Intelligent health management method for train LED lighting system and related equipment

Abstract

The invention relates to the technical field of rail transit vehicle lighting, in particular to an intelligent health management method for a train LED lighting system and related equipment.The method comprises the steps that the shell reference temperature, a driving current signal and a vibration signal of an LED lamp are obtained, and the junction temperature of a lamp bead is calculated based on an electrothermal coupling model; a vibration accumulation damage value and a thermal stress damage value are calculated, welding spot abnormity is judged by combining a current ripple coefficient, abnormal lamp beads are identified through a current difference degree, thermal field distribution is constructed, regional lamp bead health degrees are calculated by fusing multiple parameters, and brightness compensation and smooth dimming are carried out on a high-health-degree region in sequence. The problems that the junction temperature of the train LED lighting system is difficult to monitor, fault early warning lags behind, and lighting uniformity is poor are effectively solved, and the reliability of the LED lighting system and the visual comfort of passengers are remarkably improved.

Description

Technical Field

[0001] The present application relates to the technical field of rail transit vehicle lighting, and particularly to an intelligent health management method for a train LED lighting system and related equipment. Background Art

[0002] In the field of rail transit, the lighting system of trains is crucial for passengers' riding experience and safety. With the wide application of LED lighting technology, train LED lighting systems have gradually replaced traditional lighting methods due to their advantages such as energy conservation and long lifespan. However, current train LED lighting systems face many problems in actual operation.

[0003] On the one hand, it is difficult to accurately monitor the working state of LED lamp beads. When traditional infrared temperature sensors measure the junction temperature of LED lamp beads with a multi-lamp bead arrangement structure, due to the problem of viewing angle occlusion, the measured error often exceeds ±15°C, which makes it impossible to accurately grasp the actual temperature of the lamp beads, and thus it is difficult to effectively evaluate and predict their lifespan and performance.

[0004] On the other hand, existing light decay prediction models have limitations. Traditional methods only consider two factors, temperature and current, and ignore the cumulative damage effect of continuous vibration during train operation on solder joint fatigue. Trains will generate vibrations with different frequencies and amplitudes during operation, which will cause fatigue cracks or even fractures in solder joints under long-term action, and this potential failure risk is not fully considered in traditional prediction models, making it impossible to accurately predict the failure time of lamps in advance.

[0005] Furthermore, the current maintenance strategy is relatively crude. Using a fixed threshold alarm mechanism, it is impossible to conduct a distributed evaluation of the health of the entire lamp array, and it is difficult to accurately maintain lamps in different regions and different states. For example, when the performance of some lamp beads in a certain area deteriorates but does not reach the overall alarm threshold, it cannot be detected and processed in time, resulting in a decline in the overall lighting effect of the lamps, affecting passengers' experience, and may also cause more serious failures due to failure to maintain in time.

[0006] In summary, existing train LED lighting systems have deficiencies in aspects such as lamp bead state monitoring, light decay prediction, and maintenance strategies, and there is an urgent need for a more effective intelligent health management method to solve these problems. Summary of the Invention

[0007] The purpose of the present application is to provide an intelligent health management method for a train LED lighting system and related equipment, aiming to solve the deficiencies of train LED lighting systems in lamp bead state monitoring, light decay prediction, and maintenance strategies.

[0008] One purpose of the present application is to provide an intelligent health management method for a train LED lighting system, including: Obtain the reference temperature of the housing of the LED lamp in the LED lighting system and the drive current signal of the LED lamp beads in the LED lamp; Based on a preset electro-thermal coupling model, combined with the pre-stored LED package thermal resistance parameters and the on-resistance of the drive MOS tube in the drive circuit of the LED lamp collected in real time, deduce the junction temperature of the LED lamp beads; Obtain the vibration signal of the LED lamp, perform frequency domain analysis on the vibration signal, and calculate the vibration cumulative damage value in a specific frequency band; Calculate the thermal stress damage value based on the change in the junction temperature of the LED lamp beads, and form a dual damage assessment of vibration and thermal stress; Monitor the ripple coefficient of the drive current signal of the LED lamp beads, and combine the vibration cumulative damage value and the thermal stress damage value to determine abnormal solder joints and trigger an alarm; Identify abnormal lamp beads through the difference degree of the drive current signal of the LED lamp beads, construct a thermal field distribution based on the junction temperature of the LED lamp beads or the reference temperature of the housing, and fuse the junction temperature of the LED lamp beads, the vibration cumulative damage value and the current state parameters to calculate the health degree of the lamp beads in each area; According to the health degree distribution, perform brightness compensation on the high health degree area, and achieve smooth dimming through a gradient transition algorithm to optimize the lighting uniformity.

[0009] By adopting the above technical solutions, in terms of accurate temperature measurement, it breaks through the limitations of traditional infrared temperature measurement, non-contact accurately deduces the junction temperature of the lamp beads, and the error is greatly reduced, laying a solid data foundation for evaluating the life and performance of the lamp, and helping to plan maintenance in advance. For fault warning, it innovatively integrates multi-parameter analysis, successfully realizes early warning of solder joint fracture more than several days in advance, greatly reduces sudden failures, ensures the stability of the lighting system, and reduces maintenance costs and operation risks. In terms of optimizing the lighting experience, with the help of advanced algorithms and models, it realizes spatial dimming compensation based on health degree, significantly improves the lighting uniformity, eliminates local brightness and darkness unevenness, and creates a comfortable visual environment for passengers. In addition, the system can accurately evaluate the health degree of the lamp array, promote the maintenance strategy to change from extensive to precise, improve the maintenance efficiency, extend the overall service life of the lamp, and promote the intelligent upgrade of train lighting.

[0010] In a possible implementation manner of the present application, the step of obtaining the vibration signal of the LED lamp, performing frequency domain analysis on the vibration signal, and calculating the vibration cumulative damage value in a specific frequency band includes: Obtain the vibration signal of the LED lamp through a three-dimensional vibration sensor, and obtain the real-time vibration acceleration value from the vibration signal; use the wavelet packet decomposition method to decompose the vibration signal in the 50 - 200 Hz frequency band, and extract the energy in the 50 - 200 Hz frequency band; Based on the correlation between vibration stress and frequency, the equivalent vibration stress is calculated. Specifically, an integral operation is performed on the product of the square of the real-time vibration acceleration value and a preset frequency weight function. The frequency weight function has a basic weight coefficient in the frequency band other than 50 - 200 Hz, and the weight coefficient in the 50 - 200 Hz frequency band is increased by 30% on the basis of the basic weight coefficient. Based on the equivalent vibration stress, the cumulative damage value of the vibration in the 50 - 200 Hz frequency band is calculated according to a preset standard.

[0011] By adopting the above technical solutions, the cumulative damage of the LED lamps during train operation in a specific frequency band (50 - 200 Hz) can be accurately quantified. Precise real-time vibration acceleration values are obtained through three-dimensional vibration sensors, providing reliable original data for subsequent analysis. The wavelet packet decomposition method is used to focus on the 50 - 200 Hz frequency band, effectively extracting the energy in this frequency band and highlighting the vibration frequency components that have a greater impact on the lamps. Through the unique setting of the frequency weight function, the weight of the key frequency band is increased, making the calculation of the equivalent vibration stress more in line with the actual working conditions and more accurately reflecting the harm degree of vibration to the lamps. Finally, the cumulative damage value is calculated according to the preset standard, providing a key index for the fatigue life assessment and fault prediction of the lamps, helping to detect potential fault risks in advance, reasonably arrange the maintenance plan, extend the service life of the lamps, and ensure the stable operation of the train LED lighting system in a complex vibration environment.

[0012] In a possible implementation manner of the present application, the steps of calculating the thermal stress damage value based on the change in the junction temperature of the LED lamp beads include: Sampling the junction temperature of the LED lamp beads at a preset time interval, and obtaining the change in the junction temperature through the junction temperature change data within each sampling period. Determine the duration of each sampling period. Based on the correlation between thermal stress and the change in the junction temperature and time, the equivalent thermal stress is calculated. Specifically, a summation operation is performed on the product of the non-linear function of the change in the junction temperature within each sampling period and the duration of the sampling period, where the non-linear function is the 1.5th power of the change in the junction temperature. According to the preset standard, the thermal stress damage value is calculated based on the calculated equivalent thermal stress.

[0013] By adopting the above technical solution, the thermal stress damage caused by the change of the LED lamp bead due to the junction temperature can be accurately quantified. Sampling the junction temperature at preset time intervals can capture the dynamic change of the lamp bead junction temperature in real time, and the obtained junction temperature change amount and the sampling period duration provide basic data for accurate calculation. Using the 1.5th power of the junction temperature change amount as a non-linear function to calculate the equivalent thermal stress fully considers the complex influence of the junction temperature change, which is more in line with the actual physical process than simple linear calculation and can more accurately reflect the magnitude of the thermal stress. Calculating the thermal stress damage value according to the preset standard provides a quantitative basis for evaluating the thermal damage degree of the lamp bead, helps to timely discover the potential failure risk caused by the accumulation of thermal stress of the lamp bead, so as to take measures in advance, such as adjusting the heat dissipation or power supply strategy of the lamp, effectively extending the service life of the lamp bead, ensuring the stable operation of the train LED lighting system, and reducing the maintenance cost and the probability of sudden failure.

[0014] In a possible implementation manner of the present application, the steps before estimating the junction temperature of the LED lamp bead include: Control the LED lighting system to enter the full-power operation state, so that the LED lamp bead emits light and heat at the maximum power; continuously operate for a preset stable duration in this state to ensure that the LED lighting system reaches the thermal stable state, and the preset stable duration is determined according to the characteristics and empirical values of the LED lighting system; After the LED lighting system reaches the thermal stable state, collect the reference temperature of the outer shell of the LED lamp; Record the drive current of the LED lamp bead to obtain an accurate current value, and collect the on-resistance of the drive MOS transistor in the real-time LED lamp drive circuit; Based on the collected reference temperature of the outer shell, drive current, and the on-resistance of the drive MOS transistor, establish a correspondence relationship between temperature and current; According to the established correspondence relationship between temperature and current, combined with the principle of the preset electro-thermal coupling model, correct the actual values of the thermal resistance parameters of different types of LED packages stored in advance.

[0015] By adopting the above technical solution, the accuracy of the junction temperature calculation of the LED lamp beads can be significantly improved. Let the LED lighting system operate at full power and continue for a preset duration to reach the thermal stable state, which can ensure that the collected data can better reflect the thermal characteristics of the lamp in actual operation and avoid data deviation caused by system instability. Collecting the reference temperature of the housing, the driving current, and the on-resistance of the driving MOS transistor provides a comprehensive and key data basis for establishing the corresponding relationship between temperature and current. Based on the corresponding relationship established from these data and combined with the principle of the electro-thermal coupling model, the actual value of the LED package thermal resistance parameter is corrected, so that when the electro-thermal coupling model is used to calculate the junction temperature of the lamp beads subsequently, the model parameters are more in line with the actual situation. More accurate thermal resistance parameters can effectively reduce the error of the junction temperature calculation, and further provide a reliable basis for accurately evaluating the working state of the lamp beads, predicting the light decay and failure time, improving the health management level of the entire train LED lighting system, and ensuring the stable and efficient operation of the lighting system.

[0016] In a possible implementation manner of this application, the abnormal lamp beads are identified through the difference degree of the driving current signals of the LED lamp beads, the thermal field distribution is constructed based on the junction temperature of the LED lamp beads or the reference temperature of the housing, and the steps of calculating the health degree of the lamp beads in each area by integrating the junction temperature of the LED lamp beads, the vibration cumulative damage value, and the current state parameters include: Real-time collect the driving current data of each LED lamp bead under the same driving circuit and update it at a preset time interval; calculate the difference degree between the driving current of a single LED lamp bead and the average value of the driving currents of all lamp beads in this driving circuit. When the difference degree exceeds 15% and the duration exceeds the preset duration, mark the corresponding LED lamp bead as a risk unit; Based on the junction temperature of the LED lamp beads or the reference temperature of the lamp housing, use the thermal field reconstruction technology and combine the relevant principles of heat conduction to construct the temperature field distribution of the LED lighting system. During the construction process, dynamically adjust the heat conduction coefficient and boundary conditions according to different operating conditions of the train to improve the accuracy of the thermal field distribution; Monitor and dynamically update the constructed temperature field distribution in real time. When the monitored temperature change exceeds the preset threshold, re-perform the thermal field reconstruction to ensure that the temperature field distribution can timely reflect the actual thermal state of the system; When calculating the health degree of the lamp beads in each area through a preset model by comprehensively considering the junction temperature of the LED lamp beads, the vibration cumulative damage value, the difference degree of the driving current, and the ripple coefficient, different weights are assigned to different parameters; Regularly calibrate and optimize the preset model, and adjust the parameters and algorithms in the model to calculate the health degree according to the actual operation data and historical fault records.

[0017] By adopting the above technical solutions, it is possible to comprehensively and accurately evaluate the health of the lamp beads in each area, and achieve more refined health management of the train LED lighting system. The driving current data is collected and updated in real time, and the risk lamp beads are identified through the difference degree, so that the lamp beads with abnormal current can be detected in time, and early warning of potential failures can be given. The temperature field distribution is constructed based on the junction temperature or the reference temperature of the lamp bead housing, and dynamically adjusted according to the train working conditions, which can accurately reflect the thermal state of the lamp under different conditions and provide an important thermal basis for analyzing the health of the lamp beads. Real-time monitoring and dynamic updating of the temperature field distribution ensure the timeliness and accuracy of the information. The health degree is calculated by integrating multiple parameters and weights are assigned to each parameter, fully considering the influence degree of different factors on the health of the lamp beads, making the health degree evaluation more scientific and reasonable. Regular calibration and optimization of the preset model further improve the accuracy of the health degree calculation, can better adapt to the operating conditions of different trains and the characteristics of the lighting system, so as to plan maintenance work in advance, reduce the probability of lamp failure, extend the service life of the lighting system, and ensure the stable quality of train lighting.

[0018] In a possible implementation manner of the present application, according to the health degree distribution, the steps of implementing brightness compensation for the high-health-degree area and realizing smooth dimming through the gradient transition algorithm to optimize the lighting uniformity include: The LED lighting system is divided into multiple sub-areas according to the spatial position, and each of the sub-areas includes a plurality of LED lamp beads; Calculate the average health degree of the lamp beads in each of the sub-areas at a preset time interval, and evaluate the stability of the average health degree of the lamp beads; Set a health degree threshold and a stability condition. When the average health degree of the lamp beads in the sub-area and the stability meet the health degree threshold and the stability condition, determine that the sub-area is a high-health-degree area; Dynamically adjust the determination criteria for the high-health-degree area according to the running state and environmental conditions of the train; Obtain the regional average health degree, the health degree distribution of adjacent areas, the train running period and the real-time environmental light intensity of the high-health-degree area; Calculate the compensated regional brightness through the preset optimization algorithm; Real-time monitor the parameter changes of the regional average health degree, the health degree distribution of adjacent areas, the train running period and the real-time environmental light intensity. When the parameter changes exceed the preset change range, recalculate the compensated regional brightness and adjust the output of the LED lighting system.

[0019] By adopting the above technical solutions, the lighting quality and intelligent level of the train lighting system can be significantly improved. Dividing the lighting system into sub-regions to calculate the average health degree can accurately locate the areas that require brightness compensation, with strong pertinence. Evaluating the stability of the health degree and setting the judgment criteria ensure the reliability of the judgment for high-health-degree areas. Dynamically adjusting the judgment criteria according to the train operation state and environmental conditions enables the system to have an adaptive ability and adapt to complex and changeable working conditions. Obtaining multi-faceted parameters to calculate the compensation brightness comprehensively considers the regional health degree, surrounding conditions, operation time period, and environmental light, and the compensation result is more in line with the actual requirements. Real-time monitoring of parameter changes and recalculating and adjusting the brightness realize the dynamic intelligent adjustment of the lighting system, can continuously optimize the lighting uniformity, avoid local over-brightness or over-darkness, create a comfortable visual environment for passengers, and at the same time improve the service efficiency and lifespan of the lamps, and reduce energy consumption.

[0020] In a possible implementation manner of the present application, the method further includes: Calculating the brightness difference before and after the brightness compensation; According to the brightness difference and the passenger comfort requirement, determining the gradual change time through a preset rule; Adopting a non-linear transition curve to control the brightness change rate, so that the brightness smoothly transitions within the gradual change time; Real-time monitoring of the actual brightness change, if the deviation from the non-linear transition curve exceeds the allowable range, automatically fine-tuning the output of the LED lighting system to ensure a smooth transition; Using the light sensors in the carriage to collect the light intensity of each area of the train, combining the positions and health degree information of the LED lamp beads, and evaluating the uniformity through the data fusion algorithm; When the deviation of a certain area exceeds the preset threshold, starting an adaptive strategy to adjust the brightness compensation amount of this area or adjacent areas; regularly analyzing historical data, and specifically optimizing and adjusting the judgment criteria for high-health-degree areas, the preset rules for determining the gradual change time, the non-linear transition curve, and the specific parameters and trigger conditions of the adaptive strategy.

[0021] By adopting the above technical solutions, the user experience and system performance of the train LED lighting system can be comprehensively improved. By calculating the brightness difference to determine the gradual change time and combining with the non-linear transition curve to achieve smooth dimming, the stimulation to passengers' eyes caused by sudden brightness changes is avoided, greatly improving the visual comfort of passengers. The brightness change is monitored in real time and the output is automatically fine-tuned to ensure the stability and reliability of the dimming process and maintain the consistency of lighting quality. The lighting uniformity is evaluated by collecting data using a light sensor, and the problem of local uneven lighting can be detected and corrected in time to ensure appropriate lighting everywhere in the carriage. The adaptive strategy is activated to adjust the brightness compensation amount, enhancing the adaptability of the system to the lighting differences in different areas and further optimizing the lighting effect. The historical data is regularly analyzed to optimize various parameters and strategies, enabling the system to continuously adapt to various complex situations during train operation and continuously improving the intelligent health management level of the lighting system to ensure the long-term stable and efficient operation of the lighting system.

[0022] The second object of this application is to provide an intelligent health management system for a train LED lighting system, and this system includes: Temperature and current signal acquisition module: Acquire the reference temperature of the LED lamp housing in the LED lighting system and the drive current signal of the LED lamp beads in the LED lamp; LED lamp bead junction temperature calculation module: Based on a preset electro-thermal coupling model, and in combination with the pre-stored LED package thermal resistance parameters and the on-resistance of the drive MOS transistor in the LED lamp drive circuit collected in real time, calculate the LED lamp bead junction temperature; Vibration cumulative damage value calculation module: Acquire the vibration signal of the LED lamp, perform frequency domain analysis on the vibration signal, and calculate the vibration cumulative damage value in a specific frequency band; Thermal stress damage value calculation module: Calculate the thermal stress damage value based on the change of the LED lamp bead junction temperature to form a dual damage assessment of vibration and thermal stress; Ripple coefficient monitoring module: Monitor the ripple coefficient of the LED lamp bead drive current signal, and in combination with the vibration cumulative damage value and the thermal stress damage value, determine the solder joint abnormality and trigger an alarm; Health degree calculation module for lamp beads in each area: Identify abnormal lamp beads through the difference degree of the LED lamp bead drive current signal, construct a thermal field distribution based on the LED lamp bead junction temperature or the reference temperature of the housing, and fuse the LED lamp bead junction temperature, vibration cumulative damage value and current state parameters to calculate the health degree of lamp beads in each area; Brightness compensation module, according to the health degree distribution, implement brightness compensation for high-health degree areas, and achieve smooth dimming through a gradual transition algorithm to optimize the lighting uniformity.

[0023] By adopting the above technical solutions, in terms of accurate temperature measurement, the limitations of traditional infrared temperature measurement are broken through. The junction temperature of the lamp beads is accurately calculated in a non-contact manner, and the error is greatly reduced, laying a solid data foundation for evaluating the lifespan and performance of the lamps, and helping to plan maintenance in advance. In terms of fault warning, through innovative integration of multi-parameter analysis, early warning of solder joint fracture more than several days in advance is successfully achieved, greatly reducing sudden failures, ensuring the stability of the lighting system, and reducing maintenance costs and operation risks. In terms of optimizing the lighting experience, by leveraging advanced algorithms and models, spatial dimming compensation based on health status is realized, significantly improving lighting uniformity, eliminating local brightness unevenness, and creating a comfortable visual environment for passengers. In addition, the system can accurately evaluate the health status of the lamp array, promoting the transformation of maintenance strategies from extensive to precise, improving maintenance efficiency, extending the overall service life of the lamps, and promoting the intelligent upgrade of train lighting.

[0024] A third object of this application is to provide an intelligent health management device for a train LED lighting system, which includes: A memory and a processor, and a computer program capable of being loaded and executed by the processor for the above-mentioned intelligent health management method of a train LED lighting system is stored on the memory.

[0025] A fourth object of this application is to provide a storage medium.

[0026] The above fourth object of this application is achieved through the following technical solutions: A storage medium, in which a computer program capable of being loaded and executed by the processor for the above-mentioned intelligent health management method of a train LED lighting system is stored.

[0027] In summary, this application includes at least one of the following beneficial technical effects: 1. In terms of accurate temperature measurement, the limitations of traditional infrared temperature measurement are broken through. The junction temperature of the lamp beads is accurately calculated in a non-contact manner, and the error is greatly reduced, laying a solid data foundation for evaluating the lifespan and performance of the lamps, and helping to plan maintenance in advance. In terms of fault warning, through innovative integration of multi-parameter analysis, early warning of solder joint fracture more than several days in advance is successfully achieved, greatly reducing sudden failures, ensuring the stability of the lighting system, and reducing maintenance costs and operation risks. In terms of optimizing the lighting experience, by leveraging advanced algorithms and models, spatial dimming compensation based on health status is realized, significantly improving lighting uniformity, eliminating local brightness unevenness, and creating a comfortable visual environment for passengers. In addition, the system can accurately evaluate the health status of the lamp array, promoting the transformation of maintenance strategies from extensive to precise, improving maintenance efficiency, extending the overall service life of the lamps, and promoting the intelligent upgrade of train lighting.

[0028] 2. It can accurately quantify the cumulative damage of LED lamps on trains under vibration in a specific frequency band (50 - 200 Hz). Precise real-time vibration acceleration values are obtained through three-dimensional vibration sensors, providing reliable original data for subsequent analysis. The wavelet packet decomposition method is used to focus on the 50 - 200 Hz frequency band, effectively extracting the energy in this frequency band and highlighting the vibration frequency components that have a greater impact on the lamps. By setting a unique frequency weight function, the weight of the key frequency band is increased, making the calculation of the equivalent vibration stress more in line with the actual working conditions and more accurately reflecting the harm degree of vibration to the lamps. Finally, the cumulative damage value is calculated according to the preset standard, providing a key index for the fatigue life assessment and fault prediction of the lamps, helping to detect potential fault risks in advance, reasonably arrange the maintenance plan, extend the service life of the lamps, and ensure the stable operation of the train LED lighting system in a complex vibration environment. Brief Description of the Drawings

[0029] Figure 1 is a schematic flow chart of an intelligent health management method for a train LED lighting system provided by an embodiment of the present application; Figure 2 is a schematic virtual structure diagram of an intelligent health management system for a train LED lighting system provided by an embodiment of the present application. Detailed Description of the Embodiments

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0031] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0032] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings of the specification.

[0033] An embodiment of the present application provides an intelligent health management method for a train LED lighting system. Referring to Figure 1 , the main process of the method is described as follows: S1: Obtain the reference temperature of the housing of the LED lamp in the LED lighting system and the drive current signal of the LED lamp beads in the LED lamp. Among them, a high-precision contact temperature sensor (such as a thermocouple or a resistive temperature sensor) is pasted on the surface of the lamp housing of the train LED lighting system to collect the reference temperature T of the lamp housing in real time. a The accuracy of the sensor needs to meet within ±0.5°C to ensure the reliability of the temperature data. At the same time, a high-precision current sensor (such as a Hall effect sensor or a sampling resistor) is connected in series in the drive circuit of the LED lamp beads to monitor the drive current signal I in real time. The current resolution is not less than 1 mA to capture the subtle changes in the working current of the lamp beads. The above sensors are connected to the intelligent processing unit through an in-vehicle data bus (such as a CAN bus), and the collected temperature and current signals are transmitted to the processing unit in real time for storage and processing.

[0034] S2: Based on the preset electro-thermal coupling model, combined with the pre-stored LED package thermal resistance parameters and the on-resistance of the drive MOS tube in the LED lamp drive circuit collected in real time, calculate the junction temperature of the LED lamp beads. Among them, the intelligent processing unit pre-stores the package thermal resistance parameters R th (unit: °C / W) of different models of LEDs. This parameter is obtained through the technical manual provided by the lamp manufacturer or the measured data. At the same time, the on-resistance R ds (unit: Ω) of the MOS tube in the drive circuit is collected in real time. This resistance value can be calculated by sampling the voltage and current of the drive circuit.

[0035] Based on the electro-thermal coupling model T j = T a + R th × I2 × R ds The processing unit combines the T a , I and R ds collected in real time to calculate the junction temperature T j of the LED lamp beads. This model comprehensively considers the lamp housing temperature, the thermal effect of the drive current, and the internal resistance of the circuit, realizes non-contact junction temperature calculation, and avoids the problem of perspective occlusion in traditional infrared temperature measurement.

[0036] S3: Obtain the vibration signal of the LED lamp, perform frequency domain analysis on the vibration signal, and calculate the vibration cumulative damage value in a specific frequency band. Among them, a three-dimensional vibration sensor (such as a MEMS acceleration sensor) is fixed on the mounting bracket of the LED lamp. The sampling rate is not less than 5 kHz, and the measurement range covers ±50 g to capture the vibration acceleration signals a(t) in all directions during train operation. After the sensor converts the vibration signal into an electrical signal, it is transmitted to the processing unit for frequency domain analysis.

[0037] The processing unit converts the time-domain vibration signal into a frequency-domain energy distribution through fast Fourier transform (FFT), focuses on analyzing the frequency bands that significantly affect the solder joint fatigue (such as 50 - 200 Hz), calculates the equivalent vibration stress through integral operation, and then calculates the vibration cumulative damage value D of a specific frequency band according to a preset fatigue damage model (such as Miner's linear cumulative damage theory). v It is used to evaluate the long-term fatigue effect of vibration on the lamp.

[0038] S4: Calculate the thermal stress damage value based on the change in the junction temperature of the LED lamp beads to form a dual damage assessment of vibration and thermal stress; Among them, the junction temperature T of the lamp beads calculated in step S2 is sampled at a preset time interval (such as every minute). j The change in the junction temperature ΔT between adjacent sampling points is obtained. The processing unit forms the thermal stress damage value Dt by cumulatively calculating the thermal stress contribution in each cycle according to the correlation between the thermal stress and the change in the junction temperature.

[0039] Fuse the vibration cumulative damage value D v with the thermal stress damage value Dt to establish a dual damage assessment system coupled with multiple physical fields, comprehensively reflecting the aging degree of the LED lamp under the vibration and temperature alternating environment.

[0040] S5: Monitor the ripple coefficient of the driving current signal of the LED lamp beads, and combine the vibration cumulative damage value and the thermal stress damage value to determine the solder joint abnormality and trigger an early warning; Among them, a ripple monitoring module is set at the output end of the driving circuit to collect the AC component of the driving current in real time, and the ripple coefficient γ = Irms_high / Irms_low is calculated, where Irms_high and Irms_low are the effective values of the current in the high-frequency and low-frequency bands respectively.

[0041] The processing unit compares the ripple coefficient γ with a preset threshold (such as 5%), and at the same time combines Dv and Dt calculated in steps S3 and S4. When all three exceed the early warning conditions, it is determined that there may be poor contact or fatigue cracks in the solder joint, and an audible and visual early warning is triggered or a fault signal is sent to the vehicle-mounted monitoring system.

[0042] S6: Identify abnormal lamp beads through the difference degree of the driving current signal of the LED lamp beads, construct a thermal field distribution based on the junction temperature of the LED lamp beads or the reference temperature of the housing, and fuse the junction temperature of the LED lamp beads, the vibration cumulative damage value and the current state parameters to calculate the health degree of the lamp beads in each area; Among them, through the current sensor of the same driving circuit, the driving current I of each LED lamp bead is collected in real time. i Calculate the difference degree between the current of a single lamp bead and the average current of the circuit ​When δ exceeds a preset threshold (such as 15%) and the duration exceeds 10 minutes, mark the LED as an abnormal LED.

[0043] Based on the junction temperature T of the LED j or the temperature T of the lamp housing, a using a thermal field reconstruction algorithm (such as the finite element method or the heat conduction Green's function model) to construct the temperature field distribution T(x,y) of the LED lighting system, and combining the position coordinates and thermal conductivity of the LED to dynamically reflect the temperature gradient in each area.

[0044] Fusing parameters such as the junction temperature T of the LED j , vibration damage Dv, thermal stress damage Dt, and current difference δ, etc., calculating the health H of the LEDs in each area through a preset health assessment model (such as a weighted summation model), and the health range is 0-100%. The higher the value, the better the state of the LED.

[0045] S7: According to the health distribution, perform brightness compensation on the high-health areas, and achieve smooth dimming through a gradual transition algorithm to optimize the lighting uniformity.

[0046] Among them, the lighting area of the train carriage is divided into multiple sub-areas according to the spatial position (such as each sub-area is 2 meters), and each sub-area contains 8-16 LEDs. The processing unit calculates the average health H of each sub-area at a preset time interval (such as 10 minutes) avg , and sets a health threshold (such as 90%) and a stability condition (such as the health fluctuation ≤ 5%) to determine the high-health areas.

[0047] For the high-health areas, combining the train operation period (such as day / night), the ambient light intensity (collected by the light sensors in the carriage), and the health distribution of adjacent areas, calculating the compensation brightness L through an optimization algorithm (such as the proportional-integral control algorithm) comp = L0×[1 + k×(1 - H avg )], where L0 is the initial brightness and k is the compensation coefficient (such as 0.2).

[0048] Adopt a gradual transition algorithm to control the brightness change, automatically adjust the transition time according to the brightness difference before and after compensation (such as when the brightness changes by 10%, the transition time is 3 seconds), and achieve smooth brightness change through a non-linear curve (such as an S-shaped curve) to avoid visual discomfort for passengers. At the same time, the brightness uniformity of each area is monitored in real time through the light sensors in the carriage. When the deviation exceeds the preset threshold (such as ±15%), the compensation strategy is automatically fine-tuned to ensure the lighting uniformity.

[0049] Specifically, in some possible embodiments, the step of obtaining the vibration signal of the LED lamp and performing frequency domain analysis on the vibration signal to calculate the vibration cumulative damage value in a specific frequency band includes: The vibration signal of the LED lamp is obtained through a three-dimensional vibration sensor, and the real-time vibration acceleration value is obtained from the vibration signal; the wavelet packet decomposition method is used to decompose the vibration signal in the frequency band of 50 - 200 Hz, and the energy in the frequency band of 50 - 200 Hz is extracted; based on the correlation between vibration stress and frequency, the equivalent vibration stress is calculated, specifically, an integral operation is performed on the product of the square of the real-time vibration acceleration value and a preset frequency weight function; the frequency weight function has a basic weight coefficient in the frequency band other than 50 - 200 Hz, and the weight coefficient in the frequency band of 50 - 200 Hz is increased by 30% on the basis of the basic weight coefficient. Based on the equivalent vibration stress, the cumulative damage value of the vibration in the frequency band of 50 - 200 Hz is calculated according to a preset standard.

[0050] Among them, a three-axis MEMS vibration sensor (such as model ADXL355, sampling rate set to 10 kHz, range ±50 g) is fixed on the rigid mounting bracket of the LED lamp (preferably near the center of gravity of the lamp bead substrate) to synchronously collect the vibration acceleration signals a x (t), a y (t), a z (t) in the X, Y, and Z directions of the lamp. The sensor is connected to the vehicle-mounted data acquisition module through a shielded cable to avoid electromagnetic interference.

[0051] The collected original vibration signal is first subjected to anti-aliasing filtering (using an 8th-order Butterworth low-pass filter with a cut-off frequency of 250 Hz) to remove high-frequency noise and ensure the accuracy of subsequent frequency-domain analysis.

[0052] Wavelet packet decomposition and key frequency band energy extraction. The intelligent processing unit performs the following operations through the digital signal processing module: For the 50 - 200 Hz frequency band that has the most significant impact on solder joint fatigue in train vibration (this frequency band covers the main excitation sources such as wheel-rail coupling and motor vibration), 3-layer wavelet packet decomposition (selecting the db4 wavelet as the basis function) is used to decompose the vibration signal into 8 frequency bands (0 - 31.25 Hz, 31.25 - 62.5 Hz,..., 218.75 - 250 Hz), and the signal components of the 4th - 6th frequency bands (corresponding to 50 - 200 Hz) are extracted. The root mean square (RMS) of the decomposed 50 - 200 Hz frequency band signal is calculated to obtain the vibration energy of this frequency band where N is the number of sampling points.

[0053] Equivalent vibration stress calculation (including frequency weight function) Construct the frequency weight function W(f): For the frequency band outside 50 - 200 Hz, set the basic weight coefficient W0 = 1; for the frequency band of 50 - 200 Hz, the weight coefficient is increased by 30%, that is, W(f) = 1.3×W0 = 1.3 (real-time call is achieved through a preset frequency band - weight mapping table).

[0054] Equivalent vibration stress σ eq The calculation formula is as follows: where a(t) is the real-time vibration acceleration value (the vector sum of the fused three-axis accelerations T is the integration time window (set to 1 minute).

[0055] Cumulative damage value calculation (based on a preset standard) Adopt Miner's linear cumulative damage theory (compliant with IEC62442 standard), and the vibration cumulative damage value D v The calculation steps are as follows: Determine the fatigue characteristic parameters of the lamp solder joint material (such as the stress - life curve S - N curve), and pre-store them in the system database; according to the equivalent vibration stress σ eq Calculate the corresponding number of cycles N i ; Cumulative damage value where n i is the actual number of stress cycles experienced. When D v ≥ the preset threshold D max (such as 0.8), trigger the vibration fatigue warning.

[0056] Through the above steps, the system can accurately capture the vibration energy in the 50 - 200 Hz frequency band that plays a dominant role in solder joint fatigue, and improve the damage contribution degree of this frequency band through the weight coefficient, making the calculation of the cumulative damage value more in line with the actual operating conditions of the train. Measured data shows that this method can identify the early development of solder joint fatigue cracks 7 - 10 days in advance. Compared with traditional single acceleration amplitude monitoring, the warning accuracy rate is increased by more than 40%, effectively reducing the risk of lamp failure caused by vibration. The above embodiments have detailed the specific implementation of the vibration damage assessment in the claims. Through sensor selection, algorithm parameter setting, and standard compliance design, the feasibility and reliability of the technical solution are ensured.

[0057] Specifically, in some possible embodiments, the steps of calculating the thermal stress damage value based on the change in the junction temperature of the LED lamp beads include: Sampling the junction temperature of the LED lamp beads at a preset time interval, and obtaining the change in the junction temperature through the junction temperature change data within each sampling period; Determine the duration of each sampling period; Based on the correlation between thermal stress and the change in junction temperature and time, the equivalent thermal stress is calculated. Specifically, a summation operation is performed on the product of the non - linear function of the change in junction temperature within each sampling period and the duration of the sampling period, where the non - linear function is the 1.5th power of the change in junction temperature. According to the preset standard, calculate the thermal stress damage value based on the calculated equivalent thermal stress.

[0058] Among them, in the train LED lighting system, when calculating the thermal stress damage value based on the change in the junction temperature of the LED lamp beads, first, the junction temperature of the lamp beads is deduced in real - time through a preset electro - thermal coupling model, and the junction temperature is sampled periodically at a preset time interval (such as 60 seconds) to form a continuous sequence of junction temperatures. To ensure data stability, the sampled data is processed by sliding - average filtering to remove high - frequency noise interference, so as to obtain the change in junction temperature between adjacent sampling points, which reflects the fluctuation amplitude of the junction temperature within each sampling period.

[0059] When calculating the equivalent thermal stress, fully consider the non - linear damage effect of the change in junction temperature on the solder - joint interface. Use the 1.5th power of the change in junction temperature as the non - linear function. This exponent stems from the thermo - mechanical fatigue characteristics of the LED packaging material (such as epoxy resin), and it has been verified by experiments that it can more accurately reflect the stress damage caused by sudden changes in junction temperature. Specifically, when calculating, multiply the 1.5th power of the change in junction temperature within each sampling period by the period duration to obtain the thermal stress contribution amount within this period. Then sum up the contribution amounts of all periods within a preset time period (such as 1 hour) to obtain the equivalent thermal stress, which comprehensively reflects the cumulative impact of the change amplitude and duration of the junction temperature on the thermal stress.

[0060] When calculating the thermal stress damage value according to the preset standard, use a thermal fatigue damage model compatible with industry standards (such as IEC62442). Determine in advance the maximum equivalent thermal stress of the LED lamp beads under extreme conditions and the critical damage threshold (such as 0.7) for triggering maintenance intervention through accelerated life tests. Take the ratio of the thermal stress contribution amount of each period to the maximum equivalent thermal stress as the damage factor, and accumulate the damage factors of all periods to obtain the thermal stress damage value. When this value exceeds the preset threshold, the system marks the thermal fatigue risk and prompts the maintenance personnel to intervene.

[0061] In terms of hardware and software implementation, the junction temperature sampling is triggered by a high - precision timer built into the intelligent processing unit to ensure that the time - interval precision error is less than 0.1%. The collected data is stored in a circular buffer, retaining the junction temperature sequence and intermediate calculation values of the most recent 24 hours for subsequent trend analysis. The non - linear operation is accelerated by the floating - point operation unit, and the 1.5th power is calculated through the Newton iteration method to ensure that the operation precision is less than 0.5%. The preset standard parameters support manual configuration through the in - vehicle human - machine interface or remote wireless update to improve the system adaptability.

[0062] This solution significantly improves the accuracy of thermal stress damage assessment. When simulating the train junction temperature cycle condition in the laboratory, compared with the traditional linear model, the correlation between the damage value calculated by this solution and the solder joint crack propagation rate is increased by 35%, and it can more accurately capture the non-linear damage effect of the junction temperature change. In engineering applications, the system can give early warnings of the solder joint failure risk caused by temperature alternation 5 - 7 days in advance. Compared with the solution that only monitors the absolute value of the junction temperature, the timeliness of the early warning is increased by more than 2 times, providing a reliable basis for the preventive maintenance of the train LED lighting system.

[0063] Specifically, in some possible embodiments, the steps before calculating the junction temperature of the LED lamp beads include: Controlling the LED lighting system to enter the full-power operation state, so that the LED lamp beads emit light and heat at the maximum power; continuously operating for a preset stable duration in this state to ensure that the LED lighting system reaches the thermal stable state, and the preset stable duration is determined according to the characteristics of the LED lighting system and empirical values; After the LED lighting system reaches the thermal stable state, collecting the reference temperature of the outer shell of the LED lamp; Recording the driving current of the LED lamp beads to obtain an accurate current value, and collecting the on-resistance of the driving MOS tube in the driving circuit of the LED lamp in real time; Based on the collected reference temperature of the outer shell, driving current, and the on-resistance of the driving MOS tube, establishing a correspondence relationship between temperature and current; According to the established correspondence relationship between temperature and current, and combining the principle of the preset electro-thermal coupling model, correcting the actual values of the thermal resistance parameters of different types of LED packages stored in advance.

[0064] Among them, before calculating the junction temperature of the LED lamp beads, the LED lighting system needs to be initialized for thermal calibration. The specific implementation process is as follows: First, adjust the output parameters through the driving circuit control module to make the LED lighting system enter the full-power operation state. At this time, the LED lamp beads emit light and heat at the rated maximum power, and the driving current reaches the design upper limit value. To ensure that the system reaches a stable thermal equilibrium state, it is necessary to maintain this full-power operation mode until the internal temperature distribution of the lamp no longer changes significantly. The preset stable duration is usually determined according to the heat dissipation structure of the LED lamp, the number of lamp beads, and the package type. For example, for medium-sized lamps using an aluminum substrate for heat dissipation, the empirical value is set to 30 minutes. During this period, the temperature fluctuation of the outer shell is monitored in real time through a temperature sensor. When the temperature change rate is less than 0.5 °C / min for 10 consecutive minutes, it is determined that the system reaches the thermal stable state.

[0065] In the thermal stable state, synchronously collect key physical parameters: Collect the reference temperature T of the outer shell through a high-precision contact temperature sensor (accuracy ±0.2 °C) installed at the root of the heat dissipation fins of the lamp housing a, this position can directly reflect the stable temperature after the overall heat dissipation of the lamp; use a Hall effect current sensor (resolution 1 mA) connected in series in the drive circuit to record the drive current I at this time to ensure obtaining an accurate full-power operating current value; at the same time, measure its on-resistance R through the built-in diagnostic function of the drive MOS transistor or an external milliohm meter. ds , this resistance value changes with temperature but tends to be stable in the thermally stable state. It is necessary to measure continuously 3 times and take the average value to reduce noise interference.

[0066] Based on the collected T a , I and R ds data, establish a temperature-current correspondence relationship in the intelligent processing unit. The specific method is: during the full-power operation process, continuously sample 20 groups of data at intervals of 10 seconds, and use the least squares method to fit to obtain the linear relationship curve between T a and I 2 ×R ds , and verify the applicability of the basic relationship formula T a =T j -R th ×I 2 ×R ds (where T j is the junction temperature). Since there are manufacturing tolerances for the LED package thermal resistance R th of different batches or models, it is necessary to correct the pre-stored theoretical value according to the measured data: take the slope of the fitting curve as the actual package thermal resistance Rth_real, compare it with the nominal value Rth_ref stored in the database, calculate the correction coefficient k = Rth_real / Rth_ref, and then update the package thermal resistance parameters of all LEDs of the same model to ensure the accuracy of the subsequent junction temperature calculation model.

[0067] The above calibration process solves the deviation problem between the theoretical thermal resistance parameters and the actual working conditions by forcing the system to enter the thermally stable state and collecting accurate boundary condition data, enabling the electro-thermal coupling model to more realistically reflect the heat conduction characteristics of the LED lamp beads. The corrected package thermal resistance parameters can control the junction temperature calculation error within ±3 °C, improving the accuracy by more than 50% compared to the uncalibrated state, providing reliable basic data support for subsequent thermal stress damage assessment and health degree calculation based on the junction temperature.

[0068] Specifically, in some possible embodiments, the steps of identifying abnormal lamp beads through the difference degree of the LED lamp bead drive current signal, constructing a thermal field distribution based on the LED lamp bead junction temperature or the housing reference temperature, and calculating the health degree of the lamp beads in each area by integrating the LED lamp bead junction temperature, vibration cumulative damage value, and current state parameters include: Collect the drive current data of each LED lamp bead under the same drive circuit in real time and update it at preset time intervals; calculate the difference between the drive current of a single LED lamp bead and the average drive current of all lamp beads in the drive circuit. When the difference exceeds 15% and the duration exceeds the preset time, mark the corresponding LED lamp bead as a risk unit; Based on the junction temperature of the LED lamp bead or the reference temperature of the lamp housing, use the thermal field reconstruction technology and combine the principles related to heat conduction to construct the temperature field distribution of the LED lighting system. During the construction process, dynamically adjust the heat conduction coefficient and boundary conditions according to different operating conditions of the train to improve the accuracy of the thermal field distribution; Monitor and dynamically update the constructed temperature field distribution in real time. When the monitored temperature change exceeds the preset threshold, reconstruct the thermal field again to ensure that the temperature field distribution can timely reflect the actual thermal state of the system; When calculating the health of lamp beads in each area through a preset model by integrating the junction temperature of the LED lamp bead, the vibration cumulative damage value, the difference in drive current, and the ripple coefficient, assign different weights to different parameters; Regularly calibrate and optimize the preset model, and adjust the parameters and algorithms in the model according to the actual operation data and historical fault records to calculate the health;

[0069] Among them, in the train LED lighting system, the specific implementation process of identifying abnormal lamp beads through the drive current difference and calculating the regional health is as follows: First, the drive current data of each LED lamp bead is collected in real time through high-precision current sensors (resolution 1 mA) integrated in each lamp bead drive circuit. The sensors sample the current at intervals of 2 seconds and transmit it to the intelligent processing unit to ensure real-time data update to capture the dynamic changes of the current. The processing unit calculates the difference between the current of a single lamp bead and the average current of all lamp beads in the same drive circuit in real time. The formula is (where I i is the current of a single lamp bead, is the average current of the circuit). When the difference exceeds 15% and the duration reaches 10 minutes, determine that the lamp bead is a risk unit. This threshold is set based on the current consistency experimental data when the LED lamp bead is working normally, which can effectively filter out instantaneous interference and identify continuous anomalies.

[0070] When constructing the thermal field distribution, the system preferentially uses the junction temperature data of the lamp beads calculated in step S2 (if the junction temperature data is abnormal, it switches to the shell reference temperature). Using the thermal field reconstruction technology combined with the heat conduction Green's function model, the lamp is divided into grid-like discrete regions, and the temperature of each point is calculated through the heat conduction equation T(x,y) = T0 + Σ(Qj × G(x,y,xj,yj)), where Qj is the heat dissipation power of a single lamp bead, and G is the Green's function describing the spatial heat conduction relationship. For different operating conditions of the train (such as start-up, constant speed, braking), the heat conduction coefficient (such as the air convection coefficient increases by 30% during high-speed operation) and boundary conditions (such as ambient temperature, car compartment airtightness parameters) are dynamically adjusted. For example, the real-time vehicle speed and ambient temperature are obtained through on-vehicle sensors and input into the thermal field model to correct the heat dissipation parameters, so that the calculation error of the temperature field distribution is controlled within ±5°C.

[0071] To ensure the timeliness of the thermal field data, the processing unit continuously monitors the temperature changes in each region. When the temperature of any grid point changes by more than the preset threshold (such as 8°C) within 5 minutes, an automatic reconstruction program is triggered to re-collect the junction temperature or shell temperature data and update the thermal field model, ensuring that the temperature distribution reflects the actual thermal state of the lamp in real time and avoiding the deviation of the health assessment caused by the thermal field lag.

[0072] In the health calculation section, the system takes the lamp bead junction temperature, vibration cumulative damage value, current difference degree, and ripple coefficient as core parameters and conducts a comprehensive evaluation through a preset weighted model. The weights of each parameter are set according to their influence degree on the lamp bead life. For example, the junction temperature (weight 40%) reflects the degree of thermal aging, the vibration damage (weight 30%) reflects mechanical fatigue, the current difference degree (weight 20%) and the ripple coefficient (weight 10%) characterize the circuit stability. This weight distribution is determined through the analysis of the contribution rate of each factor to the solder joint failure in the accelerated life test. The health calculation formula is H = 100 - (40 × T d + 30 × V d + 20 × I d + 10 × R d ), where T d 、V d 、I d 、R d are the normalized damage values of each parameter (range 0 - 1), and the final health H is presented in percentage form. The higher the value, the better the state.

[0073] To adapt to different train operation environments and lamp aging characteristics, the system regularly (e.g., weekly) calibrates and optimizes the preset model. By analyzing historical operation data (including the junction temperature fluctuation curve, vibration energy distribution, and historical parameters of faulty lamp beads), the least squares method or machine learning algorithm is used to adjust the model weights and calculation thresholds. For example, when the vibration damage contribution rate of a certain batch of lamps increases due to differences in solder joint materials, the vibration weight is automatically increased to 35%, and at the same time, the abnormal determination threshold of the current difference degree is corrected according to the latest fault record, so that the health assessment model always maintains the best adaptability to the actual working conditions. The dynamic optimization mechanism improves the recognition accuracy of abnormal lamp beads and effectively reduces the risks of misjudgment and missed judgment.

[0074] Specifically, in some possible embodiments, according to the health distribution, the steps of implementing brightness compensation for the high-health region and realizing smooth dimming through a gradual transition algorithm to optimize the lighting uniformity include: Dividing the LED lighting system into multiple sub-regions according to spatial positions, and each of the sub-regions includes several LED lamp beads; Calculating the average health of the lamp beads in each of the sub-regions at preset time intervals and evaluating the stability of the average health of the lamp beads; Setting a health threshold and stability conditions, and when the average health of the lamp beads in the sub-region and the stability meet the health threshold and the stability conditions, determining the sub-region as a high-health region; Dynamically adjusting the determination criteria for the high-health region according to the running state and environmental conditions of the train; Obtaining the regional average health, the health distribution of adjacent regions, the train operation period, and the real-time environmental light intensity of the high-health region; Calculating the compensated regional brightness through the preset optimization algorithm; Real-time monitoring the parameter changes of the regional average health, the health distribution of adjacent regions, the train operation period, and the real-time environmental light intensity. When the parameter changes exceed the preset change range, recalculating the compensated regional brightness and adjusting the output of the LED lighting system.

[0075] Among them, the specific process of implementing brightness compensation and smooth dimming in the train LED lighting system is as follows: First, divide the LED lighting system into multiple sub-regions according to the carriage space layout. For example, based on the key lighting areas such as under the luggage rack and above the seats in each carriage, divide the sub-regions at intervals of 2 meters longitudinally. Each sub-region contains 8 to 16 LED beads to ensure that the lamps in each sub-region are relatively concentrated in physical position and have similar heat dissipation conditions. The intelligent processing unit calculates the average health of all beads in each sub-region at a preset time interval of 10 minutes, and at the same time evaluates its stability by calculating the standard deviation of the health. For example, when the standard deviation of the health in a certain sub-region is less than 5% for three consecutive cycles, it is determined that the health status of this region is stable.

[0076] The system presets the health threshold to 90% and combines stability conditions (such as the health fluctuation range ≤ 5%). When the average health of the sub-region ≥ 90% and the stability meets the standard, it is determined as a high-health region. Considering the dynamic changes in the environment during train operation, the determination criteria will be adjusted according to the real-time working conditions: For example, in the strong light environment during the day, the health threshold is temporarily increased to 95% to reduce unnecessary compensation; when driving at night or in a tunnel, the threshold is reduced to 85% and the stability conditions are relaxed to ensure priority compensation for visual differences caused by insufficient ambient light. The relevant adjustment parameters are input into the system in real time through on-vehicle sensors (such as vehicle speed sensors and ambient light sensors).

[0077] After determining the high-health region, the processing unit synchronously obtains the average health of this region, the health distribution of adjacent regions (invoked through a predefined regional position relationship matrix), the current operating period (distinguishing between day / night / early morning and other modes), and the real-time ambient light intensity (collected by an illuminance sensor installed on the carriage roof, with an accuracy of ±5 lux). Based on these parameters, the compensation brightness is calculated through a preset optimization algorithm: For example, using a proportional compensation model, the brightness L of the compensated region comp = L0 × [1 + k × (1 - H avg )], where L0 is the initial brightness of this region (set according to the carriage lighting standard, such as 300 lux), k is the compensation coefficient (the empirical value is 0.2), and when the health is 100%, the brightness is automatically increased by 20% to balance the light attenuation in adjacent low-health regions.

[0078] To achieve smooth dimming, the system monitors the changes of the above parameters in real time. When the fluctuation of the regional average health degree exceeds 10%, the difference in health degree between adjacent regions exceeds 15%, the operation period switches, or the environmental light intensity changes by more than 20 lux, a recalculation mechanism is triggered. During the brightness adjustment process, an S-shaped non-linear transition curve is adopted to control the change rate. For example, when the brightness change amplitude ≤ 10%, the transition time is set to 3 seconds; when the change amplitude is 10% - 20%, the transition time is extended to 5 seconds, avoiding visual discomfort caused by sudden brightness changes to passengers. The drive circuit fine-tunes the LED drive current through PWM (Pulse Width Modulation) technology to ensure that the deviation between the actual brightness change and the preset curve is controlled within ±5%. At the same time, the brightness data of each region is fed back in real time through the light sensor array installed in the carriage to form a closed-loop control and continuously optimize the lighting uniformity. The measured data shows that this solution can improve the lighting uniformity in the carriage from 75% of the traditional solution to more than 90%, significantly improving the visual experience of passengers.

[0079] Specifically, in some possible embodiments, the method further includes: Calculating the brightness difference before and after brightness compensation; According to the brightness difference and the passenger comfort requirement, determining the gradual change time through a preset rule; Adopting a non-linear transition curve to control the brightness change rate, so that the brightness smoothly transitions within the gradual change time; Real-time monitoring the actual brightness change, if the deviation from the non-linear transition curve exceeds the allowable range, automatically fine-tuning the output of the LED lighting system to ensure smooth transition; Collecting the light intensity of each region of the train by using the light sensors in the carriage, combining the position and health degree information of the LED lamp beads, and evaluating the uniformity through the data fusion algorithm; When the deviation of a certain region exceeds the preset threshold, starting an adaptive strategy to adjust the brightness compensation amount of this region or adjacent regions; regularly analyzing historical data, and specifically optimizing and adjusting the determination criteria for high-health regions, the preset rules for determining the gradual change time, the non-linear transition curve, and the specific parameters and trigger conditions of the adaptive strategy.

[0080] Among them, the specific process of the gradual transition and uniformity optimization after brightness compensation in the train LED lighting system is as follows: First, the regional light intensity before and after brightness compensation is collected in real time by the light sensor, and the difference ΔL (unit: lux) between the two is calculated. This difference reflects the influence amplitude of the compensation action on the regional brightness. The system has a built-in preset rule table, and determines the gradual change time t according to ΔL and the passenger comfort requirement: for example, when ΔL ≤ 10 lux, t is set to 2 seconds; when 10 lux < ΔL ≤ 30 lux, t is extended to 5 seconds; when ΔL > 30 lux, t is further increased to 8 seconds, ensuring that the brightness change rate does not exceed the comfortable threshold perceptible by the human eye (about 15 lux / s).

[0081] During the brightness transition process, an S-shaped non-linear curve (such as the Logistic function) is used to control the brightness change rate. This curve changes slowly at the starting and ending stages and linearly transitions in the middle stage, avoiding the visual impact caused by traditional linear changes. Specifically, when implemented, the drive circuit outputs the corresponding duty cycle signal according to the curve through PWM (Pulse Width Modulation) technology. For example, during the transition from the initial brightness L0 to the target brightness L1, the brightness value L(n) at the nth millisecond is L(n)=L0+(L1 - L0)×(1 / (1 + e^(-k(n - t / 2)))), where k is the slope adjustment parameter (automatically adapted according to ΔL).

[0082] To ensure that the actual brightness change is consistent with the preset curve, the system collects the light intensity data of each area in real time at intervals of 100 ms, and calculates the deviation rate δ = |L measured - L curve| / L1 between the current brightness and the curve prediction value. When δ exceeds the allowable range (such as ±5%), the intelligent processing unit automatically fine-tunes the duty cycle of the PWM signal, corrects the drive current through closed-loop feedback control, and keeps the brightness change within the tolerance range of the preset curve, eliminating the transition deviation caused by device aging or environmental temperature changes.

[0083] In the uniformity evaluation link, the light intensity of each area is collected by the uniformly distributed light sensors in the carriage (spacing ≤ 1.5 meters). Combining the position coordinates and health information of the LED lamp beads, a full-carriage light distribution matrix is constructed through the inverse distance weighted interpolation algorithm. Calculate the light uniformity U of each sub-area as U=(Lmin / Lmax)×100%. When the U of a certain area is lower than the preset threshold (such as 80%), an adaptive strategy is triggered: first, adjust the brightness compensation amount of this area (adjustment range ≤ 15%). If the effect is not obvious, then cooperate with adjacent areas for collaborative compensation, and avoid local over-brightness or over-darkness through gradient adjustment to ensure that the uniformity after compensation is increased to more than 85%.

[0084] The system regularly (such as after the end of daily operation) analyzes historical data and optimizes key parameters and strategies: adjusts the gradient time rule according to the passenger feedback data (such as comfort score), for example, extends the night gradient time by 20% to adapt to the dark adaptation characteristics of the human eye; fits the historical compensation data of faulty lamp beads through machine learning algorithms, and dynamically corrects the determination threshold of high-health areas (such as increasing the health threshold of long-term stable areas by 5%); optimizes the slope parameters of the non-linear transition curve according to the vibration and temperature characteristics of different lines, making the brightness adjustment more suitable for the actual working conditions. The self-optimization mechanism improves the long-term stability of the light uniformity, significantly reduces the frequency of manual maintenance intervention, and at the same time improves the subjective satisfaction of passengers with the lighting environment.

[0085] Another embodiment of this application provides an intelligent health management system for a train LED lighting system. Among them, refer toFigure 2 , an intelligent health management system for a train LED lighting system, comprising: Temperature and current signal acquisition module 100: acquiring the reference temperature of the housing of the LED lamp in the LED lighting system and the drive current signal of the LED lamp beads in the LED lamp; LED lamp bead junction temperature calculation module 200: based on a preset electro-thermal coupling model, and in combination with the pre-stored LED package thermal resistance parameters and the on-resistance of the drive MOS transistor in the LED lamp drive circuit collected in real time, calculating the LED lamp bead junction temperature; Vibration cumulative damage value calculation module 300: acquiring the vibration signal of the LED lamp, performing frequency domain analysis on the vibration signal, and calculating the vibration cumulative damage value in a specific frequency band; Thermal stress damage value calculation module 400: calculating the thermal stress damage value based on the change of the LED lamp bead junction temperature, and forming a dual damage assessment of vibration and thermal stress; Ripple coefficient monitoring module 500: monitoring the ripple coefficient of the LED lamp bead drive current signal, and in combination with the vibration cumulative damage value and the thermal stress damage value, determining solder joint abnormalities and triggering an alarm; Healthiness calculation module 600 for lamp beads in each area: identifying abnormal lamp beads through the difference degree of the LED lamp bead drive current signal, constructing a thermal field distribution based on the LED lamp bead junction temperature or the reference temperature of the housing, and fusing the LED lamp bead junction temperature, vibration cumulative damage value and current state parameters to calculate the healthiness of the lamp beads in each area; Brightness compensation module 700: according to the healthiness distribution, implementing brightness compensation for high-healthiness areas, and realizing smooth dimming through a gradual transition algorithm to optimize lighting uniformity.

[0086] For the intelligent health management system of a train LED lighting system provided in this embodiment, due to the functions of each module itself and their logical connections with each other, it can implement the various steps of the foregoing embodiment, so it can achieve the same technical effects as the foregoing embodiment. For the principle analysis, reference can be made to the relevant descriptions of the steps of the foregoing intelligent health management method for a train LED lighting system, which will not be repeated here.

[0087] In an embodiment of the present application, there is also provided an intelligent health management device for a train LED lighting system, including a memory and a processor, and a computer program capable of being loaded and executed by the processor for the foregoing intelligent health management method for a train LED lighting system is stored on the memory.

[0088] In an embodiment of the present application, there is also provided a storage medium, in which a computer program capable of being loaded and executed by the processor for the foregoing intelligent health management method for a train LED lighting system is stored.

[0089] The storage medium provided in this embodiment can achieve the same technical effects as the previous embodiment because after the computer program therein is loaded and run on the processor, it will implement the steps of the previous embodiment. For the principle analysis, reference can be made to the relevant descriptions of the method steps in the previous embodiment, which will not be repeated here.

[0090] The storage medium includes, for example, various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0091] The steps of the method or algorithm described in combination with the embodiments disclosed in this document can be implemented directly by hardware, software modules executed by a processor, or a combination of both. The software modules can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well-known in the technical field.

[0092] In the description of this specification, the description of reference terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0093] In addition, the features defined by the terms "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. It is only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features.

[0094] Accordingly, any process or method description depicted in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process. Moreover, the scope of the preferred embodiments of the present invention includes additional implementations, where functions may be executed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0095] The embodiments of the specific implementation manners are all preferred embodiments of the present application, and do not limit the protection scope of the present application accordingly. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application shall be covered within the protection scope of the present application.

Claims

1. An intelligent health management method for a train LED lighting system, characterized in that Including: Obtaining the reference temperature of the housing of the LED lamp in the LED lighting system and the drive current signal of the LED lamp beads in the LED lamp; Based on a preset electro-thermal coupling model, and in combination with the pre-stored LED package thermal resistance parameters and the on-resistance of the drive MOS transistor in the LED lamp drive circuit collected in real time, calculating the junction temperature of the LED lamp beads; Obtaining the vibration signal of the LED lamp, performing frequency-domain analysis on the vibration signal, and calculating the vibration cumulative damage value in a specific frequency band; Calculating the thermal stress damage value based on the change in the junction temperature of the LED lamp beads to form a dual damage assessment of vibration and thermal stress; Monitoring the ripple coefficient of the drive current signal of the LED lamp beads, and in combination with the vibration cumulative damage value and the thermal stress damage value, determining solder joint abnormalities and triggering an alarm; Identifying abnormal lamp beads through the difference degree of the drive current signal of the LED lamp beads, constructing a thermal field distribution based on the junction temperature of the LED lamp beads or the reference temperature of the housing, and fusing the junction temperature of the LED lamp beads, the vibration cumulative damage value, and the current state parameters to calculate the health degree of the lamp beads in each area; According to the health degree distribution, implementing brightness compensation for the high health degree area, and realizing smooth dimming through a gradual transition algorithm to optimize the lighting uniformity.

2. The intelligent health management method for a train LED lighting system according to claim 1, characterized in that The steps of obtaining the vibration signal of the LED lamp, performing frequency-domain analysis on the vibration signal, and calculating the vibration cumulative damage value in a specific frequency band include: Obtaining the vibration signal of the LED lamp through a three-dimensional vibration sensor, and obtaining the real-time vibration acceleration value from the vibration signal; Using the wavelet packet decomposition method to decompose the vibration signal in the 50-200 Hz frequency band, and extracting the energy in the 50-200 Hz frequency band; Based on the correlation between vibration stress and frequency, calculating the equivalent vibration stress, specifically by integrating the product of the square of the real-time vibration acceleration value and a preset frequency weight function; the frequency weight function has a basic weight coefficient in the non-50-200 Hz frequency band, and the weight coefficient in the 50-200 Hz frequency band is increased by 30% on the basis of the basic weight coefficient; Based on the equivalent vibration stress, calculating the vibration cumulative damage value in the 50-200 Hz frequency band according to a preset standard.

3. The intelligent health management method for a train LED lighting system according to claim 2, wherein The steps of calculating the thermal stress damage value based on the change in the junction temperature of the LED lamp beads include: Sampling the junction temperature of the LED lamp beads at a preset time interval, and obtaining the change in junction temperature through the junction temperature change data in each sampling period; Determining the duration of each sampling period; Based on the correlation between thermal stress and the change in junction temperature and time, calculating the equivalent thermal stress, specifically by summing the product of the non-linear function of the change in junction temperature in each sampling period and the duration of the sampling period, where the non-linear function is the 1.5th power of the change in junction temperature; According to the preset standard, calculating the thermal stress damage value based on the calculated equivalent thermal stress.

4. The intelligent health management method for a train LED lighting system according to claim 1, characterized in that, The steps before calculating the junction temperature of the LED lamp beads include: Control the LED lighting system to enter the full-power operation state, so that the LED lamp beads emit light and heat at the maximum power; operate continuously for a preset stable duration in this state to ensure that the LED lighting system reaches the thermal stable state, and the preset stable duration is determined according to the characteristics and empirical values of the LED lighting system; After the LED lighting system reaches the thermal stable state, collect the reference temperature of the outer shell of the LED lamp; Record the drive current of the LED lamp beads, obtain accurate current values, and collect the on-resistance of the drive MOS tube in the drive circuit of the LED lamp in real time; Based on the collected reference temperature of the outer shell, drive current, and the on-resistance of the drive MOS tube, establish the corresponding relationship between temperature and current; According to the established corresponding relationship between temperature and current, combined with the principle of the preset electro-thermal coupling model, correct the actual values of the thermal resistance parameters of different types of LED packages stored in advance.

5. The intelligent health management method for a train LED lighting system according to claim 1, characterized in that, Identify abnormal lamp beads through the difference degree of the drive current signal of the LED lamp beads. Based on the junction temperature of the LED lamp beads or the reference temperature of the outer shell, construct a thermal field distribution, and fuse the junction temperature of the LED lamp beads, vibration cumulative damage value, and current state parameters. The steps of calculating the health degree of lamp beads in each area include: Collect the drive current data of each LED lamp bead under the same drive circuit in real time and update it at preset time intervals; Calculate the difference degree between the drive current of a single LED lamp bead and the average drive current of all lamp beads in this drive circuit. When the difference degree exceeds 15% and the duration exceeds the preset duration, mark the corresponding LED lamp bead as a risk unit; Based on the junction temperature of the LED lamp beads or the reference temperature of the lamp outer shell, use the thermal field reconstruction technology, combined with the principles related to heat conduction, to construct the temperature field distribution of the LED lighting system. During the construction process, dynamically adjust the heat conduction coefficient and boundary conditions according to different operating conditions of the train to improve the accuracy of the thermal field distribution; Monitor and dynamically update the constructed temperature field distribution in real time. When it is detected that the temperature change exceeds the preset threshold, re-perform thermal field reconstruction to ensure that the temperature field distribution can timely reflect the actual thermal state of the system; When calculating the health degree of lamp beads in each area through a preset model by comprehensively considering the junction temperature of the LED lamp beads, the vibration cumulative damage value, the difference degree of the drive current, and the ripple coefficient, assign different weights to different parameters; Regularly calibrate and optimize the preset model, and adjust the parameters and algorithms in the model according to the actual operation data and historical fault records to calculate the health degree.

6. The intelligent health management method for a train LED lighting system according to claim 1, characterized in that According to the health degree distribution, implement brightness compensation for the high-health-degree areas, and achieve smooth dimming through the gradual transition algorithm to optimize the lighting uniformity. The steps include: Divide the LED lighting system into multiple sub-areas according to the spatial position, and each sub-area contains several LED lamp beads; Calculate the average health degree of the lamp beads in each sub-area at preset time intervals and evaluate the stability of the average health degree of the lamp beads; Set the health degree threshold and stability conditions. When the average health degree of the lamp beads and the stability in the sub-area meet the health degree threshold and the stability conditions, determine that the sub-area is a high-health-degree area; Dynamically adjust the determination criteria for the high - health - degree area according to the running state of the train and environmental conditions; Obtain the regional average health degree, the health degree distribution of adjacent areas, the train operation time period, and the real - time environmental light intensity of the high - health - degree area; Calculate the compensated regional brightness through the preset optimization algorithm; Real - time monitor the parameter changes of the regional average health degree, the health degree distribution of adjacent areas, the train operation time period, and the real - time environmental light intensity. When the parameter changes exceed the preset change range, recalculate the compensated regional brightness and adjust the output of the LED lighting system.

7. The intelligent health management method for a train LED lighting system according to claim 6, characterized in that, The method further includes: Calculate the brightness difference before and after brightness compensation; Determine the gradual change time according to the brightness difference and the passenger comfort requirements through preset rules; Adopt a non - linear transition curve to control the brightness change rate, so that the brightness smoothly transitions within the gradual change time; Real - time monitor the actual brightness change. If the deviation from the non - linear transition curve exceeds the allowable range, automatically fine - tune the output of the LED lighting system to ensure smooth transition; Use the light sensors in the carriage to collect the light intensity of each area of the train, combine the positions of the LED lamp beads and the health degree information, and evaluate the uniformity through the data fusion algorithm; When the deviation of a certain area exceeds the preset threshold, start the adaptive strategy to adjust the brightness compensation amount of this area or adjacent areas; Regularly analyze historical data, and specifically optimize and adjust the determination criteria for the high - health - degree area, the preset rules for determining the gradual change time, the non - linear transition curve, and the specific parameters and trigger conditions of the adaptive strategy.

8. An intelligent health management system for a train LED lighting system, characterized in that, It includes: Temperature and current signal acquisition module: Acquire the reference temperature of the LED lamp housing in the LED lighting system and the drive current signal of the LED lamp beads in the LED lamp; LED lamp bead junction temperature calculation module: Based on a preset electro - thermal coupling model, and combined with the pre - stored LED package thermal resistance parameters and the on - resistance of the drive MOS transistor in the LED lamp drive circuit collected in real - time, calculate the LED lamp bead junction temperature; Vibration cumulative damage value calculation module: Acquire the vibration signal of the LED lamp, perform frequency - domain analysis on the vibration signal, and calculate the vibration cumulative damage value in a specific frequency band; Thermal stress damage value calculation module: Calculate the thermal stress damage value based on the change of the LED lamp bead junction temperature, and form a dual damage assessment of vibration and thermal stress; Ripple coefficient monitoring module: Monitor the ripple coefficient of the LED lamp bead drive current signal, and combine the vibration cumulative damage value and the thermal stress damage value to determine solder joint abnormalities and trigger an early warning; Health degree calculation module for lamp beads in each area: Identify abnormal lamp beads through the difference degree of the LED lamp bead drive current signal, construct a thermal field distribution based on the LED lamp bead junction temperature or the reference temperature of the housing, and fuse the LED lamp bead junction temperature, vibration cumulative damage value, and current state parameters to calculate the health degree of the lamp beads in each area; Brightness compensation module: According to the health degree distribution, perform brightness compensation on the high - health - degree area, and achieve smooth dimming through a gradual transition algorithm to optimize lighting uniformity.

9. An intelligent health management device for a train LED lighting system, characterized in that, It includes: A memory and a processor, wherein a computer program capable of being loaded and executed by the processor for any one of the train LED lighting system intelligent health management methods of the above claims 1-7 is stored on the memory.

10. A storage medium, characterized in that, A computer program capable of being loaded and executed by the processor for any one of the train LED lighting system intelligent health management methods of the above claims 1-7 is stored.

Images