Laser processing monitoring method and apparatus

Abstract

The present invention relates to a laser processing monitoring method and apparatus, and includes a step of irradiating a target object with a laser according to a processing start signal of the target object, a step of receiving reflected light reflected by the target object from the laser, a step of checking a reflection amount pattern for the reflected light, and a step of detecting whether a hole generated by the laser irradiation is defective based on the reflection amount pattern, and is also applicable as other embodiments.

Description

Technical Field

[0001] The present invention relates to a laser processing monitoring method and apparatus.

[0002] The present invention is derived from research conducted as part of the Information and Communication Broadcasting New Talent Cultivation Project of the following Korea Research Foundation. [Problem Specific Number] 171116160332 [Problem Number] 2020-0-01741-003 [Related Ministry Name] Ministry of Science and ICT [Problem Management (Specialized) Agency Name] Korea Research Foundation [Research Project Name] Information and Communication Broadcasting New Talent Cultivation (R&D) [Research Problem Name] Grand ICT Research Center [Contribution Rate] 1 / 2 [Research Performing Institution Name] Korea University Industry-Academia Cooperation Foundation [Research Period] 2022.01.01~2022.12.31 [Problem Specific Number] 2022001410 [Problem Number] RS-2022-00141076 [Related Ministry Name] Ministry of SMEs and Startups [Problem Management (Specialized) Agency Name] Technology Information Promotion Agency for SMEs (TIPA) [Research Project Name] Smart Manufacturing Innovation Technology Development Project [Research Problem Name] Development of an Integrated Solution for Edge Infrastructure Equipment Status Monitoring through Demonstration in the Electrical and Electronic Parts Industry [Contribution Rate] 1 / 2 [Research Performing Institution Name] (Reshenie) Co., Ltd. [Research Period] 2022.04.29~2025.12.31

Background Art

[0003] In microfabrication processes such as semiconductor and display manufacturing, laser devices are widely used for substrate surface materials, via hole machining, or specific pattern formation. For this purpose, technologies for processing laser beams into special forms, shaping the spatial form of lasers into lines, surfaces, etc., that is, technologies for maintaining the spatial intensity of the beam in a specific form or minimizing the transition width of the edge part have been developed.

[0004] However, such precisely shaped laser beams can have their beam characteristics deformed by environmental factors, etc., before being irradiated onto the imaging surface, i.e., the target, which is different from the initial recipe setting. There are problems where abnormalities occur in the processed articles. In particular, when using ultra-precision laser processing equipment, the processing quality inspection is mostly carried out after the process is completed. Thus, there are problems such as difficulty in correcting defective molds or reworking, and high costs associated with mold disposal.

Summary of the Invention

Problems to be Solved by the Invention

[0005] An embodiment of the present invention for solving such conventional problems is to provide a laser processing monitoring method and apparatus capable of confirming processing defects in real time according to the type of target during target processing.

[0006] Another embodiment of the present invention is to provide a laser processing monitoring method and apparatus capable of omitting a separate defect inspection process for defect inspection by confirming processing defects in real time during target processing.

Means for Solving the Problems

[0007] The laser processing monitoring method according to an embodiment of the present invention includes: a step of irradiating a target object with a laser according to a processing start signal of the target object; a step of receiving reflected light by which the laser is reflected by the target object; a step of checking a reflection amount pattern for the reflected light; and a step of detecting whether or not a hole generated by the laser irradiation is defective based on the reflection amount pattern.

[0008] Also, the step of detecting whether or not a hole is defective includes: a step of comparing a pre-stored learning pattern with the confirmed reflection amount pattern; and a step of confirming that a defect has occurred in the hole when the learning pattern and the reflection amount pattern differ by a threshold value or more.

[0009] The method further includes a step of storing the learning pattern in consideration of the type of the target object.

[0010] The method further includes a step of storing the learning pattern in consideration of the interval between holes and the diameter of the holes.

[0011] Also, the step of storing the learning pattern is a step of storing the learning pattern generated using an autoencoder in an unsupervised learning method.

[0012] The laser processing monitoring apparatus includes: a processing unit including a light source unit that irradiates a target object with a laser and a light receiving unit that receives reflected light by which the laser is reflected by the target object; and a control unit that controls the processing unit to irradiate the laser, checks a reflection amount pattern for the reflected light received by the light receiving unit, and detects whether or not a hole generated by the laser irradiation is defective based on the reflection amount pattern.

[0013] Further, the control unit is characterized in that when the difference between the pre-stored learning pattern and the confirmed reflection amount pattern is equal to or greater than a threshold value, it is confirmed that a defect has occurred in the hole.

[0014] It further includes a memory, and the control unit is characterized in that it stores the learning pattern in the memory in consideration of the type of the target object.

[0015] Further, the control unit is characterized in that it stores the learning pattern in the memory in consideration of the interval between the holes and the diameter of the holes.

[0016] Further, the control unit is characterized in that it generates the learning pattern using an autoencoder with an unsupervised learning method.

Advantages of the Invention

[0017] As described above, the laser processing monitoring method and apparatus according to the present invention can confirm processing defects in real time according to the type of the target object during target object processing and omit a separate defect inspection process, so that it is possible to confirm the defect type and characteristics according to the type of the target object, and there is an effect that the time required for the defect inspection process can be minimized.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Embodiments for Carrying Out the Invention

[0019] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The detailed description disclosed below together with the accompanying drawings is intended to illustrate exemplary embodiments of the present invention and does not represent the only embodiments in which the present invention can be implemented. In the drawings, parts not related to the description may be omitted for clarity of the present invention, and the same reference numerals are assigned to the same or similar components throughout the specification.

[0020] FIG. 1 is a diagram showing an electronic device for laser processing monitoring according to an embodiment of the present invention.

[0021] Referring to FIG. 1, an electronic device 100 according to the present invention may include a communication unit 110, a processing unit 120, an input unit 130, a display unit 140, a memory 150, and a control unit 160. The processing unit 120 may include a light source unit 121 and a light receiving unit 122.

[0022] The communication unit 110 can receive a learning pattern (hereinafter referred to as learning data) from an external server through communication with the external server (not shown). For this purpose, the communication unit 110 can perform wireless communication such as 5G (5 th generation communication), LTE-A (Long Term Evolution-Advanced), LTE (Long Term Evolution), Wi-Fi (Wireless Fidelity) with the external server.

[0023] At this time, the learning data is learning data in which processing conditions and a learning pattern are mapped, and may be generated using an AI algorithm stored in an external server. The processing conditions are conditions for the type of target object (for example, metal, ceramic, etc.), the thickness of the target object, the interval between via holes, the diameter of the via holes, and the number of via holes. The learning pattern may be a pattern generated by learning the reflection amount pattern when the target object is normally processed according to the processing conditions.

[0024] Note that when the target object is processed normally, the external server can apply the normal reflection amount pattern obtained by the electronic device 100 as an input to the AI algorithm. In addition, the external server can apply the environment set during the processing of the target object, that is, the processing conditions for the type of the target object, the thickness of the target object, the interval between via holes, the diameter of the via holes, the number of via holes, etc., and the reflection amount pattern as inputs to the AI algorithm to generate an autoencoder model, which is an artificial neural network trained in an unsupervised learning method, and can use this to generate learning data in which the processing conditions and the learning pattern are mapped.

[0025] The processing unit 120 can include a light source unit 121 for irradiating a laser beam (hereinafter generally referred to as a laser) and a light receiving unit 122 for receiving the reflected light. The laser irradiated from the light source unit 121 is reflected by the target object to generate reflected light, and the light receiving unit 122 receives the reflected light generated by being reflected by the target object. At this time, the light source unit 121 can be formed with a plurality of light sources so as to be able to irradiate according to the number of via holes included in the processing conditions. The light receiving unit 122 generates received light data using the received reflected light and provides this to the control unit 160. At this time, the light receiving unit 122 can mean a photodiode sensor or the like.

[0026] The input unit 130 generates input data corresponding to the input of an operator who operates the electronic device 100. The input unit 130 can include at least one input means of a key pad, a dome switch, a touch panel, a touch key, and a button.

[0027] The display unit 140 outputs output data by the operation of the electronic device 100. For this purpose, the display unit 140 may include a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, a micro electro mechanical systems (MEMS) display, and an electronic paper display. The display unit 140 may be coupled to the input unit 130 and implemented as a touch screen.

[0028] The memory 150 stores the operation program of the electronic device 100. The memory 150 may store an algorithm for generating a reflection amount pattern based on the received data generated by the light receiving unit 122, and an algorithm for generating learning data with the reflection amount pattern. The memory 150 may store learning data in which a learning pattern when the target object is normally processed is mapped according to processing conditions such as the type of the target object, the thickness of the target object, the interval between via holes, the diameter of the via holes, and the number of via holes. At this time, the learning pattern means a pattern in which a reflection amount pattern when the target object is normally processed according to the processing conditions is learned.

[0029] The control unit 160 can generate learning data through tests and store it in the memory 150, and can also store the learning data received from an external server in the memory 150 through communication with the external server. To generate learning data, the control unit 160 can call the AI algorithm stored in the memory 150 and apply the normal reflection amount pattern obtained when the target object is normally processed as the input of the AI algorithm. In addition, the control unit 160 can apply the environment set during target object processing, that is, the processing conditions for the type of target object, the thickness of the target object, the interval between via holes, the diameter of the via holes, and the number of via holes, etc., as the input of the AI algorithm, and can generate an autoencoder model, which is an artificial neural network trained in an unsupervised learning method.

[0030] The control unit 160 controls the processing unit 120 to irradiate the target object with a laser. The control unit 160 checks the reflection amount pattern using the received light data generated by the light receiving unit 122 that has received the reflected light reflected by the target object. The control unit 160 can detect the presence or absence of defects generated during target object processing by laser irradiation based on the confirmed reflection amount pattern.

[0031] More specifically, when the control unit 160 receives a processing start signal for processing the target object with a laser from the input unit 130, it controls the light source unit 121 with the processing start signal to irradiate the target object with a laser. At this time, the processing start signal can include processing conditions for the type of target object to be processed, the thickness of the target object, the interval between via holes, the diameter of the via holes, and the number of via holes, etc. The laser irradiated from the light source unit 121 is reflected by the target object to generate reflected light. The generated reflected light is received by the light receiving unit 122, and the light receiving unit 122 generates received light data based on the reflected light. The light receiving unit 122 provides the confirmed received light data to the control unit 160.

[0032] The control unit 160 checks the reflection amount pattern for the reflected light using the received light reception data. At this time, the reflected light is the largest when the laser is first irradiated on the target object, and the reflected light is the smallest when the processing such as the via hole on the target object is completed. That is, the magnitude of the reflected light can gradually decrease as time elapses after the laser irradiates the target object. The control unit 160 can check the reflection amount pattern for the reflected light generated as time elapses after the laser irradiates the target object.

[0033] The control unit 160 calls the learning data pre-stored in the memory 150. The control unit 160 compares the confirmed reflection amount pattern with the learning pattern included in the called learning data. The control unit 160 checks the learning pattern mapped under the same processing conditions as those included in the processing start signal among the learning data. At this time, the control unit 160 can compare the confirmed reflection amount pattern with the called learning pattern. If the difference between the two reflection amount patterns is equal to or greater than the threshold value, the control unit 160 determines that a defect has been detected during the target object processing and can display this on the display unit 140.

[0034] FIG. 2 is a flowchart for explaining a method of performing laser processing monitoring according to an embodiment of the present invention.

[0035] Referring to FIG. 2, in step 201, if the control unit 160 receives a processing start signal for processing the target object with a laser from the input unit 130, it proceeds to step 203. If the processing start signal is not received, it waits for the reception of the processing start signal. At this time, the processing start signal can include processing conditions for the type of the target object to be processed, the thickness of the target object, the interval between the via holes, the diameter of the via holes, and the number of the via holes, etc.

[0036] In step 203, the control unit 160 controls the light source unit 121 according to the received processing start signal to irradiate the target object with a laser. In step 205, the control unit 160 receives and checks the light reception data generated by the light receiving unit 122. More specifically, the laser irradiated on the target object is reflected by the target object to generate reflected light, and the light receiving unit 122 generates light reception data based on the received reflected light and provides it to the control unit 160.

[0037] In step 207, the control unit 160 checks the reflection amount pattern for the reflected light using the received light reception data. At this time, the reflected light is the largest when the laser is first irradiated on the target object, and the reflected light is the smallest when the processing of a via hole or the like on the target object is completed. That is, the magnitude of the reflected light can gradually decrease as time elapses after the laser irradiates the target object. The control unit 160 can check the reflection amount pattern for the reflected light generated as time elapses after the laser irradiates the target object.

[0038] In step 209, the control unit 160 calls the learning data pre-stored in the memory 150. At this time, the control unit 160 can call the learning data having the same processing conditions as the type of the target object to be processed, the thickness of the target object, the interval between via holes, the diameter of the via holes, and the number of via holes included in the processing start signal received in step 201. In step 211, the control unit 160 compares the reflection amount pattern confirmed in step 207 with the learning pattern included in the learning data called in step 209.

[0039] If the difference between the reflection amount pattern confirmed in step 207 and the called learning pattern in the comparison result of step 211 is greater than or equal to the threshold value, the control unit 160 performs step 213, and if the difference is less than the threshold value, the process can be terminated. In step 213, the control unit 160 determines that a defect has been detected during the target object processing and can display this on the display unit 140.

[0040] The embodiments of the present invention disclosed in this specification and the drawings are merely specific examples presented to easily explain the technical content of the present invention and assist in the understanding of the present invention, and are not intended to limit the scope of the present invention. Therefore, the scope of the present invention should be interpreted such that all changes or modified forms derived based on the technical idea of the present invention, in addition to the embodiments disclosed herein, belong to the scope of the present invention.

Claims

1. irradiating the target object with a laser according to a processing start signal of the target object; receiving reflected light reflected by the laser by the target object; confirming a reflection amount pattern for the reflected light; detecting whether a hole generated by the laser irradiation is defective based on the reflection amount pattern, characterized by comprising: a laser processing monitoring method.

2. The step of detecting whether the hole is defective is comparing a pre-stored learning pattern with the confirmed reflection amount pattern; confirming that a defect has occurred in the hole when the learning pattern and the reflection amount pattern differ by a threshold or more, characterized by the laser processing monitoring method according to claim 1.

3. further comprising storing the learning pattern in consideration of the type of the target object, characterized by the laser processing monitoring method according to claim 2.

4. further comprising storing the learning pattern in consideration of the interval between the holes and the diameter of the holes, characterized by the laser processing monitoring method according to claim 3.

5. The step of storing the learning pattern is storing the learning pattern generated using an autoencoder in an unsupervised learning method, characterized by the laser processing monitoring method according to claim 4.

6. a processing unit including a light source unit that irradiates a target object with a laser and a light receiving unit that receives reflected light reflected by the laser by the target object; a control unit that controls the processing unit to irradiate the laser, confirms a reflection amount pattern for the reflected light received by the light receiving unit, and detects whether a hole generated by the laser irradiation is defective based on the reflection amount pattern, characterized by comprising: a laser processing monitoring device.

7. The control unit compares a pre-stored learning pattern with the confirmed reflection amount pattern and confirms that a defect has occurred in the hole when they differ by a threshold or more, characterized by the laser processing monitoring device according to claim 6.

8. further comprising a memory, The control unit stores the learning pattern in the memory in consideration of the type of the target object, characterized by the laser processing monitoring device according to claim 7.

9. The control unit The laser processing monitoring device according to claim 8, wherein the control unit stores the learning pattern in the memory in consideration of the interval between the holes and the diameter of the holes. **Claim 10** The laser processing monitoring device according to claim 9, wherein the control unit generates the learning pattern using an autoencoder with an unsupervised learning method.

Images