switch, and read the electrical signals output by each independent photosensitive channel; based on the comparison result of the electrical signals of each independent photosensitive channel with a preset threshold, it determines the photosensitive channel number or position corresponding to the junction point between the blocked area and the unblocked area, and uses this as the current junction point position; it compares the current junction point position with the historical junction point positions, and configures the selection state of the channel selection switch according to the comparison result, so that the initial light transmittance of the sensing device is consistent with the preset benchmark; it selects and activates independent photosensitive channels in a time-division manner via the channel selection switch, reads the electrical signals of the independent photosensitive channels, and performs damage detection and / or wear detection on the blade based on the electrical signals.

[0059] The blade condition detection system provided in this embodiment has the same or corresponding technical features as the blade condition detection method described above. The embodiments of the blade condition detection method have been described in detail above, and the embodiments of the blade condition detection system will not be repeated here, and the effects are the same as above.

[0060] The present invention has provided a detailed description of a blade condition detection method and system. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to the present invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the present invention.

[0061] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A blade condition detection method characterized by, A main control unit is applied to a sensing device for blade condition monitoring. The sensing device includes a transmitter and a receiver arranged opposite to each other, with a working distance between the transmitter and the receiver for accommodating the blade. The receiver includes a photosensitive element and a signal receiving circuit board. The photosensitive element has multiple independent photosensitive channels arranged along the length of a long strip-shaped detection light curtain. Each independent photosensitive channel is used to receive light at a corresponding position in the light curtain and output an electrical signal individually. The signal receiving circuit board is equipped with a channel selection switch, and the output terminal of each independent photosensitive channel is electrically connected to the corresponding input terminal of the channel selection switch; the method includes: Each independent photosensitive channel is selected in a time-division manner by the channel selection switch, and the electrical signal output by each independent photosensitive channel is read. Based on the comparison results between the electrical signals of each independent photosensitive channel and the preset threshold, the photosensitive channel number or position corresponding to the junction point between the occluded area and the unoccluded area is determined as the current junction point position; The current handover point position is compared with the historical handover point position, and the selection state of the channel selection switch is configured according to the comparison result so that the initial light transmittance of the sensing device is consistent with the preset reference. The independent photosensitive channel is selected in a time-division manner by the channel selection switch, the electrical signal of the independent photosensitive channel is read, and the blade is subjected to damage detection and / or wear detection based on the electrical signal.

2. The blade condition detection method according to claim 1, characterized by, The step of configuring the selection state of the channel selection switch according to the comparison result includes: The change in the position of the blade is determined based on the positional deviation between the current handover point position and the historical handover point position. Based on the change in position, determine the independent photosensitive channel that needs to be opened, and control the channel selection switch to open the determined independent photosensitive channel.

3. The blade condition detection method according to claim 1, characterized by, The wear detection of the blade based on the electrical signal includes: During the rotation of the blade with the spindle, electrical signals from multiple sampling points within one revolution of the blade are collected. The average electrical signal of the current cycle is obtained by averaging the electrical signals of the sampling points within a cycle. If the difference between the average electrical signal of the current loop and a preset value is greater than a preset difference, it is determined that the blade has abnormal wear.

4. The blade condition detection method according to claim 3, characterized by, Determining the preset value includes: The average electrical signal of each circle in multiple target circles is continuously acquired; wherein, the multiple target circles are multiple circles preceding the current circle; The average value is calculated for the average electrical signals corresponding to the plurality of target circles to obtain the preset value; The step of determining that the blade has abnormal wear when the difference between the average electrical signal of the current loop and a preset value is greater than a first preset difference includes: If the difference between the average electrical signal of the current cycle and the preset value is greater than the first preset difference, record the current number of consecutive exceedances. If the number of consecutive exceedances exceeds the threshold, it is determined that the blade has abnormal wear. If the current consecutive number of exceedances is less than or equal to the exceedance threshold, then return to the step of collecting electrical signals from multiple sampling points within one revolution of the blade.

5. The blade condition detection method according to claim 3, characterized in that, The step of detecting the wear amount of the blade based on the electrical signal includes: The change in the first electrical signal of a target independent photosensitive channel within a preset time period is obtained; wherein, the target independent photosensitive channel is the independent photosensitive channel corresponding to the maximum change in the second electrical signal; the second electrical signal change is the change in the electrical signal of each independent photosensitive channel determined by comparing the current electrical signal obtained at the detection time with the reference electrical signal corresponding to the independent photosensitive channel for each independent photosensitive channel. Based on the pre-calibrated correspondence between the electrical signal change and the light-shielding width change of the target independent photosensitive channel, the current light-shielding width change corresponding to the first electrical signal change is determined. The change in the current light-shielding width is determined as the current wear amount of the blade.

6. The blade condition detection method according to any one of claims 3 to 5, characterized in that, The damage detection of the blade based on the electrical signal includes: During the rotation of the blade with the spindle, an electrical signal is collected for at least two rotations of the blade. If the difference between the electrical signals at the same sampling position in at least two consecutive cycles is greater than a second preset difference, it is determined that the blade is damaged.

7. The blade condition detection method according to claim 6, characterized in that, After determining that the blade is damaged, the process also includes: Record the current number of consecutive breaks; If the number of consecutive breaks is greater than the breakage number threshold, then the blade is determined to be damaged. If the number of consecutive breaks is less than or equal to the breakage number threshold, then return to the step of collecting the electrical signal of the blade rotating at least two revolutions.

8. The blade condition detection method according to claim 6, characterized in that, After obtaining the average electrical signal of the current rotation, before detecting the relationship between the difference between the average electrical signal of the current rotation and a preset value, and / or, after acquiring the electrical signal of the blade rotating at least two rotations, before detecting the relationship between the difference between the electrical signals at the same sampling position in at least two consecutive rotations and a second preset difference, the method further includes: The average electrical signal and / or the electrical signal of at least two turns are subjected to frequency domain filtering, and a bandpass filter is used to retain the frequency band component where the blade characteristic signal is located and to filter out noise interference. And / or, perform nonlinear filtering on the average electrical signal and / or the electrical signal of at least two turns to remove impulse noise; The electrical signal after the frequency domain filtering and / or the nonlinear filtering is used as the noise-reduced electrical signal.

9. The blade condition detection method according to claim 1, characterized in that, After the step of time-division multiplexing the independent photosensitive channel via the channel selection switch and reading the electrical signal of the independent photosensitive channel, and before the step of performing damage detection and / or wear detection on the blade based on the electrical signal, the method further includes: If a continuous abnormality in the electrical signal of a single independent photosensitive channel is detected, it is determined that the independent photosensitive channel is being interfered with, and the interfered independent photosensitive channel is turned off by the channel selection switch. The step of performing damage detection and / or wear detection on the blade based on the electrical signal includes: The blade is subjected to damage detection and / or wear detection based on the electrical signals from other independent photosensitive channels that are not turned off.

10. A blade condition detection system, characterized in that, The system includes a sensing device for blade status monitoring and a main control unit. The sensing device includes a transmitter and a receiver arranged opposite each other, with a working distance between the transmitter and receiver for accommodating the blade. The receiver includes a photosensitive element and a signal receiving circuit board. The photosensitive element has multiple independent photosensitive channels arranged along the length of an elongated detection light curtain. Each independent photosensitive channel receives light from a corresponding position in the light curtain and outputs an electrical signal individually. The signal receiving circuit board is equipped with a channel selection switch, and the output terminal of each independent photosensitive channel is electrically connected to the corresponding input terminal of the channel selection switch. The main control unit is used to select each of the independent photosensitive channels in a time-division manner through the channel selection switch, read the electrical signals output by each of the independent photosensitive channels, and determine the photosensitive channel number or position corresponding to the junction point between the occluded area and the unoccluded area based on the comparison result of the electrical signals of each of the independent photosensitive channels with a preset threshold, and use it as the current junction point position. The current handover point position is compared with the historical handover point position. Based on the comparison result, the selection state of the channel selection switch is configured so that the initial light transmittance of the sensing device is consistent with the preset benchmark. The independent photosensitive channel is selected in a time-division manner through the channel selection switch, and the electrical signal of the independent photosensitive channel is read. The blade is subjected to damage detection and / or wear detection based on the electrical signal.

Images