Infrared heating lamp explosion-proof laser-induced experimental device and experimental method

Abstract

The application discloses an infrared heating lamp explosion-proof laser-induced experiment device and experiment method, and belongs to the technical field of equipment detection. The device comprises the following steps: installing at least three to-be-tested bulbs into the experiment device, laser-inducing each to-be-tested bulb, and obtaining a first data set; screening characteristic electrical mutation signals and thermal response mutation signals occurring when the bulbs break and performing synchronous determination; extracting multi-modal data segments before and after the occurrence of characteristic synchronous events, and establishing a characteristic synchronous event database; performing characteristic training and learning based on the characteristic synchronous event database, and performing risk assessment on the synchronous events of each to-be-tested bulb. In the implementation process of the technical scheme of the application, the to-be-tested bulbs are laser-induced and data is collected, synchronous determination is performed according to the time coupling relationship of the electrical signals and the thermal response mutation signals, so as to determine the risk level of the breakage of the to-be-tested bulbs.

Description

Technical Field

[0001] This application relates to the field of equipment testing technology, specifically to an explosion-proof laser-induced experimental device and method for infrared heating lamps. Background Technology

[0002] Halogen infrared heating lamps are widely used in semiconductor EPI, CVD, RTP and other processes. They have the advantages of fast heating and stable and reliable operation, and are therefore widely used.

[0003] Halogen infrared heating lamps experience filament burnout at the end of their lifespan. At the moment of filament burnout, an electric arc gas is generated inside the lamp. This gas rapidly heats up and expands under high current, which can easily cause the bulb to explode and break. Since these halogen infrared heating lamps are used in semiconductor equipment, an explosion can damage semiconductor equipment components and chip materials. Therefore, in existing technologies, it is usually necessary to conduct induced tests on halogen infrared heating lamps, such as using laser induction, to simulate the electric arc and gas expansion process at the moment of filament burnout, in order to identify high-risk breakage samples in advance.

[0004] However, existing laser-induced experimental methods mostly rely on a single signal to determine whether a bulb has broken, lacking collaborative analysis of multi-modal signals such as electrical, thermal response, and acoustic emission, resulting in a high misjudgment rate. In addition, it is difficult to accurately capture weak signs of breakage during the experiment, making it impossible to achieve early risk warning and making it difficult to accurately classify and dynamically predict breakage risks.

[0005] Therefore, it is necessary to provide an explosion-proof laser-induced experimental device and method for infrared heating lamps to solve the above problems.

[0006] It should be noted that the information disclosed in this background section is only for understanding the background technology of this application concept, and therefore may include information that does not constitute prior art. Summary of the Invention

[0007] Based on the aforementioned problems in the existing technology, the problem to be solved by this application is to provide an explosion-proof laser-induced experimental device and method for infrared heating lamps, which obtains multimodal parameters under the condition of breakage by conducting laser-induced experiments on the lamp under test, so as to facilitate subsequent monitoring.

[0008] The technical solution adopted by this application to solve its technical problem is: a method for conducting explosion-proof laser-induced experiments with an infrared heating lamp, comprising:

[0009] At least three light bulbs to be tested are installed in the experimental setup, and each light bulb is laser-induced. The first data set is then acquired using a data acquisition device.

[0010] Based on the first dataset, characteristic electrical and thermal response mutation signals that occur when a light bulb breaks are selected, and synchronous determination is made based on the temporal coupling relationship between the characteristic electrical and thermal response mutation signals.

[0011] Based on the synchronization determination results, multimodal data segments before and after the occurrence of characteristic synchronization events are extracted, and a characteristic synchronization event database is established. The electrical, thermal response and acoustic emission data within the time segment corresponding to each synchronization event are packaged and stored, and the event type and occurrence time are labeled.

[0012] Feature training and learning are performed based on a feature synchronization event database, and a risk assessment is conducted on the synchronization events of each light bulb under test to determine its risk level of breakage under laser-induced conditions.

[0013] In the implementation of the technical solution of this application, the risk level of the bulb under test is determined by laser induction and data acquisition, and synchronous judgment is made based on the time coupling relationship between electrical signal and thermal response mutation signal.

[0014] Furthermore, the first data set includes the following data:

[0015] Data on changes in electrical parameters of the bulb before and after laser induction; the time point at which the filament melts and the corresponding cumulative laser irradiation time; surface temperature distribution data of the filament at the moment of melting; image or video data of the bulb at the moment of melting or breaking; acoustic signals or micro-pressure changes generated during laser induction; and data on the deformation and crack propagation process of the bulb casing under thermal stress.

[0016] Furthermore, the synchronization determination process includes the following steps:

[0017] Timeline unification and sampling synchronization are performed, and the original data from all sources are timestamped according to a unified time base.

[0018] The mutation points of each type of data in the first dataset are extracted, and the temporal proximity analysis is performed on these mutation points. The synchronous events caused by filament burnout and their corresponding data are filtered to generate a sequence of characteristic synchronous events.

[0019] Based on the feature synchronization event sequence, the time-domain and frequency-domain feature parameters of each modal signal before and after the abrupt change moment are further extracted, including the current drop slope, the rise time of the infrared radiation peak, and the energy distribution of the acoustic emission signal main frequency. Multimodal feature vectors are constructed and normalized.

[0020] Furthermore, the abrupt change is manifested as a step change in signal amplitude or a sharp increase in derivative. The sudden drop in current and voltage fluctuation in electrical signals are detected by a differential algorithm based on a sliding window or a wavelet transform method. The instantaneous jump in infrared radiation intensity in thermal response data is identified, and the burst pulses of high-frequency components in acoustic emission signals are marked. Subsequently, time proximity analysis is performed on various abrupt changes.

[0021] Furthermore, if electrical abrupt changes, thermal response abrupt changes, and acoustic signal bursts occur simultaneously within a millisecond time window, it is determined to be a synchronization event caused by filament burnout, and these data are collected in chronological order to form a sequence of characteristic synchronization events.

[0022] Furthermore, the feature synchronization event database supports index queries by time range and event type.

[0023] Furthermore, when a combination of patterns with a steep current drop slope, short infrared response rise time, and high-intensity acoustic emission main frequency energy is identified, it is marked as a high-risk precursor to rupture.

[0024] Furthermore, after feature training and feature learning, the risk level of the light bulb breaking is determined. By calculating the similarity between the feature vector of the light bulb under the same excitation conditions and the high-risk events already marked in the database, Mahalanobis distance or cosine similarity is used for quantitative comparison to determine whether it is a high-risk breaking event.

[0025] An explosion-proof laser-induced experimental system for infrared heating lamps includes:

[0026] The data acquisition module is used to install at least three test bulbs into the experimental device, perform laser induction on each test bulb, and acquire the first data set through the acquisition device.

[0027] The synchronization determination module is used to filter out the characteristic electrical change signal and thermal response change signal that occur when the bulb breaks based on the first data set, and to make a synchronization determination based on the time coupling relationship between the characteristic electrical signal and the thermal response change signal.

[0028] The multimodal data fragment extraction module is used to extract multimodal data fragments before and after the occurrence of characteristic synchronization events based on the synchronization determination results, and to establish a characteristic synchronization event database. It packages and stores the electrical, thermal response and acoustic emission data within the time segment corresponding to each synchronization event, and marks the event type and occurrence time.

[0029] The risk assessment module is used to train and learn features based on the feature synchronization event database, and to assess the risk of each light bulb under test for synchronization events, determining the risk level of breakage under laser-induced conditions.

[0030] The beneficial effects of this application are: the infrared heating lamp explosion-proof laser-induced experimental device and experimental method provided by this application, by laser-inducing the light bulb under test and collecting data, and by synchronously judging the time coupling relationship between electrical signal and thermal response change signal, the risk level of the light bulb under test being broken can be determined.

[0031] In addition to the purposes, features, and advantages described above, this application has other purposes, features, and advantages. A further detailed description of this application will be provided below with reference to the figures. Attached Figure Description

[0032] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0033] Figure 1 This is a schematic diagram of the experimental method for explosion-proof laser-induced infrared heating lamps according to this application;

[0034] Figure 2 This is a schematic diagram of the module structure of an explosion-proof laser-induced experimental system for infrared heating lamps according to this application;

[0035] Figure 3 This is a schematic diagram of the structure of an explosion-proof laser-induced experimental device for infrared heating lamps according to this application;

[0036] The following are the labeling elements in the figure:

[0037] 1. First slide; 2. Second slide; 3. Fixing plate; 4. Mounting slot; 11. Ball screw; 12. Mounting platform. Detailed Implementation

[0038] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0039] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0040] Example 1: As Figure 1As shown, this application provides a laser-induced explosion-proof experimental method for infrared heating lamps. This method is mainly applied to the induction experiment of infrared heating lamps. Infrared heating lamps are devices used to generate infrared radiation and play a heating role in the semiconductor EPI process. However, there is a risk of cracking due to local overheating. Therefore, it is necessary to simulate extreme energy input through laser induction and monitor the infrared heating lamp to determine whether there are safety hazards such as thermal cracking, laser-induced explosion, and fragmentation. Specifically, the method includes the following steps:

[0041] Step 101: Install at least three test bulbs into the experimental setup, and perform laser induction on each test bulb, and acquire the first data set through the acquisition device;

[0042] When conducting laser-induced experiments on infrared heating lamps, a test device is required. This device is used to fix the lamp under test and provide a controllable and stable laser irradiation environment. After the lamp under test is installed, it is laser-induced. Under the action of laser irradiation, the filament inside the lamp under test will melt due to high temperature, thereby simulating the situation when the lamp reaches the end of its life in actual use and accelerating the aging process of the lamp.

[0043] The experimental setup can simultaneously install multiple light bulbs, and each bulb is tested under rated voltage. Laser induction is achieved using a high-power infrared laser. For each bulb under test, laser induction is applied individually, that is, one or more high-power infrared laser beams are used to locally or directionally irradiate specific areas of the bulb (such as the filament support, glass seal, weak areas on the bulb surface, etc.) to simulate filament overheating, melting, or localized thermal stress concentration in a short time, thereby accelerating its aging process and inducing potential breakage behavior. In this embodiment, at least three bulbs under test are required (preferably 3 to 10) to obtain sufficient sample data.

[0044] The laser used is a high-power infrared laser source, with a wavelength range of 800nm ​​to 1500nm (to match the radiation absorption characteristics of the infrared heating lamp). The laser power density is controlled between 1W / cm² and 50W / cm². The spot size and irradiation time are adjusted according to experimental needs, and the laser-induced intensity and action time received by each bulb are kept consistent to ensure the consistency of experimental conditions and the comparability of results.

[0045] During the experiment, the electrical parameters of each test bulb were collected when it melted, and these parameters were used as the first data set. The data collection devices included, but were not limited to, high-speed cameras, electrical parameter acquisition devices, infrared thermal imagers, acoustic emission sensors, etc. Multi-dimensional data of each test bulb during the laser induction process and at the moment of melting were collected in real time. The collected data together constituted the first data set.

[0046] The first dataset includes the following data:

[0047] The data includes changes in electrical parameters of the light bulb before and after laser induction, such as changes in voltage, current, power, and resistance; the time point at which the filament melts and the corresponding cumulative laser irradiation time; surface temperature distribution data of the filament at the moment of melting, which is collected by an infrared thermal imager; image or video data of the light bulb at the moment of melting or breaking, which is collected by a high-speed camera; acoustic signals or micro-pressure changes generated during laser induction, which are collected by acoustic emission sensors and pressure sensors; and data on the deformation and crack propagation process of the bulb casing under thermal stress.

[0048] By collecting the above data, it is possible to apply consistent laser-induced effects to multiple infrared heating lamp samples under the same experimental conditions, and obtain data on changes in electrical characteristics, thermal response characteristics, and structural integrity when laser energy is input near the end of their lifespan. This data serves as the basis for subsequent multi-dimensional feature analysis and lifespan prediction model construction.

[0049] Step 102: Based on the first dataset, filter out the characteristic electrical change signal and thermal response change signal that occur when the light bulb breaks, and make a synchronous determination based on the time coupling relationship between the characteristic electrical signal and the thermal response change signal;

[0050] The first dataset contains not only electrical parameters but also multimodal data such as thermal response, acoustics, and mechanical deformation. These multimodal data are usually correlated in the time dimension. Especially at the moment of filament melting, the sudden change in electrical parameters is often accompanied by a sudden change in thermal radiation intensity and acoustic emission signals caused by structural rupture. However, due to differences in sensor sampling frequencies and signal transmission paths, there may be slight time offsets between the various modal signals. In this case, it is not possible to directly determine the precise time and cause of filament melting or structural rupture at the raw data level. Therefore, it is also necessary to perform synchronous determination based on the temporal coupling relationship between characteristic electrical signals and thermal response abrupt change signals. Specifically, this synchronous determination process includes the following steps:

[0051] Timeline unification and sampling synchronization are performed, and the original data from all sources are timestamped according to a unified time base.

[0052] The raw data from all sources are timestamped according to a unified time base, and the interpolation resampling method is used to align data sequences with different sampling rates to the same time axis, thereby eliminating time offsets caused by differences in sampling rates and hardware, and ensuring that the data of each modality are aligned in the time dimension.

[0053] The mutation points of each type of data in the first dataset are extracted, and the temporal proximity analysis is performed on these mutation points. The synchronous events caused by filament burnout and their corresponding data are filtered to generate a sequence of characteristic synchronous events.

[0054] After time alignment, the various data in the first dataset can be used to extract abrupt changes. These abrupt changes are usually manifested as step changes in signal amplitude or sharp increases in derivative. Therefore, the sudden drop in current and voltage fluctuation in electrical signals can be detected by differential algorithms based on sliding windows or wavelet transform methods, the instantaneous jump in infrared radiation intensity in thermal response data can be identified, and the burst pulses of high-frequency components in acoustic emission signals can be marked. Then, time proximity analysis is performed on various abrupt changes. If electrical abrupt changes, thermal response abrupt changes, and acoustic signal bursts occur simultaneously within a millisecond time window, they are determined to be synchronous events caused by filament burnout. These data are then collected in chronological order to form a sequence of characteristic synchronous events.

[0055] Based on the feature synchronization event sequence, the time-domain and frequency-domain feature parameters of each modal signal before and after the abrupt change moment are further extracted, including the current drop slope, the rise time of the infrared radiation peak, and the energy distribution of the acoustic emission signal main frequency. Multimodal feature vectors are constructed and normalized.

[0056] Based on the feature synchronization event sequence, key features of each modal signal before and after the mutation are extracted, such as the current drop slope, the rise time of the infrared radiation peak, and the energy proportion of the acoustic emission main frequency band. Multidimensional feature vectors are constructed, and then the dimensions of each feature parameter are unified by the normalization method to eliminate the weight bias caused by the difference in signal amplitude, thereby providing subsequent standardized input.

[0057] Step 103: Based on the synchronization determination result, extract multimodal data segments before and after the occurrence of the characteristic synchronization event, and establish a characteristic synchronization event database. Package and store the electrical, thermal response and acoustic emission data within the time segment corresponding to each synchronization event, and label the event type and occurrence time.

[0058] After synchronization determination, a feature synchronization event database can be established based on the multimodal data segments in the feature synchronization event sequence. The current, voltage, infrared thermal image and acoustic emission signals in the time segment corresponding to each synchronization event related to filament melting or breaking are extracted, packaged and stored, and the event type is labeled as "filament melting" or "filament breaking", as well as the specific timestamp of the event, which is convenient for subsequent model training or fault mode comparison. At the same time, it supports indexing and querying by time range and event type to improve data analysis efficiency.

[0059] Step 104: Perform feature training and learning based on the feature synchronization event database, and conduct risk assessment on the synchronization events of each light bulb under test to determine the risk level of breakage under laser-induced conditions.

[0060] After establishing a database of characteristic synchronous events, it is also necessary to train the model and learn the features based on the known data in the database, and to perform pattern recognition and labeling on high-risk events. For example, a combination of steep current drop slope, short infrared response rise time and high-intensity acoustic emission main frequency energy can be identified and marked as a high-risk rupture precursor. Alternatively, when a slow current drop, a steady rise in infrared radiation and a weak acoustic emission signal are identified, it can be determined as a low-risk fuse failure mode.

[0061] After feature training and feature learning, it is also necessary to determine the risk level of bulb breakage. Specifically, this method calculates the similarity between the feature vector of the bulb under test under the same excitation conditions and the high-risk events marked in the database. It uses Mahalanobis distance or cosine similarity for quantitative comparison. If the current drop slope is steeper than the threshold, the infrared peak rise time is shorter than the typical fuse precursor, and the proportion of high-frequency acoustic emission energy is abnormal, it is determined to be a high-risk breakage event, outputs the corresponding risk level warning, and triggers the graded alarm mechanism.

[0062] When the similarity index exceeds the preset threshold multiple times in a row, and the consistency of multimodal features deviates from the normal range, the risk level will be automatically increased, and an early warning signal will be sent to the control system to start the protection program, cut off the laser excitation source and record the experimental parameters to avoid equipment damage.

[0063] After the aforementioned process, as the amount of data increases, during subsequent bulb use, it is only necessary to collect its electrical parameters, thermal response, and acoustic emission signals in real time. The trained model can then dynamically match the feature synchronization event pattern to achieve continuous monitoring and risk prediction of the filament status, thereby reducing the impact of bulb breakage on the equipment.

[0064] Example 2: Figure 2 As shown, this application also proposes an infrared heating lamp explosion-proof laser-induced experimental system, which operates the experimental method in Embodiment 1. The system includes:

[0065] The data acquisition module is used to install at least three test bulbs into the experimental device, perform laser induction on each test bulb, and acquire the first data set through the acquisition device.

[0066] The synchronization determination module is used to filter out the characteristic electrical change signal and thermal response change signal that occur when the bulb breaks based on the first data set, and to make a synchronization determination based on the time coupling relationship between the characteristic electrical signal and the thermal response change signal.

[0067] The multimodal data fragment extraction module is used to extract multimodal data fragments before and after the occurrence of characteristic synchronization events based on the synchronization determination results, and to establish a characteristic synchronization event database. It packages and stores the electrical, thermal response and acoustic emission data within the time segment corresponding to each synchronization event, and marks the event type and occurrence time.

[0068] The risk assessment module is used to train and learn features based on the feature synchronization event database, and to assess the risk of each light bulb under test for synchronization events, determining the risk level of breakage under laser-induced conditions.

[0069] Example 3: Figure 3 As shown, this application also proposes an explosion-proof laser-induced experimental device for infrared heating lamps. This device is used to fix an infrared laser source and can adjust the position of the infrared laser source to achieve directional irradiation of different bulbs. The experimental device includes a first slide 1, on which a ball screw 11 is provided. The ball screw 11 is connected to a drive motor (not shown in the figure), and the ball screw 11 is rotatably connected to a mounting platform 12. Under the drive of the drive motor, the mounting platform 12 can reciprocate along the axial direction of the ball screw 11.

[0070] The mounting platform 12 is provided with a second slide 2, which has the same structure as the first slide 1. A fixing plate 3 is slidably provided on the second slide 2. Under the cooperative action of the first slide 1 and the second slide 2, the fixing plate 3 can be adjusted in the horizontal and vertical directions.

[0071] The side of the fixing plate 3 is provided with a mounting groove 4, which is used to install the laser lamp holder. The laser lamp holder is fixed with an infrared laser source. When it is necessary to adjust the irradiation position of the laser source, the ball screw 11 is rotated by the drive motor, so that the mounting platform 12 moves along the first slide 1. At the same time, the position of the fixing plate 3 on the second slide 2 is adjusted to achieve precise positioning of the laser source in the horizontal and vertical directions.

[0072] The structure of the laser lamp holder can refer to existing technology and will not be described in detail in this embodiment.

[0073] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for an explosion-proof laser-induced experiment using an infrared heating lamp, characterized in that: The application relates to a method for evaluating the risk of lamp rupture under laser induction. The method comprises the following steps: At least three to-be-tested bulbs are installed into an experimental device, and laser induction is carried out on each to-be-tested bulb respectively, and a first data set is obtained through a collection device; Based on the first data set, characteristic electrical mutation signals and thermal response mutation signals occurring when the bulb ruptures are screened out, and synchronous determination is carried out according to the time coupling relationship of the characteristic electrical signals and the thermal response mutation signals; Based on the synchronous determination result, multi-modal data segments before and after the occurrence of a characteristic synchronous event are extracted, and a characteristic synchronous event database is established, the electrical, thermal response and acoustic emission data in the time segment corresponding to each synchronous event are packed and stored, and the event type and occurrence time are labeled; 2. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 1, characterized in that: Based on the characteristic synchronous event database, characteristic training and learning are carried out, and the risk of rupture of each to-be-tested bulb under laser induction is evaluated to determine the risk level of rupture. The first data set comprises the following data:

3. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 1, characterized in that: Electrical parameter change data of the to-be-tested bulb before and after the start of laser induction; time nodes of filament fusing and corresponding laser irradiation cumulative time; surface temperature distribution data of the filament at the moment of fusing; image or video data of the bulb at the moment of fusing or rupturing; acoustic signals or micro-pressure change data generated during laser induction; and deformation and crack propagation process data of the bulb shell under thermal stress. The synchronous determination process comprises the following steps: Time axis unification and sampling synchronization are carried out, and all original data from different sources are time-stamped corrected according to a unified time reference; Mutation points of various data in the first data set are extracted respectively, time proximity analysis is carried out on the mutation points, synchronous events caused by filament fusing and corresponding data are screened out, and a characteristic synchronous event sequence is generated; 4. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 3, characterized in that: Based on the characteristic synchronous event sequence, time domain and frequency domain characteristic parameters of various modal signals before and after the mutation moment are further extracted, including current drop slope, infrared radiation peak rise time, acoustic emission signal main frequency energy distribution, a multi-modal feature vector is constructed and normalized.

5. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 4, characterized in that: The mutation points are step changes in signal amplitude or sharp rises in derivative, current sudden drop and voltage fluctuation in the electrical signal are detected through a difference algorithm based on a sliding window or a wavelet transform method, instantaneous jump of infrared radiation intensity in the thermal response data is identified, and burst pulses of high-frequency components in the acoustic emission signal are marked, and then time proximity analysis is carried out on various mutation points.

6. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 1, characterized in that: If the electrical mutation, thermal response mutation and acoustic signal burst simultaneously appear in a millisecond-level time window, the synchronous event caused by filament fusing is determined, and the data are collected in time sequence to form a characteristic synchronous event sequence.

7. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 1, characterized in that: The characteristic synchronous event database supports index query according to time range and event type. When a combination mode with steep current drop slope, short infrared response rise time and high-intensity acoustic emission main frequency energy is identified, it is marked as a high-risk rupture precursor.

8. The explosion-proof laser-induced experiment method of an infrared heating lamp according to claim 1, characterized in that: After the feature training and feature learning are performed, the risk level of the bulb breakage is judged, the similarity of the feature vector of the to-be-tested bulb under the same excitation condition and the high-risk event labeled in the database is calculated, the Mahalanobis distance or cosine similarity is used for quantitative comparison, and it is determined whether it is a high-risk breakage event.

9. An explosion-proof laser-induced experiment system of an infrared heating lamp, used for implementing the explosion-proof laser-induced experiment method of the infrared heating lamp according to any one of claims 1 to 8, characterized in that: The method comprises the steps of: a data acquisition module is configured to install at least three to-be-tested bulbs into an experimental device, perform laser induction on each to-be-tested bulb, and obtain a first data set through a collection device; a synchronous judgment module is configured to filter out characteristic electrical mutation signals and thermal response mutation signals occurring when the bulb breaks based on the first data set, and perform synchronous judgment according to the time coupling relationship of the characteristic electrical signals and the thermal response mutation signals; a multi-modal data segment extraction module is configured to extract multi-modal data segments before and after the characteristic synchronous event based on the synchronous judgment result, establish a characteristic synchronous event database, pack and store the electrical, thermal response and acoustic emission data in the time segment corresponding to each synchronous event, and label the event type and occurrence time; a risk assessment module is configured to perform feature training and learning based on the characteristic synchronous event database, and perform risk assessment on the synchronous event of each to-be-tested bulb to judge the risk level of the bulb breakage under the laser induction condition.

10. An explosion-proof laser-induced experiment apparatus of an infrared heating lamp, characterized in that: An infrared heating lamp explosion-proof laser induction experiment method is used to implement the method according to any one of claims 1 to 8.

Images