An adaptive dimming method for multi-color LED devices based on IC control

Abstract

The present invention belongs to the technical field of LED dimming. The present invention provides an adaptive dimming method for a multi-color LED device based on IC control, comprising: identifying a driving current fluctuation period, determining whether stable periodic LED driving current fluctuations occur, determining whether the power supply ripple interferes with the PWM signal output through the duty cycle change of the IC's PWM signal and the phase relationship of the power supply ripple, and if interference occurs, determining whether the power supply ripple interferes with the LED driving current by being superimposed on the IC's PWM signal through fluctuation frequency consistency analysis of the LED driving current and the power supply ripple, and waveform synchronization analysis of the LED driving current and the PWM signal, and if interference occurs, determining the next driving current fluctuation period, predicting the ripple voltage peak and ripple voltage trough, and dynamically pre-compensating the PWM duty cycle of the ripple phase. The present invention solves the problem of switching power supply ripple interfering with the IC's PWM output, resulting in brightness fluctuations of multi-color LEDs.

Description

Technical Field

[0001] The present invention belongs to the technical field of LED dimming, and in particular to an adaptive dimming method for a multi-color LED device based on IC control. Background Art

[0002] In the field of LED lighting and displays, PWM (pulse width modulation) dimming technology based on integrated circuits (ICs) is widely used in smart homes, automotive ambient lighting, stage lighting, and other applications due to its ability to achieve high-precision brightness and color control. However, the ripple voltage output by the switching power supply (such as 50 / 100Hz periodic fluctuations) can easily be superimposed on the IC's PWM drive signal, causing periodic fluctuations in the LED drive current, which in turn can lead to problems such as brightness "breathing" and color drift. For example, when dimming RGB lamps, warm white light may appear orange and cool white light may appear bluish due to ripple interference, seriously affecting the consistency of light color and user experience. Existing technologies tend to ignore the monitoring of switching power supply ripple interference, and in terms of suppression, they mostly stay at passive filtering (such as LC circuits) or single-dimensional compensation (such as fixed phase adjustment), lacking a systematic processing of the entire process of "fluctuation detection-phase analysis-path identification-intelligent prediction-dynamic compensation". Its core defects are: no quantitative fault diagnosis system has been established, making it difficult to accurately locate ripple interference; lack of intelligent prediction capabilities for ripple dynamic characteristics, making it impossible to achieve advanced compensation; the compensation strategy is not deeply coupled with phase correlation, and has poor adaptability.

[0003] To this end, the present invention provides an adaptive dimming method for a multi-color LED device based on IC control. Summary of the Invention

[0004] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0005] The technical solution adopted by the present invention to solve the technical problem is: an adaptive dimming method for a multi-color LED device based on IC control, comprising:

[0006] During the dimming management cycle, based on the LED driving current of the multi-color LED device, the driving current fluctuation period is identified, and the stability analysis of the driving current fluctuation period is performed to determine whether the multi-color LED device has stable periodic LED driving current fluctuation during the dimming process;

[0007] If stable periodic LED drive current fluctuations occur, the phase relationship between the IC's PWM signal duty cycle and the power ripple can be used to determine whether the power ripple is interfering with the PWM signal output.

[0008] If the power ripple interferes with the PWM signal output, the consistency of the fluctuation frequency of the LED drive current and the power ripple, as well as the waveform synchronization of the LED drive current and the PWM signal, can be used to determine whether the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal.

[0009] If the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal, the next drive current fluctuation period is determined. The ripple voltage peak sequence and ripple voltage trough sequence are obtained by power ripple extraction. The ripple voltage peak and ripple voltage trough sequence within the next drive current fluctuation period are predicted using the LSTM model.

[0010] According to the predicted ripple voltage peak and ripple voltage trough in the next driving current fluctuation period, dynamic pre-compensation of the PWM duty cycle of the ripple phase is performed.

[0011] Furthermore, the method of identifying the driving current fluctuation period is:

[0012] The LED driving current of the multi-color LED device is obtained and compared with the preset LED driving current. The driving current fluctuation period is obtained by taking the deviation of the LED driving current from the preset LED driving current as the starting time point and the recovery of the LED driving current to the preset LED driving current after deviating from the preset LED driving current as the ending time point.

[0013] Furthermore, the process of performing stability analysis during the driving current fluctuation period is as follows:

[0014] In the dimming management cycle, the duration corresponding to each driving current fluctuation period is obtained and integrated to obtain a driving fluctuation period sequence, and the interval duration between adjacent driving current fluctuation periods is obtained and integrated to obtain a driving fluctuation interval sequence;

[0015] Calculate the coefficient of variation of the driving fluctuation period series and the driving fluctuation interval series respectively, and sum them up to output the driving fluctuation value;

[0016] If the driving fluctuation value meets the requirements, it means that stable periodic LED driving current fluctuation occurs.

[0017] Furthermore, the process of determining whether the power ripple interferes with the PWM signal output is as follows:

[0018] Calculate the phase difference between the power supply ripple signal and the PWM duty cycle change at several different time points during the driving current fluctuation period. Through phase difference analysis, obtain the fixed phase compliance value and the non-fixed phase deviation value, perform deviation processing, and output the phase relationship value;

[0019] If the phase relationship value meets the requirement, it means that the power supply ripple interferes with the PWM signal output.

[0020] Furthermore, the phase difference analysis process is:

[0021] By comparing the phase differences, the compliant phase differences and the non-compliant phase differences are identified;

[0022] Counting the proportion of the number of phase differences that meet the requirements, and obtaining a fixed phase compliance value;

[0023] Based on the non-coinciding phase difference, the absolute deviation of each non-coinciding fixed phase difference from the fixed phase difference is calculated, and the average calculation is performed to obtain the mean of the phase absolute deviation. The ratio of the mean of the phase absolute deviation to the fixed phase difference is calculated to obtain the non-fixed phase deviation value.

[0024] Furthermore, the phase difference comparison process is:

[0025] If the phase difference is within the phase difference standard range, the phase difference is marked as a compliant phase difference;

[0026] If the phase difference is not within the phase difference standard range, the phase difference is marked as a non-compliant phase difference.

[0027] Furthermore, the process of determining whether the power ripple interferes with the LED drive current by being superimposed on the PWM signal of the IC is as follows:

[0028] Synchronously collect LED drive current I LED (t) and power supply ripple voltage V ripple (t), where t represents the acquisition time;

[0029] LED drive current I LED (t) and power supply ripple voltage V ripple (t) Perform FFT Fourier transform to extract the LED driving current main frequency f LED and power supply ripple voltage main frequency f ripple ;

[0030] Calculating frequency deviation f=f LED -f ripple ;

[0031] Synchronously collect PWM duty cycle signal D(t) and LED drive current I LED (t);

[0032] Calculate the PWM duty cycle signal D(t) and LED drive current I LED The cross-correlation coefficient R of (t) DI :

[0033] , where Cov represents covariance, Decibel represents the standard deviation of the PWM duty cycle signal D(t) and the LED drive current I LED Standard deviation of (t);

[0034] If the frequency deviation f is less than or equal to the frequency deviation threshold f threshold , and the mutual correlation coefficient R DI Greater than or equal to the cross-correlation coefficient threshold R threshold , it means that the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal.

[0035] Furthermore, the next driving current fluctuation period is determined as follows:

[0036] The driving fluctuation period series and the driving fluctuation interval series are averaged to obtain the average driving fluctuation duration and the average driving fluctuation interval duration;

[0037] Determine the start time of the next driving current fluctuation period according to the current time point and the average driving fluctuation interval duration;

[0038] Determining the end time point of the next driving current fluctuation period according to the start time point of the next driving current fluctuation period and the average duration of the driving fluctuation;

[0039] The next driving current fluctuation period is determined according to the start time point and the end time point of the next driving current fluctuation period.

[0040] Furthermore, the process of predicting the ripple voltage peak and the ripple voltage trough in the next driving current fluctuation period is as follows:

[0041] The ripple voltage peaks and ripple voltage troughs on the power supply ripple are extracted respectively, and are integrated in sequence according to the time sequence to obtain the ripple voltage peak sequence and the ripple voltage trough sequence;

[0042] A time difference sequence is obtained according to the ripple voltage peak sequence and the ripple voltage trough sequence, wherein the time difference sequence includes an adjacent peak time difference sequence and an adjacent trough time difference sequence, the adjacent peak time difference sequence is composed of the time intervals between adjacent peaks, and the adjacent trough time difference sequence is composed of the time intervals between adjacent troughs;

[0043] Input features: time difference series and amplitude series;

[0044] Output target: time difference of the next peak / trough and amplitude of the next peak / trough;

[0045] Time difference series prediction: Use the pre-trained LSTM model to predict the time difference between the next peak / trough;

[0046] Amplitude sequence prediction: Use pre-trained LSTM to predict the amplitude of the next peak / trough;

[0047] According to the predicted time difference, the time of the next peak / valley is accumulated and the prediction operation is continued until the end of the driving current fluctuation period, thereby completing the prediction of the ripple voltage peak and ripple voltage trough in the next driving current fluctuation period.

[0048] Furthermore, the process of dynamic pre-compensation of the PWM duty cycle of the ripple phase is as follows: obtaining the ripple peak value of the ripple voltage peak in the next driving current fluctuation period, calculating the ripple voltage V ripple (t)=V peak *sin(2πft), where V peak is the ripple peak value, f is the ripple frequency, t is the ripple peak moment, sin(2πft) is the sine representation, and 2πft is the phase angle of the sine function;

[0049] Calculate the PWM duty cycle pre-compensation amount: ΔD(t)=V ripple (t) / V DD ×K, where V DD Expressed as IC power supply voltage, K is the compensation coefficient;

[0050] The phase relationship between the PWM duty cycle change and the ripple is determined by the phase difference standard range. If there is a positive correlation, the duty cycle is increased when the ripple voltage is at the peak and the duty cycle is reduced when the ripple voltage is at the trough according to the PWM duty cycle pre-compensation amount. If there is a negative correlation, reverse compensation is performed.

[0051] The beneficial effects of the present invention are as follows: within the dimming management cycle, according to the LED driving current of the multi-color LED device, the driving current fluctuation period is identified, and the stability analysis of the driving current fluctuation period is performed to determine whether the multi-color LED device has a stable periodic LED driving current fluctuation phenomenon during the dimming process. If a stable periodic LED driving current fluctuation phenomenon occurs, it is determined whether the power supply ripple interferes with the PWM signal output through the duty cycle change of the IC's PWM signal and the phase relationship between the power supply ripple. If the power supply ripple interferes with the PWM signal output, it is determined whether the power supply ripple interferes with the LED driving current by being superimposed on the IC's PWM signal through the fluctuation frequency consistency analysis of the LED driving current and the power supply ripple and the waveform synchronization analysis of the LED driving current and the PWM signal. If the power ripple interferes with the LED drive current by being superimposed on the PWM signal of the IC, the next drive current fluctuation period is determined, and a ripple voltage peak sequence and a ripple voltage trough sequence are obtained by extracting the power ripple. The ripple voltage peak and the ripple voltage trough in the next drive current fluctuation period are predicted in combination with the LSTM model. Based on the predicted ripple voltage peak and ripple voltage trough in the next drive current fluctuation period, dynamic pre-compensation of the PWM duty cycle of the ripple phase is performed to eliminate the superimposed influence of the power ripple on the LED drive current when the multi-color LED device is dimmed. The present invention solves the problem that the switching power supply ripple interferes with the PWM output of the IC, resulting in brightness fluctuations of the multi-color LED (such as the color "breathing" phenomenon of RGB lamps during dimming), thereby improving the dimming quality of the multi-color LED. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The present invention will be further described below with reference to the accompanying drawings.

[0053] Figure 1 This is a flowchart of the steps of an IC-controlled adaptive dimming method for a multi-color LED device according to an embodiment of the present invention;

[0054] Figure 2 This is a logic judgment diagram for power supply ripple interference in an adaptive dimming method for a multi-color LED device based on IC control according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0056] See also Figure 1 As shown, an adaptive dimming method for a multi-color LED device based on IC control according to an embodiment of the present invention includes the following steps:

[0057] Step 1: During the dimming management cycle, based on the LED drive current of the multi-color LED device, identify the drive current fluctuation period and perform stability analysis during the drive current fluctuation period to determine whether the multi-color LED device exhibits stable periodic LED drive current fluctuation during the dimming process;

[0058] In step 1, the dimming management cycle is the basic time unit for controlling LED brightness during dimming. Current sampling, oscilloscope observation, and spectrum analysis can be used to detect driving current fluctuations to distinguish between normal dimming phenomena and abnormal faults.

[0059] In step 1, based on the LED driving current of the multi-color LED device, the method of identifying the driving current fluctuation period is as follows:

[0060] During the dimming management cycle, an LED driving current of the multi-color LED device is obtained and compared with a preset LED driving current. A driving current fluctuation period is obtained by taking a deviation of the LED driving current from the preset LED driving current as a starting time point and taking a recovery of the LED driving current to the preset LED driving current after the deviation from the preset LED driving current as an end time point. Specifically, the driving current fluctuation period represents a time period during which the LED driving current of the multi-color LED device deviates from the preset LED driving current and then recovers to the preset LED driving current.

[0061] In step 1, the process of performing stability analysis during the driving current fluctuation period is as follows:

[0062] In the dimming management cycle, the duration corresponding to each driving current fluctuation period is obtained and integrated to obtain a driving current fluctuation period sequence;

[0063] In the dimming management cycle, the interval lengths between adjacent driving current fluctuation periods are obtained and integrated to obtain a driving current fluctuation interval sequence;

[0064] Calculate the coefficient of variation of the driving fluctuation period series and the driving fluctuation interval series respectively, sum them up, and output the driving fluctuation value;

[0065] Preferably, the driving fluctuation value is compared with a driving fluctuation threshold;

[0066] If the driving fluctuation value is greater than or equal to the driving fluctuation threshold, it means that there is no stable periodic LED driving current fluctuation phenomenon;

[0067] If the driving fluctuation value is less than the driving fluctuation threshold, it indicates that stable periodic LED driving current fluctuation occurs;

[0068] It should be noted that the drive fluctuation value is calculated by the coefficient of variation of the drive fluctuation period sequence and the drive fluctuation interval sequence. The coefficients of variation of the drive fluctuation period sequence and the drive fluctuation interval sequence respectively reflect the stability of each fluctuation duration of the LED drive current and the stability of the interval duration between current fluctuations. The smaller the drive fluctuation value, the more stable the LED drive current fluctuation duration and the more stable the fluctuation interval.

[0069] It can be understood that by calculating the coefficient of variation of the drive fluctuation period sequence and the interval sequence (step 1), the stability of the current fluctuation is converted into a quantifiable drive fluctuation value, which is more objective than traditional oscilloscope observation. For example, when the drive fluctuation value is less than the threshold, it can be clearly determined that the fluctuation has stable periodicity, avoiding the misinterpretation of random noise as ripple interference;

[0070] Step 2: If stable periodic LED drive current fluctuations occur, determine whether the power supply ripple interferes with the PWM signal output by comparing the duty cycle of the IC's PWM signal with the phase relationship of the power supply ripple.

[0071] In step 2, the power supply ripple signal (channel 1) and the PWM signal (channel 2) are displayed on the oscilloscope screen at the same time. The oscilloscope's "phase measurement" function is used to calculate the phase difference between the power supply ripple signal and the PWM duty cycle change at several different time points during the drive current fluctuation period.

[0072] Compare the phase difference with a standard range of phase differences;

[0073] If the phase difference is within the phase difference standard range, the phase difference is marked as a compliant phase difference;

[0074] If the phase difference is not within the phase difference standard range, the phase difference is marked as a non-compliant phase difference;

[0075] It should be noted that the power supply ripple interferes with the PWM signal output. Therefore, during the driving current fluctuation period, the phase difference between the power supply ripple signal and the PWM duty cycle change will remain in a fixed range rather than fluctuating randomly. For example, the data fluctuation is ≤±5°, and the phase difference is close to 0° or 180° (such as 0°±3°, 180°±4°). It should also be noted that 0° or 180° represents a fixed phase difference. There are two fixed phase differences, which are determined by the phase relationship between the PWM duty cycle change and the ripple: positive correlation: the rising edge of the power supply ripple corresponds to an increase in the PWM duty cycle (phase difference 0°), and negative correlation: the rising edge of the ripple corresponds to a decrease in the PWM duty cycle (phase difference 180°).

[0076] Counting the proportion of the number of phase differences that meet the requirements, and obtaining a fixed phase compliance value;

[0077] Based on the non-coinciding phase difference, the absolute deviation of each non-coinciding fixed phase difference from the fixed phase difference is calculated, and the average calculation is performed to obtain the mean of the phase absolute deviation. The ratio of the mean of the phase absolute deviation to the fixed phase difference is calculated to obtain the non-fixed phase deviation value;

[0078] Calculate the difference between the fixed phase coincidence value and the non-fixed phase deviation value to obtain the phase relationship value;

[0079] Preferably, the phase relationship value is compared with a phase relationship threshold;

[0080] If the phase relationship value is greater than or equal to the phase relationship threshold, it means that the power supply ripple interferes with the PWM signal output;

[0081] If the phase relationship value is less than the phase relationship threshold, it indicates that the power supply ripple does not interfere with the PWM signal output. Other causes of LED drive current fluctuations should be analyzed, including but not limited to: feedback loop response speed analysis, LED forward voltage drop discreteness analysis, and analog control circuit temperature drift analysis.

[0082] It is understandable that the purpose of obtaining the phase relationship value and determining whether the power supply ripple interferes with the PWM signal output is:

[0083] Function 1: The phase relationship value can be used to identify whether power supply ripple interferes with the PWM signal output when LED drive current fluctuation occurs in multi-color LED devices. This is beneficial for preliminary analysis of the cause of the LED drive current fluctuation, making it easier to more accurately identify the root cause of the LED drive current fluctuation.

[0084] Function 2: The phase relationship value can be used to identify whether power ripple is interfering with the PWM signal output when LED drive current fluctuations occur in multi-color LED devices. After the power ripple interferes with the PWM signal output, further analysis can be performed to determine whether the power ripple is interfering with the LED drive current by being superimposed on the IC's PWM signal. This helps to effectively and quickly identify the root cause of LED drive current fluctuations. In addition, by quantifying the phase relationship value, it can quickly distinguish current fluctuations caused by power ripple interference from other causes (such as feedback loops, LED voltage drop discreteness, etc.);

[0085] Function 3: If it is subsequently determined that the power supply ripple interferes with the LED drive current by being superimposed on the IC's PWM signal, the phase relationship value is the basis for subsequent ripple phase compensation (step 5). The compensation algorithm needs to dynamically adjust the PWM duty cycle based on the phase difference distribution;

[0086] Step 3: If the power ripple interferes with the PWM signal output, determine whether the power ripple interferes with the LED drive current by superimposing it on the IC's PWM signal. This is done by analyzing the consistency of the fluctuation frequency between the LED drive current and the power ripple, as well as the waveform synchronization between the LED drive current and the PWM signal.

[0087] It can be understood that the phase correlation between ripple and PWM is converted into a specific value by calculating the difference between the fixed phase coincidence value and the non-fixed phase deviation value (phase relationship value). For example, when the phase relationship value is greater than or equal to the threshold, it can be determined that ripple interference with PWM exists, avoiding misjudgment of phase fluctuations as other factors (such as feedback loop delay).

[0088] like Figure 2 As shown in Figure 2, in step 3, the consistency analysis of the fluctuation frequency of the LED drive current and the power supply ripple is:

[0089] Use a high-speed ADC to synchronously acquire the LED drive current I LED (t) and power supply ripple voltage V ripple (t), where t represents the acquisition time;

[0090] LED drive current I LED (t) and power supply ripple voltage V ripple (t) Perform FFT Fourier transform to extract the LED driving current main frequency f LED and power supply ripple voltage main frequency f ripple ;

[0091] Calculating frequency deviation f=f LED -f ripple ;

[0092] like Figure 2 As shown, in step 3, the waveform synchronization analysis of the LED drive current and the PWM signal is as follows:

[0093] Synchronously collect PWM duty cycle signal D(t) and LED drive current I LED (t);

[0094] Calculate the PWM duty cycle signal D(t) and LED drive current I LED The cross-correlation coefficient R of (t) DI :

[0095] , where Cov represents covariance, They represent the standard deviation of the PWM duty cycle signal D(t) and the LED drive current I LED Standard deviation of (t);

[0096] like Figure 2 As shown, in step 3, the method for determining whether the power supply ripple interferes with the LED drive current by being superimposed on the IC's PWM signal is as follows:

[0097] If the frequency deviation f is less than or equal to the frequency deviation threshold f threshold , and the mutual correlation coefficient R DI Greater than or equal to the cross-correlation coefficient threshold R threshold , it means that the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal. Conversely, it means that the power ripple does not interfere with the LED drive current by being superimposed on the IC's PWM signal.

[0098] It can be understood that frequency consistency analysis (step 3) and cross-correlation coefficient calculations ensure that the coupling path between the ripple interference and the PWM signal is accurately identified and coordinated with dynamic compensation;

[0099] Step 4: If the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal, the next drive current fluctuation period is determined. The ripple voltage peak sequence and ripple voltage trough sequence are obtained by extracting the power ripple. The ripple voltage peak and ripple voltage trough sequence within the next drive current fluctuation period are predicted using the LSTM model.

[0100] In step 4, the next driving current fluctuation period is determined as follows:

[0101] The driving fluctuation period series and the driving fluctuation interval series are averaged to obtain the average driving fluctuation duration and the average driving fluctuation interval duration;

[0102] Determine the start time of the next driving current fluctuation period according to the current time point and the average driving fluctuation interval duration, that is: the start time of the next driving current fluctuation period = the current time point + the average driving fluctuation interval duration;

[0103] Determine the end time point of the next driving current fluctuation period according to the start time point of the next driving current fluctuation period and the average duration of the driving fluctuation, that is, the end time point of the next driving current fluctuation period = the start time point of the next driving current fluctuation period + the average duration of the driving fluctuation;

[0104] Determining the next driving current fluctuation period according to the starting time point and the ending time point of the next driving current fluctuation period;

[0105] In step 4, the ripple voltage peak sequence and the ripple voltage trough sequence are obtained by extracting the power ripple, and the ripple voltage peak and the ripple voltage trough in the next driving current fluctuation period are predicted as follows:

[0106] Obtaining the power supply ripple, extracting the ripple voltage peak and ripple voltage trough on the power supply ripple respectively, and integrating them in sequence according to the time sequence to obtain the ripple voltage peak sequence and the ripple voltage trough sequence;

[0107] Based on the ripple voltage peak sequence and the ripple voltage trough sequence, the ripple voltage peak and the ripple voltage trough in the next driving current fluctuation period are predicted respectively in combination with the LSTM model;

[0108] Exemplarily, the process of predicting the ripple voltage peak and ripple voltage trough in the next driving current fluctuation period in combination with the LSTM model includes:

[0109] A time difference sequence is obtained according to the ripple voltage peak sequence and the ripple voltage trough sequence, wherein the time difference sequence includes an adjacent peak time difference sequence and an adjacent trough time difference sequence. Specifically, the adjacent peak time difference sequence includes the time intervals of all adjacent peaks, and the adjacent trough time difference sequence includes the time intervals of all adjacent troughs;

[0110] Input features: time difference series (time intervals between adjacent peaks / troughs) and amplitude series (amplitudes of peaks or troughs);

[0111] Output target: time difference of the next peak / trough and amplitude of the next peak / trough;

[0112] Time difference series prediction: Use the pre-trained LSTM model to predict the time difference between the next peak / trough;

[0113] Amplitude sequence prediction: Use pre-trained LSTM to predict the amplitude of the next peak / trough;

[0114] According to the predicted time difference, the time of the next peak / valley is accumulated and the prediction operation is continued until the driving current fluctuation period ends, thereby completing the prediction of the ripple voltage peak and ripple voltage trough in the next driving current fluctuation period;

[0115] It is understandable that predicting the driving current fluctuation period (step 4) and performing compensation in advance can reduce the system response delay. For example, when the automatic sensing street light is blocked by clouds, the brightness adjustment delay is shortened to avoid visual discomfort to the human eye.

[0116] Step 5: Based on the predicted ripple voltage peak and ripple voltage trough during the next driving current fluctuation period, dynamic pre-compensation of the PWM duty cycle of the ripple phase is performed to eliminate the superimposed effect of the power supply ripple on the LED driving current when the multi-color LED device is dimmed;

[0117] In step 5, the process of dynamic pre-compensation of the PWM duty cycle of the ripple phase is as follows:

[0118] Get the peak value of the ripple voltage peak during the next driving current fluctuation period and calculate the ripple voltage V ripple (t)=V peak *sin(2πft), where V peak is the ripple peak value, f is the ripple frequency, t is the ripple peak moment, sin(2πft) is the sine representation, and 2πft is the phase angle of the sine function;

[0119] Calculate the PWM duty cycle pre-compensation amount: ΔD(t)=V ripple (t) / V DD ×K, where V DD Expressed as IC power supply voltage, K is the compensation coefficient (usually 0.8-1.2);

[0120] If the PWM duty cycle change is positively correlated with the phase relationship of the ripple, then according to the PWM duty cycle pre-compensation amount, when the ripple voltage is at the peak (positive maximum value), the duty cycle is increased to increase the output energy to offset the positive fluctuation of the ripple. When the ripple voltage is at the trough (negative maximum value), the duty cycle is reduced to offset the negative fluctuation of the ripple.

[0121] If the PWM duty cycle change is negatively correlated with the phase relationship of the ripple, then when the ripple voltage is at the peak (positive maximum), the duty cycle is reduced, and when the ripple voltage is at the trough (negative maximum), the duty cycle is increased;

[0122] It can be understood that dynamically adjusting the compensation direction (increasing / decreasing the duty cycle at the peak) based on the positive / negative correlation characteristics is more accurate than a fixed compensation strategy;

[0123] The technical solution of the embodiment of the present invention is as follows: within the dimming management cycle, according to the LED driving current of the multi-color LED device, the driving current fluctuation period is identified, and the stability analysis of the driving current fluctuation period is performed to determine whether the multi-color LED device has a stable periodic LED driving current fluctuation phenomenon during the dimming process; if the stable periodic LED driving current fluctuation phenomenon occurs, then the duty cycle change of the IC's PWM signal and the phase relationship of the power supply ripple are used to determine whether the power supply ripple interferes with the PWM signal output; if the power supply ripple interferes with the PWM signal output, then the consistency analysis of the fluctuation frequency of the LED driving current and the power supply ripple and the waveform synchronization analysis of the LED driving current and the PWM signal are used to determine whether the power supply ripple interferes with the LED driving current by being superimposed on the IC's PWM signal Interference: If the power ripple interferes with the LED driving current by being superimposed on the PWM signal of the IC, the next driving current fluctuation period is determined, and the ripple voltage peak sequence and the ripple voltage trough sequence are obtained by extracting the power ripple. The ripple voltage peak and the ripple voltage trough in the next driving current fluctuation period are predicted in combination with the LSTM model. According to the predicted ripple voltage peak and ripple voltage trough in the next driving current fluctuation period, the PWM duty cycle of the ripple phase is dynamically pre-compensated to eliminate the superimposed influence of the power ripple on the LED driving current when the multi-color LED device is dimmed. The present invention solves the problem that the switching power supply ripple interferes with the PWM output of the IC, resulting in brightness fluctuations of the multi-color LED (such as the color "breathing" phenomenon of RGB lamps during dimming), thereby improving the dimming quality of the multi-color LED.

[0124] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for adaptive dimming of a multi-color LED device based on IC control, characterized by: include: During the dimming management cycle, based on the LED driving current of the multi-color LED device, the driving current fluctuation period is identified, and the stability analysis of the driving current fluctuation period is performed to determine whether the multi-color LED device has stable periodic LED driving current fluctuation during the dimming process; If stable periodic LED drive current fluctuations occur, the phase relationship between the IC's PWM signal duty cycle and the power ripple can be used to determine whether the power ripple is interfering with the PWM signal output. If the power ripple interferes with the PWM signal output, the consistency of the fluctuation frequency of the LED drive current and the power ripple, as well as the waveform synchronization of the LED drive current and the PWM signal, can be used to determine whether the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal. If the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal, the next drive current fluctuation period is determined. The ripple voltage peak sequence and ripple voltage trough sequence are obtained by power ripple extraction. The ripple voltage peak and ripple voltage trough within the next drive current fluctuation period are predicted using the LSTM model. According to the predicted ripple voltage peak and ripple voltage trough in the next driving current fluctuation period, dynamic pre-compensation of the PWM duty cycle of the ripple phase is performed.

2. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 1, characterized in that: The method for identifying the driving current fluctuation period is: The LED driving current of the multi-color LED device is obtained and compared with the preset LED driving current. The driving current fluctuation period is obtained by taking the deviation of the LED driving current from the preset LED driving current as the starting time point and the recovery of the LED driving current to the preset LED driving current after deviating from the preset LED driving current as the ending time point.

3. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 1, characterized in that: The process of performing stability analysis during the driving current fluctuation period is as follows: In the dimming management cycle, the duration corresponding to each driving current fluctuation period is obtained and integrated to obtain a driving fluctuation period sequence, and the interval duration between adjacent driving current fluctuation periods is obtained and integrated to obtain a driving fluctuation interval sequence; Calculate the coefficient of variation of the driving fluctuation period series and the driving fluctuation interval series respectively, and sum them up to output the driving fluctuation value; If the driving fluctuation value meets the requirements, it means that stable periodic LED driving current fluctuation occurs.

4. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 1, characterized in that: The process of determining whether the power ripple interferes with the PWM signal output is as follows: Calculate the phase difference between the power supply ripple signal and the PWM duty cycle change at several different time points during the driving current fluctuation period. Through phase difference analysis, obtain the fixed phase compliance value and the non-fixed phase deviation value, perform deviation processing, and output the phase relationship value; If the phase relationship value meets the requirement, it means that the power supply ripple interferes with the PWM signal output.

5. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 4, characterized in that: The phase difference analysis process is as follows: By comparing the phase differences, the compliant phase differences and the non-compliant phase differences are identified; Counting the proportion of the number of phase differences that meet the requirements, and obtaining a fixed phase compliance value; Based on the non-coinciding phase difference, the absolute deviation of each non-coinciding fixed phase difference from the fixed phase difference is calculated, and the average calculation is performed to obtain the mean of the phase absolute deviation. The ratio of the mean of the phase absolute deviation to the fixed phase difference is calculated to obtain the non-fixed phase deviation value.

6. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 5, characterized in that: The phase difference comparison process is: If the phase difference is within the phase difference standard range, the phase difference is marked as a compliant phase difference; If the phase difference is not within the phase difference standard range, the phase difference is marked as a non-compliant phase difference.

7. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 1, characterized in that: The process of determining whether the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal is as follows: Synchronously collect LED drive current I LED (t) and power supply ripple voltage V ripple (t), where t represents the acquisition time; LED drive current I LED (t) and power supply ripple voltage V ripple (t) Perform FFT Fourier transform to extract the LED driving current main frequency f LED and power supply ripple voltage main frequency f ripple ; Calculating frequency deviation f=f LED -f ripple ; Synchronously collect PWM duty cycle signal D(t) and LED drive current I LED (t); Calculate the PWM duty cycle signal D(t) and LED drive current I LED The cross-correlation coefficient R of (t) DI : , where Cov represents covariance, Decibel represents the standard deviation of the PWM duty cycle signal D(t) and the LED drive current I LED Standard deviation of (t); If the frequency deviation f is less than or equal to the frequency deviation threshold f threshold , and the mutual correlation coefficient R DI Greater than or equal to the cross-correlation coefficient threshold R threshold , it means that the power ripple interferes with the LED drive current by being superimposed on the IC's PWM signal.

8. The IC-controlled adaptive dimming method for a multi-color LED device according to claim 1, wherein: The next driving current fluctuation period is determined as follows: The driving fluctuation period series and the driving fluctuation interval series are averaged to obtain the average driving fluctuation duration and the average driving fluctuation interval duration; Determine the start time of the next driving current fluctuation period according to the current time point and the average driving fluctuation interval duration; Determining the end time point of the next driving current fluctuation period according to the start time point of the next driving current fluctuation period and the average duration of the driving fluctuation; The next driving current fluctuation period is determined according to the start time point and the end time point of the next driving current fluctuation period.

9. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 8, characterized in that: The process of predicting the ripple voltage peak and the ripple voltage trough in the next driving current fluctuation period is as follows: The ripple voltage peaks and ripple voltage troughs on the power supply ripple are extracted respectively, and are integrated in sequence according to the time sequence to obtain the ripple voltage peak sequence and the ripple voltage trough sequence; A time difference sequence is obtained according to the ripple voltage peak sequence and the ripple voltage trough sequence, wherein the time difference sequence includes an adjacent peak time difference sequence and an adjacent trough time difference sequence, the adjacent peak time difference sequence is composed of the time intervals between adjacent peaks, and the adjacent trough time difference sequence is composed of the time intervals between adjacent troughs; Input features: time difference series and amplitude series; Output target: time difference of the next peak / trough and amplitude of the next peak / trough; Time difference series prediction: Use the pre-trained LSTM model to predict the time difference between the next peak / trough; Amplitude sequence prediction: Use pre-trained LSTM to predict the amplitude of the next peak / trough; According to the predicted time difference, the time of the next peak / valley is accumulated and the prediction operation is continued until the end of the driving current fluctuation period, thereby completing the prediction of the ripple voltage peak and ripple voltage trough in the next driving current fluctuation period.

10. The method for adaptive dimming of a multi-color LED device based on IC control according to claim 9, characterized in that: The process of performing dynamic pre-compensation of the PWM duty cycle of the ripple phase is as follows: Get the peak value of the ripple voltage peak during the next driving current fluctuation period and calculate the ripple voltage V ripple (t)=V peak *sin(2πft), where V peak is the ripple peak value, f is the ripple frequency, t is the ripple peak moment, sin(2πft) is the sine representation, and 2πft is the phase angle of the sine function; Calculate the PWM duty cycle pre-compensation amount: ΔD(t)=V ripple (t) / V DD ×K, where V DD Expressed as IC power supply voltage, K is the compensation coefficient; The phase relationship between the PWM duty cycle change and the ripple is determined by the phase difference standard range. If there is a positive correlation, the duty cycle is increased when the ripple voltage is at the peak and the duty cycle is reduced when the ripple voltage is at the trough according to the PWM duty cycle pre-compensation amount. If there is a negative correlation, reverse compensation is performed.

Images