Small-spacing LED array driving waveform optimization method and device

Abstract

The invention relates to the technical field of LED array driving, and discloses a small-spacing LED array driving waveform optimization method and device. The method comprises the following steps: carrying out spectral characteristic acquisition on pixel points in a small-spacing LED array, and constructing a pixel point luminescence characteristic mathematical model; performing parameter operation on the pixel point luminescence characteristic mathematical model to obtain array display parameters, and constructing a display quality evaluation function according to the array display parameters; performing waveform compensation on the original PWM driving signal to generate a complementary driving waveform pair, executing an orthogonal test based on the complementary driving waveform pair, and establishing a characteristic incidence matrix of driving parameters and display effects; the optimal driving parameter combination is calculated according to the characteristic incidence matrix and the display quality evaluation function, the LED array driving signal is generated according to the optimal driving parameter combination, high-precision control over the driving waveform is achieved, and the stability and reliability of the driving signal are guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of LED array driving technology, and in particular to a method and device for optimizing driving waveforms of a small-pitch LED array. Background Art

[0002] In the process of driving small-pitch LED arrays, due to the continuous reduction of pixel pitch, the crosstalk effect between adjacent pixels becomes increasingly significant, which not only reduces the clarity of the displayed image, but also affects the overall performance of the display system.

[0003] The traditional PWM drive method has many problems when dealing with small-pitch LED arrays: the rising and falling edges of the driving waveform are prone to overshoot and ringing, resulting in waveform distortion; secondly, the simultaneous conduction of adjacent pixels will cause power supply ripple and cause crosstalk effect; thirdly, due to the lack of an effective driving waveform optimization mechanism, it is difficult to achieve high grayscale control accuracy while ensuring display uniformity. Summary of the invention

[0004] The present invention provides a method and device for optimizing a driving waveform of a small-pitch LED array. The present invention realizes high-precision control of the driving waveform and ensures the stability and reliability of the driving signal.

[0005] In a first aspect, the present invention provides a method for optimizing a driving waveform of a small-pitch LED array, the method comprising: Collect the spectral characteristics of the pixels in the small-pitch LED array and build a mathematical model of the pixel luminescence characteristics; Performing parameter calculation on the mathematical model of the pixel light emission characteristics to obtain array display parameters, and constructing a display quality evaluation function according to the array display parameters; Performing waveform compensation on the original PWM drive signal to generate a complementary drive waveform pair, and performing an orthogonal test based on the complementary drive waveform pair to establish a characteristic correlation matrix between drive parameters and display effects; An optimal driving parameter combination is calculated according to the characteristic association matrix and the display quality evaluation function, and an LED array driving signal is generated according to the optimal driving parameter combination.

[0006] In a second aspect, the present invention provides a small-pitch LED array driving waveform optimization device, the small-pitch LED array driving waveform optimization device comprising: The acquisition module is used to collect the spectral characteristics of the pixels in the small-pitch LED array and build a mathematical model of the luminous characteristics of the pixels; A construction module, used to perform parameter calculation on the mathematical model of the pixel light emission characteristics to obtain array display parameters, and to construct a display quality evaluation function according to the array display parameters; A test module, used to perform waveform compensation on the original PWM drive signal, generate a complementary drive waveform pair, and perform an orthogonal test based on the complementary drive waveform pair to establish a characteristic correlation matrix between drive parameters and display effects; A generating module is used to calculate an optimal driving parameter combination according to the characteristic association matrix and the display quality evaluation function, and generate an LED array driving signal according to the optimal driving parameter combination.

[0007] In the technical solution provided by the present invention, by establishing a complete mathematical model of pixel luminous characteristics and combining temperature characteristic compensation, an accurate description of the display characteristics of the LED array is achieved. By adopting a design scheme of complementary drive waveform pairs, positive and negative compensation pulses are injected into row and column scanning signals to effectively suppress voltage spikes and crosstalk effects in the driving process, thereby improving the uniformity of the display image. A characteristic correlation matrix establishment method based on orthogonal testing is introduced, and an accurate mapping relationship between drive parameters and display effects is established through systematic parameter testing and data analysis. An improved particle swarm optimization algorithm is designed, and the convergence efficiency and solution quality of the optimization process are improved through chaotic sequence initialization and adaptive weight adjustment. A drive signal generation scheme combining a digital phase-locked loop and digital pulse width modulation is adopted to achieve high-precision control of the drive waveform and ensure the stability and reliability of the drive signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0009] Figure 1 A schematic diagram of a process for optimizing a driving waveform of a small-pitch LED array provided in an embodiment of the present application; Figure 2 A schematic block diagram of the structure of a small-pitch LED array driving waveform optimization device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0011] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change based on actual conditions.

[0012] It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0013] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0014] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0015] See also Figure 1 , Figure 1 A flow chart of a method for optimizing a driving waveform of a small-pitch LED array provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method for optimizing driving waveforms of a small-pitch LED array provided in an embodiment of the present application includes steps S100 to S600.

[0016] Step S100, collecting spectral characteristics of pixels in a small-pitch LED array and constructing a mathematical model of the luminous characteristics of the pixels; It is understandable that the execution subject of the present invention may be a small pitch LED array driving waveform optimization device, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0017] Specifically, the ambient temperature of the small-pitch LED array is detected. The luminous characteristics of the LED are significantly affected by the temperature. The change in temperature will cause the drift of the spectral wavelength and the fluctuation of the brightness. The working ambient temperature of the LED array is monitored in real time by a temperature sensor or an infrared temperature measuring device, and the corresponding temperature acquisition data is obtained. The current sampling interval is divided based on the temperature distribution characteristics, so that the subsequent current input can more finely match the working characteristics of the LED at different temperatures to improve the accuracy of the model. Different driving currents are input to each pixel in the LED array in turn according to the set interval range, so as to observe the spectral response of the pixel under different current conditions. A high-precision spectrum analyzer is used to measure the luminous characteristics of each pixel, and the spectral data is recorded, including spectral wavelength data and spectral intensity data. The spectral wavelength data reflects the main luminous wavelength of the LED, while the spectral intensity data characterizes the luminous brightness of the pixel under a specific current. Since the spectral characteristics of the LED will change under different driving currents, the measurement is performed within the full current range to ensure the integrity and accuracy of the spectral data. The main wavelength extraction operation is performed on the spectral wavelength data to determine the main luminous wavelength of the LED under different driving currents. The dominant wavelength is determined by calculating the center of gravity or peak position of the spectrum curve, and the current-wavelength characteristic data is established under different currents. This data describes how the dominant wavelength of the LED drifts with the change of the driving current. At the same time, for the spectral intensity data, an integral operation is performed to calculate the total light output under different current conditions, that is, the current-brightness characteristic data, which can reflect the change of the luminous brightness of the LED with the driving current. When constructing the mathematical model of luminous characteristics, mathematical expressions are established for the current-wavelength characteristic data and the current-brightness characteristic data respectively. Based on the current-wavelength characteristic data, a wavelength characteristic polynomial function is established, and the trend of the dominant wavelength changing with the current is fitted by selecting an appropriate polynomial order, and the coefficient matrix of the polynomial is calculated using the least squares fitting algorithm to obtain the wavelength characteristic coefficient matrix. Similarly, for the current-brightness characteristic data, a brightness characteristic polynomial function is established, and the brightness characteristic coefficient matrix is ​​solved by the least squares fitting operation. A three-dimensional characteristic surface is established based on the wavelength characteristic coefficient matrix and the brightness characteristic coefficient matrix, so as to mathematically describe the luminous characteristics of the LED pixel under different current inputs. This three-dimensional characteristic surface takes current as the independent variable and wavelength and brightness as the response variables. A mathematical model of the pixel luminescence characteristics is constructed by the surface fitting method, so that it can accurately reflect the luminescence behavior of the LED under different driving conditions.

[0018] Step S200, performing parameter calculation on the mathematical model of pixel light emission characteristics to obtain array display parameters, and constructing a display quality evaluation function according to the array display parameters; Specifically, the mathematical model of the pixel luminous characteristics is parameterized to obtain the key display parameters of the array, wherein the brightness mean data is the core indicator for measuring the uniformity of the pixel luminous characteristics. Therefore, it is necessary to perform numerical integration operations on the mathematical model to calculate the brightness mean data of each pixel, and based on these data, calculate the brightness uniformity of the entire array. The brightness uniformity coefficient is calculated by determining the ratio of the maximum brightness value to the minimum brightness value, which can effectively reflect whether the brightness distribution of the LED array in different areas is uniform. Based on the brightness uniformity analysis, the waveform response parameters in the mathematical model of the pixel luminous characteristics are Fourier transformed to obtain the spectral component data of the driving signal. The Fourier transform can convert the time domain information of the PWM waveform into frequency domain information, so that the system can analyze whether there are higher harmonic components in the driving signal, and evaluate the waveform distortion rate by calculating the harmonic distortion factor. The level of waveform distortion rate directly affects the visual display effect of the LED. Excessive distortion rate can cause flickering, color deviation or grayscale instability, so it needs to be strictly controlled during the optimization process. Since there is optical crosstalk between the pixels of the small-pitch LED array, that is, the light radiation of adjacent pixels will affect each other's chromaticity and brightness performance, an optical crosstalk model between pixels is established based on the mathematical model of pixel luminous characteristics, and the model is subjected to finite element analysis to simulate the light energy propagation path between different pixels and calculate the influence range and attenuation characteristics of optical crosstalk. Through finite element analysis, the crosstalk attenuation ratio is obtained, which can quantitatively characterize the degree of optical interference between adjacent pixels. A weight matrix is ​​constructed based on the brightness uniformity coefficient, waveform distortion rate and crosstalk attenuation ratio, and the eigenvalue decomposition operation is used to solve the optimal weight coefficient. The construction process of the weight matrix is ​​based on the actual needs of the display system and the visual sensitivity of the human eye to ensure that the influence weights of different parameters are reasonably distributed in the comprehensive evaluation. The role of eigenvalue decomposition is to extract the main influencing factors, so that the optimization process can be more stable and reduce the influence of redundant parameters. The brightness uniformity coefficient, waveform distortion rate and crosstalk attenuation ratio are normalized to obtain standardized evaluation indicators. The standardized evaluation indicators are linearly weighted using the weight coefficient to obtain the initial evaluation score. The initial evaluation score is transformed using the hyperbolic tangent function to enhance the stability of the evaluation score and ensure that it varies within a certain range. The hyperbolic tangent function has good nonlinear mapping characteristics, mapping the input value to a bounded interval, so that the final display quality evaluation function is more in line with the subjective perception of the human eye. The evaluation function is used to drive the objective function of waveform optimization, thereby guiding the adjustment of the PWM drive signal, so that the LED array can reach the optimal state in terms of brightness uniformity, waveform stability and optical crosstalk suppression, thereby significantly improving the overall visual effect of the LED display.

[0019] Step S300, performing waveform compensation on the original PWM driving signal to generate a complementary driving waveform pair, and performing an orthogonal test based on the complementary driving waveform pair to establish a characteristic correlation matrix between driving parameters and display effects; Specifically, the edge detection is performed on the original PWM drive signal to accurately determine the rising edge time sequence and the falling edge time sequence of the signal, and the duty cycle data of the drive signal is calculated based on these timing information. Since the duty cycle directly affects the grayscale performance, brightness stability and power consumption of the LED, the duty cycle data must be accurately measured during the PWM signal optimization process, and the dynamic adjustment capability under different driving conditions must be ensured. The original PWM drive signal is segmented based on the duty cycle data to achieve more precise drive control. The PWM signal is divided into a pre-charge segment waveform, a main drive segment waveform and a stable segment waveform, and the timing synchronization between the three waveform segments is ensured. The pre-charge segment is used to adjust the transient response of the LED, the main drive segment is used to achieve the main light emission control, and the stable segment is used to maintain the stability of the signal to avoid the decline of the display effect due to the fluctuation of the power supply or drive characteristics. Through reasonable timing control, the current response of the LED pixel in different driving stages is optimized to improve the display uniformity and dynamic contrast. When performing PWM signal compensation, the amplitude of the pre-charge segment waveform is modulated, so as to generate a positive compensation pulse in the row scanning signal to improve the response characteristics of the LED at the moment of lighting. An inverted negative compensation pulse is generated based on the waveform parameters of the positive compensation pulse to form a complementary effect during signal transmission. The matching degree of the positive compensation pulse and the negative compensation pulse directly affects the luminous uniformity of the LED, so it is necessary to accurately control their amplitude, duration and phase relationship so that the nonlinear distortion of the driving waveform can be effectively suppressed during the current transmission process. In order to ensure that the compensation signal can play a correct role, a compensation signal transfer function is established based on the positive compensation pulse and the negative compensation pulse, and the transfer function is analyzed in the frequency domain using Laplace transform to obtain the spectrum characteristics of the compensation signal. Through spectrum analysis, the main harmonic components of the compensation signal and the introduced high-frequency noise are identified. On this basis, a crosstalk suppression filter is constructed, which is used to weaken unnecessary high-frequency harmonic components and optimize the transmission characteristics of the compensation signal in the driving system to reduce color shift or brightness fluctuation caused by signal interference. Under the action of the filter, the positive compensation pulse and the negative compensation pulse are filtered to obtain a corrected compensation pulse pair, and the corrected compensation pulse pair is superimposed on the pre-charge segment of the row scanning signal and the column scanning signal respectively to generate a complementary driving waveform pair. The role of the complementary driving waveform pair is to provide corresponding current regulation at different stages of LED lighting, making the LED's luminous characteristics more stable and reducing the color shift phenomenon in high grayscale areas. By optimizing the rising and falling edges of the driving waveform, the electromagnetic interference that may be generated during the driving process is reduced, and the overall display performance of the LED array is improved. Orthogonal testing is performed on the complementary driving waveform pair within a preset voltage range and a preset time range to systematically analyze the impact of different driving parameters on the display effect. Orthogonal testing is an experimental design method that can effectively evaluate the combined impact of multiple parameters with a smaller number of tests.By setting different PWM duty cycles, compensation signal amplitudes, filtering parameters and other factors, a series of orthogonal test experiments are performed, and key indicators such as brightness uniformity, grayscale stability, and color consistency of the LED array are measured and analyzed. Based on the experimental data of the orthogonal test, a characteristic correlation matrix of driving parameters and display effects is established, which can quantify the display performance under different driving conditions. By analyzing the characteristic correlation matrix, the key driving parameters that affect the display effect are found, and the compensation strategy of the PWM signal is adjusted according to different display requirements, so as to achieve the best driving effect of the LED array.

[0020] The amplitude and pulse width of the compensation pulse are modeled and used as orthogonal test factors to construct a systematic test scheme. Since the amplitude of the compensation pulse has a direct impact on the brightness and response speed of the LED pixel, it is reasonably divided within the target voltage range and the amplitude interval is divided into N test points to ensure that the test covers the entire driving range. At the same time, the pulse width determines the duration and energy transfer characteristics of the signal, so it is also necessary to divide it into M test points within the target time range to obtain N×M groups of test parameters. An orthogonal test table is generated based on the N×M groups of parameters, and the test table is used to set the parameters of the complementary driving waveform pair to generate the corresponding experimental waveform. The role of the orthogonal test table is to reduce the number of experiments while ensuring that all combinations of key parameters can be effectively covered. Therefore, uniformity and representativeness must be taken into account during the design. By setting the compensation pulse amplitude and pulse width of the PWM signal according to the orthogonal test table, N×M groups of experimental waveforms are obtained, and each group of experimental waveforms represents a possible configuration of a driving signal. During the experiment, waveform analysis was performed on key dynamic characteristic parameters such as the current rising edge change rate, overshoot amplitude, and crosstalk current of each group of experimental waveforms to quantify the electrical characteristics of each group of waveforms. The current rising edge change rate determines the response speed of the LED and directly affects the stability of high grayscale display. Therefore, a high-speed oscilloscope or current sensor is required to record the change of the rising edge during the test. At the same time, the measurement of the overshoot amplitude is crucial because the overshoot phenomenon will cause the instability of the driving signal, which will cause uneven brightness or chromaticity shift. In order to ensure the display quality of the LED array, the size of the crosstalk current is measured to evaluate the impact of different waveform configurations on adjacent pixels, and methods to reduce crosstalk are found through data analysis. After completing the measurement of the dynamic characteristic parameters, a multidimensional characteristic space is constructed based on these parameters and the aforementioned display quality evaluation function, so that all test data can be analyzed within a unified mathematical framework. In order to extract the main influencing factors in the data, the multidimensional characteristic space is subjected to principal vector decomposition to obtain a set of eigenvectors. The main function of principal vector decomposition is to extract the main change mode in the data and remove redundant information to improve the efficiency and accuracy of the optimization calculation. The mapping relationship between driving parameters and display effects is established according to the characteristic vector group to form an optimization model. The least squares fitting calculation is used to optimize the parameters of the model and reduce the noise interference in the experimental data. The result of the least squares fitting generates a correlation coefficient matrix, which is used to describe the specific impact of different driving parameters on the display effect. The singular value decomposition operation is performed on the correlation coefficient matrix to remove redundant information and ensure that the mapping relationship between driving parameters and display effects has high interpretability and stability. Generate a characteristic correlation matrix of driving parameters and display effects.

[0021] Step S400: Calculate an optimal driving parameter combination according to the characteristic association matrix and the display quality evaluation function, and generate an LED array driving signal according to the optimal driving parameter combination.

[0022] Specifically, the target optimization function is constructed based on the characteristic association matrix to ensure that the optimization process can accurately reflect the display quality requirements of the LED array. Since the optimization of the driving waveform involves multiple interrelated parameters, when constructing the optimization target, factors such as brightness uniformity, waveform distortion rate, and optical crosstalk suppression effect are comprehensively considered, and a unified optimization target function is formed through a weighted strategy. In order to ensure the feasibility of the solution, the constraint conditions are set for the target optimization function, and the amplitude range and pulse width range of the compensation pulse are used as the boundaries of the particle position, thereby limiting the target search space, so that the optimization search is carried out within a reasonable range, and avoiding the emergence of solutions that exceed physical constraints. The target search space is initialized with particles, an initial particle swarm is generated, and the initial state of the particle swarm is optimized. Since conventional random initialization leads to uneven particle distribution, which in turn affects the optimization efficiency, during the particle initialization process, a chaotic sequence is constructed to perturb the initial particle swarm to enhance the diversity of the particle swarm and avoid the optimization process from falling into a local optimal solution. Through this process, an initial solution space with a good search distribution is formed, and the fitness value of each initial particle is calculated based on the display quality evaluation function to evaluate the pros and cons of different initial solutions. In order to improve the stability and convergence speed of the optimization process, the adaptive inertia weight of the particles in the initial solution space is calculated. The role of the inertia weight is to adjust the movement trend of the particles in the search space, and the adaptive inertia weight dynamically adjusts its weight according to the fitness value of the particles, so as to maintain a strong global search ability in the early stage of optimization, and gradually converge to the local search in the later stage of optimization. This mechanism can effectively improve the convergence accuracy of the optimization algorithm and reduce invalid iterations in the search process. After calculating the inertia weight of the particle swarm, the particle swarm is updated based on the motion parameters, and the local search operator is introduced to adjust the particle position to form the next generation of candidate solution sets. The core of the speed update operation is to adjust the movement direction of the particles according to the historical optimal solution of the particles and the global optimal solution of the population, so that they gradually converge to the optimal solution area. At the same time, the local search operator can enhance the local exploration ability of the optimization algorithm to ensure that the optimization results can be refined and adjusted in the local optimal area while meeting the global search requirements, so as to obtain a better solution set. After the candidate solution set is generated, it is constrained to ensure that all solutions meet the physical constraints. For particles that exceed the search boundary, the Lagrange multiplier method is used to adjust them back to the feasible solution space and ensure that no invalid solutions are generated during the optimization process. The solution set after constraint processing is defined as the valid solution set, and it is sorted based on the display quality evaluation function to select the optimal solution. From the sorted valid solution set, the optimal particle position parameter is extracted, and the parameter is decoded and converted into a compensation pulse parameter to form the optimal driving parameter combination. These optimal parameters can maximize the brightness uniformity of the LED array, reduce waveform distortion, and reduce optical crosstalk, thereby improving the overall performance of the LED display system.In order to ensure that the optimal driving parameters can be accurately used for the generation of PWM signals, a high-precision clock signal is generated through a digital phase-locked loop, and the clock signal is input into a digital pulse width modulator to generate an LED array driving signal that meets the optimization requirements.

[0023] The optimal driving parameter combination is digitally encoded so that it can adapt to the frequency synthesis and phase control mechanism of the digital phase-locked loop (PLL). The amplitude parameter of the compensation pulse is converted into a frequency division coefficient so that it can control the output frequency of the PLL, and the pulse width parameter is converted into a phase control word to adjust the phase accumulation process of the PLL to obtain the complete digital phase-locked loop control parameters. A phase accumulator is constructed based on the digital phase-locked loop control parameters to ensure that the PLL can accurately generate a clock signal. The core function of the phase accumulator is to continuously accumulate the input phase increment and convert it into a sinusoidal signal through a numerically controlled oscillator. Therefore, during the phase accumulation process, a sinusoidal mapping operation is performed on the accumulated output to obtain a reference signal for the phase frequency detector. The reference signal is compared with the feedback signal in phase frequency to evaluate the phase-locked state of the PLL. The phase frequency comparator calculates the phase error between the reference signal and the feedback signal and generates an error signal, which will be input to the digital loop filter for filtering. The role of the loop filter is to smooth the phase error, reduce high-frequency noise, and output a control voltage, which will be used as the input signal of the digital control oscillator to accurately adjust the output frequency of the oscillator, thereby ensuring that the PLL can work stably and generate a clock signal that meets the requirements of the LED array drive. In the PLL system, the digital control oscillator is the core component for generating the reference clock signal. It adjusts the oscillation frequency according to the control voltage of the loop filter to ensure that the output signal is consistent with the target frequency. In order to improve the stability and flexibility of the clock signal, the reference clock signal is divided to obtain the feedback clock signal, thereby forming a complete phase-locked control loop in the PLL closed-loop system. The role of the frequency division process is to adjust the period of the clock signal to adapt it to different LED drive requirements and ensure that the final PWM signal can meet the response characteristics and display quality requirements of the LED. After generating a stable reference clock signal, a counter of the digital pulse width modulation unit (DPWM) is established based on the signal to ensure accurate control of the PWM signal. In the DPWM module, the function of the counter is to continuously accumulate the reference clock signal and determine the duty cycle of the PWM signal according to the count value. Therefore, during the counting process, the count value is compared with the duty cycle data of the compensation pulse to generate a pulse width modulation waveform. The accuracy and stability of the pulse width modulation waveform determine the grayscale display effect of the LED pixel point and directly affect the contrast and dynamic response capability of the display screen. In order to ensure that the PWM signal can drive the LED array, the PWM signal is level-converted and current-amplified through the driver circuit to match the operating voltage and current requirements of the LED. The driver circuit is composed of a power MOSFET or a constant current driver chip, which can convert the low-power PWM control signal into a high-current drive signal suitable for the LED array and ensure the linearity of the drive current to reduce the color deviation and uneven brightness of the LED display.Through digital signal processing and hardware optimization, a stable and efficient LED array drive signal is generated, so that the small-pitch LED display reaches the optimal state in terms of brightness uniformity, response speed and color accuracy, thereby improving the overall display effect and system reliability.

[0024] In the embodiment of the present invention, by establishing a complete mathematical model of pixel luminous characteristics and combining temperature characteristic compensation, an accurate description of the display characteristics of the LED array is achieved. By adopting a design scheme of complementary drive waveform pairs, by injecting positive and negative compensation pulses into row and column scanning signals, the voltage spikes and crosstalk effects in the driving process are effectively suppressed, and the uniformity of the display screen is improved. A characteristic correlation matrix establishment method based on orthogonal testing is introduced, and an accurate mapping relationship between driving parameters and display effects is established through systematic parameter testing and data analysis. An improved particle swarm optimization algorithm is designed, and the convergence efficiency of the optimization process and the quality of the solution are improved through chaotic sequence initialization and adaptive weight adjustment. A drive signal generation scheme combining a digital phase-locked loop and digital pulse width modulation is adopted to achieve high-precision control of the drive waveform and ensure the stability and reliability of the drive signal.

[0025] In a specific embodiment, the process of executing step S100 may specifically include the following steps: Detect the ambient temperature of the small-pitch LED array to obtain temperature acquisition data, and divide the current sampling interval based on the temperature acquisition data; The driving current is sequentially input into each pixel point in the small-pitch LED array according to the current sampling interval, and the spectrum data of each pixel point is collected by the spectrum analyzer to obtain the spectrum wavelength data and spectrum intensity data; Performing a main wavelength extraction operation on the spectrum wavelength data to obtain current-wavelength characteristic data, and performing an integration operation on the spectrum intensity data to obtain current-brightness characteristic data; A wavelength characteristic polynomial function is established based on the current-wavelength characteristic data, and a least square fitting operation is performed on the wavelength characteristic polynomial function to obtain a wavelength characteristic coefficient matrix; A brightness characteristic polynomial function is established based on the current-brightness characteristic data, and a least square fitting operation is performed on the brightness characteristic polynomial function to obtain a brightness characteristic coefficient matrix; A three-dimensional characteristic surface is established based on the wavelength characteristic coefficient matrix and the brightness characteristic coefficient matrix to generate a mathematical model of pixel point luminescence characteristics.

[0026] Specifically, a high-precision temperature sensor is used to monitor the working environment temperature of the LED display in real time. The light-emitting characteristics of the LED, including the spectral wavelength and brightness response, will shift with changes in temperature, and this shift directly affects the color consistency and brightness uniformity of the display. Assume that the data obtained by the temperature measurement is ( ),in Represents the temperature values ​​of different pixels at different times. By statistically analyzing these temperature data, the range of temperature changes can be determined. , and divide the current sampling interval based on this range, so that the current input can match the luminous characteristics of the LED at different temperatures. Different drive currents are input in sequence according to the divided current sampling intervals ( ) to each pixel in the small-pitch LED array, and use a spectrum analyzer to collect the corresponding spectral data, including spectral wavelength data and spectral intensity data ,in Represents the temperature and drive current The main emission wavelength is Represents the spectral intensity distribution under the corresponding conditions. After obtaining the spectral wavelength data, the main wavelength extraction operation is performed, and the spectral centroid method is used to calculate the main luminescence wavelength:

[0027] in, For a specific wavelength The calculation formula can effectively extract the main emission wavelength of the LED and reflect the influence of driving current and temperature on the spectrum drift. At the same time, in order to obtain the brightness characteristic data of the LED, the spectrum intensity data is integrated to calculate the total luminous flux:

[0028] in, Represents the temperature and drive current This integral calculation can accurately describe the change of LED luminous intensity with current and temperature, and is used for subsequent brightness compensation optimization. Finally, a wavelength characteristic polynomial function is established to fit the data relationship, using a quadratic or cubic polynomial:

[0029] in, are the coefficients to be determined, representing the effect of current and temperature on wavelength. In order to determine these coefficients, a least squares fit operation is performed on the polynomial, the goal of which is to minimize the error function:

[0030] The optimal solution can be obtained by matrix solving method, and the wavelength characteristic coefficient matrix is ​​obtained:

[0031] Similarly, for the brightness characteristic data , establish the brightness characteristic polynomial function:

[0032] in, is the brightness characteristic fitting coefficient, and the least square method is used for optimization to solve the error function:

[0033] Finally, the brightness characteristic coefficient matrix is ​​obtained: ; Based on the wavelength characteristic coefficient matrix And the brightness characteristic coefficient matrix Construct a three-dimensional characteristic surface that is driven by the current and temperature is the independent variable, with wavelength and brightness As the response variable, a mathematical model of pixel luminescence characteristics is formed: .

[0034] In a specific embodiment, the process of executing step S200 may specifically include the following steps: Perform numerical integration on the mathematical model of pixel luminous characteristics to obtain the average brightness data of each pixel, and calculate the ratio of the maximum brightness value to the minimum brightness value based on the average brightness data to obtain the brightness uniformity coefficient; Performing Fourier transformation on the waveform response parameters in the mathematical model of the pixel luminous characteristics to obtain spectrum component data, and calculating the harmonic distortion factor based on the spectrum component data to obtain the waveform distortion rate; An optical crosstalk model between pixels is established based on the mathematical model of pixel luminescence characteristics, and a finite element analysis is performed on the optical crosstalk model to obtain the crosstalk attenuation ratio; A weight matrix is ​​constructed based on the brightness uniformity coefficient, waveform distortion rate and crosstalk attenuation ratio, and an eigenvalue decomposition operation is performed on the weight matrix to obtain a weight coefficient; Normalize the brightness uniformity coefficient, waveform distortion rate and crosstalk attenuation ratio to obtain a standardized evaluation index, and linearly weight the standardized evaluation index according to the weight coefficient to obtain an initial evaluation score; The initial evaluation score is transformed by a hyperbolic tangent function to obtain a display quality evaluation function.

[0035] Specifically, based on the mathematical model of pixel luminescence characteristics The brightness function expressed , that is, in the coordinate The pixel points at different driving currents and ambient temperature In order to calculate the average brightness of each pixel, the driving time of the entire pixel is Perform numerical integration:

[0036] in, Represents the pixel point in the driving cycle The average brightness within Indicates the brightness of the pixel point over time. In order to measure the overall brightness uniformity of the LED array, find the maximum brightness value in the pixel array and minimum brightness value And calculate the brightness uniformity coefficient :

[0037] in, The value range is between (0,1]. When it approaches 1, it means that the brightness uniformity of the entire LED array is good. When it is small, it means that there is a large brightness unevenness. In order to analyze the waveform response characteristics of the pixel point, the waveform response parameters in the mathematical model of the pixel point luminescence characteristics are Fourier transformed to obtain the spectral component data of the signal. Assume that the PWM driving signal of the pixel point is , then its Fourier transform It is expressed by the following formula:

[0038] in, is the frequency component, Represents the amplitude distribution of the driving signal at different frequencies. In order to measure the degree of waveform distortion of the signal, the harmonic distortion factor needs to be calculated. , which is defined as:

[0039] in, Representative The harmonic components, represents the fundamental component. When it is smaller, it means that the distortion of the PWM signal is smaller, so the light output of the LED is more stable; conversely, a larger This will cause problems such as flickering of the display screen and unstable grayscale. When analyzing the optical crosstalk problem of the LED array, an optical crosstalk model is established based on the mathematical model of the pixel luminescence characteristics. Optical crosstalk mainly comes from the diffusion of light radiation from adjacent pixels, so the finite element analysis method is used to simulate the optical energy transmission between pixels. Assume that the optical power distribution of adjacent pixels is , then the basic equation of the optical crosstalk model is:

[0040] in, is the light wave propagation constant, represents the Laplace operator. Through finite element solution, the crosstalk attenuation ratio is obtained : ; in, Represents the optical power of the original pixel. A smaller value indicates that the optical crosstalk has less impact, while a larger value indicates that the optical crosstalk has less impact. This may cause color unevenness on the LED display. Based on the brightness uniformity coefficient , waveform distortion rate and crosstalk attenuation ratio Constructing the weight matrix To characterize the influence of each parameter on display quality. Assuming the weight matrix for:

[0041] Through the eigenvalue decomposition operation, the weight coefficient is solved : ; in, is the feature vector, is the corresponding weight coefficient. The weight coefficient calculated based on the eigenvalue is used for subsequent quality evaluation calculation. In order to standardize the various evaluation indicators, the brightness uniformity coefficient, waveform distortion rate and crosstalk attenuation ratio are normalized, and the normalized variable is defined :

[0042] in, Represents the original data, and Represent the minimum and maximum values ​​of the data respectively. After normalization, the weight coefficient is used to linearly weight the standardized evaluation index to obtain the initial evaluation score : ; The initial evaluation score is nonlinearly transformed through the hyperbolic tangent function to obtain the final display quality evaluation function : ; in, The role of is to compress the evaluation value into the range of (-1,1) and enhance the ability to distinguish extreme values.

[0043] In a specific embodiment, the process of executing step S300 may specifically include the following steps: Perform edge detection on the original PWM drive signal to obtain a rising edge time sequence and a falling edge time sequence, and calculate the duty cycle data of the drive signal according to the rising edge time sequence and the falling edge time sequence; The original PWM drive signal is segmented based on the duty cycle data to obtain a pre-charge segment waveform, a main drive segment waveform and a stable segment waveform, and the pre-charge segment waveform, the main drive segment waveform and the stable segment waveform are synchronized in time sequence; Amplitude modulate the pre-charge segment waveform to generate a positive compensation pulse in the row scanning signal, and generate an inverted negative compensation pulse according to the waveform parameters of the positive compensation pulse; A compensation signal transfer function is established based on a positive compensation pulse and a negative compensation pulse, and a frequency domain analysis is performed on the compensation signal transfer function through Laplace transform to obtain a compensation signal spectrum; A crosstalk suppression filter is constructed according to the compensation signal spectrum, and the positive compensation pulse and the negative compensation pulse are filtered to obtain a corrected compensation pulse pair, and the corrected compensation pulse pair is respectively superimposed on the pre-charge segment of the row scanning signal and the column scanning signal to generate a complementary driving waveform pair; An orthogonal test is performed on the complementary driving waveform pair within a preset voltage range and a preset time range to establish a characteristic correlation matrix between driving parameters and display effects.

[0044] Specifically, the PWM signal is sampled in the time domain to obtain the rising edge time sequence and the falling edge time sequence. Assume that the period of the PWM signal is , and its timing function is expressed as: ; in, and Represent the time sequence of rising edge and falling edge respectively, is the high level of the PWM signal, In order to calculate the duty cycle data of the PWM signal, the high level duration in each cycle is calculated. And normalize to get the duty cycle :

[0045] The duty cycle data is used to further optimize the PWM signal to match the response characteristics of the LED pixel. After obtaining the duty cycle of the PWM signal, the signal is segmented to obtain the pre-charge segment waveform, the main drive segment waveform, and the stable segment waveform. These three stages correspond to the LED pre-charge, main drive process, and signal stabilization stage, respectively, so they need to be synchronized in timing to ensure a smooth transition of the current input in each stage. The segmented waveform is expressed as: ); in, , and is the duration of the pre-charge, main drive and stabilization phases, In order to optimize the response characteristics of the LED, amplitude modulation is introduced in the pre-charge stage to generate a positive compensation pulse in the row scanning signal: ; in, is the amplitude of the compensation pulse, is the time constant, is the duration of the compensation pulse. Since the compensation pulse causes additional signal disturbance, an inverted negative compensation pulse is generated. To offset the corresponding interference: ; These two complementary pulses are used to adjust the driving response of the LED and reduce high-frequency noise. In order to analyze the impact of the compensation pulse on the system, the transfer function of the compensation signal is established and Laplace transform is performed on it to convert from time domain to frequency domain representation:

[0046] in, is the Laplace transform variable. By calculating the spectrum response of the compensation signal: ; The amplitudes of different frequency components are obtained to construct a crosstalk suppression filter to eliminate high-frequency noise. The transfer function of the crosstalk suppression filter is expressed as:

[0047] in, is the cutoff frequency of the filter. The compensation pulse pair after filtering is: ; This signal is used to adjust the row scan and column scan pre-charge stages of the LED array, thereby generating a complementary drive waveform pair:

[0048] ; After generating a complementary drive waveform pair, within the preset voltage range and preset time range The key to orthogonal testing is to construct a test matrix to convert the amplitude of the compensation pulse into and time constant As a test factor, define the test point:

[0049]

[0050] in, , . Orthogonal testing produces Each set of test parameters is used for driving experiments and measuring the brightness uniformity, grayscale response and crosstalk intensity of LEDs. The experimental data is organized into a characteristic correlation matrix: ; in, Represents the display effect evaluation value of each set of driving parameters. By calculating the optimal parameter group, the driving characteristics of the PWM signal are optimized to improve the brightness uniformity and visual effect of the LED display.

[0051] Among them, after the complementary driving waveform pair is generated and before the orthogonal test is performed, it also includes: performing time domain discretization processing on the complementary driving waveform pair, dividing the driving cycle into N preset time segments, and obtaining a discrete sampling sequence; marking the key waveform feature points in the discrete sampling sequence, dividing the driving waveform into a steady-state interval and a transient interval, and obtaining partitioned sampling data; constructing an adaptive integral model based on the partitioned sampling data, using a variable step-length integral algorithm for the transient interval, and using a fixed step-length integral algorithm for the steady-state interval, to obtain a hybrid integral parameter; numerically reconstructing the complementary driving waveform pair according to the hybrid integral parameter, and using a high-order integral algorithm to perform fine-grained calculation on the transient interval to obtain to the optimized waveform data; an error evaluation function is established based on the optimized waveform data, the integral error of each interval is obtained by recursive calculation, and compared with the preset error threshold to obtain an error distribution matrix; the integral step size is adaptively adjusted according to the error distribution matrix, the calculation step size is refined in the error exceeding threshold area, and the step size is kept unchanged in the error reaching standard area to obtain the corrected integral parameter; the complementary drive waveform pair is iteratively optimized using the corrected integral parameter until the integral error of all intervals is less than the preset threshold to obtain the final drive waveform data; the final drive waveform data is smoothed by a digital filter to generate the optimized complementary drive waveform pair.

[0052] In a specific embodiment, the process of performing an orthogonal test on the complementary driving waveform pair within a preset voltage range and a preset time range to establish a characteristic correlation matrix between driving parameters and display effects may specifically include the following steps: An orthogonal test factor is constructed for the compensation pulse amplitude and pulse width of the complementary driving waveform pair, the compensation pulse amplitude is divided into N test points within the target voltage range, and the pulse width is divided into M test points within the target time range, so as to obtain N×M groups of test parameters; An orthogonal test table is generated based on N×M groups of test parameters, and parameters of the complementary driving waveform pair are set according to the orthogonal test table to obtain N×M groups of experimental waveforms; The current rising edge change rate, overshoot amplitude and crosstalk current of N×M groups of experimental waveforms are analyzed to obtain the dynamic characteristic parameters of each group of waveforms; Based on the dynamic characteristic parameters and the display quality evaluation function, a multi-dimensional characteristic space is constructed, and a principal vector decomposition operation is performed on the multi-dimensional characteristic space to obtain a characteristic vector group; The mapping relationship between driving parameters and display effects is established according to the characteristic vector group, the least square fitting calculation is performed on the mapping relationship to obtain the correlation coefficient matrix, and the singular value decomposition operation is performed on the correlation coefficient matrix to generate the characteristic correlation matrix between driving parameters and display effects.

[0053] Specifically, an orthogonal test factor is constructed for the compensation pulse amplitude and pulse width of the complementary drive waveform pair, and the voltage amplitude range and pulse width range of the compensation pulse are determined to ensure the coverage of the test. The target range of the compensation pulse amplitude is set to , which is divided into test points, and the amplitude of each test point is expressed as:

[0054] in, Indicates Similarly, the target width range of the compensation pulse is set to , and divided into test points, the pulse width of each test point is:

[0055] in, Indicates The compensation pulse width of each test point is constructed by combining the amplitude and width of the compensation pulse. A set of test parameters is formed to form a complete experimental test matrix. The orthogonal test table is generated by grouping parameters, and the parameters of the complementary drive waveform pair are set through the orthogonal test table to generate the corresponding experimental waveform. Orthogonal test is an optimization experimental method that can obtain the optimal parameter combination within a limited number of tests. It is represented by the following matrix: ; Each row represents a set of test parameters used to control the amplitude and pulse width of the complementary drive waveform so that the experimental tests are evenly distributed throughout the parameter space. The experimental waveforms are analyzed to evaluate their dynamic characteristics. The rising edge change rate of the current in the experimental waveform is measured. , Overshoot Amplitude and crosstalk current . Set the instantaneous current of the waveform , then the rising edge change rate It is expressed as:

[0056] in, It is the rising edge time point of the waveform. Overshoot amplitude is the maximum amplitude of the transient response of the signal that exceeds the stable current value and is defined as: ; in, is the current value after the signal stabilizes. It is the current component caused by the interference between adjacent pixels, and the calculation formula is:

[0057] in, is the parasitic current of the adjacent pixel, is the current of the main driving pixel. After obtaining the dynamic characteristic parameters of all experimental waveforms, a multidimensional characteristic space is constructed based on these parameters and the display quality evaluation function. Each data point in the characteristic space is defined as for: ; The matrix representation of the entire feature space is: ; In order to reduce the dimension of the data and extract the main features, the principal vector decomposition operation is performed on the feature space, that is, Perform eigenvalue decomposition: ; in, is the eigenvector, is the corresponding eigenvalue. By calculating the eigenvector corresponding to the maximum eigenvalue, the main influencing factors are obtained and used to establish the mapping relationship between the driving parameters and the display effect. The least squares fitting calculation is performed on the mapping relationship to obtain the correlation coefficient matrix . Set the drive parameter matrix And display effect matrix : ; in, is the display quality evaluation value under each set of driving parameters. Then the goal of the least squares fitting calculation is to solve So that: ; Calculated by the least squares method: ; Finally, the correlation coefficient matrix is ​​obtained , which is used to describe the quantitative relationship between driving parameters and display effects. Singular value decomposition is performed to eliminate redundant information and improve the stability of calculation. The mathematical expression of singular value decomposition is:

[0058] in, and are the left and right singular vector matrices, It is a diagonal matrix whose diagonal elements are singular values. By screening larger singular values, the influence of noise is removed, and the final characteristic correlation matrix of driving parameters and display effects is generated: .

[0059] Among them, after generating the characteristic association matrix of driving parameters and display effects and before calculating the optimal driving parameter combination, it also includes: performing feature analysis on the characteristic association matrix, extracting the main influencing factors, establishing the objective function of parameter optimization, and determining the optimization constraints according to the physical characteristics of the LED array to obtain the initial optimization model; constructing a dual variable iterative structure based on the initial optimization model, mapping the amplitude optimization and timing optimization of the driving parameters to the original variable space and the dual variable space respectively, and obtaining the solution space mapping matrix; setting a dynamic trigger threshold for the solution space mapping matrix, establishing an adaptive sampling strategy according to the parameter sensitivity, and obtaining the parameter update trigger condition; performing a powerful ball search based on the parameter update trigger condition, and calculating the optimal driving parameter combination before calculating the optimal driving parameter combination; and obtaining the optimal driving parameter combination after calculating the optimal driving parameter combination. Calculate the gradient direction and step size of each point in the solution space, construct the search trajectory, and obtain the candidate solution sequence; perform convergence analysis on the candidate solution sequence, establish the iteration termination criterion, and calculate the convergence speed of the optimization process according to the criterion to obtain the convergence characteristic curve; dynamically adjust the search step size according to the convergence characteristic curve, increase the step size in the fast convergence area, and reduce the step size in the fine search area to obtain an adaptive step size sequence; update the parameters of the Powerball search process based on the adaptive step size sequence, and determine whether to perform the update operation by judging the trigger condition to obtain the optimized iterative sequence; input the optimized iterative sequence into the original dual iterator to perform alternating optimization in the parameter space until the convergence condition is met, and generate the preprocessed characteristic association matrix.

[0060] In a specific embodiment, the process of executing step S400 may specifically include the following steps: Based on the characteristic correlation matrix, a target optimization function is constructed, and constraints are set for the target optimization function. The compensation pulse amplitude range and pulse width range are used as particle position boundaries to obtain the target search space. Initialize the target search space with particles, construct a chaotic sequence to perturb the initial particle group, obtain the initial solution space, and calculate the fitness value of each initial particle based on the display quality evaluation function; The adaptive inertia weight of particles in the initial solution space is calculated, and the inertia weight is dynamically adjusted according to the fitness value to obtain the motion parameters of the particle swarm; Based on the motion parameters, the velocity of the particle swarm is updated, and the local search operator is introduced to update the particle position to obtain the candidate solution set after iteration; The candidate solution set is subjected to constraint check, and the out-of-bounds particles are processed by Lagrange multiplier method to obtain the valid solution set, which is then sorted according to the display quality evaluation function. Extract the optimal particle position parameters from the effective solution set, decode and convert the optimal particle position parameters into compensation pulse parameters, and generate the optimal driving parameter combination; The optimal driving parameter combination is used to generate a clock signal through a digital phase-locked loop, and the clock signal is input into a digital pulse width modulator to generate an LED array driving signal.

[0061] Specifically, using the feature association matrix The objective optimization function is established to characterize the influence of driving parameters on display quality. is the optimization variable, where To compensate the pulse amplitude, To compensate for the pulse width, the optimization objective is defined as maximizing the display quality evaluation function : ; In order to make the optimization search space reasonable, constraints are imposed on the target optimization function. The range of the compensation pulse amplitude is set to , the pulse width range is , then the search space Defined as: ; After establishing the search space, the particle swarm optimization algorithm is used to optimize the solution. Initialize the particles in the target search space and set the particle swarm size to , each particle The initial positions are randomly distributed in Within, that is: ; in, Represents in the interval In order to improve the diversity of initial particles, a chaotic sequence is used to perturb the particle swarm. Using Logistic chaotic mapping: ; in, is the chaos parameter (usually 3.99), is the chaotic sequence value, which is used to adjust the distribution of initial particles to avoid early convergence to the local optimal solution. After initializing the particle swarm, calculate the fitness value of each particle , which is the display quality evaluation function :

[0062] Based on the fitness value, the particle swarm is adaptively inertia-weighted to dynamically adjust the search behavior of the particles. The following adaptive formula is used:

[0063] in, and are the maximum and minimum values ​​of the inertia weight, respectively. and are the maximum and minimum fitness values ​​in the current particle swarm. Adaptive inertia weights can improve convergence speed while ensuring global search in the early stages and local optimization in the later stages. After updating the inertia weights, the particle swarm is updated with velocity. Speed Calculated by the following formula:

[0064] in, is the acceleration factor, is a random number, It is a particle The best historical position, is the global optimal position. The particle position is updated as:

[0065] In order to enhance the local search capability, a local search operator is introduced after each iteration to fine-tune the candidate solution set. Use Gaussian perturbation:

[0066] in, represents a mean of 0 and a variance of Gaussian noise. Constraint checks are performed on the candidate solution set to ensure that all particles are located in the legal search space For out-of-bounds particles, the Lagrange multiplier method is used for correction, and the out-of-bounds objective function is set as: ; Solution Make The boundary conditions are met and the corrected effective solution set is obtained. In the effective solution set, the particles are sorted according to the fitness value and the optimal particle position is selected. As the final compensation pulse parameter. The parameter is decoded into the optimal drive parameter combination and used to generate the drive signal. The clock signal is generated through a digital phase-locked loop (PLL), where the input clock Through the divider Generate output clock :

[0067] Among them, the frequency division coefficient By compensating the pulse width The clock signal is input into a digital pulse width modulator (PWM), where the pulse width To set the duty cycle :

[0068] The PWM signal is converted into an LED array drive signal through the drive circuit : ; Through the above optimization process, the amplitude and width of the compensation pulse are adaptively adjusted to achieve the best LED display effect.

[0069] In a specific embodiment, the process of executing the step of generating a clock signal through a digital phase-locked loop by combining the optimal driving parameters, and inputting the clock signal into a digital pulse width modulator to generate an LED array driving signal may specifically include the following steps: Digitally encode the optimal driving parameter combination, convert the compensation pulse amplitude parameter into a frequency division coefficient, convert the pulse width parameter into a phase control word, and obtain the digital phase-locked loop control parameter; A phase accumulator is constructed based on the control parameters of a digital phase-locked loop, and a sine mapping operation is performed on the accumulated output of the phase accumulator to obtain a reference signal of a phase frequency detector; Performing phase frequency comparison on the reference signal and the feedback signal, filtering the comparison result through a digital loop filter, and obtaining a control voltage of the digital controlled oscillator; The control voltage is input into the digital controlled oscillator for frequency modulation to generate a reference clock signal, and the reference clock signal is frequency-divided to obtain a feedback clock signal; A counter of a digital pulse width modulation unit is established based on a reference clock signal, and a comparison operation is performed on the count value of the counter and the duty cycle data of the compensation pulse to obtain a pulse width modulation waveform, and the pulse width modulation waveform is subjected to level conversion and current amplification through a driver circuit to generate an LED array drive signal.

[0070] Specifically, the optimal driving parameter combination is digitally encoded to adapt to the control logic of the digital phase-locked loop. Assume that the optimal compensation pulse amplitude is , the optimal pulse width is , then first convert the compensation pulse amplitude into the PLL frequency division coefficient , and convert the pulse width parameter into a phase control word The calculation of the frequency division coefficient is based on the target output frequency With input reference frequency Relationship, satisfying:

[0071] in, Must be an integer, and quantization is performed by finding the nearest integer value. Need to match pulse width , the phase step is controlled by the phase accumulator, and the calculation formula is:

[0072] in, is the clock period, is the bit width of the phase accumulator, so that The value range is , which is used to control the phase increment. A phase accumulator is constructed based on the digital phase-locked loop control parameters. The function of this accumulator is to accumulate the phase increment with the current phase value to control the phase change. The calculation formula is:

[0073] in, represents the current accumulated phase value, and Control phase step. Since the output value of the phase accumulator is A sine mapping operation is required to obtain the reference signal of the phase frequency detector, which is expressed as:

[0074] in, is the amplitude of the oscillating signal, ensuring that the reference signal is used for subsequent phase-frequency comparison. With feedback signal Perform phase-frequency comparison. Calculate phase error : ; The error signal enters the digital loop filter, which is used to smooth the phase error and remove high-frequency noise. The transfer function of the loop filter is usually a second-order IIR filter:

[0075] in, and are the proportional and integral gains respectively, Control the filter time constant, the filtered signal Used to control the digital controlled oscillator to generate a modulated output frequency. Control voltage Frequency modulation is performed through a digitally controlled oscillator (DCO), and the output frequency of the DCO Determined by the control voltage: ; in, is the center frequency of the DCO, is the gain factor of the DCO. The reference clock signal output by the DCO The feedback clock signal is generated by a programmable divider:

[0076] This feedback signal is used for closed-loop control of the phase-locked loop to ensure stable operation of the system. After obtaining a stable reference clock signal, a counter of the digital pulse width modulation (PWM) unit is established based on the signal to achieve duty cycle control of the PWM signal. The initial value of the counter is set to 0, and it increments in each clock cycle until it reaches the maximum value. Time zero:

[0077] The counter value is compared with the duty cycle data of the compensation pulse to generate a PWM waveform:

[0078] in, is the high level of the PWM signal, which is used to control the LED drive current. The PWM waveform is converted and amplified by the drive circuit to generate the drive signal for the LED array. Assume that the gain of the driver is , output drive current for: .

[0079] See also Figure 2 , Figure 2 A schematic block diagram of the structure of a small-pitch LED array driving waveform optimization device 200 provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, a small pitch LED array driving waveform optimization device 200 includes: The acquisition module 210 is used to acquire the spectral characteristics of the pixels in the small-pitch LED array and construct a mathematical model of the luminous characteristics of the pixels; A construction module 220 is used to perform parameter calculation on the mathematical model of the pixel light emission characteristics to obtain array display parameters, and to construct a display quality evaluation function according to the array display parameters; The test module 230 is used to perform waveform compensation on the original PWM drive signal, generate a complementary drive waveform pair, and perform an orthogonal test based on the complementary drive waveform pair to establish a characteristic correlation matrix between the drive parameters and the display effect; The generating module 240 is used to calculate the optimal driving parameter combination according to the characteristic correlation matrix and the display quality evaluation function, and generate the LED array driving signal according to the optimal driving parameter combination.

[0080] Through the cooperation of the above components, by establishing a complete mathematical model of pixel luminescence characteristics and combining temperature characteristic compensation, the accurate description of the display characteristics of the LED array is achieved. By adopting the design of complementary driving waveform pairs and injecting positive and negative compensation pulses into the row and column scanning signals, the voltage spikes and crosstalk effects in the driving process are effectively suppressed, and the uniformity of the display image is improved. The characteristic correlation matrix establishment method based on orthogonal testing is introduced. Through systematic parameter testing and data analysis, the accurate mapping relationship between driving parameters and display effects is established. An improved particle swarm optimization algorithm is designed. Through chaotic sequence initialization and adaptive weight adjustment, the convergence efficiency of the optimization process and the quality of the solution are improved. The driving signal generation scheme combining digital phase-locked loop and digital pulse width modulation is adopted to achieve high-precision control of the driving waveform and ensure the stability and reliability of the driving signal.

[0081] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0082] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0083] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for optimizing driving waveform of a small-pitch LED array, characterized in that: include: Collect the spectral characteristics of the pixels in the small-pitch LED array and build a mathematical model of the pixel luminescence characteristics; Performing parameter calculation on the mathematical model of the pixel light emission characteristics to obtain array display parameters, and constructing a display quality evaluation function according to the array display parameters; Performing waveform compensation on the original PWM drive signal to generate a complementary drive waveform pair, and performing an orthogonal test based on the complementary drive waveform pair to establish a characteristic correlation matrix between drive parameters and display effects; An optimal driving parameter combination is calculated according to the characteristic association matrix and the display quality evaluation function, and an LED array driving signal is generated according to the optimal driving parameter combination.

2. The method for optimizing driving waveform of a small-pitch LED array according to claim 1, characterized in that: The spectral characteristics of the pixels in the small-pitch LED array are collected to construct a mathematical model of the luminous characteristics of the pixels, including: Detecting the ambient temperature of the small-pitch LED array to obtain temperature acquisition data, and dividing the current sampling interval based on the temperature acquisition data; Inputting the driving current into each pixel point in the small-pitch LED array in sequence according to the current sampling interval, collecting spectrum data of each pixel point through a spectrum analyzer, and obtaining spectrum wavelength data and spectrum intensity data; Performing a main wavelength extraction operation on the spectrum wavelength data to obtain current-wavelength characteristic data, and performing an integration operation on the spectrum intensity data to obtain current-brightness characteristic data; Establishing a wavelength characteristic polynomial function based on the current-wavelength characteristic data, performing a least squares fitting operation on the wavelength characteristic polynomial function to obtain a wavelength characteristic coefficient matrix; Establishing a brightness characteristic polynomial function based on the current-brightness characteristic data, performing a least squares fitting operation on the brightness characteristic polynomial function to obtain a brightness characteristic coefficient matrix; A three-dimensional characteristic surface is established based on the wavelength characteristic coefficient matrix and the brightness characteristic coefficient matrix to generate a mathematical model of pixel point luminescence characteristics.

3. The method for optimizing driving waveform of a small-pitch LED array according to claim 1, characterized in that: The performing parameter calculation on the mathematical model of the pixel light emission characteristics to obtain array display parameters, and constructing a display quality evaluation function according to the array display parameters, comprises: Performing numerical integration operation on the mathematical model of the pixel luminous characteristic to obtain the average brightness data of each pixel, and calculating the ratio of the maximum brightness value to the minimum brightness value according to the average brightness data to obtain the brightness uniformity coefficient; Performing Fourier transformation on the waveform response parameters in the mathematical model of the pixel point luminous characteristics to obtain spectrum component data, and calculating the harmonic distortion factor according to the spectrum component data to obtain the waveform distortion rate; Establishing an optical crosstalk model between pixels according to the mathematical model of the pixel light emission characteristics, performing finite element analysis on the optical crosstalk model to obtain a crosstalk attenuation ratio; Constructing a weight matrix based on the brightness uniformity coefficient, the waveform distortion rate and the crosstalk attenuation ratio, and performing an eigenvalue decomposition operation on the weight matrix to obtain a weight coefficient; Normalizing the brightness uniformity coefficient, the waveform distortion rate, and the crosstalk attenuation ratio to obtain a standardized evaluation index, and linearly weighting the standardized evaluation index according to the weight coefficient to obtain an initial evaluation score; The initial evaluation score is transformed by a hyperbolic tangent function to obtain a display quality evaluation function.

4. The method for optimizing driving waveform of a small-pitch LED array according to claim 1, characterized in that: The waveform compensation is performed on the original PWM driving signal to generate a complementary driving waveform pair, and an orthogonal test is performed based on the complementary driving waveform pair to establish a characteristic correlation matrix between driving parameters and display effects, including: Performing edge detection on the original PWM driving signal to obtain a rising edge time sequence and a falling edge time sequence, and calculating duty cycle data of the driving signal according to the rising edge time sequence and the falling edge time sequence; Based on the duty cycle data, the original PWM drive signal is segmented to obtain a pre-charge segment waveform, a main drive segment waveform and a stable segment waveform, and the pre-charge segment waveform, the main drive segment waveform and the stable segment waveform are time-synchronized; Amplitude modulating the pre-charging segment waveform to generate a positive compensation pulse in the row scanning signal, and generating an inverted negative compensation pulse according to the waveform parameters of the positive compensation pulse; Establishing a compensation signal transfer function based on the positive compensation pulse and the negative compensation pulse, and performing frequency domain analysis on the compensation signal transfer function through Laplace transform to obtain a compensation signal spectrum; A crosstalk suppression filter is constructed according to the compensation signal spectrum, the positive compensation pulse and the negative compensation pulse are filtered to obtain a corrected compensation pulse pair, and the corrected compensation pulse pair is respectively superimposed on the pre-charge segments of the row scanning signal and the column scanning signal to generate a complementary driving waveform pair; An orthogonal test is performed on the complementary driving waveform pair within a preset voltage range and a preset time range to establish a characteristic correlation matrix between driving parameters and display effects.

5. The method for optimizing driving waveform of a small-pitch LED array according to claim 4, characterized in that: The performing of orthogonal testing on the complementary driving waveform pair within a preset voltage range and a preset time range to establish a characteristic correlation matrix between driving parameters and display effects includes: Constructing orthogonal test factors for the compensation pulse amplitude and pulse width of the complementary drive waveform pair, dividing the compensation pulse amplitude into N test points within a target voltage range, and dividing the pulse width into M test points within a target time range, to obtain N×M groups of test parameters; Generate an orthogonal test table based on the N×M groups of test parameters, and set parameters of the complementary drive waveform pair according to the orthogonal test table to obtain N×M groups of experimental waveforms; Performing waveform analysis on the current rising edge change rate, overshoot amplitude and crosstalk current of the N×M groups of experimental waveforms to obtain dynamic characteristic parameters of each group of waveforms; Constructing a multidimensional characteristic space based on the dynamic characteristic parameter and the display quality evaluation function, and performing a principal vector decomposition operation on the multidimensional characteristic space to obtain a feature vector group; A mapping relationship between driving parameters and display effects is established according to the characteristic vector group, a least square fitting calculation is performed on the mapping relationship to obtain a correlation coefficient matrix, and a singular value decomposition operation is performed on the correlation coefficient matrix to generate a characteristic correlation matrix between driving parameters and display effects.

6. The method for optimizing driving waveform of a small-pitch LED array according to claim 1, characterized in that: The calculating the optimal driving parameter combination according to the characteristic association matrix and the display quality evaluation function, and generating the LED array driving signal according to the optimal driving parameter combination, comprises: Constructing a target optimization function based on the characteristic association matrix, setting constraints on the target optimization function, taking the compensation pulse amplitude range and the pulse width range as the particle position boundary, and obtaining a target search space; Initializing particles in the target search space, constructing a chaotic sequence to perturb the initial particle group to obtain an initial solution space, and calculating the fitness value of each initial particle based on the display quality evaluation function; Performing adaptive inertia weight calculation on the particles in the initial solution space, dynamically adjusting the inertia weight according to the fitness value, and obtaining motion parameters of the particle swarm; Based on the motion parameters, a velocity update operation is performed on the particle swarm, and a local search operator is introduced to update the particle position to obtain an iterative candidate solution set; Performing constraint checks on the candidate solution set, processing out-of-bounds particles by Lagrange multiplier method to obtain a valid solution set, and sorting the valid solution set according to the display quality evaluation function; Extracting optimal particle position parameters from the effective solution set, decoding and converting the optimal particle position parameters into compensation pulse parameters, and generating an optimal driving parameter combination; The optimal driving parameter combination is used to generate a clock signal through a digital phase-locked loop, and the clock signal is input into a digital pulse width modulator to generate an LED array driving signal.

7. The method for optimizing driving waveform of a small-pitch LED array according to claim 6, characterized in that: The method of generating a clock signal by combining the optimal driving parameters through a digital phase-locked loop, and inputting the clock signal into a digital pulse width modulator to generate an LED array driving signal comprises: Digitally encoding the optimal driving parameter combination, converting the compensation pulse amplitude parameter into a frequency division coefficient, converting the pulse width parameter into a phase control word, and obtaining a digital phase-locked loop control parameter; A phase accumulator is constructed based on the digital phase-locked loop control parameter, and a sine mapping operation is performed on the accumulated output of the phase accumulator to obtain a reference signal of a phase frequency detector; Performing phase frequency comparison on the reference signal and the feedback signal, filtering the comparison result through a digital loop filter to obtain a control voltage of the digital controlled oscillator; Inputting the control voltage into a digitally controlled oscillator for frequency modulation to generate a reference clock signal, and performing frequency division processing on the reference clock signal to obtain a feedback clock signal; A counter of a digital pulse width modulation unit is established based on the reference clock signal, and a comparison operation is performed on the count value of the counter and the duty cycle data of the compensation pulse to obtain a pulse width modulation waveform, and the pulse width modulation waveform is subjected to level conversion and current amplification through a driver circuit to generate an LED array drive signal.

8. A device for optimizing driving waveform of a small-pitch LED array, characterized in that: The method for optimizing driving waveform of a small-pitch LED array according to any one of claims 1 to 7 comprises: The acquisition module is used to collect the spectral characteristics of the pixels in the small-pitch LED array and build a mathematical model of the luminous characteristics of the pixels; A construction module, used to perform parameter calculation on the mathematical model of the pixel light emission characteristics to obtain array display parameters, and to construct a display quality evaluation function according to the array display parameters; A test module, used to perform waveform compensation on the original PWM drive signal, generate a complementary drive waveform pair, and perform an orthogonal test based on the complementary drive waveform pair to establish a characteristic correlation matrix between drive parameters and display effects; A generating module is used to calculate an optimal driving parameter combination according to the characteristic association matrix and the display quality evaluation function, and generate an LED array driving signal according to the optimal driving parameter combination.

Images