Fault detection and protection method and system for LED drive circuit

Abstract

The invention discloses a fault detection and protection method and system for an LED drive circuit, and belongs to the technical field of circuit fault detection and protection. The method comprises the following steps: acquiring operating parameters of each branch in the LED driving circuit in real time, determining a temperature correction coefficient according to a deviation condition between an LED junction temperature and a rated temperature in junction temperature data, obtaining a dynamic overcurrent threshold adaptive to a current working condition in combination with branch health degree data of a previous round, calculating weighted values of current deviation, voltage deviation and a temperature change rate, and calculating a dynamic overcurrent threshold value adaptive to the current working condition according to the weighted values. The method comprises the steps of obtaining a fault confidence coefficient by combining a coupling relation between current and temperature, calculating a branch health degree according to the fault confidence coefficient and an operation compliance score, and executing preset hierarchical protection according to the branch health degree. Balance of accuracy, timeliness and efficiency of fault detection and protection of the LED drive circuit is realized, and the adaptability and continuous optimization capability of the system are further enhanced.

Description

Technical Field

[0001] This invention relates to the field of circuit fault detection and protection technology, and in particular to a fault detection and protection method and system for LED driver circuits. Background Technology

[0002] LED driver circuits are the core components of LED lighting and display systems. Their stability directly determines the lifespan of LEDs and the safety of the system. With the expansion of LED application scenarios (such as smart streetlights and industrial lighting), driver circuits need to cope with complex operating conditions (such as temperature fluctuations and power grid interference). Therefore, the accuracy and timeliness of fault detection and protection have become crucial.

[0003] Existing LED driver fault detection technologies often employ fixed overcurrent thresholds without considering the influence of LED junction temperature and branch health status. This results in enhanced LED current tolerance at low temperatures, but overly strict thresholds can lead to false triggering of protection. Conversely, at high temperatures, LED current tolerance decreases, but overly lenient thresholds make it difficult to prevent overcurrent burnout. Furthermore, existing technologies rely heavily on single electrical parameters for fault detection, neglecting voltage, temperature change rate, and the correlation between multiple parameters. This leads to missed detections during short-circuit faults (voltage drops suddenly but current does not exceed the fixed threshold), and during open-circuit faults (current is zero but voltage surges), making single-current detection insufficient. In cases of latent faults, abnormal temperature change rate occurs, but current and voltage show no significant deviation, resulting in delayed identification. Additionally, existing protection actions are based on single measurement results, lacking historical health status accumulation and feedback, making early warning difficult. Moreover, after disconnecting the faulty branch, the current in the healthy branch is not optimized, leading to insufficient system brightness.

[0004] Therefore, there is an urgent need for a detection and protection method that integrates multiple parameters, dynamic thresholds, and closed-loop feedback to overcome the limitations of existing technologies. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that the existing technology has the disadvantage of being difficult to adapt to the nonlinear characteristics and dynamic working conditions of LEDs, resulting in poor protection effect. To this end, we propose a fault detection and protection method and system for LED driving circuits.

[0006] In a first aspect, one embodiment of the present invention provides a fault detection and protection method for an LED driving circuit, the method comprising the following steps: Real-time acquisition of operating parameters of each branch in the LED driver circuit, including current data, voltage data, junction temperature data, and the health status data of the branch in the previous round; Based on the deviation between the LED junction temperature and the rated temperature in the junction temperature data, a temperature correction coefficient is determined, and the rated current of the LED branch is adjusted in conjunction with the previous round of branch health data to obtain a dynamic overcurrent threshold suitable for the current operating conditions. Based on the current data, voltage data, and junction temperature data of the LED branch, the weighted values ​​of current deviation, voltage deviation, and temperature change rate are calculated, and the fault confidence is obtained by combining the coupling relationship between current and temperature. Based on the current data, voltage data, and junction temperature data, the operational compliance score of the LED branch is determined, and the branch health is calculated based on the fault confidence and operational compliance score. Based on the branch health status, a pre-set graded protection is performed, and the branch health status is included in the previous round of branch health status data to update the dynamic overcurrent threshold for the next round.

[0007] Preferably, the current data includes rated current and real-time current; The voltage data includes real-time voltage and rated voltage; The junction temperature data includes the LED junction temperature, rated temperature, and maximum allowable junction temperature of the LED. The previous round branch health data includes the previous round branch health.

[0008] Preferably, the specific method for obtaining the dynamic overcurrent threshold is as follows: Based on the degree of difference between the LED junction temperature and the rated temperature, a temperature difference characteristic is determined, and based on the temperature difference characteristic and the temperature correction coefficient, a temperature correction term is determined. Retrieve the health status of the branch in the previous round and use it as the historical correction coefficient; Multiply the rated current of the LED branch by the temperature correction term and the historical correction factor to obtain the dynamic overcurrent threshold of the current branch.

[0009] Preferably, the specific process for determining the temperature correction coefficient is as follows: If the LED junction temperature is less than or equal to the rated temperature, it is determined to be a low-temperature operating condition, and the temperature correction factor is determined to be the low-temperature current tolerance factor for relaxing the threshold at low temperatures, and the temperature correction term is greater than 1. If the LED junction temperature is greater than the rated temperature, it is determined to be a high-temperature operating condition, and the temperature correction coefficient is determined to be the high-temperature current limiting coefficient for tightening the threshold at high temperatures, and the temperature correction term is less than 1.

[0010] Preferably, the specific method for obtaining the fault confidence level is as follows: The current deviation rate is determined based on the degree of deviation between the real-time current and the dynamic overcurrent threshold. The voltage deviation rate is determined based on the degree of deviation between the real-time voltage and the rated voltage; Obtain the time interval and LED junction temperature of two sampling times for each branch, divide the LED junction temperature by the time interval to determine the junction temperature change rate, and divide the junction temperature change rate by the maximum allowable junction temperature of the LED to determine the normalized temperature rise rate term. Calculate the first difference between the real-time current and the dynamic overcurrent threshold, then calculate the second difference between the LED junction temperature and the rated temperature, and multiply the first and second differences to obtain the numerator of the correlation between the current anomaly and the temperature anomaly. Calculate the third difference between the maximum allowable junction temperature of the LED and the rated temperature, and then multiply the third difference with the rated current to obtain the denominator. Divide the numerator by the denominator to obtain the coupling term reflecting the coupling relationship between current and temperature. The branch fault confidence is obtained by weighted summation of the current deviation rate, the voltage deviation rate, the normalized temperature rise rate, and the coupling term.

[0011] Preferably, the specific method for obtaining the health status of the branch is as follows: Based on the aforementioned fault confidence level, the health risk offset value is obtained; Multiply the ratio of the real-time current to the dynamic overcurrent threshold by a preset current weighting coefficient to obtain the current compliance item. Multiply the ratio of the real-time voltage to the rated voltage by a preset voltage weighting coefficient to obtain the voltage compliance item; Based on the normalized temperature rise rate term and the preset temperature weighting coefficient, determine the compliance sub-item of the temperature change rate; The operational compliance score is obtained by adding the current compliance item, the voltage compliance item, and the temperature change rate compliance item together. The branch health score is obtained by multiplying the health risk offset value by the operational compliance score.

[0012] Preferably, the graded protection specifically includes: If the health level of the branch is ≥0.8, then the current branch is determined to be in a healthy state and should proceed with normal operation. If 0.6 ≤ the health status of the branch < 0.8, then the current branch is determined to be in a sub-healthy state, and an early warning is issued and the sampling frequency is increased. If the health status of the branch is less than 0.6, the current branch is determined to be in a fault state. The faulty branch is disconnected, the current of the healthy branch is adjusted, and the fault log is recorded. Specifically, when the branch health is less than 0.6, the branch health is only recorded in the fault log and is not used to update the dynamic overcurrent threshold for the next round.

[0013] Preferably, the calculation formula for adjusting the current in the healthy branch is as follows: ; In the formula: I new JK represents the update current of other branches, I represents the real-time current of each branch, and JK represents the update current of other branches. 总和 For the overall health of the branch roads, JK fault This refers to the health status of the faulty branch.

[0014] Secondly, embodiments of the present invention also provide a fault detection and protection system for an LED driving circuit, including a memory, a processor, and a computer program stored in the memory, characterized in that the processor executes the computer program to implement the steps of the above-described method.

[0015] The technical effects and advantages of this invention are as follows: In this invention, the temperature correction coefficient of the dynamic overcurrent threshold is adjusted by the deviation between the LED junction temperature and the rated temperature to reflect the current tolerance capability of the LED at different temperatures. This allows the threshold to be relaxed at low temperatures to avoid false triggering, and the threshold to be tightened at high temperatures for early protection. Furthermore, based on this correction, the threshold is further corrected using past health data of the branch (the threshold of the sub-healthy branch is further tightened), so that the protection threshold fits the real-time operating conditions and is adapted to the dynamic tolerance characteristics of the LED.

[0016] In this invention, three fundamental features are comprehensively considered: current deviation reflecting overload / undercurrent, voltage deviation reflecting open circuit / short circuit, and temperature change rate reflecting latent faults. Furthermore, the correlation between current and temperature is taken into account (e.g., when the current exceeds the threshold and the temperature rises sharply, the fault confidence is enhanced), and the weight of each feature is balanced. This allows for the comprehensive capture of fault signs and reduces the omission / false detection of a single parameter (e.g., the coupled feature of voltage drop + temperature rise during a short circuit can quickly identify the fault), thereby improving the timeliness and efficiency of fault detection and protection.

[0017] In this invention, the health status of a branch is quantified by combining the probability of failure and operational compliance, and the current health status is used as the basis for the next round of threshold correction. When a branch is in a sub-healthy state, an early warning is given (indicator light flashing / data reporting), and when a branch is faulty, it is quickly isolated (current is cut off). After disconnecting the faulty branch, the current is dynamically adjusted based on the health status of the remaining branches to ensure the system brightness. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the steps of the fault detection and protection method for this LED driver circuit. Detailed Implementation

[0019] The present invention will now be described in further detail with reference to the accompanying drawings and preferred embodiments.

[0020] Reference Figure 1 As shown, the present invention provides a technical solution: a fault detection and protection method for an LED driver circuit, the method comprising the following steps: Step S001: Real-time acquisition of operating parameters of each branch in the LED driver circuit, including current data, voltage data, junction temperature data, and the health status data of the branch in the previous round.

[0021] This step is the initial step of fault detection and protection. The operating parameters of each branch are collected in real time by sensors (such as current sensors, voltage sensors, NTC or integrated temperature sensors) deployed on each branch of the drive circuit. Among the operating parameters, the current data is used to reflect the load status of the branch, the voltage data is used to reflect the stability of the power supply circuit, the junction temperature data is used to reflect the thermal status of the LED chip, and the previous branch health data is the historical health assessment result, which is used for closed-loop feedback.

[0022] The sampling frequency can be set according to system requirements, for example, once every 10 milliseconds. The collected data needs to be filtered (such as moving average filtering) to eliminate noise interference and ensure data reliability.

[0023] This completes the real-time collection of operating parameters for each branch, providing a data foundation for subsequent dynamic threshold calculation and health assessment.

[0024] Step S002: Based on the deviation between the LED junction temperature and the rated temperature in the junction temperature data, determine the temperature correction coefficient, and adjust the rated current of the LED branch in combination with the branch health data from the previous round to obtain a dynamic overcurrent threshold suitable for the current operating conditions.

[0025] In this step, the specific method for obtaining the dynamic overcurrent threshold is as follows: Based on the degree of difference between the LED junction temperature and the rated temperature, the temperature difference characteristics are determined. Then, based on these characteristics and the temperature correction coefficient, a temperature correction term is determined. The previous branch health status is retrieved and used as the historical correction coefficient. The rated current of the LED branch is multiplied by the temperature correction term and the historical correction coefficient to obtain the current branch's dynamic overcurrent threshold. The specific calculation process is as follows: When W j ≤W ref At that time, the dynamic overcurrent threshold I th The calculation is as follows: ; When W j >W ref At that time, the dynamic overcurrent threshold I th The calculation is as follows: ; In the formula: Ith The dynamic overcurrent threshold, i.e., the maximum allowable current in the current branch, I rated The rated current of the LED branch is W. ref For the rated temperature, W j LED junction temperature (real-time measurement, obtained via NTC or chip sensor), JK prev The previous branch health status is defined, with a value ranging from 0 to 1. k1 is the temperature correction coefficient - low-temperature current tolerance coefficient (e.g., 0.001 / ℃, meaning the threshold increases by 0.1% for every 1℃ below the rated temperature), and k2 is the temperature correction coefficient - high-temperature current limitation coefficient (e.g., 0.002 / ℃, meaning the threshold decreases by 0.2% for every 1℃ above the rated temperature). k1 and k2 are temperature correction coefficients under different operating conditions, and their specific determination process is as follows: When W j ≤W ref If the condition is low-temperature, it is determined to be a low-temperature operating condition. The temperature correction factor is set to the low-temperature current tolerance factor k1, which relaxes the threshold at low temperatures. Its essence is to avoid false triggering at low temperatures, and the temperature correction term ( () greater than 1; If W j >W ref If so, it is determined to be a high-temperature operating condition, and the temperature correction factor is determined to be the high-temperature current limiting factor k2 that tightens the threshold at high temperatures. Its essence is junction temperature derating protection, and the temperature correction term ( () Less than 1.

[0026] Understandably, the dynamic overcurrent threshold I th After calculation in step S002 of this invention, the value is no longer fixed, but adaptively adjusted according to junction temperature and historical health status, thereby enhancing the LED current tolerance at low temperatures and thus relaxing the dynamic overcurrent threshold I. th To avoid false protection at low temperatures; at high temperatures, LEDs are prone to overheating and damage, therefore the dynamic overcurrent threshold I is tightened. th This allows for early prevention, while also incorporating the health status of the previous branch from the previous round of health data (JK). prev The secondary correction ensures that branches that previously exhibited sub-optimal performance have stricter protection thresholds in subsequent operations, demonstrating the system's memory and early warning capabilities.

[0027] At this point, the dynamic overcurrent threshold I is complete. th The calculation provides a current judgment benchmark for subsequent fault confidence and health assessment.

[0028] Step S003: Based on the current data, voltage data and junction temperature data of the LED branch, calculate the weighted value of the current deviation, voltage deviation and temperature change rate, and obtain the fault confidence level by combining the coupling relationship between current and temperature.

[0029] In this step, the specific method for obtaining the fault confidence level is as follows: The current deviation rate is determined based on the degree of deviation between the real-time current and the dynamic overcurrent threshold. The voltage deviation rate is determined based on the degree of deviation between the real-time voltage and the rated voltage. The time interval and LED junction temperature at two sampling moments are obtained for each branch. The LED junction temperature is divided by the time interval to determine the junction temperature change rate. The junction temperature change rate is then divided by the maximum allowable junction temperature of the LED to determine the normalized temperature rise rate. The first difference between the real-time current and the dynamic overcurrent threshold is calculated. The second difference between the LED junction temperature and the rated temperature is then calculated. The first and second differences are multiplied to obtain the numerator of the correlation between current anomaly and temperature anomaly. The third difference between the maximum allowable junction temperature of the LED and the rated temperature is calculated. The third difference is then multiplied by the rated current to obtain the denominator. The numerator is divided by the denominator to obtain the coupling term reflecting the coupling relationship between current and temperature. The current deviation rate, voltage deviation rate, normalized temperature rise rate, and coupling term are weighted and summed to obtain the branch fault confidence. The specific calculation process is as follows: ; In the formula: GZD is the fault confidence level, I is the real-time current (measured value) of the branch, and DY is the real-time voltage (measured value) of the branch. rated The rated voltage of the branch is ΔW, where ΔW is the LED junction temperature (W) of the LED branch. j The junction temperature change at two consecutive sampling times, where Δt is the time interval between the two sampling times, ΔW / Δt is the junction temperature change rate, and ΔW max For the maximum allowable junction temperature change rate, W max The maximum allowable junction temperature of the LED is represented by w1, w2, w3, and w4, which are all weighting coefficients, and w1+w2+w3+w4=1. For example, w1, w2, w3, and w4 can all be set to 0.25, or the weights can be reasonably allocated according to the influence of specific factors.

[0030] Understandably, the fault confidence level (GZD) is a multi-dimensional fault probability index that integrates current overload / undercurrent, voltage anomalies (such as short-circuit voltage sag and open-circuit voltage sag), abnormal temperature changes, and current-temperature coupling effects, as well as the current deviation rate (…). Primarily detects overcurrent or undercurrent faults; voltage deviation rate ( ) is used to identify short circuits or open circuits; the normalized temperature rise rate term ( This can detect latent thermal faults; coupling terms ( This strengthens fault diagnosis when abnormal current and abnormal temperature occur simultaneously, such as in cases of overcurrent and severe temperature rise, and the coupling term (( It can also eliminate misjudgment interference: for example, when power grid fluctuations cause current to rise but the temperature is normal, the coupling term (( The value is negative to reduce the fault confidence; during a short circuit, the current exceeds the threshold and the temperature rises sharply, and the coupling term (( The confidence level is strengthened by setting the initial value to positive. Then, by weighted summation, fault signs with different characteristics are merged into a unified confidence value, which improves the comprehensiveness and accuracy of fault identification.

[0031] In addition, the fault confidence level GZD ranges from 0 to 1: If the fault confidence GZD≈0: all features are normal, and the probability of fault is almost 0; If 0 < fault confidence level GZD < 0.5: individual characteristics are abnormal (such as slight current deviation caused by power grid fluctuations), and the probability of fault is low; If 0.5 ≤ Fault Confidence GZD < 0.8: Multiple abnormal characteristics (such as current deviation + slow temperature rise) indicate a moderate probability of failure. If the fault confidence level GZD ≥ 0.8C: all features conform to the fault mode (e.g., all four items → 1 during a short circuit), the fault probability is close to 100%.

[0032] In short, the higher the GZD value of the fault confidence score, the more complete the "chain of evidence" for the fault, and the higher the probability of the fault.

[0033] This completes the quantitative calculation of the fault confidence level (GZD) for each branch, providing a risk basis for health assessment.

[0034] Step S004: Determine the operational compliance score of the LED branch based on the current data, voltage data, and junction temperature data, and calculate the branch health based on the fault confidence and operational compliance score.

[0035] In this step, the specific method for obtaining the branch health status is as follows: Based on the fault confidence level, a health risk mitigation value is obtained. The ratio of real-time current to dynamic overcurrent threshold is multiplied by a preset current weighting coefficient to obtain the current compliance item. The ratio of real-time voltage to rated voltage is multiplied by a preset voltage weighting coefficient to obtain the voltage compliance item. Based on the normalized temperature rise rate and a preset temperature weighting coefficient, a temperature change rate compliance item is determined. The current compliance item, voltage compliance item, and temperature change rate compliance item are added together to obtain the operational compliance score. The health risk mitigation value is multiplied by the operational compliance score to obtain the branch health level. The specific calculation process is as follows: ; In the formula: JK is the health of the branch (the value ranges from 0 to 1, and the smaller the value, the worse the health status), w5, w6 and w7 are all weight coefficients, and w5+w6+w7=1. For example, w5=0.4, w6=0.3, w7=0.3 or the same as in step S003, the weights are reasonably allocated according to the influence of specific factors.

[0036] Understandably, the branch health score JK is a composite indicator that integrates fault risk (reflected by fault confidence level GZD) and operational compliance (reflected by the degree to which current, voltage, and temperature rise rate are close to the ideal state). The (1-GZD) term converts the fault probability into a health risk discount factor - health risk offset value. When the fault confidence level GZD is high, this factor ((1-GZD)) is small and will significantly reduce the health score. The operational compliance score reflects the branch's "excellent" level in terms of current utilization, voltage stability, and temperature rise stability. Among them, the current compliance sub-item... Used to assess the compliance of actual current with dynamic thresholds—the closer the ratio is to 1, the better the current meets real-time protection requirements; voltage compliance sub-item. Used to assess the compliance of actual voltage with respect to rated value—the closer the ratio is to 1, the better the voltage meets design operating requirements; temperature change rate compliance item. The normalized rate of temperature rise term is used to assess thermal stability. The smaller (closer to 0), the more 1 is subtracted from that term. The closer the result is to 1, the higher the compliance. Multiplying the two results penalizes the risk of failure and rewards stable and efficient operation, thus obtaining a quantitative value that can comprehensively and dynamically reflect the overall health level of the branch.

[0037] This completes the calculation of the health status of each branch road, providing a direct basis for implementing graded protection decisions.

[0038] Step S005: Perform the pre-set graded protection based on the branch health status, and include the branch health status in the previous round of branch health status data to update the dynamic overcurrent threshold for the next round.

[0039] Specifically, the graded protection system includes: If the branch health level is ≥0.8, the current branch is considered to be in a healthy state and should proceed with normal operation. If 0.6 ≤ branch health < 0.8, the current branch is determined to be in a sub-healthy state, and an early warning is issued and the sampling frequency is increased. If the branch health level is <0.6, the current branch is determined to be in a fault state. When the faulty branch is disconnected, the current of the healthy branch is adjusted and the fault log is recorded, the specific formula for calculating the current of the healthy branch is as follows: ; In the formula: I new For the renewal current of other branches, JK 总和 The total branch health score is the sum of the branch health scores JK of all branches (including faulty branches) in the LED driver circuit. fault For the health status of the faulty branch, The goal is to match the increase in current in healthy branches with their respective health levels (branches with higher health levels can withstand more current). Understandably, the tiered protection mechanism achieves a gradient response from early warning to isolation. The sub-health warning provides maintenance personnel with an intervention window to prevent the fault from escalating; rapid isolation of faulty branches prevents damage propagation; and when the branch health JK < 0.6, the branch health JK is only recorded in the fault log and is not used to update the dynamic overcurrent threshold I in the next round. th The system will cut off the current to the faulty branch via the MOSFET, causing the faulty branch to stop working. No sampling data will be available in subsequent cycles (current, voltage, and temperature are all 0), therefore the dynamic overcurrent threshold I does not need to be substituted. th The calculation is performed using the formula. For healthy branches where the branch health JK ≥ 0.8 or 0.6 ≤ branch health JK < 0.8, the branch health JK is used as the branch health JK from the previous round. prev Substitute dynamic overcurrent threshold I th In the calculation formula, the dynamic overcurrent threshold I is adjusted. th This achieves closed-loop optimization. Among them, the dynamic adjustment of the current in healthy branches is a highlight. By redistributing power at the system level and utilizing the redundancy of the remaining healthy branches, it maximizes the overall brightness of the system and improves the availability and resilience of the system. Its closed-loop feedback mechanism (health status is used to update the threshold for the next round) enables the system to have learning and adaptive capabilities, and can continuously optimize its protection strategy based on the historical performance of the branches.

[0040] Furthermore, the branch health JK result mentioned above will be used as the branch health JK for the previous round. prev Feedback to dynamic overcurrent threshold I th In the calculation formula, a negative feedback loop is formed: The branch health JK of the faulty branch decreases → dynamic overcurrent threshold I th More stringent → Fault confidence level GZD is more likely to increase → Branch health level JK is further reduced → Ultimately triggering disconnection action; The health status of the healthy branch JK is stable → dynamic overcurrent threshold I th Maintaining a reasonable level → ensuring the protection threshold is not excessively tightened, thus guaranteeing system efficiency.

[0041] This completes a full cycle of fault detection, assessment, protection, and feedback.

[0042] Based on the same inventive concept as the above method, this embodiment of the invention also provides a fault detection and protection system for an LED driver circuit, including a memory, a processor, and a computer program stored in the memory. When the processor executes the computer program, it implements the above-mentioned fault detection and protection method for an LED driver circuit.

[0043] It should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should also be within the scope of protection of this invention.

Claims

1. A fault detection and protection method for an LED driver circuit, characterized in that, The method includes the following steps: Real-time acquisition of operating parameters of each branch in the LED driver circuit, including current data, voltage data, junction temperature data, and the health status data of the branch in the previous round; Based on the deviation between the LED junction temperature and the rated temperature in the junction temperature data, a temperature correction coefficient is determined, and the rated current of the LED branch is adjusted in conjunction with the previous round of branch health data to obtain a dynamic overcurrent threshold suitable for the current operating conditions. Based on the current data, voltage data, and junction temperature data of the LED branch, the weighted values ​​of current deviation, voltage deviation, and temperature change rate are calculated, and the fault confidence is obtained by combining the coupling relationship between current and temperature. Based on the current data, voltage data, and junction temperature data, the operational compliance score of the LED branch is determined, and the branch health is calculated based on the fault confidence and operational compliance score. Based on the branch health status, a pre-set graded protection is performed, and the branch health status is included in the previous round of branch health status data to update the dynamic overcurrent threshold for the next round.

2. The method for fault detection and protection of an LED driver circuit according to claim 1, characterized in that, The current data includes rated current and real-time current; The voltage data includes real-time voltage and rated voltage; The junction temperature data includes the LED junction temperature, rated temperature, and maximum allowable junction temperature of the LED. The previous round branch health data includes the previous round branch health.

3. The fault detection and protection method for an LED driver circuit according to claim 2, characterized in that, The specific method for obtaining the dynamic overcurrent threshold is as follows: Based on the degree of difference between the LED junction temperature and the rated temperature, a temperature difference characteristic is determined, and based on the temperature difference characteristic and the temperature correction coefficient, a temperature correction term is determined. Retrieve the health status of the branch in the previous round and use it as the historical correction coefficient; Multiply the rated current of the LED branch by the temperature correction term and the historical correction factor to obtain the dynamic overcurrent threshold of the current branch.

4. The fault detection and protection method for an LED driver circuit according to claim 3, characterized in that, The specific process for determining the temperature correction coefficient is as follows: If the LED junction temperature is less than or equal to the rated temperature, it is determined to be a low-temperature operating condition, and the temperature correction factor is determined to be the low-temperature current tolerance factor for relaxing the threshold at low temperatures, and the temperature correction term is greater than 1. If the LED junction temperature is greater than the rated temperature, it is determined to be a high-temperature operating condition, and the temperature correction coefficient is determined to be the high-temperature current limiting coefficient for tightening the threshold at high temperatures, and the temperature correction term is less than 1.

5. The fault detection and protection method for an LED driver circuit according to claim 2, characterized in that, The specific method for obtaining the fault confidence level is as follows: The current deviation rate is determined based on the degree of deviation between the real-time current and the dynamic overcurrent threshold. The voltage deviation rate is determined based on the degree of deviation between the real-time voltage and the rated voltage; Obtain the time interval and LED junction temperature of two sampling times for each branch, divide the LED junction temperature by the time interval to determine the junction temperature change rate, and divide the junction temperature change rate by the maximum allowable junction temperature of the LED to determine the normalized temperature rise rate term. Calculate the first difference between the real-time current and the dynamic overcurrent threshold, then calculate the second difference between the LED junction temperature and the rated temperature, and multiply the first and second differences to obtain the numerator of the correlation between the current anomaly and the temperature anomaly. Calculate the third difference between the maximum allowable junction temperature of the LED and the rated temperature, and then multiply the third difference with the rated current to obtain the denominator. Divide the numerator by the denominator to obtain the coupling term reflecting the coupling relationship between current and temperature. The branch fault confidence is obtained by weighted summation of the current deviation rate, the voltage deviation rate, the normalized temperature rise rate, and the coupling term.

6. The fault detection and protection method for an LED driver circuit according to claim 4, characterized in that, The specific method for obtaining the health status of the branch is as follows: Based on the aforementioned fault confidence level, the health risk offset value is obtained; Multiply the ratio of the real-time current to the dynamic overcurrent threshold by a preset current weighting coefficient to obtain the current compliance item. Multiply the ratio of the real-time voltage to the rated voltage by a preset voltage weighting coefficient to obtain the voltage compliance item; Based on the normalized temperature rise rate term and the preset temperature weighting coefficient, determine the compliance sub-item of the temperature change rate; The operational compliance score is obtained by adding the current compliance item, the voltage compliance item, and the temperature change rate compliance item together. The branch health score is obtained by multiplying the health risk offset value by the operational compliance score.

7. The fault detection and protection method for an LED driver circuit according to claim 1, characterized in that, The hierarchical protection specifically refers to: If the health level of the branch is ≥0.8, then the current branch is determined to be in a healthy state and should proceed with normal operation. If 0.6 ≤ the health status of the branch < 0.8, then the current branch is determined to be in a sub-healthy state, and an early warning is issued and the sampling frequency is increased. If the health status of the branch is less than 0.6, the current branch is determined to be in a fault state. The faulty branch is disconnected, the current of the healthy branch is adjusted, and the fault log is recorded. Specifically, when the branch health is less than 0.6, the branch health is only recorded in the fault log and is not used to update the dynamic overcurrent threshold for the next round.

8. The method for fault detection and protection of an LED driving circuit according to claim 7, characterized in that, The specific formula for calculating the current of the healthy branch is as follows: ； In the formula: I new JK represents the update current of other branches, I represents the real-time current of each branch, and JK represents the update current of other branches. 总和 For the overall health of the branch roads, JK fault This refers to the health status of the faulty branch.

9. A fault detection and protection system for an LED driver circuit, characterized in that, The method includes a memory, a processor, and a computer program stored on the memory, characterized in that the processor executes the computer program to implement the steps of any one of claims 1-8.

Images