Automobile lamp with structural health monitoring function

Abstract

The application provides an automobile lamp with a structural health detection function, comprising a lamp, the lamp comprising an outer light distribution mirror, a shell, an internal light-emitting component, and a sealing component connecting the outer light distribution mirror and the shell and forming a sealed cavity, further comprising: an elastic wave sensor array comprising a plurality of elastic wave sensor units, emitting elastic wave signals as an excitation source and receiving elastic wave signals as a receiver; a signal processing module extracting time domain characteristics and frequency domain characteristics of the received elastic wave signals; a health assessment module calculating a lamp structure health index according to deviations between the time domain characteristics and the frequency domain characteristics and pre-labeled reference characteristics; a defect type identification module inputting a multi-dimensional feature vector formed by the time domain characteristics and the frequency domain characteristics into a trained machine learning classification model and obtaining a defect type of the lamp structure; and a life prediction module calculating a remaining service life of the lamp according to the lamp structure health index and a pre-set attenuation model.

Description

Technical Field

[0001] This invention relates to the field of automotive lighting technology, and more specifically, to an automotive lighting fixture with structural health detection function. Background Technology

[0002] Automotive lighting is a critical component for safe vehicle operation, serving important functions such as illumination, signal indication, and aesthetic decoration. As automotive lighting structures become increasingly complex, the reliability challenges faced by the external lens and housing of automotive lights are also becoming more severe. External lens materials for automotive lights are typically made of PMMA or PC, while the housings are usually made of ABS or PP. During long-term use, these materials are affected by the following factors: ultraviolet radiation leading to material aging and degradation; mechanical impact causing the initiation and propagation of microcracks; temperature cycling causing thermal fatigue and material performance degradation; and chemical corrosion leading to surface deterioration, discoloration, and decreased mechanical properties. The welded structure or sealant between the automotive light housing and the external lens may fail under long-term vibration loads and thermal cycling, allowing moisture and dust to enter the lamp interior, affecting optical performance and accelerating the corrosion of electrical components.

[0003] However, existing automotive lighting technologies have significant limitations in terms of structural integrity or structural health testing, mainly in the following aspects: 1. Outdated detection methods: Relying on periodic manual visual inspections or reactive repairs after malfunctions, real-time online detection is not possible; 2. Defect identification failure: The system cannot automatically identify the types of defects in the lighting fixture structure (such as breakage, cracks, aging, connection failure, etc.). 3. Lack of predictive capability: The remaining service life of the lighting fixture structure cannot be predicted, making predictive maintenance impossible; 4. Delayed maintenance response: By the time users discover problems such as fogging, water ingress, or decreased brightness in the lights, the lights have often already suffered serious damage.

[0004] These limitations not only affect the reliability and lifespan of automotive lighting fixtures, but may also lead to user complaints and safety hazards. Therefore, there is an urgent need for a structural health monitoring technology for automotive lighting fixtures capable of real-time detection, defect identification, and lifespan prediction. Summary of the Invention

[0005] This invention provides an automotive lamp with structural health detection function, aiming to solve the problem that existing automotive lamps cannot detect the structural health status, automatically identify defect types, and predict the remaining lifespan in real time, thereby improving the reliability, service life, and driving safety of the lamp.

[0006] To achieve the above objectives, the present invention provides an automotive lamp with structural health detection function, comprising a lamp, the lamp including an external light-emitting lens, a housing, an internal light-emitting component, and a sealing component connecting the external light-emitting lens and the housing to form a sealed cavity, and further comprising: An elastic wave sensor array includes multiple elastic wave sensor units, each of which acts as an excitation source to emit elastic wave signals and as a receiver to receive elastic wave signals. The signal processing module has its input terminal electrically connected to the output terminal of the elastic wave sensor array, and the signal processing module extracts the time-domain and frequency-domain features of the received elastic wave signal. A health assessment module, the input of which is electrically connected to the signal processing module, calculates the structural health index of the luminaire based on the deviation between the time-domain characteristics and frequency-domain characteristics and the pre-calibrated reference characteristics; A defect type identification module is provided, the input of which is electrically connected to the signal processing module. The defect type identification module inputs the multi-dimensional feature vector composed of the time domain features and frequency domain features into a trained machine learning classification model and obtains the defect type of the lamp structure. The lifespan prediction module is electrically connected to the output of the health assessment module. The lifespan prediction module calculates the remaining lifespan of the lamp based on the lamp structure health index and a preset attenuation model.

[0007] In one embodiment, the number of elastic wave sensor units is 4 to 8, and the multiple elastic wave sensor units are arranged in an equally spaced grid, with the spacing between adjacent elastic wave sensor units not exceeding 100 mm.

[0008] In one embodiment, the elastic wave sensor array includes the following operating modes: In the first mode, the elastic wave sensor array is mounted on the external light distribution lens, and the sensor units excite and receive each other. In the second mode, the elastic wave sensor array is mounted on the housing, and the sensor units excite and receive each other. In the third mode, the elastic wave sensor array is mounted on the external light distribution lens and the housing. In this mode, the sensor unit on the external light distribution lens and the sensor unit on the housing excite and receive each other.

[0009] In one embodiment, when the elastic wave sensor unit is used as an excitation source, it emits a sinusoidal pulse signal containing 4 to 6 cycles, with an excitation frequency range of 30 Hz to 70 Hz and an excitation voltage amplitude of 10 V to 16 V.

[0010] In one embodiment, the signal processing module samples the received elastic wave signal at a sampling frequency at least twice the excitation frequency, and the sampling time covers the entire excitation pulse period.

[0011] In one embodiment, the time-domain features include time of flight, signal energy, signal peak value, and number of zero crossings; the frequency-domain features include average frequency, frequency standard deviation, centroid frequency, and mean square frequency.

[0012] In one embodiment, the health assessment module calculates the structural health index of the luminaire according to the following formula: LSHI = 100-Σ(w i × |f i －f i0 |) in, f i This is the i-th feature value extracted at the current time. f i0 This is the baseline value for the i-th feature in a healthy state; w i The weight coefficients for the i-th eigenvalue satisfy... Sw i = 1 .

[0013] In one embodiment, the health assessment module further performs a grading determination based on the value of the lamp structure health index, and the grading determination specifically includes: When LSHI ≥ 90, it is classified as Grade I, i.e., healthy; When 80 ≤ LSHI < 90, it is classified as Grade II, i.e., sub-health. When 60 ≤ LSHI < 80, it is Level III, i.e., a warning. When 40 ≤ LSHI < 60, it is classified as Level IV, i.e., dangerous; When LSHI < 40, it is classified as Grade V, which is severe.

[0014] In one embodiment, the machine learning classification model is a support vector machine model, and the kernel function used by the support vector machine model is a radial basis function; The multidimensional feature vector includes the flight time, signal energy, signal peak value, number of zero crossings, average frequency, frequency standard deviation, centroid frequency, and mean square frequency. The defect type identification module identifies defect types including fracture, crack, aging, and connection failure.

[0015] In one embodiment, the lifespan prediction module calculates the remaining lifespan of the luminaire according to the following formula. LRUL : LRUL = t 0 - τ × ln(LSHI(t) / LSHI 0 ) in t 0 Design the lifespan of the lighting fixtures. t This is the time constant for the decay of the lamp's lifespan. LSHI 0 The structural health index of the luminaire under initial healthy conditions. LSHI(t) For runtime t The health index of the lighting fixture structure.

[0016] The present invention has the following beneficial effects: 1. A quantitative description of the structural health status of luminaires has been achieved. By introducing the Structural Health Index (LSHI), the deviations of time-domain and frequency-domain characteristic parameters are weighted and fused to quantify the structural health level into a continuous value. Combined with multi-level judgment criteria, this provides an accurate quantitative basis for maintenance decisions, overcoming the limitation of existing technologies that can only make qualitative judgments.

[0017] 2. Automatic identification and continuous optimization of defect types have been achieved. A support vector machine classification model is adopted, using the multi-dimensional time-frequency feature vector of the elastic wave signal as input. It can automatically identify various defect types such as fracture, crack, aging, and connection failure. Furthermore, the model supports online incremental learning, continuously improving the identification accuracy as new samples accumulate.

[0018] 3. It enables quantitative prediction of the remaining service life of lighting fixtures. Based on the structural health index decay model and combined with the time constant calibrated by accelerated aging tests, the remaining service life of lighting fixtures is dynamically calculated, providing a direct time reference for predictive maintenance and transforming the maintenance method from reactive repair to proactive prevention. Attached Figure Description

[0019] Figure 1 This is a structural schematic diagram of an automotive lamp with structural health detection function according to an embodiment of the present invention; Figure 2 In Example 1 Figure 1 A magnified view of part A; Figure 3 In Example 2 Figure 1 A magnified view of part A; Figure 4 This is a schematic diagram of the power supply and communication interface module according to an embodiment of the present invention.

[0020] Among them, 1 is the external light distribution lens, 2 is the housing, 3 is the internal light-emitting component, 4 is the sealing component, 5 is the elastic wave sensor array, 6 is the signal processing module, 7 is the health assessment module, 8 is the defect type identification module, 9 is the life prediction module, and 10 is the power supply and communication interface module. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0022] See Figure 1 , Figure 2 , Figure 4 An embodiment of this application provides an automotive lamp with structural health detection function, comprising a lamp, the lamp including an external light-emitting lens 1, a housing 2, an internal light-emitting component 3, and a sealing component 4 connecting the external light-emitting lens and the housing to form a sealed cavity, and further including: The elastic wave sensor array 5 includes multiple elastic wave sensor units, each of which acts as an excitation source to emit elastic wave signals and as a receiver to receive elastic wave signals. The signal processing module 6 has its input terminal electrically connected to the output terminal of the elastic wave sensor array 5. The signal processing module 6 extracts the time-domain and frequency-domain characteristics of the received elastic wave signal. Multiple elastic wave sensor units are arranged on the inner surface of the outer light distribution mirror and / or the inner sidewall of the housing. The health assessment module 7 is electrically connected to the signal processing module 6. The health assessment module 7 calculates the health index of the lamp structure based on the deviation between the time domain characteristics and frequency domain characteristics and the pre-calibrated reference characteristics. The defect type identification module 8 is electrically connected to the signal processing module. The defect type identification module 8 inputs the multi-dimensional feature vector composed of the time domain features and frequency domain features into the trained machine learning classification model and obtains the defect type of the lamp structure. The lifespan prediction module 9 is electrically connected to the output of the health assessment module 7. The lifespan prediction module 9 calculates the remaining lifespan of the lamp based on the lamp structure health index and a preset attenuation model.

[0023] The power supply and communication interface module 10 is electrically connected to the signal processing module 6, the health assessment module 7, the defect type identification module 8, and the life prediction module 9, respectively. The power supply and communication interface module 10 supplies power to each module and transmits the output information of the health assessment module 7, the defect type identification module 8, and the life prediction module 9 to the vehicle control terminal.

[0024] The external lens 1 is made of PMMA material, with a thickness of 3mm and a circular shape with a diameter of 105mm. The housing 2 is made of ABS material, with a thickness of 2mm. In Example 1, as... Figure 2 As shown, the external light distribution lens 1 and the housing 2 are ultrasonically welded to form a sealing assembly 4, connecting the two and enclosing a sealed cavity. The internal light-emitting assembly 3 includes a PCB board with multiple LED beads and corresponding optical sub-assemblies such as reflectors and refractive lenses, all of which can be implemented using conventional structures in the art.

[0025] The elastic wave sensor array 5 consists of four elastic wave sensor units 5a, 5b, 5c, and 5d. Each elastic wave sensor unit has a side length of 5mm and a thickness of 3mm, and possesses bidirectional electromechanical coupling characteristics. It can both generate elastic waves as an excitation source under voltage excitation and receive elastic waves and convert them into electrical signals as a receiver. In this embodiment, elastic wave sensor units 5a and 5b are uniformly attached to the inside of the black decorative area on the inner surface of the outer lens 1 at a spacing of approximately 65mm, while elastic wave sensor units 5c and 5d are uniformly attached to the inner sidewall of the housing 2 at a spacing of approximately 75mm. All elastic wave sensor units are electrically connected to the signal processing module 6 via wires on a printed circuit board.

[0026] During operation, the elastic wave sensor array 5 operates according to a preset multi-mode excitation-reception strategy, including the following three modes: Mode 1 (Testing the external lens): Using elastic wave sensor unit 5a as the excitation source and sensor unit 5b as the receiver; then using elastic wave sensor unit 5b as the excitation source and elastic wave sensor unit 5a as the receiver; each group is tested once. The elastic wave generated by the excitation mainly propagates along the material of the external lens, and the changes in the received signal can reflect whether cracks, aging or other damage have occurred inside the external lens.

[0027] Mode 2 (Shell Inspection): Elastic wave sensor unit 5c is used as the excitation source, and elastic wave sensor unit 5d is used as the receiver; then, elastic wave sensor unit 5d is used as the excitation source, and elastic wave sensor unit 5c is used as the receiver; each group is tested once. Elastic waves propagate in the shell material and are used to determine the integrity of the shell structure.

[0028] Mode 3 (Detection of Sealed Connection Area): Elastic wave sensor units 5a and 5b are used as excitation sources, and elastic wave sensor units 5c and 5d are used as receivers; elastic wave sensor units 5c and 5d are used as excitation sources, and elastic wave sensor units 5a and 5b are used as receivers; a total of 2 sets of tests; the excitation signal passes through the external light distribution lens, the external light distribution lens and the housing connection area and the housing to the receiver, and is used to detect the sealing status of the lamp structure.

[0029] In all modes, the excitation source signal uses a 5-cycle sinusoidal pulse signal with an excitation frequency of 50Hz and an excitation voltage amplitude of 13.5V. The excitation mode uses single-frequency excitation, i.e., the excitation voltage can be expressed as u(t) = 6.75 × sin(2π × 50 × t), 0 ≤ t ≤ 10ms. The reception and sampling of the elastic wave is controlled by the signal processing module 6, with the sampling frequency set to 100Hz (i.e., twice the excitation frequency) and the sampling time lasting 10ms, covering the entire excitation pulse cycle.

[0030] Signal processing module 6 is a general-purpose programmable controller, located on a PCB board inside the lamp. Its input terminals are electrically connected to the output terminals of each elastic wave sensor unit, and it is responsible for extracting features from the sampled elastic wave signals.

[0031] In one embodiment, the extracted time-domain features include flight time. f 1 and signal energy f 2 Frequency domain characteristics include average frequency. f 3 and frequency standard deviation f 4 The sampling window width for feature extraction is set to 30ms, with the excitation signal emission time as the sampling start point, to ensure that all feature information of elastic wave propagation is fully captured.

[0032] The health assessment module 7 and the signal processing module 6 share the same programmable controller. This module pre-stores reference characteristic values, which are time-domain and frequency-domain characteristic values ​​obtained from the same model of lamps in perfect, defect-free condition at the time of manufacture, calibrated under the same test conditions to obtain the time-of-flight reference values. f 10 Signal energy reference value f 20 Average frequency reference value f 30 and frequency standard deviation benchmark f 40 .

[0033] The health assessment module 7 obtains the currently extracted time-domain and frequency-domain feature values ​​from the signal processing module 6 in real time, and calculates the Lamp Structure Health Index (LSHI) according to the following formula: LSHI = 100-Σ(w i × |f i －f i0 |) in, f i This is the i-th feature value extracted at the current time. f i0 This is the baseline value for the i-th feature in a healthy state; w i The weight coefficients for the i-th eigenvalue satisfy... Sw i = 1 , f i This is the i-th feature value extracted at the current time. f i0 The baseline value of the i-th feature in a healthy state is obtained through initial calibration or statistical learning. w i The weight coefficients for the i-th eigenvalue satisfy... Sw i = 1 The weights are determined based on the sensitivity of each feature to structural damage.

[0034] Preferably, the weight of the time-of-flight feature is 0.3, the weight of the signal energy feature is 0.25, the weight of the frequency standard deviation feature is 0.25, and the weight of the average frequency feature is 0.2.

[0035] In one embodiment, LSHI can be calculated using the following formula: LSHI=100-(0.3×|f 1 - f 10 |+0.25×|f 2 - f 20 |+0.2×|f 3 - f 30 |+0.25×|f 4 - f 40 |) The LSHI value ranges from 0 to 100, with higher values ​​indicating healthier structures. After obtaining the LSHI value, the health assessment module 7 further classifies the structure according to preset thresholds, with the following classification criteria: Level I (Healthy): LSHI ≥ 90, at this point the luminaire structure is intact and no maintenance is required; Level II (Sub-health): 80 ≤ LSHI < 90, indicating slight structural changes. It is recommended to pay attention, but it does not affect the use of the product. Level III (Warning): 60 ≤ LSHI < 80 indicates possible initial damage, and professional examination is recommended; Level IV (Danger): 40 ≤ LSHI < 60, indicating that the structural damage is quite serious and repairs should be arranged immediately; Level V (Severe): LSHI < 40. At this point, structural integrity has been significantly compromised, and relevant components need to be replaced to ensure safety.

[0036] The defect type identification module 8 is another independent or shared programmable controller, mounted on the PCB board. This module incorporates a machine learning classification model based on Support Vector Machine (SVM), using a Radial Basis Function (RBF) kernel to accommodate the nonlinear boundaries of defect classification. During model training, for four types of defects—fracture, crack, aging, and connection failure—at least 150 known samples are collected for each defect, and an 8-dimensional feature vector is extracted as input. The 8-dimensional feature vector specifically includes: Four time-domain characteristics: flight time, signal energy, signal peak value, and number of zero crossings; Four frequency domain characteristics: average frequency, frequency standard deviation, centroid frequency, and mean square frequency.

[0037] After supervised training, the SVM model can automatically output the defect type identification result based on the input feature vector.

[0038] In practical use, after each elastic wave excitation-reception and feature extraction, the defect type identification module 8 feeds the 8-dimensional feature vector into the trained SVM model. The model outputs the corresponding defect type; if normal, it outputs "no defect." This module also has online learning and updating capabilities: when new defect samples are detected during subsequent vehicle operation and confirmed manually, the feature vectors and labels of the new samples can be added to the training set to incrementally train the SVM model, thereby continuously improving the identification accuracy.

[0039] The lifespan prediction module 9 also shares the same programmable controller as the health assessment module 7. It predicts the remaining lifespan (LRUL) of the luminaire based on an exponential decay model, calculated using the following formula: LRUL = t 0 - τ × ln(LSHI(t) / LSHI 0 ) In the formula, LSHI(t) The current state of the lamp after operating for time t. LSHI value; LSHI 0 In the initial healthy state LSHI value.

[0040] in, LRUL Units and t 0 and t Consistent (if) t 0 and t In terms of days, then LRUL (Measured in days).

[0041] Lighting life decay time constant t The rate of structural degradation is reflected and determined through calibration: accelerated aging tests are conducted on the same batch of lamps under laboratory conditions to obtain... LSHI The attenuation curve was obtained by fitting. t The average value is taken as the typical value; Preferred t Reference value: Under normal operating conditions, t The value is approximately t 0 Under harsh working conditions, t The value may be shortened to t 0 Half of it.

[0042] In this embodiment, the calibration value is 100; t 0 For the design lifespan of the luminaire, this embodiment uses 5 years (i.e., 1825 days); τ is the luminaire's lifespan decay time constant, reflecting the rate of structural degradation. This constant value was obtained through accelerated aging tests on the same batch of luminaires in the laboratory, specifically: testing the luminaires under combined conditions of ultraviolet irradiation, temperature cycling, and vibration, and conducting periodic inspections. LSHI The attenuation curve was plotted, and the τ value was obtained by fitting it using the least squares method, with the average value of the batches taken. Under normal operating conditions, the τ value is approximately equal to the design life. t 0 Under harsh operating conditions such as high temperature, high humidity, and strong vibration, the τ value may shorten to [value missing]. t 0 Half of it. For example, in this embodiment, under normal operating conditions, it is taken as... τ=5 If the remaining lifespan is [number] years, the formula for calculating the remaining lifespan simplifies to: LRUL = 5-5×ln(LSHI(t) / 100) (Unit: year) When calculated LRULWhen the value is less than a certain preset threshold, the system can trigger a replacement warning.

[0043] The power supply and communication interface module 10 connects to the vehicle wiring harness via a 4-pin connector, where PIN1 is the DC power supply positive terminal, PIN2 is ground, PIN3 is the CAN bus high-order signal line, and PIN4 is the CAN bus low-order signal line. The internal power management unit of this module converts the 12V or 24V DC power supplied by the vehicle into a stable voltage required by each processing module, and simultaneously converts the communication level to the CAN protocol level. The processing results of the signal processing module 6, health assessment module 7, defect type identification module 8, and life prediction module 9, including the current LSHI value, classification level, defect type, and LRUL predicted value, are encapsulated into messages and transmitted to the vehicle control unit via the CAN bus at a frequency of 1Hz. Upon receiving this information, the vehicle control unit can display it on the in-vehicle central control screen and send it to the owner's mobile app via the vehicle network, pushing lighting health status and warning prompts to achieve predictive maintenance.

[0044] This embodiment also provides another implementation scheme, Embodiment 2, whose basic structure is the same as that of Embodiment 1, the difference being that the external lens 1 is made of PC material and the housing 2 is made of PP material. Figure 3 As shown, the sealing assembly 4 is formed by applying structural adhesive to the mating surface between the outer light distribution lens 1 and the housing 2 and then curing it. The adhesive layer thickness is approximately 3 mm. The arrangement of the elastic wave sensor array 5 is slightly adjusted according to the shape of the housing, but the principle of uniform distribution with equal spacing is maintained. The number of sensor units remains 4, and the dimensions are the same. The working principles of the operating mode, excitation parameters, signal processing, health assessment, defect identification, and life prediction modules are completely consistent with those in Example 1. The communication interface still uses a 4-pin connector to transmit data to the vehicle control unit via the CAN bus.

[0045] Because the sealing method has been changed from welding to adhesive application, the characteristics of elastic wave attenuation and mode conversion after passing through the adhesive layer differ from those of the welded structure in Mode 3 (detecting the sealed connection area). Therefore, the reference characteristic value in this embodiment... f 10 ～ f 40 Calibration needs to be performed separately based on the sealant application status. In addition, the acoustic properties (such as sound velocity and attenuation coefficient) of PC and PP materials differ from those of PMMA and ABS, and the corresponding flight time and signal energy reference values ​​will also be different, but this does not change the logic and algorithm framework of the entire detection method.

[0046] In the description of this application, it should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. For ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0047] It should be noted that, in this application, 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. It should also be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with reference to certain examples may be combined in other examples.

[0048] Furthermore, it should be noted that, unless otherwise explicitly specified and limited, the terms "connection" and "driving" used in the description of this application should be interpreted broadly. They can refer to direct connections, connections through an intermediate medium, or relationships within two elements. Those skilled in the art can understand their specific meaning in this application based on the specific circumstances.

[0049] The embodiments described above are merely further illustrations of the present invention and are not intended to limit the present invention in any other way. The present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding modifications and changes based on the present invention, but all such modifications and changes should fall within the protection scope of the present invention.

Claims

1. An automotive lamp with structural health detection function, comprising a lamp, the lamp including an external light-emitting lens, a housing, an internal light-emitting component, and a sealing component connecting the external light-emitting lens and the housing to form a sealed cavity, characterized in that, Also includes: An elastic wave sensor array includes multiple elastic wave sensor units, each of which acts as an excitation source to emit elastic wave signals and as a receiver to receive elastic wave signals. The signal processing module has its input terminal electrically connected to the output terminal of the elastic wave sensor array, and the signal processing module extracts the time-domain and frequency-domain features of the received elastic wave signal. A health assessment module, the input of which is electrically connected to the signal processing module, calculates the structural health index of the luminaire based on the deviation between the time-domain characteristics and frequency-domain characteristics and the pre-calibrated reference characteristics; A defect type identification module is provided, the input of which is electrically connected to the signal processing module. The defect type identification module inputs the multi-dimensional feature vector composed of the time domain features and frequency domain features into a trained machine learning classification model and obtains the defect type of the lamp structure. The lifespan prediction module is electrically connected to the output of the health assessment module. The lifespan prediction module calculates the remaining lifespan of the lamp based on the lamp structure health index and a preset attenuation model.

2. The automotive lighting fixture with structural health detection function according to claim 1, characterized in that, The number of elastic wave sensor units is 4 to 8, and the multiple elastic wave sensor units are arranged in an equally spaced grid, with the spacing between adjacent elastic wave sensor units not exceeding 100mm.

3. The automotive lamp with structural health detection function according to claim 1, characterized in that, The elastic wave sensor array includes the following operating modes: In the first mode, the elastic wave sensor array is mounted on the external light distribution lens, and the sensor units excite and receive each other. In the second mode, the elastic wave sensor array is mounted on the housing, and the sensor units excite and receive each other. In the third mode, the elastic wave sensor array is mounted on the external light distribution lens and the housing. In this mode, the sensor unit on the external light distribution lens and the sensor unit on the housing excite and receive each other.

4. The automotive lamp with structural health detection function according to claim 1, characterized in that, When the elastic wave sensor unit is used as an excitation source, it emits a sinusoidal pulse signal containing 4 to 6 cycles, with an excitation frequency range of 30Hz to 70Hz and an excitation voltage amplitude of 10V to 16V.

5. The automotive lamp with structural health detection function according to claim 4, characterized in that, The signal processing module samples the received elastic wave signal at a sampling frequency at least twice the excitation frequency, and the sampling time covers the entire excitation pulse period.

6. The automotive lamp with structural health detection function according to claim 1, characterized in that, The time-domain features include flight time, signal energy, signal peak value, and number of zero crossings; the frequency-domain features include average frequency, frequency standard deviation, centroid frequency, and mean square frequency.

7. The automotive lamp with structural health detection function according to claim 6, characterized in that, The health assessment module calculates the structural health index of the lighting fixture according to the following formula: LSHI = 100-Σ(w i × |f i －f i0 |) in, f i This is the i-th feature value extracted at the current time. f i0 This is the baseline value for the i-th feature in a healthy state; w i The weight coefficients for the i-th eigenvalue satisfy... Σw i = 1 .

8. The automotive lamp with structural health detection function according to claim 7, characterized in that, The health assessment module also performs a grading determination based on the value of the lamp structure health index. The grading determination specifically includes: When LSHI ≥ 90, it is classified as Grade I, i.e., healthy; When 80 ≤ LSHI < 90, it is classified as Grade II, i.e., sub-health. When 60 ≤ LSHI < 80, it is Level III, i.e., a warning. When 40 ≤ LSHI < 60, it is classified as Level IV, i.e., dangerous; When LSHI < 40, it is classified as Grade V, which is severe.

9. The automotive lamp with structural health detection function according to claim 6, characterized in that, The machine learning classification model is a support vector machine model, and the kernel function used in the support vector machine model is a radial basis function; The multidimensional feature vector includes the flight time, signal energy, signal peak value, number of zero crossings, average frequency, frequency standard deviation, centroid frequency, and mean square frequency. The defect type identification module identifies defect types including fracture, crack, aging, and connection failure.

10. The automotive lamp with structural health detection function according to claim 7, characterized in that, The lifespan prediction module calculates the remaining lifespan of the lamps according to the following formula. LRUL : LRUL = t 0 - τ × ln(LSHI(t) / LSHI 0 ) in t 0 Design the lifespan of the lighting fixtures. τ This is the time constant for the decay of the lamp's lifespan. LSHI 0 The structural health index of the luminaire under initial healthy conditions. LSHI(t) For runtime t The health index of the lighting fixture structure.

Images