A compensation control method for light-emitting stability of a copper wire lamp in a low-temperature environment

Abstract

This application relates to the field of semiconductor lighting driving control technology, and discloses a method for compensating for the luminous stability of a copper wire lamp under low-temperature conditions. This method injects a diagnostic current into the copper wire lamp during the turn-off time slot of a single-cycle driving sequence to obtain the transient voltage decay sequence of the light-emitting device. Based on this sequence, it decouples and calculates the equivalent ambient temperature and the mechanical stress deviation caused by the cold shrinkage of the packaging material. When the mechanical stress deviation exceeds a safety threshold, it calculates the corresponding active heat injection waveform parameters. Based on these parameters, it performs charge equivalence conversion, corrects the loss in the steady-state luminous time of the current cycle, and reconstructs and outputs a composite driving waveform including the heat injection and steady-state luminous stages. This invention improves the structural reliability and luminous stability of the device without adding external sensors by detecting and utilizing Joule heating to release low-temperature mechanical compressive stress through electrical signal quantization and Joule heating.

Description

Technical Field

[0001] This invention relates to the field of semiconductor lighting drive control technology, specifically a method for compensating for the luminous stability of copper wire lamps in low-temperature environments. Background Technology

[0002] Bare-chip copper wire lamps encapsulated with epoxy resin are widely used in various lighting scenarios due to their compact structure and high flexibility. However, when operating in low-temperature environments, the encapsulation material undergoes physical contraction due to thermal expansion and contraction, resulting in mechanical compressive stress on the internal light-emitting chip. This compressive stress induces a piezoresistive effect, altering the dynamic internal resistance and forward voltage drop of the light-emitting device, causing a shift in luminous power under constant current drive. Simultaneously, continuous physical compression can easily cause fatigue or even breakage of the internal micro-leads, reducing the device's lifespan.

[0003] Existing technologies addressing luminous decay caused by low temperatures typically involve adding external heating elements or altering the overall driving current amplitude. Adding external heating elements is limited by the small physical size of copper wire lamps and significantly increases circuit hardware costs. Simply adjusting the driving current amplitude to compensate for heating the light-emitting device alters the total charge received by the device in a single driving cycle, causing energy fluctuations in the luminous output. These fluctuations manifest macroscopically as abrupt changes in brightness and flickering. Current driving control technologies fail to separately detect ambient temperature and packaging stress, making it impossible to maintain smooth and stable optical output while eliminating physical compressive stress, and thus difficult to balance the structural safety and visual consistency of the light-emitting device in low-temperature environments. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for predicting electricity meter faults based on edge computing. It solves the problem that when copper wire lamps encapsulated with materials such as epoxy resin operate in low-temperature environments, the encapsulation material shrinks due to cold and generates mechanical compressive stress on the internal light-emitting chip. This stress induces a piezoresistive effect, causing a change in the dynamic internal resistance of the light-emitting device, and there is a risk of open circuit due to compression of the internal leads.

[0005] The first aspect of this invention provides a method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions, comprising the following steps: During the turn-off time slot of a single-cycle drive sequence, a diagnostic current is injected into the copper wire lamp, and the transient voltage decay sequence across the light-emitting device is obtained. Based on the transient voltage decay sequence, the ambient temperature equivalent of the current light-emitting device and the mechanical stress deviation caused by the cold shrinkage of the packaging material are decoupled and calculated. When the mechanical stress deviation exceeds the set safety threshold, the corresponding active heat injection waveform parameters are calculated based on the deviation. Based on the active heat injection waveform parameters, charge equivalence conversion is performed, the steady-state emission time of this cycle is corrected for loss, and the output is reconstructed to include the composite driving waveform containing the heat injection stage and the steady-state emission stage.

[0006] Furthermore, the steps of injecting diagnostic current into the copper wire lamp and obtaining the transient voltage decay sequence specifically include: turning off the time slot at the end of the PWM cycle, turning on the microampere-level diagnostic current source to inject diagnostic current into the lamp string; after the voltage stabilizes, acquiring the steady-state reference voltage drop at this time; cutting off the diagnostic current, and continuously acquiring the discharge process from the two ends of the light-emitting device to the residual zero bias voltage at a preset sampling rate to obtain the transient voltage decay sequence.

[0007] Furthermore, the process of decoupling the calculation of the ambient temperature equivalent includes: extracting the preset lamp bead voltage-temperature coefficient; calculating the voltage difference between the steady-state reference voltage drop and the standard room temperature voltage drop; and dividing the voltage difference by the voltage-temperature coefficient to obtain the ambient temperature equivalent of the current light-emitting device.

[0008] Furthermore, the process of decoupling the calculation of mechanical stress deviation includes: based on the ambient temperature equivalent, querying a preset reference time constant mapping table to obtain the theoretical reference time constant of the light-emitting device under the current temperature and stress-free state; performing exponential decay fitting on the transient voltage decay sequence to extract the measured time constant; and calculating the difference between the measured time constant and the theoretical reference time constant to obtain the mechanical stress deviation.

[0009] Furthermore, the measured time constant is the time corresponding to when the voltage value in the transient voltage decay sequence drops to 36.8% of the difference between the initial steady-state reference voltage drop and the residual zero bias voltage.

[0010] Furthermore, the active thermal injection waveform is a trapezoidal current waveform, and its waveform parameters include the leader ramp-up time, the target peak current, and the constant energy injection region maintenance time.

[0011] Furthermore, the process of calculating the active heat injection waveform parameters includes: mapping the required transient Joule thermal energy target in the forward direction based on the magnitude of the mechanical stress deviation; calculating the leading ramp-up time to limit inductor overshoot by combining the parasitic inductance parameters of the power supply line; and jointly solving the target peak current and the constant energy injection region maintenance time based on the transient Joule thermal energy target and the thermal capacity characteristics of the light-emitting device.

[0012] Furthermore, the amplitude of the target peak current is set to be 2 to 4 times the rated steady-state luminous current of the light-emitting device, and less than the transient fusing current threshold of the gold wire inside the light-emitting device.

[0013] Furthermore, the process of performing charge equivalence conversion and time loss correction includes: integrating the active heat injection waveform in the time domain to calculate the total charge injected during the heat injection stage; dividing the total charge by the rated steady-state luminous current of the light-emitting device to obtain the equivalent steady-state luminous time; and subtracting the equivalent steady-state luminous time from the originally set target luminous time to obtain the corrected actual steady-state luminous time.

[0014] Furthermore, the specific timing actions of reconstructing the output composite drive waveform are as follows: First, output an active heat injection waveform to soften the packaging material using Joule heating to release mechanical compressive stress; then control the drive current amplitude to decrease and clamp it to the rated steady-state luminous current, entering a steady-state luminous stage with a duration equal to the actual steady-state luminous time; finally, cut off the output current and enter the periodic turn-off time slot.

[0015] Furthermore, the timing of the diagnostic current injection, the timing of the transient voltage sampling, and the timing of the composite drive waveform output are all completed sequentially in a closed loop within a single PWM high-frequency cycle.

[0016] Furthermore, the copper wire lamp is formed by encapsulating the bare LED chip with epoxy resin. The mechanical stress deviation represents the intensity of the piezoresistive effect of the epoxy resin on the LED chip lattice under low-temperature shrinkage.

[0017] A second aspect of the present invention provides a luminous stability compensation control system for copper wire lamps in low-temperature environments, comprising: The diagnostic sampling module is used to inject diagnostic current into the copper wire lamp during the turn-off time slot of a single-cycle drive sequence and obtain the transient voltage decay sequence across the light-emitting device. The parameter decoupling module is used to decouple and calculate the ambient temperature equivalent of the current light-emitting device and the mechanical stress deviation caused by the cold shrinkage of the packaging material based on the transient voltage decay sequence. The waveform calculation module is used to calculate the corresponding active heat injection waveform parameters based on the deviation when the mechanical stress deviation exceeds the set safety threshold. The reconstructed output module is used to perform charge equivalence conversion based on the active heat injection waveform parameters, correct the loss of steady-state emission time in this cycle, and reconstruct the output of a composite driving waveform that includes the heat injection stage and the steady-state emission stage.

[0018] Furthermore, the diagnostic sampling module includes a micro-current source circuit and an ADC acquisition circuit, with the ADC acquisition circuit connected in parallel with the copper wire lamp series circuit.

[0019] Furthermore, the reconfigurable output module includes a high-frequency PWM generator and a voltage-controlled constant current source circuit, and is configured to switch the drive current amplitude from the peak value of the hot injection stage to the rated value of the steady-state light emission stage within a set response time.

[0020] This invention provides a method for compensating and controlling the luminous stability of copper wire lamps in low-temperature environments. It has the following beneficial effects: 1. This invention injects diagnostic current during the turn-off time slot of a single-cycle drive sequence and acquires the transient voltage decay sequence of the light-emitting device. It then utilizes the difference in the RC discharge time constant to decouple and extract the mechanical stress deviation caused by the cold shrinkage of the packaging material. This method converts the purely mechanical packaging compressive stress into an electrical signal for quantitative detection without adding external physical sensors or altering the original circuit hardware topology, thus controlling hardware size and system manufacturing costs.

[0021] 2. This invention calculates the active heat injection waveform based on the mechanical stress deviation obtained from decoupling, and uses the local Joule heat generated by the transient high current to soften the packaging material and release compressive stress. This heat injection control mechanism, which is dynamically implemented based on real-time stress quantification values, directly eliminates the interference of the piezoresistive effect on the chip lattice caused by the low-temperature environment, and eliminates the physical risk of internal leads breaking due to material compression, thereby improving the structural reliability and luminous maintenance rate of copper wire lamps in low-temperature environments.

[0022] 3. This invention introduces a charge equivalence conversion step while performing active heat injection. By integrating the current during the heat injection stage over time, the actual steady-state emission time within the current cycle is calculated and deducted by an equivalent amount. This processing logic maintains a constant total charge flowing through the light-emitting device within a single cycle, offsetting the additional luminous flux generated by transient heating current pulses, and avoiding sudden changes in macroscopic luminous brightness and visual flicker caused by variations in the driving current amplitude. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the control system and physical model of an embodiment of the present invention; Figure 2 This is a flowchart of the light emission control method according to an embodiment of the present invention; Figure 3 This is a diagnostic sampling timing diagram according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating the parameter decoupling logic of an embodiment of the present invention; Figure 5 This is a schematic diagram of the trapezoidal heat injection waveform calculation logic of an embodiment of the present invention; Figure 6 This is a timing reconstruction diagram of the composite drive waveform according to an embodiment of the present invention; Figure 7 This is a flowchart illustrating the fault diagnosis, fault tolerance processing, and cross-cycle resampling control of an embodiment of the present invention. Figure 8 This is a comparison diagram of transient voltage decay and stress decoupling in this invention; Figure 9This is a timing diagram of the single-cycle composite drive waveform of the present invention; Figure 10 This is the relative luminous flux decay lifetime curve for the extreme cold environment (-30℃) of the present invention.

[0024] Among them, 100 is the control system; 101 is the main control module; 102 is the conversion module; 103 is the drive module; 104 is the storage module; 200 is the light string; 201 is the light-emitting device; 202 is the wire; and 203 is the encapsulation layer. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] See attached document Figure 1 , Figure 1 This is a schematic diagram of a control system and physical model according to an embodiment of the present invention. The present invention provides a luminous stability compensation control system for copper wire lamps in low-temperature environments, comprising: a control system 100, a main control module 101, a conversion module 102, a drive module 103, and a storage module 104.

[0027] The main control module 101 serves as the core computing and processing unit, used for timing scheduling, data parsing, and driving signal calculation.

[0028] The conversion module 102 establishes a data transmission connection with the main control module 101. In terms of hardware integration, the conversion module 102 can be integrated as an on-chip analog-to-digital converter unit within the main control module 101, or as an independent analog-to-digital converter chip connected externally to the main control module 101. When the conversion module 102 is integrated within the main control module 101, the communication connection manifests as an on-chip data bus or register data reading connection. The sampling input terminal of the conversion module 102 is connected to the voltage sampling nodes at both ends of the lamp string 200 via a current limiting, voltage dividing, and clamping protection network. This network is used to acquire the analog voltage signal and convert it into a digital sequence for transmission to the main control module 101. The voltage dividing and clamping protection network is configured as a high input impedance sampling network, with its equivalent input impedance greater than a preset multiple of the equivalent dynamic internal resistance of the lamp string 200 within the transient attenuation measurement range, preferably not less than 100 times. The clamping device is in a cutoff state within the normal sampling voltage range, and its leakage current is limited to a range that does not change the transient attenuation time constant. During the production calibration phase, the main control module 101 records the equivalent load correction amount introduced by the sampling network, or deducts the fixed offset caused by the sampling network load on the attenuation curve when calculating the measured attenuation time constant.

[0029] The control input terminal of the drive module 103 is connected to the main control module 101. The drive module 103 consists of a high-bandwidth operational amplifier and a linear constant current trap circuit. The drive module 103 receives the control signal output by the main control module 101 and outputs a drive current of the corresponding waveform to the outside. The drive module 103 is equipped with a current sampling feedback branch, a maximum compliant output voltage limiting branch, and an overcurrent clamping branch, enabling it to operate in diagnostic low current mode, transient hot injection mode, and steady-state light-emitting drive mode, respectively. When the lamp string 200 is in an open circuit state, causing the constant current output to fail to be established, the output voltage of the drive module 103 is limited to a preset compliant voltage range, and the corresponding high voltage state is acquired by the conversion module 102. When the lamp string 200 is in a short circuit state, the drive module 103 limits the output current to not exceed a preset safety value through the current sampling feedback branch.

[0030] The storage module 104 is connected to the main control module 101, and contains the system's preset benchmark calibration data and multi-parameter mapping matrix.

[0031] The electrical output terminal of the control system 100 is connected to the external light string 200.

[0032] The light string 200 includes light-emitting devices 201, wires 202, and encapsulation layer 203 that are electrically connected to each other.

[0033] The conductor 202 is a bare metal wire substrate that connects each light-emitting device 201. In the electrical equivalent model, the conductor 202 is equivalent to having the line equivalent resistance and the line parasitic inductance connected in series.

[0034] The light-emitting device 201 constitutes the core of the photoelectric conversion circuit. In the transient physical equivalent model, the internal semiconductor structure of the light-emitting device 201 is equivalent to a parallel topology of dynamic internal resistance and parasitic junction capacitance. When the light string 200 includes multiple light-emitting devices 201 connected in series, the main control module 101 converts the dynamic internal resistance, parasitic junction capacitance, and parasitic parameters of the wires 202 of each light-emitting device 201 into the equivalent dynamic internal resistance at the light string level. With equivalent parasitic junction capacitance The equivalent time constant corresponding to the dominant decay mode is used as the object of subsequent decoupling calculation. When the decay curve is detected to deviate significantly from the single exponential model, the main control module 101 uses a preset goodness-of-fit threshold to determine that the sampling is invalid and re-executes the diagnosis when sampling is allowed in the future.

[0035] The encapsulation layer 203 completely covers the light-emitting device 201 and part of the conductive wire 202. Under low temperature conditions, the encapsulation layer 203 shrinks in volume due to the difference in the coefficient of thermal expansion of the materials, thereby applying mechanical compressive stress to the internal light-emitting device 201.

[0036] Mechanical compressive stress alters the depletion layer width within the light-emitting device 201 through the semiconductor piezoresistive effect, causing shifts in dynamic internal resistance and parasitic junction capacitance parameters. The control system 100 identifies and intervenes in the mechanical state by acquiring changes in these electrical parameters. Specifically, the control system 100 does not directly measure the absolute mechanical stress value within the encapsulation layer 203; instead, it defines the shift in the transient decay time constant caused by the combined dynamic internal resistance and parasitic junction capacitance as the equivalent stress deviation, and uses this equivalent stress deviation to characterize the observable electrical impact of the encapsulation compressive stress on the light-emitting device 201.

[0037] See attached document Figure 2 , Figure 2 This is a flowchart of a light emission control method according to an embodiment of the present invention. The present invention provides a method for compensating for the luminous stability of a copper wire lamp in a low-temperature environment, comprising the following steps: S100 deploys a diagnostic time window containing a flyback blanking mechanism within the drive turn-off time slot, injects diagnostic current into the lamp string 200, and obtains the steady-state reference voltage drop and transient voltage decay sequence. S200, based on the obtained steady-state reference voltage drop and transient voltage decay sequence, establishes a decoupling model to separate the ambient temperature equivalent and stress deviation; S300, when the stress deviation exceeds the threshold, calculate the trapezoidal heat injection waveform parameters for actively releasing compressive stress by combining the parasitic inductance parameter of conductor 202. S400 calculates steady-state luminescence compensation parameters based on the equivalent charge amount calculated from the ambient temperature equivalent and the trapezoidal heat injection waveform, and reconstructs and outputs a composite driving waveform containing the temporal gradient. S500: After obtaining diagnostic data in step S100, open-circuit and short-circuit boundary checks are performed first. If no hard fault lockout is triggered, stress decoupling, heat injection waveform calculation, and luminescence compensation control are performed sequentially. After completing the output of the composite drive waveform for this cycle, a heat conduction hysteresis timing mechanism is activated to lock the earliest trigger node for the next round of transient attenuation sampling. Subsequently, degradation trend diagnosis is performed based on the diagnostic data and control parameters, controlling the main control module 101 to enter the next round of closed-loop detection and execution. Among these, open-circuit and short-circuit boundary checks have a higher priority than stress decoupling, heat injection calculation, and steady-state luminescence compensation. Degradation trend diagnosis is only performed when the diagnostic data for this cycle is deemed valid and no hard fault lockout is triggered. When the heat conduction hysteresis timing mechanism has not yet released permission for the next round of sampling, the main control module 101 skips the transient attenuation sampling step and performs steady-state luminescence scheduling based on the most recent valid diagnostic result or preset safety parameters.

[0038] See attached document Figure 3 , Figure 3This is a diagnostic sampling timing diagram according to an embodiment of the present invention. In this embodiment, during the drive module 103 being in the drive shutdown time slot, step S100 specifically includes the following sub-steps: S110, when the main control module 101 detects the main light-emitting drive pulse cut-off signal, it starts the internal timer to establish the initial flyback blanking time window. After the initial flyback blanking time window ends, it controls the drive module 103 to output diagnostic current to the lamp string 200.

[0039] The conductor 202 has parasitic inductance in its electrical characteristics. At the instant the main light-emitting drive pulse is cut off, this parasitic inductance will generate a flyback voltage spike interference. To avoid this spike signal damaging the sampling device or causing data distortion, as a preferred implementation, the initial flyback blanking time window is set to be longer than the freewheeling discharge time of the parasitic inductance, typically ranging from several microseconds to tens of microseconds. As a feasible method of determination, the main control module 101 determines this based on the parasitic inductance parameters pre-stored in the storage module 104. and equivalent series resistance parameter Calculate the time constant of the line inductance The initial flyback blanking time window is set to be no less than 3. Preferably 3 Up to 5 When the system has not pre-stored and During the production calibration phase, the time it takes for the voltage spike to fall back below a preset safety threshold can be detected by using oscilloscope sampling or high-speed sampling results from the conversion module 102, and this time is written into the storage module 104 as a blanking time parameter. During this period, the main control module 101 controls the conversion module 102 to be in a sampling shielded state.

[0040] After the initial flyback blanking time window ends, the main control module 101 outputs a current setting control quantity to the drive module 103. This current setting control quantity can be an analog voltage output by the on-chip digital-to-analog converter, a pulse-width modulation signal reconstructed after low-pass filtering, or a digital control code transmitted to the digital-to-analog converter unit inside the drive module 103. The drive module 103 generates a corresponding analog control level based on this current setting control quantity, driving the internal power devices to establish a constant diagnostic current and injecting it into the lamp string 200. The diagnostic current amplitude is set below the visual emission threshold of the light-emitting device 201, typically ranging from 0.1mA to 1mA, to avoid perceptible light output during the turn-off time slot. The visual emission threshold is obtained by calibrating the brightness of the same type of light-emitting device 201 in a dark room environment, and is limited to the condition that no perceptible light output is generated even under the lowest temperature and highest luminous efficiency sample conditions; when the number of light-emitting devices 201 connected in series, the emission color, or the packaging type in the light string 200 changes, the main control module 101 calls the corresponding model's diagnostic current upper limit table in the storage module 104 to limit the amplitude. In order to fully disclose the technical solution, the applicant points out that the physical implementation of this control mechanism is not limited to a specific model of digital-to-analog converter. For the specific structure of the internal reference voltage adjustment and current negative feedback adjustment of the linear constant current drive circuit, those skilled in the art can use conventional operational amplifiers in conjunction with sampling resistors to implement it, and its specific hardware circuit construction is a well-known technology in the field, and will not be described in detail here.

[0041] S120, after the diagnostic current output is stable, the main control module 101 triggers the conversion module 102 to collect the voltage signal across the lamp string 200 as the steady-state reference voltage drop.

[0042] Due to the presence of parasitic inductance in the line, the diagnostic current needs to undergo an inductor magnetization delay to climb from zero to the set value. The main control module 101 delays until the differential rate of change of the parasitic inductance in the line approaches zero. Considering potential impedance changes in actual engineering environments, and to prevent the control algorithm from deadlocking while waiting for current stabilization, this delay process incorporates timeout forced exit logic. When the timeout forced exit logic is triggered, the main control module 101 determines that the current diagnostic sampling is invalid, abandons the subsequent parameter decoupling calculation for the current cycle, and maintains the steady-state illumination state of the previous cycle. However, if the timeout exit is not triggered and the circuit meets the steady-state conditions, the conversion module 102 performs analog-to-digital conversion on the analog voltage signals across the LED string 200, acquires and records the voltage value at this time, and defines it as the steady-state reference voltage drop. From the perspective of the internal physical principles of the device, this steady-state reference voltage drop... The semiconductor positive physical voltage drop exhibited by the light-emitting device 201 under a specific small diagnostic current was characterized. Due to the thermal vibration effect of the semiconductor lattice, this voltage drop parameter shows a negative correlation with changes in ambient temperature, thus serving as core data support for subsequent thermal characteristic decoupling.

[0043] S130, the main control module 101 controls the drive module 103 to cut off the diagnostic current, establish a secondary blanking time window, and after the secondary blanking time window ends, controls the conversion module 102 to collect the transient voltage decay sequence at both ends of the lamp string 200.

[0044] At the moment the diagnostic current is cut off, the residual inductive energy stored in the conductor 202 will cause secondary voltage oscillation. The main control module 101 uses a secondary blanking time window to span this oscillation physical interval. The time span of this secondary blanking time window is generally set to 1 to 3 microseconds to meet the anti-interference requirements. After the secondary blanking time window ends, the external power supply circuit is effectively shut off. At this time, a microscopic charge transfer process occurs inside the system. The charge stored in the parasitic junction capacitance inside the light-emitting device 201 forms an internal closed-loop discharge path through its dynamic internal resistance. The conversion module 102 continuously digitally samples the analog voltage signal across the two ends of the lamp string 200 at a fixed time interval frequency to obtain a transient voltage decay sequence containing multiple discrete voltage data points. The main control module 101 records the first valid sampled voltage after the secondary blanking time window ends, or extrapolates the initial decay voltage based on multiple valid sampled points, and defines the first valid sampled voltage or the extrapolated initial voltage as the initial decay voltage. Subsequent attenuation fitting uses the initial attenuation voltage V0 as the initial amplitude and the corresponding effective sampling start point as the time reference. When using the first valid sampled voltage after the secondary blanking time window ends, the decay time variable The first valid sampling time is taken as the zero point; when When the decay time variable is obtained by back extrapolation, The moment when the diagnostic current is cut off is taken as the zero point, and the time offset corresponding to the secondary blanking time window is used as the fitting correction. In order to ensure that a sufficient number of effective data points are captured within the short discharge cycle to support the algorithm fitting, the sampling frequency of the conversion module 102 is preferably set in the range of 1Msps to 5Msps.

[0045] The internal closed-loop discharge process follows an exponential decay law. Its theoretical decay voltage function over time is expressed as: ; in, The zero point according to The method of acquisition is determined; when When it is the first valid sampling voltage, =0 corresponds to the first valid sampling time; when When it is the reverse extrapolated voltage, =0 corresponds to the moment when the diagnostic current is cut off. The moment after the diagnostic current is cut off The transient voltage value across the light-emitting device 201; The steady-state reference voltage drop value obtained in step S120; It is a natural constant; This is the elapsed time parameter calculated from the start of the diagnostic current cutoff operation; This is the time constant parameter for the internal discharge physical model. In terms of physical dimensions, it is equivalent to the product of the current instantaneous dynamic internal resistance of the light-emitting device 201 and the parasitic junction capacitance.

[0046] The discretized transient voltage decay sequence obtained by the conversion module 102 is transmitted to the main control module 101 and stored in the storage module 104 to provide basic digital model support for subsequent multi-physics parameter decoupling operations.

[0047] See attached document Figure 4 , Figure 4 This is a flowchart of the parametric decoupling logic according to an embodiment of the present invention. In this embodiment, step S200, which establishes a decoupling model based on the acquired steady-state reference voltage drop and transient voltage decay sequence to separate the ambient temperature equivalent and stress deviation, specifically includes the following sub-steps: S210, the main control module 101 calls the preset voltage-temperature conversion model and deduces the current system's ambient temperature equivalent based on the obtained steady-state reference voltage drop.

[0048] The forward voltage drop of the internal semiconductor structure of the light-emitting device 201 exhibits a negative correlation with the change in ambient temperature. Based on this negative temperature coefficient physical characteristic, the main control module 101 reads the room temperature calibration parameters stored in the storage module 104 and substitutes them into the decoupling model to extract the ambient temperature equivalent in the pure thermal dimension. When the diagnostic current amplitude is lower than the visual emission threshold and is in the small signal detection range, the residual sensitivity of the steady-state reference voltage drop to mechanical stress is written into the storage module 104 through pre-calibration. The main control module 101 first obtains the initial value of the ambient temperature based on the steady-state reference voltage drop. When residual cross-correction is used, the system executes closed-loop iterative calculation from step S210 to step S230, using the stress deviation calculated in the current round to correct the initial value of the ambient temperature in the next iteration round until the stress deviation converges within the preset error range; or directly calls the multi-parameter mapping matrix containing voltage, temperature, and stress cross terms to perform joint lookup output. The corresponding analytical equation is specifically expressed as follows: ; in, The initial ambient temperature value is the one without residual stress correction. The forward voltage drop temperature coefficient of the light-emitting device 201 under diagnostic current is given, with dimensions of voltage / temperature, and a preferred value range of -1.5mV / ℃ to -3.0mV / ℃. When the main control module 101 directly calculates the temperature difference based on the voltage difference, the storage module 104 stores the reciprocal conversion coefficient. Its dimensions are temperature / voltage. After obtaining the equivalent stress deviation in step S230, the main control module 101 further corrects the initial value of the ambient temperature according to the preset residual cross correction coefficient, or calls the function containing... , and The multi-parameter mapping matrix outputs the final ambient temperature equivalent.

[0049] The specific values ​​are determined by the factory test of the light-emitting device 201 and written to the storage module 104 during the system initialization phase. To prevent the obtained ambient temperature equivalent from exceeding the data structure addressing range of subsequent algorithms under extreme operating conditions, the main control module 101 is configured with upper and lower limit value clamping logic for the calculated ambient temperature equivalent to prevent the system from falling into the operation overflow dead zone.

[0050] S220, the main control module 101 performs data feature processing on the transient voltage decay sequence, calculates and extracts the measured decay time constant.

[0051] In actual sampling environments, the transient voltage decay sequence acquired by the conversion module 102 is usually accompanied by high-frequency electromagnetic interference noise. Before performing feature fitting, the main control module 101 performs a moving average filter on the discrete digital sequence to smooth data fluctuations. For the specific implementation of the moving average filter for the data sequence, those skilled in the art can use a conventional window shifting method to perform smoothing calculations on the array. The underlying data filtering algorithm logic is a well-known technology in the field and will not be described in detail here.

[0052] After filtering, the main control module 101 extracts features from the data sequence based on the physical model of exponential decay of resistance and capacitance. According to the physical mechanism of capacitor discharge, the time constant characterizes the time span required for the voltage to decay to approximately 36.8% of its initial amplitude. The main control module 101 calculates the target voltage parameter when the initial amplitude of the decay decreases to a specific proportion. This specific proportion corresponds to the reciprocal of the natural constant, assuming the system has a sampling zero bias voltage or a residual bias voltage. In this case, the target voltage parameter is determined as: the initial voltage attenuation and The difference is 0.37 times, plus... That is, when step S130 uses an extrapolated initial voltage, the target voltage parameter is equivalent to (the extrapolated initial voltage - ... 0.37 times + When the secondary blanking time is much smaller than the estimated decay time constant and the resulting error is lower than the preset tolerance, the aforementioned initial decay voltage can be approximated by substituting the steady-state reference voltage drop value into the calculation. The main control module 101 traverses the filtered transient voltage decay sequence, searches for the discrete time node corresponding to the target voltage parameter, and uses the initial decay voltage obtained by reverse extrapolation. When the diagnostic current is cut off, the time span from the start of the operation to the target voltage node is defined as the measured decay time constant. When the first valid sampled voltage after the secondary blanking time window ends is used as... At that time, the time span from the effective sampling start point to the target voltage node is defined as the measured decay time constant. The time offset caused by the secondary blanking time window is treated as a fixed correction amount. Preferably, the main control module 101 can also deduct the sampling zero-bias voltage correction amount. Preferably, the main control module 101 can also deduct the sampling zero-bias voltage correction amount. Subsequently, logarithmic linear fitting was performed on multiple valid sampling points to reduce the impact of single-point noise on the measured decay time constant. The corresponding fitting relationship is as follows: ; in, This refers to the zero-bias voltage or the residual bias voltage at the end of the attenuation phase of the conversion module 102. The main control module 101 calculates the goodness of fit. When the goodness of fit is lower than a preset threshold or the number of effective sampling points is less than a preset number, the transient attenuation sequence is determined to be invalid, and stress decoupling based on this sequence is abandoned.

[0053] Due to the discrete nature of digital sampling, there is usually no absolute matching point in the sequence that is exactly equal to the target voltage parameter. To avoid the control algorithm getting stuck in the query dead zone, the main control module 101 extracts two adjacent valid sampling points on both sides of the target voltage parameter, and uses linear interpolation to establish a local straight line equation, thereby calculating the accurate measured attenuation time constant.

[0054] S230, the main control module 101 obtains the theoretical reference time constant based on the ambient temperature equivalent, and separates the stress deviation dominated by mechanical deformation through multi-physics field parameter comparison.

[0055] The dynamic internal resistance and parasitic junction capacitance of the light-emitting device 201 are affected not only by its own junction temperature but also by the mechanical compressive stress generated by the physical contraction of the encapsulation layer 203 in a low-temperature environment. This is because there is a significant difference in the coefficient of thermal expansion between the epoxy resin and the internal semiconductor substrate, and the asynchronous contraction under extremely cold conditions will generate huge mechanical compression on the semiconductor lattice. This compressive stress, through the piezoresistive effect of the semiconductor, forces a physical distortion in the width of the internal depletion layer, thereby causing a drift in the electrical time constant of the system. In order to accurately isolate the physical effects of the purely thermal dimension and quantify the stress state of the device separately, a thermal transfer matrix is ​​pre-configured in the storage module 104.

[0056] This thermal transfer matrix records the mapping relationship between discrete ambient temperature nodes and their corresponding theoretical reference time constants under standard reference conditions without external mechanical compressive stress. To obtain a pure reference thermal response, the data for this mapping matrix comes from characteristic spectra obtained by performing multi-point temperature oscillation tests on bare light-emitting devices without the encapsulation layer 203 in a variable-temperature constant-temperature chamber, or from calibration samples from the same batch with the same electrical connection structure and under conditions without external compressive stress or pre-released encapsulation stress. If a bare light-emitting device is used for calibration, the storage module 104 also writes a zero-bias correction amount introduced by the encapsulation structure to compensate for the systematic differences between the bare device and the actual encapsulated LED string.

[0057] The main control module 101 uses the ambient temperature equivalent calculated in step S210 as the index addressing variable to perform data query on the thermal transfer matrix and extract the corresponding theoretical reference time constant. If the value of the ambient temperature equivalent falls between two adjacent preset temperature nodes in the matrix, the main control module 101 performs linear interpolation to obtain the accurate theoretical reference time constant. If the ambient temperature equivalent touches the index boundary of the thermal transfer matrix, the boundary extreme value is used as the output limit to prevent memory pointer out-of-bounds errors during algorithm execution.

[0058] After completing the table lookup and comparison, the main control module 101 performs a difference stripping calculation between the obtained measured attenuation time constant and the theoretical reference time constant. The corresponding mathematical model is as follows: ; in, To calculate the equivalent stress deviation, the main control module 101 further uses the compressive stress direction calibration coefficient pre-stored in the storage module 104. Extract the effective component representing the increase in encapsulation compressive stress: ; when When it is positive, it indicates that the current transient decay time constant has shifted relative to the unstressed reference state in accordance with the increase in package stress; when When the value is non-positive, the main control module 101 will not use it as the basis for triggering the hot injection release.

[0059] This stress deviation eliminates natural parameter drift caused by ambient temperature fluctuations. Its magnitude effectively characterizes the degree of physical distortion caused by the piezoresistive effect of the current encapsulation layer 203 volume shrinkage on the internal light-emitting device 201. This parameter provides an independent physical reference for the subsequent control system 100 to determine whether to trigger active stress release intervention.

[0060] See attached document Figure 5 , Figure 5 This is a schematic diagram of the trapezoidal heat injection waveform calculation logic according to an embodiment of the present invention. In this embodiment, when the stress deviation exceeds the threshold, the step S300 of calculating the trapezoidal heat injection waveform parameters for actively releasing compressive stress, in conjunction with the parasitic inductance parameter of the conductor 202, specifically includes the following sub-steps: S310, the main control module 101 performs conditional branch judgment on the stress deviation of the decoupled output, and maps it into the target transient Joule thermal energy equivalent when the active intervention condition is triggered.

[0061] The main control module 101 extracts the stress deviation calculated in step S200 and compares it with the preset safety stress threshold in the storage module 104. This safety stress threshold represents the physical boundary of the maximum deformation tolerance of the encapsulation layer 203 without causing breakage of the gold wires inside the light-emitting device 201 or irreversible damage to the crystal lattice. Its specific value is usually pre-calibrated based on the accelerated aging and thermal shock reliability test data of the lamp string 200. As a preferred embodiment, the safety stress threshold can be set to an offset range of 5% to 10% of the corresponding system reference time constant.

[0062] If the current stress deviation is determined not to exceed the safe stress threshold, the main control module 101 identifies that the system is in a safe operating range. At this time, the system does not output additional thermal compensation waveforms, but maintains the normal steady-state light emission driving logic. If the current stress deviation is determined to exceed the safe stress threshold, the system triggers active physical intervention logic. Before determining the threshold, the main control module 101 determines the effective component representing the increase in encapsulation compressive stress in the stress deviation based on the compressive stress direction calibration coefficient pre-stored in the storage module 104. When the effective component does not exceed the safe stress threshold, thermal injection intervention is not triggered. When the effective component exceeds the safe stress threshold, only the part exceeding the safe stress threshold is mapped to the target transient Joule thermal energy equivalent.

[0063] To counteract the mechanical compression caused by the shrinkage of the encapsulation layer 203, the control system 100 induces localized thermal softening and micro-expansion of the epoxy resin material within a short time through a quantitative micro-Joule heat injection. The main control module 101 positively maps the transient Joule thermal energy equivalent target value based on the stress deviation, and the corresponding calculation equation is as follows: ; ; in, This refers to the effective intervention deviation amount after exceeding the safety stress threshold; This represents the effective component of the increase in compressive stress during encapsulation; The safety stress threshold; To calculate the target transient Joule thermal energy equivalent; This is the preset energy-equivalent stress deviation conversion factor. Only when... At that time, the main control module 101 triggers the calculation of the trapezoidal heat injection waveform.

[0064] The conversion factor is pre-determined in the laboratory based on the isobaric specific heat capacity and thermal expansion coefficient of the encapsulation layer 203 material and is solidified in the storage module 104. Its physical quantity is generally in the range of microjoules / microseconds to millijoules / microseconds. To avoid energy injection overflow due to data drift, the main control module 101 configures a hardware safety upper limit for the calculated target transient joule thermal energy equivalent. When the calculation result exceeds this upper limit, a forced amplitude throttling output is executed.

[0065] S320, the main control module 101 retrieves the intrinsic parasitic parameters of the conductor 202, calculates the climbing constraint time of the leading region in combination with the first-order dynamic response model, and establishes a feedforward compensation constraint to suppress inductive distortion.

[0066] In a conventional long-distance copper wire lamp system, the inherent equivalent resistance of the conductor 202 and the parasitic inductance of the line constitute a low-pass filter network. If the drive module 103 directly outputs a high-frequency or steeply rising square wave pulse, the parasitic inductance of the line will impede the current change, resulting in waveform distortion and generating overvoltage spikes at both ends of the device that can easily damage the semiconductor junction.

[0067] To overcome waveform distortion and impact risks caused by the distributed parameters of long conductors, the main control module 101 performs feedforward shaping on the heat injection waveform, constructing it into a trapezoidal wave structure with a gently rising slope. The main control module 101 reads the equivalent resistance parameters of the conductor 202 pre-stored in the storage module 104. Parasitic inductance parameters of the line .in, The equivalent series resistance in the hot-injection current channel includes at least the equivalent resistance of the wire 202, the equivalent resistance of the output channel of the drive module 103, the sampling resistor, and the dynamic resistance component of the light-emitting device 201 participating in the current ramp-up process in equivalent series within the injection current range. For the above two intrinsic physical parameters of the circuit, researchers can perform offline measurements on standard samples of the same specification of lamp strings using a high-frequency LCR bridge on the production line, and then burn the average measurement value as a fixed constant into the storage module 104. Based on this, the main control module 101 derives the inductance time constant of the system circuit. The equivalent relationship is as follows: ; in, This refers to the parasitic inductance of the circuit in the heat-injected current channel. This is the equivalent series resistance in the heat injection current ramp-up path.

[0068] Based on the zero-state response characteristics of a first-order inductor-resistor circuit, the current reaches approximately 95% of the steady-state target value after about three times the time constant. Therefore, the main control module 101 limits the time parameters of the trapezoidal waveform leader establishment region to meet the safety ramp-up constraint conditions, expressed by the constraint inequality: ; in, Establish the leading time zone. As a preferred method, the main control module 101 can directly use the time zone when performing control calculations. Assign a value of 3.5 to 4.0 times. This ensures a smooth transition of the current ramp. Simultaneously, the main control module 101 extracts the set target peak current from the storage module 104. The target peak current is set as a safety constant that does not exceed the fuse limit of the internal leads of the light-emitting device 201. The conventional value is between 3 and 5 times the rated operating current of the light-emitting device 201, so as to provide concentrated transient heat flow in a short time. The target peak current is also limited by the pulse safe operating area, the maximum allowable junction temperature, the maximum single-cycle injection energy specified in the datasheet of the light-emitting device 201, and the overcurrent protection threshold of the drive module 103. When any of the above limiting conditions is inconsistent with the empirical range of 3 to 5 times, the main control module 101 uses the lower safety limit value as the target peak current.

[0069] S330, the main control module 101, based on the set target peak current and the leading ramp waveform function, uses the energy equivalent equation to integrate and calculate the constant energy injection region maintenance time required to satisfy the target transient Joule heating.

[0070] The complete energy release process of the trapezoidal heat injection waveform consists of a leading ramp-up phase and a constant current sustaining phase. During the controlled descent phase after the heat injection phase, if the current is still higher than the steady-state compensation current and the duration is not negligible, the main control module 101 will also include the current integral energy corresponding to this descent phase as the descent transition energy in the heat injection energy statistics for this cycle. When the descent phase is quickly clamped to the steady-state compensation current and the duration is lower than a preset neglect threshold, this descent transition energy can be approximately ignored. Correspondingly, when the descent transition energy is included in the heat injection energy statistics, the main control module 101 calculates the constant energy injection region sustaining time. At that time, from the target transient Joule thermal energy equivalent The energy of the leading climb phase and the estimated energy of the descent transition phase are deducted; when the descent transition energy is ignored... Solving for the target's transient Joule thermal energy equivalent solely based on the energy during the leading climb phase and the constant current maintenance phase requires work to be done on the dynamic internal resistance of the light-emitting device 201 by the injected current. Therefore, the main control module 101 further constructs an energy equivalent integral equation to analyze the unknown time parameters. The specific expansion of its energy closed-loop integral equation is as follows: ; in, The equivalent heating dynamic resistance of the light-emitting device used for calculating heat injection energy; This represents the estimated heat injection energy corresponding to the controlled descent phase. When the descent phase is included in the heat injection phase... The current waveform during the descent phase is obtained by integration; when the descent phase is ignored or included in the steady-state light emission transition phase... , The target transient Joule thermal energy equivalent determined in step S310; The leader establishment time determined in step S320; The duration of the trapezoidal wave in the constant-energy injection region to be solved; The transient current function that rises with time during the leader establishment period; The target peak current constant is set. The dynamic internal resistance parameter of the light-emitting device 201; This is the integration time variable. The dynamic internal resistance parameter of the aforementioned light-emitting device 201. The equivalent slope resistance value within the injection current range can be selected based on the VI current-voltage characteristic curve of the device at the factory, and it can be pre-stored in the storage module 104 for direct retrieval.

[0071] From an engineering implementation perspective, to avoid excessive computational power consumption during real-time control by the main control module 101 performing calculus operations, the main control module 101 controls the drive module 103 to output a linear ramp current in the lead-in establishment region. Therefore, the transient current function can be equivalently replaced by... Under these approximate conditions, when the main control module 101 controls the drive module 103 to output a linear ramp current in the leader establishment region, the energy in the leader establishment region is: ; The duration of the constant energy injection zone is calculated using the following formula: ; when At that time, the main control module 101 will The actual injected energy is set to zero, and the effective peak current is reduced, the effective ramp-up endpoint is shortened, or the residual control quantity is transferred to subsequent cycles to ensure that the actual injected energy does not exceed the target transient Joule thermal energy equivalent.

[0072] After completing the maintenance time calculation, the main control module 101 verifies the total available emission time slots for the current cycle against the total duration of the trapezoidal wave. The total duration of the trapezoidal wave includes at least the preamble setup time. Duration of the constant energy injection zone In addition to the controlled fall time included in the heat injection stage; when further reconstructing the composite drive waveform, the main control module 101 will also include the actual steady-state emission time. The total duration of a single cycle is included in the verification. When the verification feedback shows that the calculated total duration of the trapezoidal wave exceeds the maximum safe injection pulse width limit allowed in a single cycle, the main control module 101... The execution cycle truncation operation postpones the allocation of any remaining unreleased energy share to the intervention scheduling task in the next luminescence cycle. When included... When the total duration of the composite drive waveform exceeds the available light emission time slot in a single cycle, the main control module 101 prioritizes compressing or rolling the heat injection task instead of extending the preset light emission cycle.

[0073] This fault-tolerance and task splitting mechanism completely eliminates the system risk of burning out devices due to excessive heat concentration caused by extremely long pulses.

[0074] See attached document Figure 6 , Figure 6 This is a timing reconstruction diagram of a composite driving waveform according to an embodiment of the present invention. In this embodiment, the step S400 of calculating steady-state emission compensation parameters based on the ambient temperature equivalent and equivalent charge, and reconstructing and outputting the composite driving waveform containing the timing gradient, specifically includes the following sub-steps: S410, the main control module 101 adaptively presets steady-state emission reference driving parameters to maintain the reference luminous flux based on the ambient temperature equivalent obtained by decoupling.

[0075] The photoelectric conversion efficiency of the light-emitting device 201 fluctuates with changes in ambient temperature. At low temperatures, the forward voltage drop, carrier recombination efficiency, and packaging stress state of the light-emitting device 201 may deviate from the reference calibration state. At temperatures higher than the reference temperature, the increased junction temperature may also lead to an increase in non-radiative recombination probability and trigger thermal quenching. Therefore, the main control module 101 uses a luminous efficacy compensation lookup table to bidirectionally correct the steady-state compensation current amplitude and reference conduction time under different temperature equivalents to maintain the reference luminous flux. To maintain macroscopic visual brightness consistency, the control system 100 needs to eliminate light emission attenuation in the purely thermal dimension. The main control module 101 extracts the decoupled ambient temperature equivalent from step S200 and uses it as an input index to query the luminous efficacy compensation lookup table pre-stored in the storage module 104. Through table lookup and comparison, the main control module 101 obtains the steady-state compensation current amplitude and reference conduction time that match the current ambient temperature equivalent. As a preferred approach, the amplitude of the steady-state compensation current is typically configured to fluctuate within the range of 80% to 120% of the rated operating current of the light-emitting device 201. For the calibration of the data in the luminous efficacy compensation comparison table, those skilled in the art can use a standard photometer with an integrating sphere to perform offline luminous flux measurements of the light-emitting device at different ambient temperatures to establish a data mapping relationship. The specific calibration method is well-known in the art and will not be elaborated upon here.

[0076] After determining the steady-state reference driving parameters, the system faces the additional light output problem caused by the actively introduced thermal injection waveform. To avoid visually perceptible flicker or brightness abrupt changes due to transient thermal intervention, the system performs an equivalent loss deduction on the emitted energy in the time domain.

[0077] S420, the main control module 101 integrates the charge amount of the trapezoidal heat injection waveform, calculates the equivalent charge amount and calculates the steady-state time loss amount to correct the actual steady-state luminescence time.

[0078] While the trapezoidal heat injection waveform injects Joule heat into the system to release compressive stress, the actual current flowing through it inevitably excites the light-emitting device 201 to generate instantaneous additional light radiation. Based on the physical characteristic that the total number of photons emitted and the total number of injected electrons are approximately proportional on a macroscopic time scale in the microscopic mechanism of light emission, the main control module 101 establishes a charge equivalent conversion model. Considering the possible difference in luminous efficiency between the heat injection current range and the steady-state luminous current range, the storage module 104 also pre-stores the luminous flux equivalent conversion coefficient corresponding to the heat injection waveform. The main control module 101 converts the actual injected charge into a luminous charge equivalent to the steady-state luminous current based on this luminous flux equivalent conversion coefficient. The main control module 101 performs current-time integration calculations on the leading region and constant-energy injection region of the trapezoidal heat injection waveform to extract the total charge consumed in the transient heat injection process. This total charge... The algebraic formula for calculation is expressed as: ; in, The actual injected charge amount generated by the trapezoidal heat injection waveform; The target peak current constant is set. Establish the leading area time; Duration of the constant energy injection zone; This refers to the additional charge generated when the controlled descent phase is included in the thermal injection phase. This applies when the descent phase is ignored or included in the steady-state luminescence transition phase. .

[0079] Considering that the luminous efficiency in the hot injection current range and the steady-state luminous emission current range are not exactly the same, the main control module 101 further calls the luminous flux equivalent conversion factor pre-stored in the storage module 104. The actual injected charge is converted into the equivalent steady-state luminescence charge. The corresponding relationship is as follows: ; in, The steady-state luminescent equivalent charge corresponding to the trapezoidal heat injection waveform; This is the equivalent conversion factor of luminous flux between the thermal injection current range and the steady-state luminous current range.

[0080] Subsequently, the main control module 101 divides the calculated total charge by the steady-state compensation current amplitude to determine the steady-state time loss that needs to be deducted within this cycle, thus establishing the actual steady-state emission time after compensation correction. The corresponding correction calculation model is as follows: ; in, This refers to the calculated actual steady-state luminescence time; The reference conduction duration obtained in step S410; This represents the total charge generated by the trapezoidal wave. The steady-state compensation current amplitude is determined in step S410.

[0081] When calculated When less than or equal to zero, the main control module 101 will The time unit is limited to the minimum hardware resolution, and the equivalent charge that cannot be deducted within the current cycle is recorded as the cross-cycle deduction margin, which is then deducted in one or more subsequent light-emitting cycles according to the preset smooth deduction rules.

[0082] Considering that under extreme stress conditions, the required stress deviation is large, and the prolonged heat injection waveform time may cause the calculated steady-state time loss to exceed the reference conduction time, the main control module 101 is configured with underflow protection logic to maintain the closed-loop algorithm logic and prevent errors in the underlying counter. When the system determines... When the calculation result is less than or equal to zero, the main control module 101 forces the actual steady-state emission time to be configured to the minimum hardware resolution time unit (e.g., 1 microsecond), and rolls over the overflow charge loss quota to subsequent emission cycles for smooth deduction.

[0083] After completing time loss correction and parameter reconstruction, the main control module 101 performs time-level hardware splicing of discrete waveform segments.

[0084] S430, the main control module 101 reconstructs the composite drive waveform containing the timing gradient and controls the drive module 103 to output drive signals sequentially within a single cycle.

[0085] The main control module 101 utilizes internal timer resources to map the independent time parameters of each stage to the underlying hardware control timing sequence. Within a complete light-emitting conduction cycle, the main control module 101 sequentially sends a sequence of instructions to the driver module 103. The driver module 103 receives external instruction signals and continuously executes three modes of driving actions on the light string 200.

[0086] As a preferred approach, the timing of this waveform reconstruction is specifically manifested as follows: starting from the beginning of the self-luminous cycle, the driving module 103 outputs a leading ramp-up current with a slope limit to the set target peak current; after the leading establishment period, the driving module 103 maintains the target peak current output for a duration equal to the constant energy injection period maintenance time; after the hot injection stage ends, the driving module 103 actively controls the current amplitude to decrease. Considering the physical constraint of the parasitic inductance freewheeling current in the long conductor, the current cannot abruptly change to the target value. The main control module 101 allows the current to fall back naturally based on the system inductance time constant until it is clamped and maintained at the steady-state compensation current amplitude, and the duration of this state maintenance is equal to the calculated actual steady-state luminous time. If the natural fallback phase is defined as the heat injection fallback phase, its corresponding energy and charge are included in the statistical results of steps S330 and S420; if the natural fallback phase is defined as the steady-state light emission transition phase, the main control module 101 deducts the time occupied by the natural fallback phase from the actual steady-state light emission time to ensure that the total light emission energy in a single cycle does not exceed the preset target. The above two processing methods are determined by the fallback phase attribution flag in the storage module 104, and only one processing method is allowed to be selected within the same control cycle to avoid repeated counting of the fallback phase energy and charge. When the fallback phase belongs to the heat injection phase, its corresponding... and The steps S330 and S420 are respectively included; when the descent phase belongs to the steady-state luminescence transition phase, the step S330 is included. and in step S420 All are zero, and from The time taken during the descent phase is deducted.

[0087] After completing the continuous output of the above single-cycle three-mode composite waveform, the drive module 103 cuts off the output current. The system then enters the next drive shutdown time slot, waiting and preparing to start the physical diagnosis or routine light emission scheduling task of the subsequent cycle.

[0088] See attached document Figure 7 , Figure 7 This is a flowchart of fault diagnosis, fault-tolerant processing, and cross-cycle resampling control according to an embodiment of the present invention. In this embodiment, after obtaining diagnostic data in each round of step S100, the control system 100 first performs open-circuit and short-circuit hard fault verification based on the steady-state reference voltage drop; only after the hard fault verification passes, does it continue to execute the stress decoupling, heat injection, and light emission compensation control in steps S200 to S400; after completing the single-cycle composite drive output, it further performs degradation trend diagnosis based on the diagnostic data and control parameters, and initiates a heat conduction hysteresis timing mechanism to determine the next round of resampling node: S510, the main control module 101 performs boundary condition verification on the steady-state reference voltage drop collected in step S100 to diagnose whether there is an electrical open circuit fault in the light string 200.

[0089] If a physical break occurs in the light string 200, such as a burnt-out gold wire inside the light-emitting device 201 or a disconnected wire 202, the diagnostic current loop will be interrupted. Under this physical condition, the voltage across the light string 200 acquired by the conversion module 102 will no longer be the normal forward voltage drop of the semiconductor device, but will be pulled up to a higher voltage level close to the system power supply rail. Accordingly, the main control module 101 will acquire the steady-state reference voltage drop... Compare with the preset open-circuit decision voltage threshold. If If the voltage exceeds the open-circuit decision threshold, an open-circuit fault is determined to have occurred in the system. As a preferred embodiment, the open-circuit decision voltage threshold can be set to 90% of the system supply voltage. The main control module 101 can also combine the constant current establishment failure flag or the output compliance voltage saturation flag of the drive module 103 for joint determination; when... When the voltage exceeds the open circuit judgment threshold and the diagnostic current fails to reach the target value within the preset establishment time, the main control module 101 confirms the open circuit fault, thereby avoiding misjudgment caused solely by transient interference.

[0090] Once an open-circuit fault is detected, the main control module 101 will immediately interrupt all subsequent stress decoupling and light-emitting drive calculations, and control the drive module 103 to enter a safe, non-driven state. This open-circuit fault detection is executed first after each round of step S100, and its execution priority is higher than that of the ambient temperature analysis, stress deviation calculation, heat injection waveform calculation, and composite drive waveform output in steps S200 to S400. At the same time, the main control module 101 sets the fault flag register inside the storage module 104 and reports the fault status code to the external management system through a reserved communication interface (such as I2C or SPI bus) for manual maintenance. This locked protection state will be maintained until the system undergoes a complete power reset.

[0091] S520, the main control module 101 performs a lower limit condition check on the steady-state reference voltage drop to diagnose whether there is an electrical short circuit fault in the light string 200.

[0092] In contrast to an open-circuit fault, if one or more light-emitting devices 201 in the lamp string 200 experience a PN junction breakdown short circuit, the total voltage drop across the lamp string 200 will be significantly lower than normal when the same diagnostic current is injected. The main control module 101 will collect the steady-state reference voltage drop. Compare with the preset short-circuit decision voltage threshold. If If the voltage drops below the short-circuit decision voltage threshold, a short-circuit fault is determined in the system. The short-circuit decision voltage threshold is adjusted based on the ambient temperature equivalent, the number of series-connected light-emitting devices, and the historical normal reference voltage drop. When the system has not yet obtained a valid ambient temperature equivalent, the main control module 101 uses the lowest safe voltage drop boundary written during the production calibration phase as a temporary threshold. To avoid false locking caused by single-sample noise, the short-circuit determination can also be configured with a number of consecutive confirmations or a voltage hysteresis interval. As a preferred embodiment, when the light string 200 includes multiple series-connected light-emitting devices 201, the short-circuit decision voltage threshold is determined jointly based on the total reference voltage drop of the normal light string under diagnostic current, the number of series-connected light-emitting devices, and the forward conduction voltage drop of a single light-emitting device under diagnostic current. For overall short circuit or multiple short circuit faults, the short circuit judgment voltage threshold can be set as a preset ratio of the lower limit of the normal total reference voltage drop; for a single light-emitting device short circuit fault, the main control module 101 judges by the amount of decrease in the current steady-state reference voltage drop relative to the historical normal reference voltage drop, or judges by the corresponding segment voltage drop anomaly when the lamp string 200 is configured with segmented sampling nodes.

[0093] For hard faults such as short circuits that can also endanger system safety, the fault tolerance mechanism is similar to that for open circuit faults. The main control module 101 also performs lockout protection actions such as drive shutdown, setting the fault flag, and reporting the fault status to prevent the risk of overcurrent caused by the continuous output of the drive module 103 under short circuit conditions and to avoid the fault range from expanding.

[0094] S530, the main control module 101 establishes a long-period observation window to perform trend analysis on key control parameters in order to diagnose whether there is a slow-degradation fault in the system.

[0095] In addition to the aforementioned open-circuit and short-circuit faults, the system may also experience gradual performance degradation due to material aging or prolonged exposure to extreme environments. To effectively detect this type of gradual degradation fault mode, the main control module 101 is equipped with a continuous anomaly counter. During each diagnostic cycle, the main control module 101 calculates the stress deviation obtained in step S230. A high-level threshold comparison is performed. If the stress deviation exceeds the preset degradation alarm threshold for multiple consecutive cycles, the internal continuous anomaly counter is incremented. When the accumulated counter value reaches the preset number of fault determinations, the system determines that a gradual degradation fault has occurred. If the stress deviation falls below the degradation alarm threshold in any diagnostic cycle, the main control module 101 clears the continuous anomaly counter to zero, or decrements it according to a preset decrement step size. To avoid misjudgment caused by jitter near the threshold, the degradation alarm threshold can also be configured with a release hysteresis threshold.

[0096] In a preferred embodiment, the degradation alarm threshold can be set to 120% of the safety stress threshold in step S310, and the number of fault determinations can be set to 10 to 50 consecutive diagnostic cycles according to reliability requirements.

[0097] Upon detecting a gradual degradation fault, the system's fault-tolerant handling mechanism differs from the immediate shutdown mechanism for hard faults. The main control module 101 does not immediately execute a shutdown lockout; instead, it prioritizes entering a derating operation mode. In this mode, the main control module 101 forcibly reduces the overall steady-state compensation current amplitude in the luminous efficacy compensation table by a fixed percentage (e.g., 20%) to reduce the overall electrical and thermal load of the system and extend its remaining service life. Simultaneously, the system reports a warning maintenance status code via the communication interface, alerting administrators that the LED string unit is in a sub-healthy state and requires preventative replacement. This derating operation mode will continue as the current operating state until the system power is reset and a health assessment is performed again.

[0098] S540, after completing the single-cycle composite drive waveform output, the main control module 101 starts a thermal conduction hysteresis timer to lock the earliest trigger time for the next round of diagnostic sampling. The duration of the thermal conduction hysteresis timer is determined based on the thermal diffusion time constant of the encapsulation layer 203, the equivalent thermal resistance from the light-emitting device 201 to the outer surface of the encapsulation layer 203, and the thermal injection energy of the previous cycle. As one possible implementation, the main control module 101 uses the actual injected energy of the previous cycle... Estimate the local temperature rise and according to the thermal diffusion time constant of the encapsulation layer 203. Set the resampling wait time: ; in, To preset the thermal stability coefficient, a value of 3 to 5 is preferred; when the actual energy injected in the previous cycle... When the energy level exceeds the preset energy threshold, the main control module 101 increases... Or extend according to the energy classification table .

[0099] If the thermal conduction hysteresis timer has not yet reached its full count, the main control module 101 prohibits the execution of the next round of transient decay sampling and only allows the maintenance of normal steady-state luminescence scheduling. If the thermal conduction hysteresis timer reaches its full count, the main control module 101 allows re-entry into step S100 to avoid repeated sampling before the local temperature transient caused by the previous heat injection has fully diffused, which could lead to misjudgment of the ambient temperature equivalent and stress deviation. During the period when the thermal conduction hysteresis timer has not reached its full count, the main control module 101 does not update. , and Instead, it uses the ambient temperature equivalent and steady-state luminescence compensation parameters obtained from the most recent effective diagnostic cycle, or performs steady-state luminescence according to the preset safety derating parameters; during this period, no new heat injection tasks are triggered, and only safety inspections are performed on the hard fault monitoring signal and the drive current feedback signal.

[0100] Specific application examples are described below: Application scenario settings: An outdoor landscape lighting project uses a 10-meter-long copper wire light string (200) containing 100 series-connected blue LED bare chips (encapsulated with epoxy resin adhesive layer 203). The system operates in an extremely cold environment of -30℃, with a standard operating frequency of 1kHz (1000μs per cycle) and a rated steady-state luminous current. .

[0101] System operation steps analysis: S100 (Diagnostic Sampling Phase): During the turn-off time slot at the beginning of a single cycle, the main control module 101 injects... The tiny diagnostic current. The steady-state reference voltage drop is measured by the conversion module 102. (Average voltage per chip: 2.65V). After disconnecting the diagnostic current, the transient voltage decay sequence was obtained at a sampling rate of 2Msps.

[0102] S200 (Multi-parameter decoupling stage): Temperature analysis: The main control module 101 calculates the equivalent temperature of the current ambient temperature based on the calibrated temperature coefficient (-2.0mV / ℃). .

[0103] Stress extraction: The theoretical reference time constant of the light string under stress-free conditions at -30℃ was obtained from the table. The measured decay sequence was fitted, and the measured time constant was extracted. The stress deviation was calculated. (This deviation indicates that the epoxy resin underwent severe shrinkage at -30°C, resulting in a significant mechanical piezoresistive effect on the chip.)

[0104] S300 (Heat Injection Waveform Calculation Stage): The safety stress threshold is set to .because This triggers active stress release.

[0105] The mapping calculation requires the injection of transient Joule thermal energy. Considering the line parasitic inductance (approximately 15 μH), the main control module 101 sets the trapezoidal heat injection waveform: leader climb time... Target peak current (Rapid heating at 3 times rated current), duration of constant energy injection zone .

[0106] S400 (Composite Waveform Reconstruction Stage): Charge Equivalent Subtraction: The equivalent charge of the trapezoidal heat injection waveform is calculated. This short pulse of high current (60mA) will also emit light. After converting it to its corresponding steady-state charge, it is calculated that approximately 60μs of emission time needs to be subtracted from the normal emission cycle.

[0107] After completing the charge equivalence conversion, the main control module 101 will calculate the actual steady-state emission time for this cycle. The correction is made to the difference after subtracting 60μs from the original emission time. In the subsequent drive output phase, the main control module 101 controls the drive module 103 to sequentially execute the following timing actions within a single cycle: first, output a 10μs-time pre-leader ramp-up current to a peak value of 60mA, and maintain it for 15μs to complete the transient hot injection operation; then, control the current amplitude to decrease and clamp it to 20mA, entering a duration of... The system first achieves steady-state light emission; then, it cuts off the output current to enter the drive shutdown time slot. Through the timing scheduling of the above composite waveforms, the system utilizes local Joule thermal transients to soften the epoxy resin and release mechanical compressive stress, while precisely offsetting the additional light radiation introduced by the thermal pulse, thus achieving a smooth brightness transition and flicker-free control in a macroscopic visual perspective.

[0108] Experimental verification and effect comparison To verify the effectiveness of the present invention, thermal shock and extreme cold luminescence maintenance rate experiments were conducted in an environmental test chamber.

[0109] Experimental conditions: Test subject: Three 10-meter / 100-lamp copper wire light strings produced in the same batch.

[0110] Environmental conditions: -40℃ to +25℃ high and low temperature cyclic shock test, and -30℃ constant temperature continuous lighting test for 1000 hours.

[0111] Comparison Groups: Control group A (traditional constant current drive): Only a fixed 20mA constant current PWM drive is used, without temperature and stress compensation.

[0112] Control group B (temperature compensation only): The drive current is adjusted only according to the reference voltage drop (traditional light effect compensation), without mechanical stress decoupling and heat injection functions.

[0113] Experimental Group C (System of this invention): Fully activate the temperature / stress decoupling, trapezoidal heat injection, and charge equivalence correction functions of the above S100-S500.

[0114] Test results explanation (with attached diagram): Appendix Figure 8 To be continued Figure 10 (Generated by MATLAB below): See attached document Figure 8 (Comparison of transient voltage decay and stress decoupling): Diagrammatic Explanation: This diagram corresponds to the physical verification in stage S200 of embodiment. The diagram shows the exponential decay curve of the voltage across the light-emitting device over time after the diagnostic current is cut off. The blue dashed line represents the theoretical decay under the "stress-free standard state," and the red solid line represents the measured decay under the "-30℃ high-pressure stress state."

[0115] Results demonstrate that, clearly, under the piezoresistive effect, changes in the device's junction capacitance and dynamic internal resistance lead to a slower discharge (increased time constant). This invention precisely capitalizes on this effect. Figure 8 The time difference between the two curves ( It achieves purely electrical non-contact detection of mechanical stress.

[0116] See attached document Figure 9 (Single-cycle composite drive waveform timing diagram): Diagrammatic Explanation: This diagram corresponds to the verification of the actual current waveform output by the drive module in the S400. The diagram shows a complete PWM cycle (partial magnification within 1000μs).

[0117] Results demonstrate that the waveforms sequentially exhibit: 1. A leading-edge micro-diagnostic current detection window; 2. A trapezoidal high-current heat injection region with anti-inductance overshoot slope (used for mitigation). Figure 8 3. The steady-state luminescence compensation region after deducting the equivalent charge time. This composite waveform confirms the complete feasibility of the scheme at the hardware level.

[0118] See attached document Figure 10 (Relative luminous flux decay lifetime curve in extremely cold environments (-30℃):) Graphical fusion explanation: This figure shows the macroscopic optical performance of three groups of samples during a 1000-hour extreme cold test.

[0119] Proof of effectiveness: The light flux of control group A (gray line) rapidly decreased to 65% due to stress-induced lattice compression and changes in low-temperature luminous efficacy, and was accompanied by some dead lamps (gold wires were broken due to stress).

[0120] Although the control group B (blue line) compensated for the temperature, it could not eliminate the decrease in internal quantum efficiency caused by physical stress, and ultimately remained at around 85%.

[0121] Experimental group C (red line, this invention) released mechanical stress by real-time softening of the encapsulating resin through microscopic heat injection, maintaining a luminous flux of over 98% and a zero dead lamp rate. This effectively broke through the technical bottleneck of "extreme cold failure" of copper wire lamps and achieved unexpected technical results.

[0122] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions, characterized in that, Includes the following steps: During the turn-off time slot of the single-cycle drive sequence, a diagnostic current is injected into the copper wire lamp, and the transient voltage decay sequence across the light-emitting device in the copper wire lamp is obtained. Based on the transient voltage decay sequence, the current ambient temperature equivalent of the light-emitting device and the mechanical stress deviation caused by the cold shrinkage of the packaging material are decoupled and calculated. When the mechanical stress deviation exceeds the set safety threshold, the corresponding active heat injection waveform parameters are calculated based on the mechanical stress deviation. Based on the active heat injection waveform parameters, charge equivalence conversion is performed, the steady-state emission time of this cycle is corrected for loss, and the composite driving waveform containing the heat injection stage and the steady-state emission stage is reconstructed and output.

2. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 1, characterized in that, The step of injecting diagnostic current into the copper wire lamp and obtaining the transient voltage decay sequence across the light-emitting device in the copper wire lamp includes: At the end of the PWM cycle, the time slot is turned off, and the diagnostic current source is turned on to inject the diagnostic current into the copper wire lamp. After the voltage stabilizes, the steady-state reference voltage drop across the light-emitting device is measured. The diagnostic current is cut off, and the discharge process from the two ends of the light-emitting device to the residual zero bias voltage is continuously collected at a preset sampling rate to obtain the transient voltage decay sequence.

3. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 2, characterized in that, The step of decoupling and calculating the current ambient temperature equivalent of the light-emitting device and the mechanical stress deviation caused by the cold shrinkage of the packaging material based on the transient voltage decay sequence includes: extracting a preset voltage-temperature coefficient; Calculate the voltage difference between the steady-state reference voltage drop and the standard room temperature voltage drop, and divide the voltage difference by the voltage-temperature coefficient to obtain the ambient temperature equivalent.

4. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 3, characterized in that, The process of decoupling the calculation of the mechanical stress deviation includes: Based on the ambient temperature equivalent, a preset reference time constant mapping table is consulted to obtain the theoretical reference time constant of the light-emitting device under the current temperature and stress-free state. The transient voltage decay sequence is fitted with an exponential decay model to extract the measured time constant. The mechanical stress deviation is obtained by calculating the difference between the measured time constant and the theoretical reference time constant. The measured time constant is the time when the voltage value in the transient voltage decay sequence drops to 36.8% of the difference between the steady-state reference voltage drop and the residual zero bias voltage.

5. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 1, characterized in that, The active thermal injection waveform is a trapezoidal current waveform, and the active thermal injection waveform parameters include the leader ramp-up time, the target peak current, and the constant energy injection region maintenance time.

6. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 5, characterized in that, The active heat injection waveform parameters calculated based on the mechanical stress deviation include: Based on the mechanical stress deviation, the desired transient Joule thermal energy target is obtained by mapping. Based on the parasitic inductance parameters of the power supply line, the leading climb time that limits inductor overshoot is calculated; Based on the transient Joule thermal energy target and the thermal capacity characteristics of the light-emitting device, the target peak current and the duration of the constant energy injection region are calculated.

7. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 6, characterized in that, The amplitude of the target peak current is set to be 2 to 4 times the rated steady-state luminous current of the light-emitting device, and less than the transient fusing current threshold of the internal leads of the light-emitting device.

8. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 1, characterized in that, The step of performing charge equivalence conversion based on the active heat injection waveform parameters, correcting the loss in the steady-state emission time of this cycle, and reconstructing the output composite driving waveform including the heat injection stage and the steady-state emission stage includes: The total charge injected during the heat injection stage is calculated by integrating the active heat injection waveform in the time domain. The equivalent steady-state luminous time is calculated by dividing the total charge by the rated steady-state luminous current of the light-emitting device. Subtracting the equivalent steady-state emission time from the originally set target emission time yields the corrected actual steady-state emission time. The active heat injection waveform is first output according to the preset timing sequence. Then, the amplitude of the drive current is controlled to decrease and clamped to the rated steady-state luminous current, entering the steady-state luminous stage with a duration equal to the actual steady-state luminous time. Finally, the output current is cut off.

9. The method for compensating and controlling the luminous stability of a copper wire lamp under low-temperature conditions according to claim 1, characterized in that, The injection timing of the diagnostic current, the acquisition timing of the transient voltage decay sequence, and the output timing of the composite drive waveform are all completed sequentially in a closed loop within a single PWM cycle. The copper wire lamp is a device formed by encapsulating LED bare chips with epoxy resin. The mechanical stress deviation is used to characterize the intensity of the piezoresistive effect of the epoxy resin on the LED bare chip under low temperature shrinkage.

10. A luminous stability compensation control system for copper wire lamps under low-temperature conditions, used to execute the luminous stability compensation control method for copper wire lamps under low-temperature conditions as described in any one of claims 1-9, characterized in that, include: The diagnostic sampling module is used to inject diagnostic current into the copper wire lamp during the turn-off time slot of a single-cycle drive sequence and obtain the transient voltage decay sequence across the light-emitting device in the copper wire lamp. The parameter decoupling module is used to decouple and calculate the current ambient temperature equivalent of the light-emitting device and the mechanical stress deviation caused by the cold shrinkage of the packaging material based on the transient voltage decay sequence. The waveform calculation module is used to calculate the corresponding active heat injection waveform parameters based on the mechanical stress deviation when the mechanical stress deviation exceeds the set safety threshold. The reconstructed output module is used to perform charge equivalence conversion based on the active heat injection waveform parameters, correct the loss of steady-state emission time in this cycle, and reconstruct the output of a composite driving waveform that includes the heat injection stage and the steady-state emission stage.

Images