LED lamp bead current control system

Abstract

The application discloses an LED lamp bead current control system and belongs to the field of LED display screens. The LED lamp bead current control system comprises a signal acquisition unit, an intelligent control unit, a PWM driving unit, a multi-stage compensation unit and a protection control unit. The application solves the problem that the prior art is difficult to simultaneously realize fast response, high-precision constant current and intelligent protection under power fluctuation, thermal drift and dynamic working conditions. Through an integrated closed-loop control system, the application realizes real-time multi-dimensional perception and intelligent dynamic adjustment of the working state of an LED module. The application can not only quickly compensate voltage mutation and accurately predict the trend of slow change, but also can perform nonlinear modeling and active correction on the electrical parameter drift caused by temperature change. Meanwhile, the application is equipped with a hierarchical over-temperature protection and an adaptive soft start mechanism, so that high-precision, high-stability and high-reliability constant current control is realized under all working conditions, and the consistency of the display quality of the whole screen is ensured.

Description

Technical Field

[0001] This invention relates to the field of LED display technology, specifically to an LED lamp bead current control system. Background Technology

[0002] With the rapid development of LED display technology, LED displays have become widely used in advertising media, stage performances, traffic guidance, monitoring and command, and commercial displays due to their advantages such as high brightness, high contrast, long lifespan, and energy efficiency. To ensure the uniformity of display, color consistency, and lifespan of LED displays, precise and stable control of their operating current is required to achieve constant current drive. However, in practical applications, the electrical and optical characteristics of LED modules are easily affected by a variety of factors.

[0003] LED displays typically consist of large display panels composed of multiple cascaded modules. Each module includes an LED array, driver IC, and control circuitry integrated on a PCB. On one hand, due to the large size of the display and the long power supply lines, line impedance can cause voltage drop differences between modules, resulting in lower supply voltages for modules farther from the power source. This leads to brightness decay and uneven display in edge areas. On the other hand, LEDs generate heat during operation, causing junction temperatures to rise. This results in thermal drift in their forward voltage drop and luminous efficiency, making it difficult for fixed-duty-cycle PWM drives to maintain long-term current stability. Furthermore, the system may face risks such as overcurrent and overtemperature shocks during start-up, shutdown, sudden load changes, or drastic changes in ambient temperature, affecting device safety. Traditional LED driver control schemes often employ simple feedback regulation, which struggles to quickly and smoothly compensate for dynamic disturbances such as line voltage drops and thermal drift. They also lack adaptive capabilities in safety mechanisms such as over-temperature protection and soft-start, easily leading to slow control response, insufficient steady-state accuracy, or overly coarse protection actions, affecting the uniformity of the entire screen display.

[0004] Therefore, since the existing requirements are not met, we have proposed an LED lamp bead current control system. Summary of the Invention

[0005] The purpose of this invention is to provide an LED lamp bead current control system, which solves the problems mentioned in the background art by integrating real-time multi-parameter acquisition, dynamic PWM control based on intelligent algorithms, active compensation mechanism combining multi-level feedforward and feedback, and multiple protection strategies with adaptive capabilities.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an LED lamp bead current control system, comprising:

[0007] The signal acquisition unit is configured to acquire the power supply voltage of the LED module, the current flowing through the module, and the temperature of the module in real time, and output the corresponding voltage signal, current signal, and temperature signal.

[0008] The intelligent control unit has a built-in preset control algorithm, which is configured to calculate the duty cycle adjustment of the PWM drive signal based on the voltage signal, current signal and temperature signal.

[0009] The PWM drive unit is configured to dynamically adjust the duty cycle of the PWM drive signal output to the LED module according to the calculated duty cycle adjustment amount, so as to realize constant current control of the LED module;

[0010] The multi-level compensation unit is configured to perform current fluctuation compensation and thermal drift compensation. The current fluctuation compensation is based on the change trend of the voltage signal and the position information of the module in the whole screen, and performs reverse compensation on the duty cycle adjustment in advance to eliminate the brightness unevenness caused by line voltage drop. The thermal drift compensation is based on the historical change data of temperature signal and current signal to perform proportional-integral compensation on the duty cycle adjustment.

[0011] The protection control unit includes an over-temperature protection module and a soft-start module;

[0012] The over-temperature protection module is configured to control the PWM drive unit to reduce the output power when the temperature signal exceeds a first preset threshold, and to turn off the PWM drive signal output when the temperature signal exceeds a second preset threshold.

[0013] The soft-start module is configured to control the duty cycle of the PWM drive unit output to increase linearly from zero to an initial set value with a first slope when the system starts or restarts, wherein the first slope is dynamically adjusted according to the temperature signal.

[0014] Furthermore, the intelligent control unit includes:

[0015] The core control module is configured to receive voltage signals, current signals, and temperature signals, and run a preset control algorithm to output an initial duty cycle adjustment command.

[0016] The signal fusion module is configured to perform time-series alignment and weighted fusion of temperature and current signals to generate characteristic parameters that reflect the dynamic thermo-electric working state of the LED module, and feed these characteristic parameters back to the core control module to correct the initial duty cycle adjustment command.

[0017] The adaptive adjustment module is configured to dynamically adjust the single update step size of the duty cycle adjustment output to the PWM drive unit based on the changing trend of the historical adjustment frequency and historical adjustment amplitude of the duty cycle adjustment.

[0018] Furthermore, the intelligent control unit introduces a mode switching mechanism, which automatically switches between a conventional constant current mode, a fast response mode, and a high stability mode based on the stability of the voltage signal, the rate of change of the temperature signal, and the ripple coefficient of the current signal.

[0019] In fast response mode, the core control module prioritizes increasing the control bandwidth, while the adaptive adjustment module temporarily increases the update step size.

[0020] In high stability mode, the core control module prioritizes reducing the control bandwidth to suppress noise, the adaptive adjustment module locks to the minimum update step size, and enables proportional-integral compensation in the multi-level compensation module as the main adjustment method.

[0021] Further, the PWM drive unit includes:

[0022] The signal isolation module is configured to receive the PWM logic signal corresponding to the duty cycle adjustment output by the intelligent control unit, and to electrically isolate the received PWM logic signal through the optocoupler integrated inside the signal isolation module to eliminate electrical noise interference and ground loop effects between the main control side and the power side. At the same time, it converts the voltage level of the received PWM logic signal into a level standard suitable for the subsequent power stage circuit driving, and outputs an isolated and level-matched drive logic signal.

[0023] The drive enhancement module is configured to receive drive logic signals, provide drive current with fast rising and falling edges, and output enhanced drive signals.

[0024] The power switch module is configured to receive the enhanced drive signal and adopts a synchronous rectification topology consisting of a power MOSFET and a freewheeling diode. It is used to generate and output the PWM drive signal to drive the LED module by turning the enhanced drive signal on or off.

[0025] Furthermore, the current fluctuation compensation in the multi-stage compensation unit includes:

[0026] The feedforward compensation module is configured to monitor the voltage signal representing the power supply voltage of the LED module in real time. When the change in voltage signal within adjacent sampling periods exceeds the preset sudden change threshold, it is determined that a voltage sudden change has occurred, and a duty cycle feedforward adjustment command is generated according to the amplitude and direction of the sudden change.

[0027] The trend prediction module is configured to store historical voltage signal values ​​for N consecutive sampling periods in a data buffer, perform linear fitting on the voltage data in the buffer, and calculate the average slope and direction of voltage change within the corresponding time period of the N sampling periods. This is used to predict the voltage fluctuation trend in the next control period. If the predicted trend is a continuous gradual increase, a gradually decreasing duty cycle pre-compensation amount is generated; if the predicted trend is a continuous gradual decrease, a gradually increasing duty cycle pre-compensation amount is generated. Here, N is an integer greater than or equal to 2.

[0028] The voltage drop compensation module is configured to calculate the voltage drop of the LED module relative to the power input terminal based on the physical position coordinates of the LED module in the display screen, the length of the power supply line, and the line impedance parameters, and generate a corresponding duty cycle compensation amount to compensate for the brightness attenuation of the far-end module caused by the line impedance.

[0029] Furthermore, the thermal drift compensation in the multi-level compensation unit includes:

[0030] The data modeling module is configured to establish an equivalent thermal model that reflects the hysteresis characteristics of LED module junction temperature changes based on historical time-series data of temperature and current signals. The equivalent thermal model simulates and outputs the predicted LED junction temperature based on real-time heating power and ambient temperature.

[0031] The adaptive compensation module is configured to dynamically adjust the proportional coefficient and integral time constant in the proportional-integral compensation based on the predicted junction temperature output by the equivalent thermal model and the real-time deviation of the current signal relative to the set value. This is used to achieve nonlinear compensation for LED forward voltage drop drift and luminous efficiency decay caused by temperature gradient.

[0032] Furthermore, the over-temperature protection module includes:

[0033] The graded response module is configured to control the PWM drive unit to reduce the output power in a stepwise manner at a first cooling rate when the temperature signal exceeds a first preset threshold, and to restore the output power in a stepwise manner at a first recovery rate after the temperature signal falls back below the first preset threshold.

[0034] The thermal accumulation monitoring module is configured to calculate in real time the cumulative time of the LED module at temperatures above the third preset threshold, and dynamically adjust the first preset threshold and the second preset threshold according to the length of the cumulative time.

[0035] Furthermore, the soft-start module includes:

[0036] The slope adaptive module is configured to dynamically determine the first slope based on the difference between the temperature signal at the moment of system startup and the rated operating temperature. When the difference is positive, the first slope is positively correlated with the magnitude of the difference.

[0037] The status feedback module is configured to monitor the rise rate of the current signal and the fall rate of the voltage signal in real time during the soft start process, and dynamically fine-tune the first slope based on the deviation between the rise rate of the current signal and the fall rate of the voltage signal and the expected rate, until the duty cycle reaches the initial set value and the current signal stabilizes at the target constant current value.

[0038] Furthermore, the signal fusion module performs time-series alignment and weighted fusion of the temperature signal and the current signal to generate characteristic parameters reflecting the dynamic thermo-electric operating state of the LED module, including:

[0039] The ripple of the current signal is smoothed by sliding window averaging filter to obtain the calibrated current signal; the temperature lag compensation time of the temperature signal is calculated based on the instantaneous rate of change of the current, and the temperature signal is compensated based on the temperature lag compensation time to obtain the compensated temperature signal.

[0040] Using the time axis of the calibrated current signal as a reference, the compensated temperature signal is linearly interpolated to generate a temperature interpolation signal; the peak point of the current in the calibrated current signal is selected as the synchronization anchor point to verify the rationality of the temperature interpolation point, and the target temperature interpolation signal is obtained after the verification is passed.

[0041] The calibrated current signal and the target temperature interpolation signal are normalized respectively to obtain the normalized current signal and the normalized target temperature signal. The normalized current signal is amplitude modulated using the normalized current signal as the carrier and the normalized target temperature signal as the modulation signal to obtain the amplitude modulated current signal.

[0042] Envelope detection is performed on the amplitude-modulated current signal to extract the envelope corresponding to the amplitude-modulated current signal;

[0043] The LED's operating stages are determined by real-time analysis of current and temperature signals using FPGA. The operating stages include a startup stage, a stabilization stage, and a shutdown stage. Different fusion weights are set for the startup stage, the stabilization stage, and the shutdown stage. Based on the fusion weights corresponding to each operating stage, the normalized current signal and the normalized target temperature signal are fused in segments to obtain the segmented fused signal corresponding to each operating stage.

[0044] Feature extraction is performed on the segmented fusion signals of the start-up phase, the stable phase, and the turn-off phase to obtain characteristic parameters reflecting the dynamic thermo-electric working state of the LED module.

[0045] The characteristic parameters reflecting the dynamic thermo-electric operating state of the LED module are fed back to the core control module to correct the initial duty cycle adjustment command.

[0046] Furthermore, the adaptive adjustment module dynamically adjusts the single update step size of the duty cycle adjustment output to the PWM drive unit based on the changing trends of the historical adjustment frequency and historical adjustment amplitude of the duty cycle adjustment, including:

[0047] According to the preset sampling frequency, the real-time adjustment frequency data and real-time adjustment amplitude data of the duty cycle adjustment amount are obtained;

[0048] Acquire historical adjustment frequency data and historical adjustment amplitude data of duty cycle adjustment within a preset time window;

[0049] Calculate the historical frequency moving average and historical amplitude moving average of the duty cycle adjustment within the preset time window;

[0050] The frequency trend coefficient of duty cycle adjustment is determined based on real-time adjustment frequency data and historical frequency sliding average.

[0051] The amplitude trend coefficient of the duty cycle adjustment is determined based on real-time adjustment amplitude data and historical amplitude sliding average.

[0052] The dynamic step size correction factor is determined based on the frequency trend coefficient and amplitude trend coefficient of the duty cycle adjustment.

[0053] ;

[0054] in, This is the dynamic step size correction factor; Frequency trend weighting factor; This is the frequency trend coefficient; The magnitude trend weighting factor; This is the amplitude trend coefficient;

[0055] The single update step size of the duty cycle adjustment is calculated based on the dynamic step size correction factor and the preset basic step size.

[0056] ;

[0057] in, The single update step size for duty cycle adjustment; Use the base step size.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] This invention uses a signal acquisition unit to perform real-time multi-dimensional sensing of voltage, current, and temperature, providing a precise data foundation for intelligent control. The intelligent control unit, based on built-in algorithms and mode switching mechanisms, can adaptively achieve an optimal balance between response speed and control stability. The PWM drive unit adopts an isolation and enhancement design to ensure the purity of the drive signal and the reliability of power output. The multi-level compensation unit achieves rapid and accurate active suppression of major interference sources through current fluctuation compensation combining feedforward and prediction, and adaptive thermal drift compensation modeling thermal hysteresis characteristics. The graded over-temperature protection and slope-adjustable soft-start module of the protection control unit can improve the robustness and startup safety of the system. This invention achieves high-precision and high-stability constant current control of the LED display module under all operating conditions, while significantly enhancing the system's adaptability and long-term operational reliability, extending the service life of the LED display, and thus ensuring the consistency of the overall display quality. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the LED lamp bead current control system of the present invention. Detailed Implementation

[0061] 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.

[0062] To address the technical challenges of achieving rapid response, high-precision constant current, and intelligent protection simultaneously under power fluctuations, thermal drift, and dynamic operating conditions, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution:

[0063] LED lamp bead current control system, including:

[0064] The signal acquisition unit is configured to acquire the power supply voltage of the LED module, the current flowing through the module, and the temperature of the module in real time, and output the corresponding voltage signal, current signal, and temperature signal. The LED module is composed of several semiconductor light-emitting diodes connected in series or in parallel.

[0065] The intelligent control unit, which is the main controller, has a built-in preset control algorithm configured to calculate the duty cycle adjustment of the PWM drive signal based on the voltage signal, current signal, and temperature signal.

[0066] The PWM drive unit, connected to the main controller, is configured to dynamically adjust the duty cycle of the PWM drive signal output to the LED module according to the calculated duty cycle adjustment amount, in order to achieve constant current control of the LED module;

[0067] The multi-level compensation unit, integrated into the main controller, is configured to perform current fluctuation compensation and thermal drift compensation. The current fluctuation compensation performs reverse compensation on the duty cycle adjustment amount in advance based on the change trend of the voltage signal and the position information of the module in the whole screen, in order to eliminate the brightness unevenness caused by line voltage drop. The thermal drift compensation performs proportional-integral compensation on the duty cycle adjustment amount based on the historical change data of temperature signal and current signal.

[0068] The protection control unit, connected to the main controller, includes an over-temperature protection module and a soft-start module;

[0069] The over-temperature protection module is configured to control the PWM drive unit to reduce the output power when the temperature signal exceeds a first preset threshold, and to turn off the PWM drive signal output when the temperature signal exceeds a second preset threshold.

[0070] The soft-start module is configured to control the duty cycle of the PWM drive unit output to increase linearly from zero to an initial set value with a first slope when the system starts or restarts, wherein the first slope is dynamically adjusted according to the temperature signal.

[0071] The technical effects of the above-mentioned technical solution are as follows: This invention provides a high-precision and high-reliability LED constant current drive solution for energy-saving lighting equipment and other electric lighting applications not included in this category. Through the signal acquisition unit, it achieves real-time acquisition of all parameters of the LED module voltage, current, and temperature, providing a precise data foundation for the intelligent control unit. This allows the main controller to quickly calculate the precise PWM duty cycle adjustment based on the built-in control algorithm, thereby driving the PWM drive unit to achieve high-precision dynamic constant current control, effectively ensuring the stability and consistency of LED operation. Simultaneously, the multi-level compensation unit suppresses interference caused by input voltage changes through current fluctuation compensation, and combines thermal drift compensation with proportional-integral correction using historical data. This invention significantly improves the system's anti-interference capability and current stability under long-term operation, which is crucial for maintaining the long-term performance of energy-saving lighting equipment. In addition, the over-temperature protection module in the protection control unit can reduce or cut off the output in stages when the temperature is abnormal to avoid LED overheating damage. The soft-start module achieves smooth start-up and reduces current surges by dynamically adjusting the rise slope according to the temperature, thus comprehensively enhancing the reliability and service life of the system. Through the synergistic effect of each unit, this invention comprehensively enhances the reliability and service life of various LED backlights and display systems composed of semiconductor light-emitting diodes. Especially in the field of large-size electronic device manufacturing, it effectively solves the problem of uneven brightness in displays caused by line voltage drop.

[0072] The intelligent control unit includes:

[0073] The core control module is configured to receive voltage signals, current signals, and temperature signals, and run a preset control algorithm to output an initial duty cycle adjustment command.

[0074] The control algorithm is an adaptive control algorithm based on a combination of improved incremental proportional-integral-derivative (PID) and fuzzy logic. In specific implementation, the control algorithm has built-in initial PID control parameters and receives voltage, current, and temperature signals from the signal acquisition unit in real time.

[0075] The control algorithm compares the acquired current signal with the system's preset target constant current value to calculate the real-time current deviation. Simultaneously, the algorithm analyzes the voltage signal to assess the potential impact of the power supply state and references the temperature signal to determine the LED's operating environment. Then, the improved incremental control algorithm uses this current deviation as the primary input to calculate the theoretically required basic duty cycle adjustment to eliminate it. The integrated fuzzy logic unit performs online real-time correction and softening of the basic adjustment output by the control algorithm based on the rate of change of the current deviation and the stability of the voltage signal to optimize dynamic response characteristics. Finally, the adjustment amount, corrected by fuzzy logic, is output as the initial duty cycle adjustment command. This process operates in real-time and continuously, ensuring that the system can respond quickly and smoothly to current deviations.

[0076] The signal fusion module is communicatively connected to the core control module and is configured to perform time-series alignment and weighted fusion of temperature signals and current signals to generate characteristic parameters that reflect the dynamic thermo-electric working state of the LED module. These characteristic parameters are then fed back to the core control module to correct the initial duty cycle adjustment command.

[0077] The specific implementation steps for time series alignment and weighted fusion of temperature and current signals are as follows:

[0078] Independent data buffers are allocated for temperature and current signals respectively. The main control cycle of the system is used as a unified time reference. The sampling points of the two signals are timestamped and interpolated to ensure that their data points on the same time axis correspond one-to-one and complete time synchronization.

[0079] The two aligned data streams are filtered to suppress noise, and the statistical characteristics of each signal within the current sliding time window, such as mean and rate of change, are calculated.

[0080] According to the preset fusion rules of the thermal-electric coupling model of the LED module, the current signal is assigned a weight related to dynamic response, and the temperature signal is assigned a weight related to thermal inertia and long-term drift. The sum of the weight coefficients of the two is a fixed constant, and the weight coefficient can be dynamically adjusted according to different working stages such as whether the system is currently in a rapid heating or high current transient.

[0081] The weighted current statistical characteristic value and temperature statistical characteristic value are combined according to the preset normalization formula to generate a dimensionless characteristic parameter value that comprehensively reflects the instantaneous electrical load and thermal accumulation state of the LED module.

[0082] An adaptive adjustment module, connected between the core control module and the PWM drive module, is configured to dynamically adjust the single update step size of the duty cycle adjustment output to the PWM drive unit based on the changing trend of the historical adjustment frequency and historical adjustment amplitude of the duty cycle adjustment.

[0083] The technical effects of the above-mentioned technical solution are as follows: The core control module, by combining an improved incremental PID and fuzzy logic adaptive control algorithm, not only achieves accurate and rapid calculation of current deviation based on traditional PID, but also uses fuzzy logic to perform real-time correction and softening of dynamic response, thereby outputting an initial duty cycle adjustment command that is both fast and smooth, thus improving the system's response speed and control stability. The signal fusion module, by aligning and weighting temperature and current signals over time, generates characteristic parameters that comprehensively reflect the dynamic thermoelectric state of the LED module and feeds them back to the core control module, thereby achieving thermo-electric coupling correction of the initial command and effectively enhancing the system's adaptability to thermal drift and environmental changes and the accuracy of control. The adaptive adjustment module, by dynamically adjusting the output update step size according to the historical adjustment frequency and amplitude, can automatically reduce the step size to improve control accuracy when the system tends to a steady state, thus maintaining a fast response during dynamic processes, thereby achieving intelligent balance and optimization of control accuracy and response speed throughout the entire operating range.

[0084] The intelligent control unit incorporates a mode switching mechanism, automatically switching between conventional constant current mode, fast response mode, and high stability mode based on the stability of the voltage signal, the rate of change of the temperature signal, and the ripple coefficient of the current signal.

[0085] In fast response mode, the core control module prioritizes increasing the control bandwidth, while the adaptive adjustment module temporarily increases the update step size.

[0086] In high stability mode, the core control module prioritizes reducing the control bandwidth to suppress noise, the adaptive adjustment module locks to the minimum update step size, and enables proportional-integral compensation in the multi-level compensation module as the main adjustment method.

[0087] In this embodiment, the ripple coefficient is a dimensionless parameter used to quantify the magnitude of the AC fluctuation component in the current signal relative to the DC average value, reflecting the stability of the current. This ripple coefficient is obtained by real-time monitoring of the current signal flowing through the LED module. Specifically, within each fixed calculation cycle, the intelligent control unit processes the current sampling data collected over a time window: it calculates the arithmetic mean of the current within that time period as the DC component, and simultaneously extracts the amplitude representation of the AC fluctuation component using an algorithm (such as calculating the root mean square error or the difference between the peak and valley values). Dividing the amplitude representation of the AC fluctuation component by the DC average value yields the current current ripple coefficient.

[0088] In this embodiment, the automatic mode switching mechanism based on the three criteria of voltage stability, temperature change rate, and current ripple coefficient is executed according to the following logic:

[0089] The intelligent control unit continuously calculates and monitors the real-time values ​​of these three parameters, comparing and logically judging them with their respective preset mode switching thresholds. When the system is in normal operating condition and all parameters are within normal range, it maintains the normal constant current mode. When a sudden increase in the current ripple coefficient or a large and rapid fluctuation in the voltage signal is detected (indicating a sudden change in the load or power supply), the system immediately switches to the fast response mode. At this time, the core control module increases the bandwidth of the control algorithm to speed up the response, and simultaneously instructs the adaptive adjustment module to temporarily increase the update step size of the duty cycle adjustment to quickly suppress disturbances. Conversely, when an extremely low current ripple coefficient, a highly stable voltage signal, but a very slow rate of change in the temperature signal is detected (indicating that the system is in a thermally stable state), the system switches to the high stability mode. At this time, the core control module prioritizes reducing the control bandwidth to filter high-frequency noise, and the adaptive adjustment module locks its update step size at a preset minimum value to prevent over-adjustment from introducing jitter. At the same time, the proportional-integral compensation based on historical data in the multi-level compensation unit is promoted to the main adjustment method, focusing on slow and precise correction of subtle thermal drift, thereby achieving extremely high steady-state accuracy. The mode switching process is a smooth transition to avoid abrupt changes in output at the switching point.

[0090] The technical effects of the above solution are as follows: By introducing an intelligent mode switching mechanism based on voltage signal stability, temperature signal change rate, and current signal ripple coefficient, the system can adaptively switch dynamically between conventional constant current mode, fast response mode, and high stability mode. In fast response mode, the core control module increases the control bandwidth and, in conjunction with the adaptive adjustment module, temporarily increases the update step size, ensuring the system's rapid tracking capability under scenarios such as load changes or startup transients. In high stability mode, the core control module reduces the control bandwidth to suppress noise, and the adaptive adjustment module locks the minimum update step size and uses proportional-integral compensation in the multi-level compensation module as the main adjustment method, which can significantly improve the system's accuracy and anti-interference capability during steady-state operation. This allows the entire system to automatically optimize the control strategy across the entire operating range based on real-time operating conditions.

[0091] The PWM drive unit includes:

[0092] The signal isolation module is configured to receive the PWM logic signal corresponding to the duty cycle adjustment output by the intelligent control unit, and to electrically isolate the received PWM logic signal through the optocoupler integrated inside the signal isolation module to eliminate electrical noise interference and ground loop effects between the main control side and the power side. At the same time, it converts the voltage level of the received PWM logic signal into a level standard suitable for the subsequent power stage circuit driving, and outputs an isolated and level-matched drive logic signal.

[0093] The drive enhancement module is configured to receive drive logic signals, provide drive current with fast rising and falling edges, and output enhanced drive signals.

[0094] The power switch module is configured to receive the enhanced drive signal and adopts a synchronous rectification topology consisting of a power MOSFET and a freewheeling diode. It is used to generate and output the PWM drive signal to drive the LED module by turning on or off according to the enhanced drive signal.

[0095] The switching frequency of the power switch module is adaptively adjusted by the intelligent control unit based on temperature and current signals, specifically as follows:

[0096] When the temperature signal is below a preset temperature threshold and the current signal is stable, the switching frequency is set to a first frequency; when the temperature signal is above the preset temperature threshold or the current signal fluctuates beyond a preset range, the switching frequency is reduced to a second frequency lower than the first frequency.

[0097] The technical effects of the above solution are as follows: The signal isolation module achieves electrical isolation between the main control side and the power side through optocouplers, effectively eliminating noise interference and ground loop effects, and completing signal level matching and conversion, thereby enhancing the system's reliability and anti-interference capability. The drive enhancement module ensures that the power switching module can respond to control signals quickly and accurately by providing a strong drive current with fast edges, thereby reducing switching losses and improving overall efficiency. The power switching module adopts a synchronous rectification topology and combines a switching frequency adjustment mechanism based on temperature and current signals. Under low-temperature stable conditions, a higher switching frequency is used to optimize response and ripple performance, while the frequency is automatically reduced under high temperature or current fluctuations to reduce switching losses and thermal stress. Thus, while ensuring drive performance, intelligent balance and optimization of power consumption, heat dissipation, and reliability are achieved.

[0098] Current fluctuation compensation in a multi-stage compensation unit includes:

[0099] The feedforward compensation module is configured to monitor the voltage signal representing the power supply voltage of the LED module in real time. When the change in the voltage signal within adjacent sampling periods exceeds a preset abrupt change threshold, a voltage abrupt change is determined to have occurred. Based on the amplitude and direction of the abrupt change, a duty cycle feedforward adjustment command is generated, specifically:

[0100] When a voltage surge is detected, the instruction instructs the PWM duty cycle to be reduced instantaneously in advance to suppress any potential current surge.

[0101] Conversely, when a voltage drop is detected, the duty cycle is instantly increased to buffer the downward trend of the current, thereby offsetting the main disturbance caused by the power supply change before the feedback loop takes effect.

[0102] The trend prediction module is configured to store historical voltage signal values ​​for N consecutive sampling periods in a data buffer, perform linear fitting on the voltage data in the buffer, and calculate the average slope and direction of voltage change within the corresponding time period of the N sampling periods. This is used to predict the voltage fluctuation trend in the next control period. If the predicted trend is a continuous gradual increase, a gradually decreasing duty cycle pre-compensation amount is generated; if the predicted trend is a continuous gradual decrease, a gradually increasing duty cycle pre-compensation amount is generated. Here, N is an integer greater than or equal to 2, and the specific value is set according to the system sampling frequency and control period.

[0103] Among them, the pre-compensation amount and the adjustment amount generated by the feedforward compensation module for instantaneous changes are logically superimposed to form a duty cycle feedforward adjustment command for the PWM drive signal, so as to realize the advanced and smooth compensation for the slow drift or regular fluctuation of the power supply voltage, and further improve the overall suppression capability of the system for input voltage changes.

[0104] The voltage drop compensation module is configured to calculate the voltage drop of the LED module relative to the power input terminal based on the physical position coordinates of the LED module in the display screen, the length of the power supply line, and the line impedance parameters, and generate a corresponding duty cycle compensation amount to compensate for the brightness attenuation of the far-end module caused by the line impedance.

[0105] The technical effects of the above solution are as follows: The feedforward compensation module, by monitoring the voltage signal in real time, can immediately generate a reverse duty cycle adjustment command when a voltage surge is detected, quickly suppressing the current surge or drop, thereby improving the system's active suppression capability and dynamic response speed against instantaneous power supply disturbances. The trend prediction module, by analyzing the linear fitting trend of historical voltage data, predicts the gradual change direction of the voltage and generates a progressive pre-compensation amount, enabling advanced smooth compensation for slow power supply drift or regular fluctuations, thereby enhancing the system's adaptability and steady-state accuracy to medium- and long-term voltage fluctuations. The compensation amounts of the two modules are logically superimposed, so that the current fluctuation compensation mechanism has both a fast feedforward response to instantaneous surges and a predictive smooth adjustment for gradual trends, thereby comprehensively improving the anti-interference capability and stability of the LED lamp bead current control system when facing complex power supply fluctuations. The voltage drop compensation module, by calculating the voltage drop value of the LED module relative to the power input terminal and generating a corresponding duty cycle compensation amount, can compensate for the brightness attenuation of the far-end module caused by line impedance, thereby eliminating the problem of uneven brightness caused by line voltage drop.

[0106] Thermal drift compensation in a multi-stage compensation unit includes:

[0107] The data modeling module is configured to establish an equivalent thermal model that reflects the hysteresis characteristics of LED module junction temperature change based on historical time-series data of temperature and current signals. The equivalent thermal model simulates and outputs the predicted LED junction temperature based on real-time heat generation power (calculated from current and voltage signals) and ambient temperature. The modeling process of the equivalent thermal model is existing technology in the field and is not the technical point of this invention, so it will not be described in detail.

[0108] The adaptive compensation module is configured to dynamically adjust the proportional coefficient and integral time constant in the proportional-integral compensation based on the predicted junction temperature output by the equivalent thermal model and the real-time deviation of the current signal relative to the set value. This is used to achieve nonlinear compensation for LED forward voltage drop drift and luminous efficiency decay caused by temperature gradient.

[0109] In this embodiment, the adaptive compensation module synchronously receives the predicted junction temperature output by the data modeling module and the current signal provided by the signal acquisition unit. The adaptive compensation module internally presets a compensation table or function reflecting the relationship between junction temperature and LED forward voltage drop drift and luminous efficiency decay. First, the adaptive compensation module substitutes the predicted junction temperature into the compensation table or function to calculate the theoretical deviation of the current setpoint caused by pure thermal effects. Next, the adaptive compensation module compares and fuses this theoretical deviation with the real-time deviation of the current signal relative to the original setpoint to obtain a comprehensive temperature-current composite deviation signal. Then, based on the magnitude and trend of this composite deviation signal, the adaptive compensation module dynamically adjusts the parameters of the subsequent proportional-integral compensator: when the composite deviation signal is large or changes rapidly, the proportional coefficient is appropriately increased and the integral time constant is decreased according to preset rules to enhance the speed of compensation; when the composite deviation signal is small and changes slowly, the proportional coefficient is decreased and the integral time constant is increased according to preset rules to improve the stability and steady-state accuracy of compensation, thereby achieving accurate adaptive compensation for thermal drift nonlinearity.

[0110] The technical effects of the above solution are as follows: The data modeling module establishes an equivalent thermal model based on historical temperature and current data, which can simulate and predict the changes in LED junction temperature in real time, thereby overcoming the shortcomings of temperature sensor detection delay and the influence of ambient temperature. This provides an accurate and timely temperature status basis for thermal drift compensation. The adaptive compensation module dynamically adjusts the proportional-integral compensation parameters according to the predicted junction temperature and real-time current deviation, which can achieve nonlinear and accurate compensation for LED forward voltage drop drift and luminous efficiency decay caused by gradual temperature changes. This significantly improves the current stability and constant current accuracy of the system during long-term operation or when the ambient temperature changes, thereby enhancing the adaptability and reliability of the entire control system.

[0111] The over-temperature protection module includes:

[0112] The graded response module is configured to control the PWM drive unit to reduce the output power in a stepwise manner at a first cooling rate when the temperature signal exceeds a first preset threshold, and to restore the output power in a stepwise manner at a first recovery rate after the temperature signal falls back below the first preset threshold.

[0113] When the temperature signal first exceeds the first preset threshold, the graded response module does not immediately perform a large-scale step power reduction. Instead, it sends a stepped power reduction command to the PWM drive unit with a first cooling rate as the step speed. This command is triggered at fixed time intervals (e.g., every 100 milliseconds). Each trigger reduces the current PWM duty cycle command value by a fixed percentage or absolute value until the temperature signal begins to show a clear downward trend. This process is closed-loop. The graded response module monitors the rate of change of the temperature signal in real time. If the temperature drop is too slow during the power reduction process, it automatically accelerates the first cooling rate, i.e., increases the... The power reduction rate is set at a maximum within each time interval; conversely, if the temperature drops too quickly, the rate is automatically slowed down to achieve smooth cooling and avoid severe impact on the light output. When the temperature signal drops below the first preset threshold, the graded response module starts the recovery logic, gradually increasing the PWM duty cycle command value in a stepwise manner at another preset recovery rate, which is usually slower than the cooling rate, until it recovers to the operating point before overheating or the system's reset safe power point. During the recovery process, the graded response module closely monitors the temperature signal, and if it detects that the temperature rises too quickly again, it immediately pauses or slows down the recovery process.

[0114] The heat accumulation monitoring module is configured to calculate in real time the cumulative time of the LED module at temperatures above the third preset threshold, and dynamically adjust the first preset threshold and the second preset threshold according to the length of the cumulative time.

[0115] The thermal accumulation monitoring module continuously integrates and calculates the total duration of the LED module at temperatures above the third preset threshold. At the same time, the thermal accumulation monitoring module dynamically adjusts the protection threshold according to a preset thermal aging model. This model defines a negative correlation between thermal accumulation time and the maximum allowable operating temperature of the LED (i.e., the first and second preset thresholds). This means that as the cumulative operating time of the LED module at higher temperatures increases, the system will initiate graded power reduction or even shutdown protection at lower temperature trigger points, thereby protecting the LED more conservatively and preventing the increase in failure rate caused by long-term thermal aging of materials. This down-adjustment process usually only recovers at a very slow rate after the temperature has been below the third preset threshold for a long time, thus simulating the thermal aging effect of LED devices.

[0116] The first, second, and third preset thresholds logically constitute a progressively advanced, functionally coupled multi-level thermal protection and aging early warning system. The first preset threshold serves as the primary over-temperature response trigger point. Upon triggering, the system initiates a graded, stepped power reduction to proactively suppress temperature rise and restore thermal balance. The second preset threshold serves as the upper limit for safety protection. Once the temperature exceeds this threshold, the PWM output is directly cut off to achieve forced shutdown protection and prevent thermal damage. The third preset threshold serves as the start threshold for thermal accumulation monitoring, used to define the high-temperature operating state. When the temperature remains above this threshold, the system accumulates the duration and dynamically lowers the first and second preset thresholds based on the accumulated duration. This allows the over-temperature protection point to gradually move forward with the LED thermal aging process, achieving preventative protection.

[0117] The technical effects of the above solution are as follows: The graded response module controls the PWM drive unit to reduce and restore the output power at a step-wise rate after the first preset temperature threshold is triggered, thereby achieving smooth and gradual adjustment of the overheating state. This avoids current surges or light flickering caused by sudden changes in output power, and improves the continuity and stability of system operation while ensuring heat dissipation. The thermal accumulation monitoring module calculates and accumulates the high-temperature working time in real time, and dynamically lowers the first and second temperature protection thresholds according to the accumulated time. This makes the over-temperature protection trigger point of the system more sensitive as the thermal fatigue of the LED module increases, thereby achieving preventive protection against long-term thermal damage to the LED and significantly enhancing the reliability and service life of the system in complex or harsh thermal environments.

[0118] The soft start module includes:

[0119] The slope adaptive module is configured to dynamically determine the first slope based on the difference between the temperature signal at the moment of system startup and the rated operating temperature. When the difference is positive, the first slope is positively correlated with the magnitude of the difference.

[0120] The slope adaptive module immediately captures the initial temperature signal output by the temperature sensor at the moment of system power-on and compares it with the LED's rated operating temperature value pre-stored in the intelligent control unit, calculating the difference between the two. Internally, the module uses a lookup table that maps the temperature difference to a soft-start slope reference value. When the difference is positive (i.e., the initial temperature is higher than the rated temperature), it indicates that the LED is already in a hot state. Based on the magnitude of the difference, the module selects a larger initial first slope value from the mapping relationship, allowing the duty cycle to rise more quickly, thereby rapidly establishing the operating current to activate the necessary cooling mechanism (e.g., if the system has active cooling). Conversely, if the difference is negative (i.e., the initial temperature is lower than the rated temperature), a smaller initial first slope value is selected to achieve a smoother start-up and reduce thermal shock. Furthermore, the slope adaptive module also considers the system's historical operating temperature before the last shutdown. In the case of a hot restart, even if the current initial temperature is lower than the rated value, a positive offset is introduced into the calculation to moderately increase the first slope to cope with possible residual thermal stress.

[0121] The status feedback module is configured to monitor the rise rate of the current signal and the fall rate of the voltage signal in real time during the soft start process, and dynamically fine-tune the first slope according to the deviation between the rise rate of the current signal and the fall rate of the voltage signal and the expected rate, until the duty cycle reaches the initial set value and the current signal stabilizes at the target constant current value.

[0122] In this process, after the slope adaptive module determines the first slope and initiates the soft-start process, the state feedback module begins operation. The state feedback module samples the current signal and the LED module's voltage signal in real-time at a high frequency. Internally, the state feedback module pre-sets expected current rise rate curves and expected voltage fall rate curves corresponding to the current first slope. During each minute period of the soft-start process, the state feedback module calculates the actual instantaneous rise rate of the current signal and the actual instantaneous fall rate of the voltage signal, comparing them with the expected rates. If the actual current rise rate is consistently higher than expected and the voltage fall rate is too slow, it indicates that the load impedance or circuit condition is causing the current to increase too quickly, posing an overshoot risk. In this case, the module generates a negative fine-tuning command to dynamically reduce the value of the first slope, slowing down the duty cycle increase. Conversely, if the actual current rise rate is too low and the voltage fall rate is too fast, it indicates that the startup process is too slow. In this case, the module generates a positive fine-tuning command to dynamically increase the value of the first slope. This fine-tuning process is continuous and closed-loop, ensuring that the actual startup trajectory is constrained within a safe and smooth range. Finally, when the duty cycle reaches the initial set value and the current signal stabilizes within a small error band near the target constant current value, the state feedback module stops fine-tuning and exits intervention, completely transferring control to the core constant current control loop.

[0123] The technical effects of the above solution are as follows: The slope adaptive module dynamically determines the initial startup slope based on the difference between the system startup temperature and the rated operating temperature. At low temperatures, a gentler slope is used to reduce thermal shock, while the startup speed is accelerated when the operating temperature is approached. This achieves a preliminary match between the startup process and the thermal state. The state feedback module monitors the current rise and voltage drop rates in real time during the soft start process and compares them with the expected values. It then dynamically fine-tunes the startup slope to ensure that the current can reach the target constant current value in a stable and controllable manner. This effectively suppresses startup current overshoot and voltage drop, thereby optimizing the speed and stability of the startup process while ensuring LED safety.

[0124] Working Principle: By acquiring voltage, current, and temperature signals in real time, the intelligent control unit uses control algorithms to calculate the precise PWM duty cycle adjustment based on these signals. The PWM drive unit then dynamically adjusts the output accordingly to achieve the core constant current control. The key lies in the collaborative work of multiple compensation units: current fluctuation compensation uses feedforward and trend prediction to pre-compensate for voltage sudden changes and gradual changes, quickly offsetting power supply disturbances; thermal drift compensation uses historical data to establish an equivalent thermal model and adaptively adjusts PI parameters to accurately compensate for electrical parameter drift caused by gradual temperature changes; and the protection control unit ensures system safety and smooth startup through graded over-temperature protection and a soft-start mechanism that dynamically adjusts the slope based on real-time status. This invention significantly improves current control accuracy, dynamic response speed, and system robustness, effectively extends the lifespan of the LED display screen, and ensures the consistency of the overall display quality.

[0125] The signal fusion module performs time-series alignment and weighted fusion of temperature and current signals to generate characteristic parameters reflecting the dynamic thermo-electric operating state of the LED module, including:

[0126] The ripple of the current signal is smoothed by sliding window averaging filter to obtain the calibrated current signal; the temperature lag compensation time of the temperature signal is calculated based on the instantaneous rate of change of the current, and the temperature signal is compensated based on the temperature lag compensation time to obtain the compensated temperature signal.

[0127] Using the time axis of the calibrated current signal as a reference, the compensated temperature signal is linearly interpolated to generate a temperature interpolation signal; the peak point of the current in the calibrated current signal is selected as the synchronization anchor point to verify the rationality of the temperature interpolation point, and the target temperature interpolation signal is obtained after the verification is passed.

[0128] The calibrated current signal and the target temperature interpolation signal are normalized respectively to obtain the normalized current signal and the normalized target temperature signal. The normalized current signal is amplitude modulated using the normalized current signal as the carrier and the normalized target temperature signal as the modulation signal to obtain the amplitude modulated current signal.

[0129] Envelope detection is performed on the amplitude-modulated current signal to extract the envelope corresponding to the amplitude-modulated current signal;

[0130] The LED's operating stages are determined by real-time analysis of current and temperature signals using FPGA. The operating stages include a startup stage, a stabilization stage, and a shutdown stage. Different fusion weights are set for the startup stage, the stabilization stage, and the shutdown stage. Based on the fusion weights corresponding to each operating stage, the normalized current signal and the normalized target temperature signal are fused in segments to obtain the segmented fused signal corresponding to each operating stage.

[0131] Feature extraction is performed on the segmented fusion signals of the start-up phase, the stable phase, and the turn-off phase to obtain characteristic parameters reflecting the dynamic thermo-electric working state of the LED module.

[0132] The characteristic parameters reflecting the dynamic thermo-electric operating state of the LED module are fed back to the core control module to correct the initial duty cycle adjustment command.

[0133] In this embodiment, the temperature hysteresis compensation time is calculated based on the instantaneous rate of change of current from the temperature signal. The temperature signal is then compensated based on this hysteresis compensation time to obtain the compensated temperature signal. The instantaneous rate of change of current is: ,in, The instantaneous rate of change of current; This is the current sample value at the current moment; This is the current sample value from the previous moment; This is the current sampling time; This is the previous sampling time. The temperature lag compensation time is: ;in, The instantaneous rate of change of current; This is the temperature lag compensation time; The temperature lag compensation coefficient is determined through LED thermal characteristic calibration experiments. It is used to quantify the influence of the current change rate on the temperature lag time and is the core parameter for calculating the temperature lag compensation time.

[0134] In this embodiment, using the time axis of the calibrated current signal as a reference, a temperature interpolation signal is generated by linear interpolation of the compensated temperature signal. Each temperature sampling point (1ms interval) corresponds to 10 current sampling points. A 10kHz temperature interpolation signal is generated using a linear interpolation formula to ensure that the sampling frequencies of the current and temperature signals are completely consistent. The interpolation calculation formula is as follows: ,in, This is a temperature interpolation signal; This is the k-th temperature sample value; This is the (k+1)th temperature sample value; The target time for interpolation; The time for the kth temperature sampling period; This is the (k+1)th temperature sampling time.

[0135] In this embodiment, the peak current point in the calibrated current signal is selected as the synchronization anchor point to verify the rationality of the temperature interpolation point. Specifically, the rationality verification is as follows: if the deviation between the temperature interpolation at the anchor point and the adjacent actual temperature sampling point is >0.3℃, the interpolation window is readjusted (expanded to 20 current sampling points), and the interpolation is repeated until the deviation is ≤0.3℃.

[0136] In this embodiment, the normalized current signal is used as the carrier signal and the normalized target temperature signal is used as the modulation signal. The normalized current signal is amplitude modulated to obtain the amplitude-modulated current signal. The modulation formula is as follows: ,in, This is the amplitude-modulated current signal; This is the current normalized signal; The target temperature is the normalized signal; k m k is the modulation coefficient. m =0.3, used to balance the weights of current and temperature. This formula reflects the negative feedback mechanism that needs to suppress current when the temperature rises, which is in line with the physical laws of LED thermal management.

[0137] In this embodiment, envelope detection is performed on the amplitude-modulated current signal to extract the envelope corresponding to the amplitude-modulated current signal. The digital envelope detection algorithm is used to extract the envelope of the amplitude-modulated current signal. The envelope simultaneously carries the time series correlation information of the dynamic change of the current and the temperature.

[0138] In this embodiment, the operating phase includes: a startup phase, a stabilization phase, and a shutdown phase; wherein, the startup phase: 0-1s after power-on, or the current rises from 0 to 0.9I. n Initially, the characteristics are a rapid increase in current and a slow increase in temperature. Stable phase: Current ≥ 0.9I. n Furthermore, the temperature fluctuation is ≤0.5℃ / s, characterized by a tendency for both current and temperature to stabilize with minor fluctuations. Turn-off section: Current ≤0.1In In the last 0-2 seconds, the characteristic is a rapid decrease in current and a slow decrease in temperature; I n This is the rated current for the LED.

[0139] In this embodiment, the normalized current signal and the normalized target temperature signal are fused in segments based on the fusion weights corresponding to each working stage to obtain the segmented fused signal corresponding to each working stage. Specifically: In the startup stage, current dominates the thermo-electric relationship, and weights are allocated accordingly. Weighted by current; Temperature weighting; =0.7、 =0.3, Start-up segment fusion signal ;in, This is the fusion signal for the startup segment; For the envelope of the modulated signal; This is the current normalized signal; The signal is normalized to the target temperature. Stable phase: thermal-electrical balance, weighted as follows: =0.5、 =0.5, stable segment fused signal ;in, For stable segment fusion signals; For the envelope of the modulated signal; This is the current normalized signal; The target temperature is the normalized signal; the off-section: temperature dominates the thermo-electric relationship, and the weighting is as follows: =0.3、 =0.7, cutoff segment fusion signal ,in, For the fusion signal of the shutdown segment; For the envelope of the modulated signal; This is the current normalized signal; This is the normalized signal for the target temperature.

[0140] The working principle and beneficial effects of the above technical solution are as follows: By performing sliding window averaging filtering on the current signal, ripple is effectively smoothed and the current signal is calibrated; hysteresis compensation is applied to the temperature signal, making it more accurately reflect the actual situation and improving signal reliability and accuracy; linear interpolation of the temperature signal is performed using the time axis of the calibrated current signal as a reference, and the rationality of the temperature interpolation point is verified through synchronous anchor points, ensuring accurate alignment of the temperature and current signals in the time series, laying the foundation for subsequent fusion processing; after normalizing the processed current and temperature signals, amplitude modulation and envelope detection are performed to extract characteristic parameters that reflect the dynamic thermo-electric working state of the LED module, comprehensively and accurately characterizing the LED's working characteristics; based on FPGA real-time analysis, different working stages of the LED are determined, and segmented signal fusion is performed with different fusion weights for each stage, enabling more detailed capture of thermo-electric characteristic changes at different stages; the characteristic parameters reflecting the dynamic thermo-electric working state of the LED module are fed back to the core control module to correct the initial duty cycle adjustment command, enabling the system to dynamically adjust the control strategy according to the actual working state of the LED, improving system stability and energy efficiency.

[0141] The adaptive adjustment module dynamically adjusts the single update step size of the duty cycle adjustment output to the PWM drive unit based on the changing trends of the historical adjustment frequency and historical adjustment amplitude of the duty cycle adjustment, including:

[0142] According to the preset sampling frequency, the real-time adjustment frequency data and real-time adjustment amplitude data of the duty cycle adjustment amount are obtained;

[0143] Acquire historical adjustment frequency data and historical adjustment amplitude data of duty cycle adjustment within a preset time window;

[0144] Calculate the historical frequency moving average and historical amplitude moving average of the duty cycle adjustment within the preset time window;

[0145] The frequency trend coefficient of duty cycle adjustment is determined based on real-time adjustment frequency data and historical frequency sliding average.

[0146] The amplitude trend coefficient of the duty cycle adjustment is determined based on real-time adjustment amplitude data and historical amplitude sliding average.

[0147] The dynamic step size correction factor is determined based on the frequency trend coefficient and amplitude trend coefficient of the duty cycle adjustment.

[0148] ;

[0149] in, This is the dynamic step size correction factor; Frequency trend weighting factor; This is the frequency trend coefficient; The magnitude trend weighting factor; This is the amplitude trend coefficient;

[0150] The single update step size of the duty cycle adjustment is calculated based on the dynamic step size correction factor and the preset basic step size.

[0151] ;

[0152] in, The single update step size for duty cycle adjustment; Use the base step size.

[0153] In this embodiment, to accurately capture the historical trend of duty cycle adjustment and suppress single-sampling interference, a time sliding window technique and a circular queue storage mechanism are introduced. The adaptive adjustment module allocates a circular queue of length M in RAM, where the size of M depends on the window time length and the sampling frequency. The adjustment frequency data and adjustment amplitude data of the duty cycle adjustment in the circular queue are used as the historical adjustment frequency data and historical adjustment amplitude data of the duty cycle adjustment, respectively. The historical frequency sliding average and historical amplitude sliding average of the duty cycle adjustment within the preset time window are calculated. Based on the real-time adjustment frequency data and the historical frequency sliding average, the frequency trend coefficient of the duty cycle adjustment is determined. ;in, This is the frequency trend coefficient; This is the frequency moving average of the most recent M duty cycle adjustments in the historical adjustment amplitude data; The average adjustment frequency for the current sampling period is used. By comparing the average frequency of the most recent M times with the average frequency of the current sampling period, the deviation of the current adjustment frequency from the historical trend is quantified. The qualitative logic that frequent adjustment equals increased system fluctuation and sparse adjustment equals system stability is transformed into a quantitative value of the frequency trend coefficient, providing a clear frequency dimension basis for step size adjustment.

[0154] Determine the amplitude trend coefficient of the duty cycle adjustment based on historical adjustment amplitude data of the duty cycle adjustment; ;in, This is the amplitude trend coefficient; This represents the moving average of the most recent M duty cycle adjustments in the historical adjustment amplitude data; This represents the average adjustment amplitude over the current sampling period. By comparing the average amplitude over the most recent M periods with the average amplitude over the current sampling period, the deviation of the current adjustment amplitude from the historical trend is quantified. This transforms the qualitative logic that a large adjustment amplitude equals increased system volatility and a small adjustment amplitude equals system stability into a quantitative value for the amplitude trend coefficient. This value is used to supplement the volatility characteristics not covered by the frequency dimension. Frequency and amplitude are the two core dimensions characterizing duty cycle adjustment volatility. Together, they can comprehensively cover the dynamic state of the system, avoiding the one-sidedness of judgment based on a single dimension.

[0155] In this embodiment, a dynamic step size correction factor is determined based on the frequency trend coefficient and amplitude trend coefficient of the duty cycle adjustment. The fluctuation characteristics of the duty cycle adjustment are jointly determined by the frequency change and amplitude change. The trend judgment in a single dimension has obvious limitations. By integrating the fluctuation characteristics of the two core dimensions into a unified dynamic step size correction factor, a comprehensive trend representation in both frequency and amplitude dimensions is achieved, ensuring that the system can accurately identify various fluctuation conditions and avoid step size adjustment deviations caused by misjudgment in a single dimension.

[0156] In this embodiment, the single update step size of the duty cycle adjustment is calculated based on a dynamic step size correction factor and a preset base step size. Using the base step size as the core, a dynamic step size correction factor is introduced, and lightweight nonlinear adaptation is achieved through square root operations. This ultimately achieves precise matching between the step size and the system's dynamic operating conditions, breaking the limitations of traditional fixed step sizes. Traditional fixed step sizes cannot adapt to system fluctuations; excessively large step sizes easily lead to overshoot, while excessively small step sizes result in slow response. The nonlinear characteristics of the square root allow the step size to contract rapidly during fluctuations and expand smoothly during stabilization, suppressing overshoot without sacrificing adjustment efficiency. The square root operation is a basic MCU operation, requiring no complex algorithm library, resulting in high computational efficiency and adapting to the real-time requirements of embedded systems.

[0157] The working principle and beneficial effects of the above technical solution are as follows: By acquiring real-time adjustment frequency and amplitude data according to a preset sampling frequency, and simultaneously acquiring historical data within a preset time window, the changes in duty cycle adjustment can be comprehensively grasped, providing a rich and accurate data foundation for subsequent analysis; the moving average of historical frequency and amplitude is calculated, and the frequency trend coefficient and amplitude trend coefficient are determined based on this, quantifying the changing trend of duty cycle adjustment, which facilitates precise dynamic adjustment in the future; the dynamic step size correction factor is determined based on the frequency trend coefficient and amplitude trend coefficient, which comprehensively considers the changing trends of duty cycle adjustment in frequency and amplitude. This system enables adaptive dynamic adjustment of the single update step size for duty cycle adjustment. Based on a dynamic step size correction factor and a preset base step size, the single update step size is calculated, allowing for flexible changes in the step size according to actual conditions. This avoids problems such as untimely or over-adjustment that may occur with a fixed step size. Adaptively adjusting the single update step size of the duty cycle adjustment allows for more precise adjustment under different operating conditions, reducing fluctuations during the adjustment process and improving system stability and reliability. Furthermore, it can adjust the step size promptly according to the changing trend of the duty cycle adjustment, enabling the system to respond more quickly and appropriately to various changes, thus improving the overall performance and efficiency of the system.

[0158] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0159] 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.

Claims

1. An LED lamp bead current control system, characterized in that, include: The signal acquisition unit is configured to acquire the power supply voltage of the LED module, the current flowing through the module, and the temperature of the module in real time, and output the corresponding voltage signal, current signal, and temperature signal. The intelligent control unit has a built-in preset control algorithm, which is configured to calculate the duty cycle adjustment of the PWM drive signal based on the voltage signal, current signal and temperature signal. The PWM drive unit is configured to dynamically adjust the duty cycle of the PWM drive signal output to the LED module according to the calculated duty cycle adjustment amount, so as to realize constant current control of the LED module; The multi-level compensation unit is configured to perform current fluctuation compensation and thermal drift compensation. The current fluctuation compensation is based on the change trend of the voltage signal and the position information of the module in the whole screen, and performs reverse compensation on the duty cycle adjustment in advance to eliminate the brightness unevenness caused by line voltage drop. The thermal drift compensation is based on the historical change data of temperature signal and current signal to perform proportional-integral compensation on the duty cycle adjustment. The protection control unit includes an over-temperature protection module and a soft-start module; The over-temperature protection module is configured to control the PWM drive unit to reduce the output power when the temperature signal exceeds a first preset threshold, and to turn off the PWM drive signal output when the temperature signal exceeds a second preset threshold. The soft-start module is configured to control the duty cycle of the PWM drive unit output to increase linearly from zero to an initial set value with a first slope when the system starts or restarts, wherein the first slope is dynamically adjusted according to the temperature signal.

2. The LED lamp bead current control system according to claim 1, characterized in that, The intelligent control unit includes: The core control module is configured to receive voltage signals, current signals, and temperature signals, and run a preset control algorithm to output an initial duty cycle adjustment command. The signal fusion module is configured to perform time-series alignment and weighted fusion of temperature and current signals to generate characteristic parameters that reflect the dynamic thermo-electric working state of the LED module, and feed these characteristic parameters back to the core control module to correct the initial duty cycle adjustment command. The adaptive adjustment module is configured to dynamically adjust the single update step size of the duty cycle adjustment output to the PWM drive unit based on the changing trend of the historical adjustment frequency and historical adjustment amplitude of the duty cycle adjustment.

3. The LED lamp bead current control system according to claim 2, characterized in that, The intelligent control unit incorporates a mode switching mechanism, automatically switching between a conventional constant current mode, a fast response mode, and a high stability mode based on the stability of the voltage signal, the rate of change of the temperature signal, and the ripple coefficient of the current signal. In fast response mode, the core control module prioritizes increasing the control bandwidth, while the adaptive adjustment module temporarily increases the update step size. In high stability mode, the core control module prioritizes reducing the control bandwidth to suppress noise, the adaptive adjustment module locks to the minimum update step size, and enables proportional-integral compensation in the multi-level compensation module as the main adjustment method.

4. The LED lamp bead current control system according to claim 1, characterized in that, The PWM drive unit includes: The signal isolation module is configured to receive the PWM logic signal corresponding to the duty cycle adjustment output by the intelligent control unit, and to electrically isolate the received PWM logic signal through the optocoupler integrated inside the signal isolation module. This is used to eliminate electrical noise interference and ground loop effects between the main control side and the power side. At the same time, it converts the voltage level of the received PWM logic signal into a level standard suitable for driving the subsequent power stage circuit, and outputs an isolated and level-matched drive logic signal. The drive enhancement module is configured to receive drive logic signals, provide drive currents on rising and falling edges, and output enhanced drive signals. The power switch module is configured to receive the enhanced drive signal and adopts a synchronous rectification topology consisting of a power MOSFET and a freewheeling diode. It is used to generate and output the PWM drive signal to drive the LED module by turning the enhanced drive signal on or off.

5. The LED lamp bead current control system according to claim 1, characterized in that, The current fluctuation compensation in the multi-level compensation unit includes: The feedforward compensation module is configured to monitor the voltage signal representing the power supply voltage of the LED module in real time. When the change in the voltage signal within adjacent sampling periods exceeds the preset sudden change threshold, it determines that a voltage sudden change has occurred and generates a duty cycle feedforward adjustment command based on the amplitude and direction of the sudden change. The trend prediction module is configured to store historical voltage signal values ​​for N consecutive sampling periods in a data buffer, perform linear fitting on the voltage data in the buffer, and calculate the average slope and direction of voltage change within the corresponding time period of the N sampling periods. This is used to predict the voltage fluctuation trend in the next control period. If the predicted trend is a continuous gradual increase, a gradually decreasing duty cycle pre-compensation amount is generated; if the predicted trend is a continuous gradual decrease, a gradually increasing duty cycle pre-compensation amount is generated. Here, N is an integer greater than or equal to 2. The voltage drop compensation module is configured to calculate the voltage drop of the LED module relative to the power input terminal based on the physical position coordinates of the LED module in the display screen, the length of the power supply line, and the line impedance parameters, and generate a corresponding duty cycle compensation amount to compensate for the brightness attenuation of the far-end module caused by the line impedance.

6. The LED lamp bead current control system according to claim 1, characterized in that, Thermal drift compensation in the multi-level compensation unit includes: The data modeling module is configured to establish an equivalent thermal model that reflects the hysteresis characteristics of LED module junction temperature changes based on historical time-series data of temperature and current signals. The equivalent thermal model simulates and outputs the predicted LED junction temperature based on real-time heating power and ambient temperature. The adaptive compensation module is configured to dynamically adjust the proportional coefficient and integral time constant in the proportional-integral compensation based on the predicted junction temperature output by the equivalent thermal model and the real-time deviation of the current signal relative to the set value. This is used to achieve nonlinear compensation for LED forward voltage drop drift and luminous efficiency decay caused by temperature gradient.

7. The LED lamp bead current control system according to claim 1, characterized in that, The over-temperature protection module includes: The graded response module is configured to control the PWM drive unit to reduce the output power in a stepwise manner at a first cooling rate when the temperature signal exceeds a first preset threshold, and to restore the output power in a stepwise manner at a first recovery rate after the temperature signal falls back below the first preset threshold. The thermal accumulation monitoring module is configured to calculate in real time the cumulative time of the LED module at temperatures above the third preset threshold, and dynamically adjust the first preset threshold and the second preset threshold according to the length of the cumulative time.

8. The LED lamp bead current control system according to claim 1, characterized in that, The soft-start module includes: The slope adaptive module is configured to dynamically determine the first slope based on the difference between the temperature signal at the moment of system startup and the rated operating temperature. When the difference is positive, the first slope is positively correlated with the magnitude of the difference. The status feedback module is configured to monitor the rise rate of the current signal and the fall rate of the voltage signal in real time during the soft start process, and dynamically fine-tune the first slope based on the deviation between the rise rate of the current signal and the fall rate of the voltage signal and the expected rate, until the duty cycle reaches the initial set value and the current signal stabilizes at the target constant current value.

9. The LED lamp bead current control system according to claim 2, characterized in that, The signal fusion module performs time-series alignment and weighted fusion of temperature and current signals to generate characteristic parameters reflecting the dynamic thermo-electric operating state of the LED module, including: The ripple of the current signal is smoothed by sliding window averaging filter to obtain the calibrated current signal; the temperature lag compensation time of the temperature signal is calculated based on the instantaneous rate of change of the current, and the temperature signal is compensated based on the temperature lag compensation time to obtain the compensated temperature signal. Using the time axis of the calibrated current signal as a reference, the compensated temperature signal is linearly interpolated to generate a temperature interpolation signal; the peak point of the current in the calibrated current signal is selected as the synchronization anchor point to verify the rationality of the temperature interpolation point, and the target temperature interpolation signal is obtained after the verification is passed. The calibrated current signal and the target temperature interpolation signal are normalized respectively to obtain the normalized current signal and the normalized target temperature signal. The normalized current signal is amplitude modulated using the normalized current signal as the carrier and the normalized target temperature signal as the modulation signal to obtain the amplitude modulated current signal. Envelope detection is performed on the amplitude-modulated current signal to extract the envelope corresponding to the amplitude-modulated current signal; The LED's operating stage is determined by real-time analysis of current and temperature signals using FPGA. The operating stage includes a startup stage, a stabilization stage, and a shutdown stage. Different fusion weights are set for the startup stage, the stabilization stage, and the shutdown stage. Based on the fusion weights corresponding to each operating stage, the normalized current signal and the normalized target temperature signal are fused in segments to obtain the segmented fused signal corresponding to each operating stage. Feature extraction is performed on the segmented fusion signals of the start-up phase, the stable phase, and the turn-off phase to obtain characteristic parameters reflecting the dynamic thermo-electric working state of the LED module. The characteristic parameters reflecting the dynamic thermo-electric operating state of the LED module are fed back to the core control module to correct the initial duty cycle adjustment command.

10. The LED lamp bead current control system according to claim 2, characterized in that, The adaptive adjustment module dynamically adjusts the single update step size of the duty cycle adjustment output to the PWM drive unit based on the changing trends of the historical adjustment frequency and historical adjustment amplitude of the duty cycle adjustment, including: According to the preset sampling frequency, the real-time adjustment frequency data and real-time adjustment amplitude data of the duty cycle adjustment amount are obtained; Acquire historical adjustment frequency data and historical adjustment amplitude data of duty cycle adjustment within a preset time window; Calculate the historical frequency moving average and historical amplitude moving average of the duty cycle adjustment within the preset time window; The frequency trend coefficient of duty cycle adjustment is determined based on real-time adjustment frequency data and historical frequency sliding average. The amplitude trend coefficient of the duty cycle adjustment is determined based on real-time adjustment amplitude data and historical amplitude sliding average. The dynamic step size correction factor is determined based on the frequency trend coefficient and amplitude trend coefficient of the duty cycle adjustment. ； in, This is the dynamic step size correction factor; Frequency trend weighting factor; This is the frequency trend coefficient; The magnitude trend weighting factor; This is the amplitude trend coefficient; The single update step size of the duty cycle adjustment is calculated based on the dynamic step size correction factor and the preset basic step size. ； in, The single update step size for duty cycle adjustment; Use the base step size.

Images