Multi-path adjusting system of LED lamp

Abstract

The invention relates to the technical field of LED lamp adjustment, in particular to a multi-path adjusting system of an LED lamp. The method comprises the following steps that: a parameter acquisition module acquires multi-path real-time light intensity parameters and performs noise covariance estimation and channel interference analysis to generate a noise matrix and a channel crosstalk matrix, a change simulation module predicts light intensity change based on the matrixes and generates a simulation LED lamp group, and a dead zone compensation module calculates a duty ratio and performs compensation. According to the method, an anti-overlapping sequence is formed, in the phase distribution module, phase quantization and peak shifting distribution are carried out on the anti-overlapping sequence to obtain a peak shifting control sequence, the channel coordination module generates a multi-path control sequence according to temperature compensation processing, and the feedback optimization module continuously adjusts an LED lamp set by collecting a feedback strong light value to achieve accurate multi-path adjustment. According to the invention, the accuracy of LED lamp adjustment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of LED lamp adjustment, and in particular to a multi-channel adjustment system for LED lamps. Background Art

[0002] In modern lighting technology, LED lamps are widely used due to their high efficiency and long life. In actual use, LED lamp groups often face problems such as uneven light intensity, insufficient adjustment accuracy and excessive energy consumption. These problems not only affect the lighting effect, but also may lead to energy waste and environmental burden. Traditional light intensity adjustment methods often rely on simple manual control or basic automation algorithms, which leads to slow response to environmental changes and difficulty in achieving real-time adjustment. Interference between light sources and signal noise will also affect the accuracy of adjustment, thereby causing a decrease in the overall lighting effect. The deficiencies of existing technologies in noise management and channel interference analysis limit the reliability of light intensity prediction and adjustment. Summary of the invention

[0003] Based on this, it is necessary to provide a multi-channel adjustment system for LED lamps to solve at least one of the above technical problems.

[0004] To achieve the above object, a multi-channel regulation system for LED lamps includes: The parameter acquisition module is used to obtain multiple channels of real-time light intensity parameters; perform noise covariance estimation on the multiple channels of real-time light intensity parameters to obtain a noise matrix; perform channel interference analysis on the multiple channels of real-time light intensity parameters to generate a channel crosstalk matrix; The change simulation module is used to predict the light intensity change of the real-time light intensity parameters based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; and perform change simulation processing according to the predicted light intensity value to obtain a simulated LED light group; The dead zone compensation module is used to calculate the multi-channel duty cycle of the simulated LED light group to obtain a basic duty sequence; and perform dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence; The phase allocation module is used to perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence; The channel coordination module is used to perform multi-channel temperature compensation on the peak-shifting control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence; The feedback optimization module is used to perform multi-channel adjustment control according to the multi-channel control sequence and collect the feedback strong light value at the same time; based on the feedback strong light value, the simulated LED light group is feedback optimized to perform multi-channel adjustment of the LED light.

[0005] The present invention improves the real-time acquisition capability of light intensity parameters, so that the system can quickly respond to environmental changes and enhance the flexibility and adaptability of adjustment. The combination of noise covariance estimation and channel interference analysis improves the ability to accurately identify signal noise and ensures the reliability and stability of data. Through the prediction of light intensity changes, forward-looking adjustment of light intensity can be achieved, and the performance of the LED lamp group is optimized. The dead zone compensation processing effectively eliminates the signal overlap problem and improves the accuracy and efficiency of multi-channel control. The application of phase peak shifting distribution technology optimizes the light emission timing of the light source and reduces the loss of light efficiency and energy consumption. Multi-channel temperature compensation ensures the consistency of light intensity under different environmental conditions and maintains the stability of the lighting effect. The feedback optimization mechanism continuously adjusts the system parameters through real-time feedback of the strong light value, thereby improving the accuracy and intelligence level of adjustment. The overall system structure is reasonably designed, and the modules collaborate smoothly, which enhances the integration and maintainability of the system.

[0006] Preferably, the parameter acquisition module is used to: obtain multi-channel real-time light intensity parameters; perform noise covariance estimation on the multi-channel real-time light intensity parameters to obtain a noise matrix; perform channel interference analysis on the multi-channel real-time light intensity parameters to generate a channel crosstalk matrix, which is specifically used for: Acquire multiple channels of real-time light parameters; perform smoothing and filtering on the multiple channels of real-time light parameters to obtain filtered light intensity parameters; Perform state space mapping on the filter intensity parameters to obtain an intensity state matrix; perform noise covariance estimation on the intensity state matrix to obtain a noise matrix; Perform multi-channel topology construction on multi-channel real-time light intensity parameters to obtain the network interference topology structure; Based on the network interference topology, channel crosstalk analysis is performed on multi-channel real-time optical intensity parameters to generate a channel crosstalk matrix.

[0007] The present invention improves the response speed and reliability of the system by accurately acquiring multi-channel real-time light intensity parameters. The smoothing filtering process effectively reduces the influence of noise on the light intensity parameters, ensuring the stability and accuracy of the data. The state space mapping optimizes the representation of light intensity data and enhances the effectiveness of subsequent analysis. The noise covariance estimation provides noise characteristic information and supports more accurate signal processing. The multi-channel topology construction improves the ability to analyze network interference and ensures the effective identification of interference relationships between different channels. The channel crosstalk matrix provides an important basis for subsequent light intensity regulation and helps to achieve more accurate control strategies.

[0008] Preferably, the change simulation module is used to: predict the light intensity change of the real-time light intensity parameter based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; perform change simulation processing according to the predicted light intensity value to obtain a simulated LED light group, specifically for: Based on the channel crosstalk matrix, the noise matrix is ​​mapped to a channel correlation matrix to obtain a crosstalk correlation matrix; Perform spectrum conversion on real-time light intensity parameters based on the crosstalk correlation matrix to generate a light intensity spectrum matrix; Based on the light intensity spectrum matrix, the light intensity change is predicted to obtain the predicted light intensity value; according to the preset basic parameters of the lamp group, the predicted light intensity value is mapped to the light intensity parameter to generate the corresponding parameters of the light intensity; The light group topology is reconstructed based on the parameters corresponding to the light intensity to obtain a simulated LED light group.

[0009] The present invention improves the responsiveness and adaptability of the LED lamp group by predicting the light intensity change of real-time light intensity parameters. The channel association mapping effectively reveals the relationship between noise and crosstalk and enhances the interference management capability of the system. The generation of the light intensity spectrum matrix makes the light intensity change analysis more comprehensive and accurate. The light intensity parameter mapping provides the lamp group with an optimized configuration based on real-time data. The lamp group topology reconstruction ensures the performance and stability of the simulated LED lamp group, which improves the accuracy and efficiency of light intensity regulation as a whole and provides a solid foundation for intelligent lighting applications.

[0010] Preferably, the dead zone compensation module is used to: perform multi-channel duty cycle calculation on the simulated LED light group to obtain a basic duty sequence; perform dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence, specifically for: Perform channel brightness equalization processing on the simulated LED light group to obtain balanced light intensity of the light group; Perform multi-channel gamma mapping on the light group to balance the light intensity and generate a balanced linear matrix; Based on the balanced linear matrix, channel interaction compensation is deduced for the simulated LED light group to obtain an interaction compensation matrix; Perform multi-dimensional duty cycle calculation on the interactive compensation matrix to generate a basic duty sequence; The edge conflict detection is performed on the basic duty sequence to obtain the timing scheduling parameters; the dead zone balance reconstruction is performed based on the timing scheduling parameters to generate the anti-overlap sequence.

[0011] The present invention ensures the accuracy of the basic duty sequence of the simulated LED lamp group through multi-channel duty cycle calculation, the dead zone compensation processing effectively avoids the light overlapping phenomenon and improves the brightness balance of the lamp group, the channel brightness balance processing enhances the overall light intensity performance of the lamp group, the balanced linear matrix generation makes the light intensity distribution more uniform, the deduction of the interactive compensation matrix provides effective compensation for the mutual influence between different channels, the edge conflict detection ensures the rationality and effectiveness of the timing scheduling, and the dead zone balance reconstruction further optimizes the operating efficiency of the lamp group, thereby improving the adjustment performance and lighting quality of the LED lamp as a whole.

[0012] Preferably, the specific steps of performing edge conflict detection on the basic duty sequence to obtain the timing scheduling parameters; and performing dead zone balance reconstruction based on the timing scheduling parameters to generate the anti-overlap sequence are: Perform multi-path edge recognition on the basic duty sequence to obtain multi-path switching boundaries; perform switching conflict detection based on the multi-path switching boundaries to generate switching conflict points; Assign priorities to the switching conflict points and obtain timing scheduling parameters; Perform scheduling simulation on the basic duty sequence based on the timing scheduling parameters to obtain the simulated scheduling behavior; Perform dead zone demand analysis on simulated scheduling behavior and generate a set of adjustment intervals; The inter-channel balance compensation is performed on the basic duty sequence according to the adjustment interval set to obtain an anti-overlap sequence.

[0013] The present invention ensures the accurate positioning of the switching boundary through multi-path edge recognition and improves the efficiency of conflict detection. The generation of switching conflict points provides key data for subsequent scheduling. Priority allocation optimizes timing scheduling parameters and enhances the flexibility and effectiveness of scheduling. Scheduling simulation provides an intuitive reference for actual behavior. Dead zone demand analysis ensures the reasonable setting of the adjustment interval. Inter-channel balance compensation achieves the uniformity of light intensity output. The generation of anti-overlapping sequences effectively avoids the phenomenon of light overlap, thereby improving the performance and adjustment accuracy of the LED lamp group as a whole.

[0014] Preferably, the phase allocation module is used to: perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence, specifically used to: Performing phase quantization processing on the anti-overlapping sequence to obtain a quantized phase sequence; Perform phase frequency statistics on the phase quantization sequence to generate the phase quantity; Perform power grouping analysis on the anti-overlap sequence based on the number of phases to obtain a phase configuration reference system; The anti-overlapping sequence is interference optimized and rearranged according to the phase configuration reference system to generate a peak-shifting control sequence.

[0015] The present invention ensures that the phase information of the anti-overlapping sequence is accurately captured through phase quantization processing. The quantized phase sequence provides basic data for subsequent analysis. Phase frequency statistics reveal the phase distribution characteristics. The generated phase quantity provides a basis for power grouping analysis. The power grouping analysis effectively optimizes the phase configuration reference system. The interference optimization rearrangement improves the light intensity stability of the anti-overlapping sequence. The generation of the peak-shifting control sequence reduces interference during light switching, improves the operating efficiency and light quality of the LED light group, and enhances the accuracy of light adjustment as a whole.

[0016] Preferably, the specific steps of performing interference optimization rearrangement on the anti-overlap sequence according to the phase configuration reference system to generate the peak shifting control sequence are: Perform channel distribution identification on the anti-overlap sequence in the phase configuration reference system to obtain a channel distribution sequence; The phase clustering region is divided according to the channel distribution sequence to obtain the phase clustering coordinates; Channels are grouped based on phase clustering coordinates to generate a channel grouping set; Performing inter-group interference evaluation on the channel grouping set to generate an inter-channel interference value; Based on the inter-channel interference value, the channel grouping set is screened for rearrangement targets to generate rearrangement target channels; The sequence of the rearrangement target channel is rearranged according to the phase configuration reference system to generate a peak shifting control sequence.

[0017] The present invention ensures that the channel management of the anti-overlapping sequence is clearer through channel distribution identification. The generation of the channel distribution sequence provides data support for the subsequent segmentation of the phase aggregation area. The determination of the phase aggregation coordinates accurately reflects the relationship between the channels. The formation of channel grouping optimizes the coordination between groups. The inter-group interference assessment provides a basis for the calculation of the inter-channel interference value, effectively identifies potential interference problems, and the screening of the rearranged target channels improves the flexibility and response speed of the system. The sequence rearrangement processing ensures the continuity and stability of the light intensity output based on the phase configuration reference system. The generation of the peak-shifting control sequence effectively avoids the overlapping and flickering of lights, which improves the adjustment accuracy and usage experience of the LED lamp group as a whole, and provides a guarantee for the reliability of the intelligent lighting system.

[0018] Preferably, the channel coordination module is used to: perform multi-channel temperature compensation on the peak-shifting control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence, specifically for: Based on the preset basic parameters of the lamp group, the thermal field gradient mapping is performed on the peak-shifting control sequence to obtain the temperature influence parameters; According to the temperature influence parameter, the channel temperature coefficient compensation is performed on the peak shifting control sequence to obtain the temperature compensation sequence of each channel; Perform compensation integration processing on the temperature compensation sequence of each channel to generate a temperature compensation sequence; Perform channel coordination processing according to the temperature compensation sequence to obtain the channel correction coefficient; perform luminous flux reconstruction based on the channel correction coefficient to generate a light efficiency compensation sequence; The light effect compensation sequence is processed in time domain synchronization to generate a multi-channel control sequence.

[0019] The present invention ensures the stability and reliability of the peak-shifting control sequence through multi-channel temperature compensation. The thermal field gradient mapping accurately reflects the influence of temperature on the performance of the lamp group, and provides a scientific basis for temperature influence parameters. Channel temperature coefficient compensation improves the consistency of light intensity of each channel. The generation of temperature compensation sequence effectively integrates the temperature characteristics of each channel. Channel coordination processing lays the foundation for the overall performance optimization of the system. The calculation of channel correction coefficient makes the luminous flux reconstruction more accurate. The generation of light efficiency compensation sequence improves the energy efficiency performance of the light source. Time domain synchronization processing ensures the coordination and consistency of the control sequence, which improves the lighting effect and utilization efficiency of the LED lamp group in different environments as a whole, and provides strong support for the stable operation of the intelligent lighting system.

[0020] Preferably, the feedback optimization module is used to: perform multi-channel adjustment control according to the multi-channel control sequence, and collect the feedback strong light value at the same time; perform feedback optimization on the simulated LED light group based on the feedback strong light value to perform multi-channel adjustment of the LED light, specifically for: Perform hardware timing conversion on multi-channel control sequences to obtain drive control signals; Perform multi-channel adjustment control according to the driving control signal and collect feedback intensity value at the same time; According to the feedback strong light value, the light intensity deviation of the simulated LED light group is compared to obtain the light intensity deviation comparison value; Feedback optimization is performed on the simulated LED lamp group based on the light intensity deviation comparison value to perform multi-channel adjustment of the LED lamp.

[0021] The present invention ensures the response speed and accuracy of the LED lamp group through multi-channel adjustment control. The hardware timing conversion effectively converts the multi-channel control sequence into a driving signal, thereby improving the stability of the system. The collection of feedback strong light values ​​provides basic data for real-time adjustment. The light intensity deviation comparison realizes the accurate comparison between the actual light intensity and the preset value. The calculation of the light intensity deviation comparison value provides a basis for subsequent optimization. The feedback optimization mechanism improves the adjustment accuracy and flexibility of the LED lamp group. The execution of multi-channel adjustment ensures the consistency of lighting effects in different environments, which enhances the performance and usage experience of the LED lamp group as a whole, and provides solid support for the adaptive adjustment of the intelligent lighting system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The figure is a structural schematic diagram of a multi-channel regulation system for an LED lamp. DETAILED DESCRIPTION

[0023] The technical method of the present invention is clearly and completely described below in conjunction with the accompanying drawings.

[0024] To achieve this, please refer to Figure 1 , a multi-channel regulation system for LED lamps comprises: The parameter acquisition module is used to obtain multiple channels of real-time light intensity parameters; perform noise covariance estimation on the multiple channels of real-time light intensity parameters to obtain a noise matrix; perform channel interference analysis on the multiple channels of real-time light intensity parameters to generate a channel crosstalk matrix; The change simulation module is used to predict the light intensity change of the real-time light intensity parameters based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; and perform change simulation processing according to the predicted light intensity value to obtain a simulated LED light group; The dead zone compensation module is used to calculate the multi-channel duty cycle of the simulated LED light group to obtain a basic duty sequence; and perform dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence; The phase allocation module is used to perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence; The channel coordination module is used to perform multi-channel temperature compensation on the peak-shifting control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence; The feedback optimization module is used to perform multi-channel adjustment control according to the multi-channel control sequence and collect the feedback strong light value at the same time; based on the feedback strong light value, the simulated LED light group is feedback optimized to perform multi-channel adjustment of the LED light.

[0025] The present invention improves the real-time acquisition capability of light intensity parameters, so that the system can quickly respond to environmental changes and enhance the flexibility and adaptability of adjustment. The combination of noise covariance estimation and channel interference analysis improves the ability to accurately identify signal noise and ensures the reliability and stability of data. Through the prediction of light intensity changes, forward-looking adjustment of light intensity can be achieved, and the performance of the LED lamp group is optimized. The dead zone compensation processing effectively eliminates the signal overlap problem and improves the accuracy and efficiency of multi-channel control. The application of phase peak shifting distribution technology optimizes the light emission timing of the light source and reduces the loss of light efficiency and energy consumption. Multi-channel temperature compensation ensures the consistency of light intensity under different environmental conditions and maintains the stability of the lighting effect. The feedback optimization mechanism continuously adjusts the system parameters through real-time feedback of the strong light value, thereby improving the accuracy and intelligence level of adjustment. The overall system structure is reasonably designed, and the modules collaborate smoothly, which enhances the integration and maintainability of the system.

[0026] In the embodiment of the present invention, refer to Figure 1 , is a schematic diagram of the structure of a multi-channel regulation system for an LED lamp of the present invention. In this example, the multi-channel regulation system for an LED lamp includes: The parameter acquisition module is used to obtain multiple channels of real-time light intensity parameters; perform noise covariance estimation on the multiple channels of real-time light intensity parameters to obtain a noise matrix; perform channel interference analysis on the multiple channels of real-time light intensity parameters to generate a channel crosstalk matrix; In this embodiment, the light intensity signal output by the LED lamp is collected in real time through a multi-channel light sensor array. The light sensor model is a TSL2591 high-precision digital light intensity sensor. Its data interface is an I2C protocol (two-wire interface). The sampling frequency is set to 500Hz. Each light intensity signal is converted into a digital signal through an ADC module (analog-to-digital converter) built in the sensor. After the collection is completed, the light intensity parameters are transmitted to the central processing unit (CPU) through the STM32F103C8T6 microcontroller for storage. The light intensity parameters are processed by the embedded data processing unit through a real-time covariance calculation method to estimate the noise distribution characteristics. The dimension of the covariance matrix is ​​determined according to the number of acquisition channels. Taking 8 channels as an example, the covariance matrix is ​​an 8×8 symmetric matrix. Subsequently, the interference characteristics of each light intensity are extracted based on the wavelet decomposition method. The crosstalk coefficient between channels is calculated through channel cross-correlation analysis to generate a channel crosstalk matrix. The crosstalk matrix is ​​also an 8×8 symmetric matrix, and the result is stored in the on-chip memory.

[0027] The change simulation module is used to predict the light intensity change of the real-time light intensity parameters based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; and perform change simulation processing according to the predicted light intensity value to obtain a simulated LED light group; In this embodiment, the noise matrix and channel crosstalk matrix are called from the parameter acquisition module, and the light intensity change prediction is taken as the goal. The light intensity change prediction model based on support vector regression (Support Vector Regression) is used, and the input parameters are set as the real-time value of each light intensity and the corresponding noise matrix and crosstalk matrix. The prediction model is implemented through the SVR toolbox in MATLAB. The model output is the predicted light intensity value of each channel at the next time step. After the prediction is completed, the predicted light intensity value is input into the simulation tool. Based on the SPICE (Simulation Program with Integrated Circuit Emphasis) platform, the change simulation of the LED light group is performed to determine the dynamic behavior of each channel under different light intensity states, and finally generate a virtual simulation LED light group model.

[0028] The dead zone compensation module is used to calculate the multi-channel duty cycle of the simulated LED light group to obtain a basic duty sequence; and perform dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence; In this embodiment, the output characteristic data of the simulated LED lamp group is obtained from the simulation module, and the duty cycle sequence of each channel is calculated based on a dimming scheme based on PWM (pulse width modulation). The duty cycle sequence is discretized based on the control period, with 1ms as the minimum time interval. According to the duty cycle sequence of each channel, the dead zone compensation of the time interval of adjacent channels is performed. The dead zone compensation is achieved by inserting a blank time period with a minimum pulse width of 0.5ms. The GPIO (general input and output) module of STM32 is used to output the compensated PWM signal, and the result is stored as an anti-overlapping sequence.

[0029] The phase allocation module is used to perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence; In this embodiment, the anti-overlapping sequence is called from the dead zone compensation module, and the frequency distribution characteristics of the anti-overlapping sequence are analyzed using a phase quantization method based on FFT (Fast Fourier Transform). The phase quantization accuracy is set to a 1-degree step of 360 degrees, and the number of phases is calculated. According to the number of phases, the anti-overlapping sequence is phase-shifted at fixed intervals. The allocation algorithm sets the initial phase offset to 10 degrees and increases it by 10 degrees each time until the allocation is completed. The allocated peak-shifting control sequence is written to the memory through the DMA (Direct Memory Access) module of the STM32.

[0030] The channel coordination module is used to perform multi-channel temperature compensation on the peak-shifting control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence; In this embodiment, the peak-shifting control sequence generated by the phase allocation module is called, and data compensation processing based on the temperature sensor is performed on each channel. The temperature sensor uses the DS18B20 single-bus digital temperature sensor with an accuracy of 0.1°C. Each sensor collects temperature data in real time and transmits it to the central processing unit through the serial port. According to the collected temperature data, the compensation factor of temperature to light intensity is calculated by linear interpolation method, and superimposed on the peak-shifting control sequence to complete the generation of the temperature compensation sequence. Subsequently, the output power of multiple channels is synchronously adjusted through a multi-channel coordination method based on an improved weighted average algorithm to generate the final multi-channel control sequence.

[0031] The feedback optimization module is used to perform multi-channel adjustment control according to the multi-channel control sequence and collect the feedback strong light value at the same time; based on the feedback strong light value, the simulated LED light group is feedback optimized to perform multi-channel adjustment of the LED light.

[0032] In this embodiment, a multi-channel control sequence is obtained from the channel coordination module, and the actual LED light group is adjusted and controlled through the driving circuit. The driving circuit adopts a constant current source driving scheme with an output power of 350mA per channel. During the adjustment and control process, the feedback light intensity value is collected in real time through a high-sensitivity light intensity sensor. The sensor model is OPT3001 with a resolution of 0.01 lx. After the feedback light intensity value is transmitted to the central processing unit, the feedback light intensity value is compared with the target light intensity value in real time through the PID (proportional integral differential) control algorithm, the error is calculated and the control sequence is dynamically adjusted to optimize the multi-channel adjustment performance of the LED lamp, and finally achieve precise adjustment of the LED light group.

[0033] In this embodiment, the parameter acquisition module is used to: obtain multiple channels of real-time light intensity parameters; perform noise covariance estimation on the multiple channels of real-time light intensity parameters to obtain a noise matrix; perform channel interference analysis on the multiple channels of real-time light intensity parameters to generate a channel crosstalk matrix, which is specifically used for: Acquire multiple channels of real-time light parameters; perform smoothing and filtering on the multiple channels of real-time light parameters to obtain filtered light intensity parameters; Perform state space mapping on the filter intensity parameters to obtain an intensity state matrix; perform noise covariance estimation on the intensity state matrix to obtain a noise matrix; Perform multi-channel topology construction on multi-channel real-time light intensity parameters to obtain the network interference topology structure; Based on the network interference topology, channel crosstalk analysis is performed on multi-channel real-time optical intensity parameters to generate a channel crosstalk matrix.

[0034] In this embodiment, a TSL2591 light intensity sensor array is used, and each sensor is independently connected to the I2C bus of the STM32F429 microcontroller. The sampling frequency is set to 1000 Hz. During the sampling process, each sensor converts the analog light intensity signal into a digital signal through its internal analog-to-digital converter, and transmits it to the central processing unit (CPU) with a 16-bit resolution. The collected digital light intensity data is stored in an external FRAM (ferroelectric memory) module for subsequent processing. In the actual arrangement, 8 light intensity sensors are installed at the light output positions of different channels, and the real-time light intensity data stored in the FRAM is smoothed and filtered. A smoothing processing method based on a moving average filtering algorithm is used. Five consecutive sampling values ​​of each light intensity parameter are used as a sliding window. The average value of the data in the sliding window is calculated to smooth out high-frequency noise. The smoothed data is stored in the internal RAM of the STM32 to reduce storage and reading delays. The width parameter of the sliding window is determined to be 5 sampling points according to experimental tests. The filtered light intensity parameters are mapped using a state space transformation method, and the state space is constructed using the MATLAB Simulink tool. Mapping model, the input parameter is the filtered light intensity data, the state space transformation is based on the system dynamic characteristics of the light intensity channel, each light intensity signal is converted from the time domain to the state space domain, and the output is the light intensity state matrix. The dimension of the light intensity state matrix is ​​determined by the number of light intensity sensors. In the case of 8 channels, the state matrix is ​​a 8×2 two-dimensional matrix, where the first column represents the current state of the light intensity signal, and the second column represents the state change rate of the light intensity signal. After the mapping is completed, the light intensity state matrix is ​​transmitted to the data analysis module. Based on the light intensity state matrix, the covariance estimation of the noise characteristics of each channel signal is performed. , a noise covariance calculation method based on recursive least squares method is adopted. In the covariance calculation process, the covariance matrix is ​​first initialized to a zero matrix, and then each row in the light intensity state matrix is ​​used as input for dynamic iterative calculation. The calculation result is an 8×8 symmetric covariance matrix. The diagonal elements of the covariance matrix represent the noise variance of each light intensity signal, and the non-diagonal elements represent the noise covariance between channels. The light intensity channel is used as a node, and the interaction relationship of the light intensity parameters is used as an edge to construct a network. The method based on the Pearson correlation coefficient is used to quantify the interaction strength between channels, and the threshold is set to 0.5 or more correlations as valid edges to generate a weighted undirected graph. The node weights in the topological structure represent the intensity of the light intensity signal, and the edge weights represent the interaction intensity between channels, thereby generating a topological structure. The generated multi-channel topological structure is used to analyze the crosstalk between channels. The shortest path analysis method based on graph theory is used to calculate the crosstalk path length between channels. At the same time, combined with the interaction weight data of multiple channels, the generalized interference index (General Interference Index) is used to evaluate the crosstalk intensity of each channel. The result of the crosstalk analysis is the channel crosstalk matrix, which is an 8×8 symmetric matrix, in which each element represents the degree of interference between two channels. .

[0035] In this embodiment, the change simulation module is used to: predict the light intensity change of the real-time light intensity parameter based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; perform change simulation processing according to the predicted light intensity value to obtain a simulated LED light group, specifically for: Based on the channel crosstalk matrix, the noise matrix is ​​mapped to a channel correlation matrix to obtain a crosstalk correlation matrix; Perform spectrum conversion on real-time light intensity parameters based on the crosstalk correlation matrix to generate a light intensity spectrum matrix; Based on the light intensity spectrum matrix, the light intensity change is predicted to obtain the predicted light intensity value; according to the preset basic parameters of the lamp group, the predicted light intensity value is mapped to the light intensity parameter to generate the corresponding parameters of the light intensity; The light group topology is reconstructed based on the parameters corresponding to the light intensity to obtain a simulated LED light group.

[0036] In this embodiment, the MATLAB tool is used to perform channel correlation mapping processing on the channel crosstalk matrix and the noise matrix. The channel crosstalk matrix (8×8 dimension) is used as the input matrix A, and the noise matrix (8×8 dimension) is used as the input matrix B. The matrix multiplication operation is performed to obtain the channel correlation matrix C. During the mapping process, each row and column of A and B are matched and calculated in turn. Each element value of the channel correlation matrix C represents the coupling degree of channel noise and crosstalk. The channel correlation matrix C is loaded and matched with the sampling sequence of the real-time light intensity parameter channel by channel. The sampling sequence is composed of the previously stored filtered light intensity data. According to the information provided, the real-time light intensity parameters are combined with the correlation matrix C to perform weight distribution calculation to generate a spectral sequence. The spectral sequence data is reorganized into an 8×16 light intensity spectrum matrix, where 8 represents the number of light intensity channels and 16 represents the spectral segment division of each channel. The spectral division range is from 380 nanometers to 780 nanometers (the visible light range of the human eye). The deep learning model LSTM (long short-term memory network) is used to predict the light intensity spectrum matrix. The LSTM model is built in the TensorFlow framework. The model input is the time series of the light intensity spectrum matrix, and the output is the unseen data of each channel. The light intensity value is obtained from the actual sampled light intensity historical data. The time window for light intensity change prediction is set to 100ms. The training data of the model is extracted from the actual sampled light intensity historical data. The GPU (graphics processing unit) is used to accelerate the calculation during the prediction process to improve the real-time performance of the model. The predicted light intensity value is parameterized based on the basic parameters of the lamp group. The basic parameters of the lamp group include the power (0.5W) of the LED chip, the driving current (20mA), the luminous efficiency (100 lumens / watt) and the radiation angle (120 degrees). The predicted light intensity value is converted into the actual LED luminous flux value by table lookup, and combined with the geometric position parameters of the lamp group to generate the specific light intensity corresponding parameters of each LED lamp. The light intensity corresponding parameters are recorded in the form of a triple, including the channel number, the light intensity value and the spatial coordinates. Based on the light intensity corresponding parameters, the topological structure of the simulated LED lamp group is constructed, and the topological reconstruction is completed using the Blender 3D modeling tool. First, the light intensity corresponding parameters are imported into Blender, and the 3D coordinate points are automatically generated by the Python script, and the light intensity value is assigned to each coordinate point as the material luminous intensity. Then, the coordinate points are connected into an overall structure according to the preset physical layout rules of the lamp group.

[0037] In this embodiment, the dead zone compensation module is used to: perform multi-channel duty cycle calculation on the simulated LED light group to obtain a basic duty sequence; perform dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence, specifically for: Perform channel brightness equalization processing on the simulated LED light group to obtain balanced light intensity of the light group; Perform multi-channel gamma mapping on the light group to balance the light intensity and generate a balanced linear matrix; Based on the balanced linear matrix, channel interaction compensation is deduced for the simulated LED light group to obtain an interaction compensation matrix; Perform multi-dimensional duty cycle calculation on the interactive compensation matrix to generate a basic duty sequence; The edge conflict detection is performed on the basic duty sequence to obtain the timing scheduling parameters; the dead zone balance reconstruction is performed based on the timing scheduling parameters to generate the anti-overlap sequence.

[0038] In this embodiment, for each channel in the simulated LED lamp group, the original light intensity matrix of the lamp beads is , the target light intensity is , by calculating the gain matrix To achieve brightness balance, the gain calculation formula is as follows: ; The gain matrix Applied to the raw light intensity matrix , and get the balanced light intensity matrix , the calculation formula is as follows: ; Assumptions The original light intensity of a channel is 120cd. is 90cd, then the gain , which is 1.33, the light intensity after equalization , which is 120cd. The gain matrix and the balance matrix are calculated through matrix operation tools (such as MATLAB or Python's NumPy library). The matrix dimension is consistent with the number of lamp channels. The balanced light intensity data is gamma mapped. In this process, the light intensity of each channel is transformed by the specified gamma coefficient. The gamma coefficient is selected by the system according to the preset brightness standard. Assuming the gamma coefficient is 2.2, the light intensity value of each channel will be adjusted according to the coefficient to make the light intensity distribution more uniform. For example, if the balanced light intensity of a channel is 80 units, after the gamma mapping conversion, the mapping result of the channel is the transformed light intensity value. The mapping values ​​of all channels are calculated and updated in turn to obtain the light intensity data after gamma mapping. The light intensity data of each channel is collected. The light intensity data of each channel is collected and recorded as a matrix. Then, the compensation coefficient of each channel is calculated. The coefficient is dynamically adjusted according to the interaction effect of adjacent channels. The goal of compensation is to eliminate mutual interference between channels. Assuming that the compensation coefficient of a channel is 1.1 and that of another channel is 0.9, the compensation coefficient of each channel will be adjusted by an algorithm during the compensation process, and finally an interactive compensation matrix is ​​generated, which contains the final compensation coefficients of all channels. After compensation, the light intensity data of each channel is updated, and the light intensity value of each channel is compared with the target light intensity value to calculate the duty cycle. Based on the compensated light intensity data and the target light intensity data, a basic duty cycle sequence is generated. For example, assuming that the compensated light intensity of a channel is 70 units and the target light intensity is 100 units, the duty cycle of the channel is 70%. After similar calculations for all channels, the basic duty cycle sequence is obtained. The generated basic duty cycle sequence is subjected to edge conflict detection to check the time interval between the rising edge and the falling edge of the duty cycle signal. If the time difference between the rising edge and the falling edge of the two signals is less than the preset dead time threshold, it is determined that there is a conflict. Assuming that the time difference between the rising edge and the falling edge of the two signals is 2 units, and the dead time threshold is set to 5 units, it is determined that a conflict occurs. The conflict information is recorded and fed back to The scheduling module reconstructs the dead zone balance based on the timing scheduling parameters, optimizes the time interval between signals by adjusting the triggering timing of the signals to avoid conflicts. If it is detected that the interval between the signals is less than the dead zone time threshold, the time interval is adjusted to meet the preset dead zone time requirement. For example, if the time interval between two signals is 2 units and the dead zone time threshold is 5 units, the triggering timing of one of the signals is adjusted to an interval of 5 units to avoid overlap. After adjustment, an anti-overlapping duty cycle sequence is generated and transmitted to the LED driver module, and finally an optimized signal is output.

[0039] In this embodiment, the specific steps of performing edge conflict detection on the basic duty sequence to obtain the timing scheduling parameters; performing dead zone balance reconstruction based on the timing scheduling parameters to generate the anti-overlap sequence are: Perform multi-path edge recognition on the basic duty sequence to obtain multi-path switching boundaries; perform switching conflict detection based on the multi-path switching boundaries to generate switching conflict points; Assign priorities to the switching conflict points and obtain timing scheduling parameters; Perform scheduling simulation on the basic duty sequence based on the timing scheduling parameters to obtain the simulated scheduling behavior; Perform dead zone demand analysis on simulated scheduling behavior and generate a set of adjustment intervals; The inter-channel balance compensation is performed on the basic duty sequence according to the adjustment interval set to obtain an anti-overlap sequence.

[0040] In this embodiment, by capturing the rising edge and falling edge positions of the signal, the switching point position of each signal is determined, and the detected switching point is recorded in the storage unit in the form of a timestamp. The edge detection module samples each signal at 100 nanoseconds through a high-speed sampler. Assuming that the sampling result of a certain signal is that the rising edge time is 10 microseconds and the falling edge time is 15 microseconds, the switching boundary is the time point of 10 microseconds and 15 microseconds. All signals are processed in the same way in turn, and finally a multi-channel switching boundary data table is generated. The data table includes the rising edge and falling edge time points of each signal. The multi-channel switching boundary data table is analyzed by the switching conflict detection module, and the rising edge and falling edge time points of each signal are compared. If it is found that the boundary time difference of adjacent signals is less than the minimum switching time point set by the system, If the switching interval threshold is exceeded, it is marked as a switching conflict point. Assuming that the minimum switching interval threshold set by the system is 5 microseconds, if the falling edge time of signal A is 20 microseconds and the rising edge time of signal B is 23 microseconds, this situation will not trigger a conflict. However, if the rising edge time of signal B is 24 microseconds, it is determined to be a switching conflict and the conflict point is recorded as 24 microseconds. All conflict points are recorded in the switching conflict point data list. Each conflict point in the switching conflict point data list is processed based on the signal priority sorting rule. The priority sorting rule is set according to parameters such as the importance of the signal and the switching frequency. The importance is defined by the priority weight matrix set by the system. The switching frequency is calculated by analyzing the number of switching times of the signal in the basic duty sequence. Assuming that the priority weight of signal A is 0.8 and that of signal B is 0.6. Then at the conflict point of 24 microseconds, signal A gets the scheduling resource first. The priority allocation module generates the timing scheduling parameters according to the calculation results, marks signal A as the priority signal, and records the delay time of signal B. The generated timing scheduling parameters include the delay adjustment values ​​of the priority signal and other signals. The basic duty cycle sequence is simulated and scheduled using the timing scheduling module. According to the priority signal and the delay adjustment value in the timing scheduling parameters, the signal switching timing at the conflict point is rearranged. The specific operation is to keep the original switching timing of the priority signal and delay the switching of other signals according to the delay adjustment value. Assuming that signal A is the priority signal, its switching timing remains unchanged, and the rising edge time of signal B is postponed from 24 microseconds to 26 microseconds. The scheduling simulation module loads the adjusted duty cycle timing into the simulation environment for operation, observes the overlap of the signals, and records the simulation results to generate a simulated scheduling behavior log, which includes the switching timing and conflict detection results of the adjusted signals. The dead zone demand analysis module is used to analyze the simulated scheduling behavior log. Compare with the minimum dead time requirement. If it is found that the adjusted signal time interval is still lower than the minimum dead time threshold, further adjustment is required. To this end, by analyzing the location and frequency of the conflict point, an adjustment interval set containing all time intervals that need to be adjusted is generated. Assuming that the minimum dead time threshold is 5 microseconds, if the falling edge time of signal A is 30 microseconds and the rising edge time of signal B is 33 microseconds, the interval is less than the dead time. The adjustment interval set records the interval interval with an adjustment range of 30 to 33 microseconds. The inter-channel balance compensation module is used to adjust the basic duty sequence according to the interval interval in the adjustment interval set. By adjusting the signal trigger time and duty cycle width, it is ensured that the time interval of all signals meets the minimum dead time requirement. Assuming that in the range of 30 to 33 microseconds, the falling edge time of signal A is adjusted to 29 microseconds and the rising edge time of signal B is adjusted to 34 microseconds to expand the time interval between the signals, the inter-channel balance compensation module processes all interval intervals that need to be adjusted in turn, and the adjusted duty sequence is the anti-overlap sequence. .

[0041] In this embodiment, the phase allocation module is used to: perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence, which is specifically used to: Performing phase quantization processing on the anti-overlapping sequence to obtain a quantized phase sequence; Perform phase frequency statistics on the phase quantization sequence to generate the phase quantity; Perform power grouping analysis on the anti-overlap sequence based on the number of phases to obtain a phase configuration reference system; The anti-overlapping sequence is interference optimized and rearranged according to the phase configuration reference system to generate a peak-shifting control sequence.

[0042] In this embodiment, the switching timing of each signal in the anti-overlapping sequence is sampled by a phase quantization module, and the phase quantization interval is set to 10 microseconds. The signal switching time is aligned with the phase quantization interval in turn. The specific operation is to align the switching timing of the signal to the nearest integer multiple according to the phase quantization interval. For example, if the switching timing of a certain signal is 35 microseconds, it is adjusted to 40 microseconds according to the quantization interval, and the adjusted time point is recorded at the same time. The recorded content includes the channel number of the signal and the adjusted switching time point. The quantized data of all signals are recorded as a quantized phase sequence, and the quantized phase sequence is stored in a table form, including The signal channel number, adjusted rising edge time, adjusted falling edge time, etc., group the signals according to the integer value of the quantized time point, and then count the number of signals corresponding to each time point. Assuming that the switching time points of some signals in the quantized phase sequence are 40 microseconds, 50 microseconds, and 60 microseconds, these three time points are counted as three phase groups respectively, and the number of signals in each phase group is used as the frequency value of the phase. At the same time, the total number of all phase groups is calculated. The frequency statistics module will generate a phase quantity data table, which includes the time point of each phase group and its corresponding signal number. The signals are grouped according to the time points in the phase quantity data table. , the power value of each group of signals is accumulated and calculated. Assuming that a phase group includes 3 signals, and the power value of each signal is 1 watt, 2 watts and 3 watts respectively, the total power of the phase group is 6 watts. The power grouping analysis module calculates all phase groups in turn, and records the total power of each phase group as the phase power value. At the same time, the phase power values ​​are sorted, and the phase group with the highest total power is marked as the priority phase to generate a phase configuration reference system. The phase configuration reference system includes the time point, total power value and priority label of each phase group. The interference optimization module is used to rearrange the anti-overlapping sequence. According to the priority label in the phase configuration reference system, The signal switching timing is adjusted to arrange the signal in the priority phase in the time period with the least switching interference. The specific operation is to reallocate the switching time points of all signals in the phase group. For example, if the time point of a priority phase group is 40 microseconds and the total power value is 6 watts, the signal switching time point in the phase group is adjusted to an adjacent interference-free time period, such as 45 microseconds. At the same time, ensure that the switching interval between signals meets the minimum switching interval requirement. After the adjustment is completed, the interference optimization module generates a peak-shifting control sequence, which contains the reallocated signal switching timing and phase group information, and is ultimately used to drive the LED lamp to achieve stable output.

[0043] In this embodiment, the specific steps of performing interference optimization rearrangement on the anti-overlap sequence according to the phase configuration reference system to generate the peak shifting control sequence are: Perform channel distribution identification on the anti-overlap sequence in the phase configuration reference system to obtain a channel distribution sequence; The phase clustering region is divided according to the channel distribution sequence to obtain the phase clustering coordinates; Channels are grouped based on phase clustering coordinates to generate a channel grouping set; Performing inter-group interference evaluation on the channel grouping set to generate an inter-channel interference value; Based on the inter-channel interference value, the channel grouping set is screened for rearrangement targets to generate rearrangement target channels; The sequence of the rearrangement target channel is rearranged according to the phase configuration reference system to generate a peak shifting control sequence.

[0044] In this embodiment, the channel information in the phase configuration reference system is parsed by the channel identification module, the signal switching time point and power value corresponding to each channel are extracted, and the switching time and power distribution of all channels are recorded in a table. For example, assuming that the switching time points of channel A are 20 microseconds, 30 microseconds, and 40 microseconds, and the powers are 2 watts, 3 watts, and 5 watts respectively, and the switching time points of channel B are 25 microseconds, 35 microseconds, and 45 microseconds, and the powers are 3 watts, 4 watts, and 6 watts respectively, the information of each channel is summarized and sorted to form a channel distribution sequence, and the time point data in the channel distribution sequence is clustered and analyzed by the regional segmentation module, and segmented using a fixed time interval, for example, the time interval is set to 10 microseconds, and the number of channels in each time period is counted The amount and total power value are recorded, and the time period where the channel density exceeds the threshold is marked as a phase aggregation area. Assuming that the time period 30 to 40 microseconds contains channel A and channel B, the total power value is 8 watts, and the channel density of this time period exceeds the preset threshold 5, then this time period is marked as a phase aggregation area, and the start and end time are recorded at the same time to form a phase aggregation coordinate table. The phase aggregation coordinate contains the start time, end time and channel density value of all aggregation areas. The channels in the phase aggregation area are classified through the grouping module, and the channels in the same area are divided into a channel group, and the time range and total power value of each channel group are recorded. For example, for the time period 30 to 40 microseconds in the phase aggregation coordinate, channel A and channel B belong to the time period. signal, then divide channel A and channel B into the same channel group, and record the time range of the channel group as 30 to 40 microseconds, and the total power value as 8 watts. After processing all the channel groups in the clustered areas in turn, a channel grouping set is generated. The channel grouping set is recorded in the form of a list. Each channel group contains the time range, channel number and total power value. The interference evaluation module is used to perform interference calculation on each pair of channel groups in the channel grouping set. The specific steps are to extract the time range of the two groups of signals, calculate the overlap ratio of the time range, and estimate the interference intensity in combination with the total power value of the channel group. Assume that the time range of channel group 1 is 30 to 40 microseconds, the total power value is 8 watts, the time range of channel group 2 is 35 to 45 microseconds, and the total power value is 10 watts. The overlap ratio is 50%, then the interference intensity is calculated to be 4 watts based on the overlap ratio and power value. The interference assessment module records the interference values ​​between all channel groups as an interference matrix. The interference matrix is ​​stored in the form of a two-dimensional array. Each element in the matrix represents the interference intensity between two channel groups. The interference values ​​in the interference matrix are sorted by the screening module, and the channel group pairs with the largest interference values ​​are extracted and marked as the target channel groups for rearrangement. Assuming that the interference intensity between channel group 1 and channel group 2 in the interference matrix is ​​4 watts, which is the maximum value among all channel group pairs, channel group 1 and channel group 2 are used as the target channel groups for rearrangement. At the same time, the signal switching time points of the target channel groups for rearrangement are further analyzed, and the signals with the largest interference impact are extracted as the target channels for rearrangement.For example, if the switching time point of signal A in channel group 1 is 35 microseconds, and the switching time point of signal B in channel group 2 is 36 microseconds, then signal A and signal B are marked as the target channels for rearrangement, and the time point interval and priority information in the phase configuration reference system are extracted. The signal switching time point of the target channel for rearrangement is moved to the time period with the lowest priority. For example, according to the priority information of the phase configuration reference system, the priority time period is 10 to 20 microseconds, and the suboptimal time period is 20 to 30 microseconds. Then the switching time of signal A is adjusted to 15 microseconds, and the switching time of signal B is adjusted to 25 microseconds. At the same time, the adjusted time points are verified to ensure that the interval between the adjusted signal switching time point and the switching time points of other signals meets the minimum interval requirement. After the adjustment is completed, the rearrangement module generates a peak shifting control sequence, which records the adjusted signal switching time points and channel numbers in a table. ,

[0045] In this embodiment, the channel coordination module is used to: perform multi-channel temperature compensation on the peak shifting control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence, specifically for: Based on the preset basic parameters of the lamp group, the thermal field gradient mapping is performed on the peak-shifting control sequence to obtain the temperature influence parameters; According to the temperature influence parameter, the channel temperature coefficient compensation is performed on the peak shifting control sequence to obtain the temperature compensation sequence of each channel; Perform compensation integration processing on the temperature compensation sequence of each channel to generate a temperature compensation sequence; Perform channel coordination processing according to the temperature compensation sequence to obtain the channel correction coefficient; perform luminous flux reconstruction based on the channel correction coefficient to generate a light efficiency compensation sequence; The light effect compensation sequence is processed in time domain synchronization to generate a multi-channel control sequence.

[0046] In this embodiment, the basic parameters of the preset lamp group are analyzed through the thermal field analysis module, and the basic parameters such as the current value, power value, and heating coefficient of the LED lamp are used as input. In combination with the time node and channel number of the peak-shifting control sequence, the thermal field gradient calculation method is used to generate the thermal field distribution map of the lamp group. The heat generation is associated with time and position and then a thermal field gradient mapping model is established. Assuming that the basic parameters of the lamp group include a current of 700 mA, a power of 5 W, and a heating coefficient of 0.02°C / W, the temperature influence parameter is calculated according to the thermal field distribution for the control signal of each channel. For example, if the temperature of the heating area of ​​a certain channel is 45°C, the corresponding temperature influence parameter is 2°C / second, and all channels are The temperature influence parameters are stored in matrix form. Each column of the matrix represents the temperature change rate of a channel. The temperature influence parameters are processed channel by channel through the temperature compensation module. The temperature coefficient compensation value of each channel is calculated using the temperature change rate of each channel. The compensation value is superimposed on the corresponding peak shifting control sequence to generate a temperature compensation sequence. Assuming that the time of a signal switching point in the peak shifting control sequence is 10 microseconds, the original power value is 2 watts, and the corresponding temperature change rate is 2°C / second, the temperature compensation value is calculated to be 0.4 watts. After adding the compensation value to the original power value, it is adjusted to 2.4 watts. After the compensation is completed, the temperature compensation sequence of each channel generated is stored in a two-dimensional table with time points as rows and channel numbers as columns. Each item represents The power value of a certain channel at a certain time point is analyzed comprehensively through the compensation integration module for the temperature compensation sequence of all channels, and the temperature compensation value and the sequence after the channel power adjustment are merged to ensure that the overall power distribution uniformity is met within the time domain. The specific implementation method is to first sum up the power values ​​of all channels at each time point to check whether they exceed the total power limit of the lamp group. For example, if the total power value of all channels at a certain moment is 15 watts and the power limit of the lamp group is 20 watts, no adjustment is required. If the total power exceeds the limit value, the power value of each channel is reduced by linear weight distribution to generate the final temperature compensation sequence. The temperature compensation sequence is indexed by the channel number and time point. The channel coordination module analyzes the temperature compensation sequence, extracts the power difference between each channel at each time point, and quantifies the power difference using the correction coefficient calculation model. Assuming that the power of channel A is 3 watts and the power of channel B is 2.8 watts at a certain time point, the power difference is 0.2 watts. The difference is converted into a correction coefficient of 0.05 according to the correction coefficient calculation formula. The correction coefficients of all time points are stored in matrix form. Each item of the matrix represents the correction coefficient of a channel at the corresponding time point. The power value of each channel is multiplied by the correction coefficient to obtain a new power value. Assuming that the power of channel A is 3 watts at a certain time point in the temperature compensation sequence and the correction coefficient is 0.05, the corrected power value is 3.15 watts, the corrected power values ​​of all channels are combined to form a light effect compensation sequence. The light effect compensation sequence uses the time point as the index, and stores the final power value of each channel in matrix form. The signal switching time at each time point is accurate to the microsecond level, and the switching time of all channels is aligned. Assuming that the switching time point of a signal in the light effect compensation sequence is 10.2 microseconds, and the switching time points of other channels are 10 microseconds and 10.3 microseconds respectively, all switching time points are uniformly adjusted to 10.2 microseconds through the interpolation algorithm. After the adjustment is completed, a multi-channel control sequence is generated. The multi-channel control sequence records the final switching time point and the corresponding power value in a table. .

[0047] In this embodiment, the feedback optimization module is used to: perform multi-channel adjustment control according to the multi-channel control sequence, and collect the feedback strong light value at the same time; perform feedback optimization on the simulated LED light group based on the feedback strong light value to perform multi-channel adjustment of the LED light, specifically for: Perform hardware timing conversion on multi-channel control sequences to obtain drive control signals; Perform multi-channel adjustment control according to the driving control signal and collect feedback intensity value at the same time; According to the feedback strong light value, the light intensity deviation of the simulated LED light group is compared to obtain the light intensity deviation comparison value; Feedback optimization is performed on the simulated LED lamp group based on the light intensity deviation comparison value to perform multi-channel adjustment of the LED lamp.

[0048] In this embodiment, each channel signal of the multi-channel control sequence is sampled and processed by the hardware timing conversion module, the time node and power value of the signal are re-encoded according to the requirements of the hardware drive, and a standardized driving signal is generated. The specific operation includes reading the signal timing of each channel in the multi-channel control sequence, dividing the time point in microseconds, and converting the power value range of different channels according to the hardware voltage level. For example, a signal with a power value of 10 watts is converted into a 3.3 volt voltage signal according to the ratio of the hardware drive voltage, and the time node is accurate to 10 microseconds. The generated driving control signal is output in the form of a timing file and transmitted to the driving hardware through an electrical interface. The generated driving control signal is loaded by the driving module to drive each channel of the simulated LED lamp group to emit light. During the lighting process, the output light intensity of each channel is collected in real time by an optical sensor. The sampling frequency of the optical sensor is set to 1 kHz. The collected data includes the light intensity value and the corresponding time node. For example, at a certain time point, the light intensity value of channel A is 500 lumens, and the light intensity value of channel B is 450 lumens. These feedback data are stored in a matrix file. Each of the matrix Each row corresponds to a time point, and each column corresponds to the light intensity value of a channel. The feedback intensity value matrix and the target intensity value matrix are read and compared row by row according to the time point. The feedback value at each time point is subtracted from the target value to obtain the deviation value. For example, if the target light intensity value at a certain time point is 520 lumens and the feedback intensity value is 500 lumens, then the light intensity deviation is -20 lumens. The light intensity deviation values ​​of all time points and channels are stored in matrix form. Each item of the matrix represents the light intensity deviation value of a channel at the corresponding time point. The light intensity deviation comparison value matrix is ​​processed by the feedback optimization module and adjusted using the optimization algorithm. The driving signal of each channel is adjusted to reduce the light intensity deviation. The specific method includes performing a weighted analysis on the light intensity deviation value and calculating the driving signal correction coefficient of each channel. For example, the light intensity deviation of a certain channel is -20 lumens, and the corresponding correction coefficient is 0.05. The correction coefficient is superimposed on the driving signal to generate a new driving control signal. The new driving control signal is transmitted to the simulated LED lamp group, and the adjusted output light intensity is re-collected and compared. This is repeated until the light intensity deviation value is reduced to less than 2 lumens, and finally the feedback optimization and multi-channel adjustment control of the simulated LED lamp group are completed.

[0049] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A multi-channel adjustment system for LED lamps, characterized in that: The multi-channel adjustment system of the LED lamp includes a parameter acquisition module, a change simulation module, a dead zone compensation module, a phase distribution module, a channel coordination module and a feedback optimization module: The parameter acquisition module is used to obtain multiple channels of real-time light intensity parameters; The noise covariance of multiple real-time light intensity parameters is estimated to obtain the noise matrix; Perform channel interference analysis on multiple real-time optical intensity parameters to generate a channel crosstalk matrix; The change simulation module is used to: predict the light intensity change of the real-time light intensity parameter based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; Performing a change simulation process according to the predicted light intensity value to obtain a simulated LED light group; The dead zone compensation module is used to: perform multi-channel duty cycle calculation on the simulated LED light group to obtain a basic duty sequence; Performing dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence; The phase allocation module is used to: perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence; The channel coordination module is used to: perform multi-channel temperature compensation on the staggered peak control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence; The feedback optimization module is used to: perform multi-channel adjustment control according to the multi-channel control sequence and collect feedback intensity values ​​at the same time; Feedback optimization is performed on the simulated LED light group based on the feedback intensity value to perform multi-channel adjustment of the LED light.

2. The multi-channel regulation system for LED lamps according to claim 1, characterized in that: The parameter acquisition module is used to: obtain multi-channel real-time light intensity parameters; perform noise covariance estimation on the multi-channel real-time light intensity parameters to obtain a noise matrix; perform channel interference analysis on the multi-channel real-time light intensity parameters to generate a channel crosstalk matrix, specifically for: Get multi-channel real-time light parameters; Perform smoothing and filtering on the data of multiple channels of real-time light parameters to obtain filtered light intensity parameters; Perform state space mapping on the filtered light intensity parameters to obtain the light intensity state matrix; Estimating the noise covariance of the light intensity state matrix to obtain the noise matrix; Perform multi-channel topology construction on multiple real-time optical intensity parameters to obtain the network interference topology structure; Based on the network interference topology, channel crosstalk analysis is performed on multi-channel real-time optical intensity parameters to generate a channel crosstalk matrix.

3. The multi-channel regulation system for LED lamps according to claim 1, characterized in that: The change simulation module is used to: predict the light intensity change of the real-time light intensity parameters based on the noise matrix and the channel crosstalk matrix to generate a predicted light intensity value; perform change simulation processing based on the predicted light intensity value to obtain a simulated LED light group, specifically for: Based on the channel crosstalk matrix, the noise matrix is ​​subjected to channel correlation mapping to obtain the crosstalk correlation matrix; Perform spectrum conversion on real-time light intensity parameters based on the crosstalk correlation matrix to generate a light intensity spectrum matrix; Light intensity change prediction is performed based on the light intensity spectrum matrix to obtain the predicted light intensity value; According to the preset basic parameters of the lamp group, the predicted light intensity value is mapped to the light intensity parameter to generate the corresponding light intensity parameter; The light group topology is reconstructed based on the parameters corresponding to the light intensity to obtain a simulated LED light group.

4. The multi-channel regulation system for LED lamps according to claim 1, characterized in that: The dead zone compensation module is used to: calculate the multi-channel duty cycle of the simulated LED light group to obtain a basic duty sequence; perform dead zone compensation processing on the basic duty sequence to generate an anti-overlap sequence, specifically for: Perform channel brightness balancing on the simulated LED light group to obtain balanced light intensity of the light group; Perform multi-channel gamma mapping on the balanced light intensity of the lamp group to generate a balanced linear matrix; Based on the balanced linear matrix, channel interaction compensation is deduced for the simulated LED light group to obtain the interaction compensation matrix; Perform multi-dimensional duty cycle calculation on the interaction compensation matrix to generate a basic duty sequence; Perform edge conflict detection on the basic duty sequence to obtain timing scheduling parameters; Dead zone balance reconstruction is performed based on timing scheduling parameters to generate an anti-overlap sequence.

5. The multi-channel regulation system for LED lamps according to claim 4, characterized in that: The specific steps of performing edge conflict detection on the basic duty sequence to obtain timing scheduling parameters; and performing dead zone balance reconstruction based on the timing scheduling parameters to generate an anti-overlap sequence are as follows: Perform multi-path edge recognition on the basic duty sequence to obtain multi-path switching boundaries; perform switching conflict detection based on the multi-path switching boundaries to generate switching conflict points; Assign priorities to the switching conflict points and obtain timing scheduling parameters; Perform scheduling simulation on the basic duty sequence based on the timing scheduling parameters to obtain simulated scheduling behavior; Perform dead zone demand analysis on simulated scheduling behavior and generate a set of adjustment intervals; Inter-channel balance compensation is performed on the basic duty sequence according to the adjustment interval set to obtain an anti-overlap sequence.

6. The multi-channel regulation system for LED lamps according to claim 1, characterized in that: The phase allocation module is used to: perform phase quantization on the anti-overlapping sequence to obtain the phase quantity; perform phase peak shifting allocation on the anti-overlapping sequence based on the phase quantity to obtain the peak shifting control sequence, specifically for: Performing phase quantization processing on the anti-overlapping sequence to obtain a quantized phase sequence; Perform phase frequency statistics on the phase quantization sequence to generate the phase quantity; Perform power grouping analysis on the anti-overlap sequence based on the number of phases to obtain a phase configuration reference system; The anti-overlap sequence is interference optimized and rearranged according to the phase configuration reference system to generate a peak-shifting control sequence.

7. The multi-channel regulation system for LED lamps according to claim 6, characterized in that: The specific steps of performing interference optimization rearrangement of the anti-overlap sequence according to the phase configuration reference system to generate the peak shifting control sequence are as follows: Perform channel distribution identification on the anti-overlap sequence in the phase configuration reference system to obtain a channel distribution sequence; The phase clustering region is divided according to the channel distribution sequence to obtain the phase clustering coordinates; Channels are grouped based on phase clustering coordinates to generate a channel grouping set; Performing inter-group interference evaluation on the channel grouping set to generate an inter-channel interference value; Performing rearrangement target screening on the channel grouping set based on the inter-channel interference value to generate rearrangement target channels; The target channel is sequence rearranged according to the phase configuration reference system to generate a peak shifting control sequence.

8. The multi-channel regulation system for LED lamps according to claim 1, characterized in that: The channel coordination module is used to: perform multi-channel temperature compensation on the staggered peak control sequence to obtain a temperature compensation sequence; perform multi-channel coordination processing according to the temperature compensation sequence to generate a multi-channel control sequence, specifically for: Based on the preset basic parameters of the lamp group, the thermal field gradient mapping of the peak control sequence is performed to obtain the temperature influence parameters; Perform channel temperature coefficient compensation on the staggered peak control sequence according to the temperature influence parameter to obtain the temperature compensation sequence of each channel; Perform compensation integration processing on the temperature compensation sequence of each channel to generate a temperature compensation sequence; Perform channel coordination processing according to the temperature compensation sequence to obtain the channel correction coefficient; reconstruct the luminous flux based on the channel correction coefficient to generate the light efficiency compensation sequence; The light effect compensation sequence is subjected to time domain synchronization processing to generate a multi-channel control sequence.

9. The multi-channel regulation system for LED lamps according to claim 1, characterized in that: The feedback optimization module is used to: perform multi-channel adjustment control according to the multi-channel control sequence, and collect feedback intensity values ​​at the same time; perform feedback optimization on the simulated LED light group based on the feedback intensity values ​​to perform multi-channel adjustment of the LED lights, specifically for: Perform hardware timing conversion on multiple control sequences to obtain drive control signals; Perform multi-channel adjustment control according to the driving control signal and collect feedback intensity value at the same time; Perform light intensity deviation comparison on the simulated LED light group according to the feedback strong light value to obtain the light intensity deviation comparison value; Feedback optimization is performed on the simulated LED lamp group based on the light intensity deviation comparison value to perform multi-channel adjustment of the LED lamp.

Images