Method and device for adaptively adjusting parameters of analog light source controller

Abstract

The present application relates to the technical field of light source controllers, and discloses a method and device for adaptively adjusting parameters of an analog light source controller. The method comprises: collecting an input voltage signal, an output current signal, an ambient temperature signal, and a channel power signal of the analog light source controller, and inputting them into a neural network model for parameter mapping to obtain a dynamic response matrix of light source parameters; performing feature dimension reduction and constructing a target optimization model; gridding the parameter space, and generating a steepest descent direction vector and a vertical search direction vector at the grid points after the division to obtain a parameter optimization direction set; iteratively updating each control parameter to obtain a parameter iteration sequence set; performing a voltage stabilization output test to obtain a plurality of candidate control parameter combinations; performing a parallel verification test to determine the optimal control parameter combination, thereby effectively suppressing the crosstalk between channels of the analog light source controller and solving the mutual interference problem when multiple channels work in parallel.

Description

Technical Field

[0001] The present application relates to the technical field of light source controllers, and in particular to a method and device for adaptively adjusting parameters of an analog light source controller. Background Art

[0002] As a key component in the optical test system, the performance of the analog light source controller directly affects the test accuracy and reliability of the entire system. Traditional analog light source controllers mainly use fixed parameter control methods, which are difficult to maintain stable output characteristics under different working environments and load conditions. Especially when multiple channels work in parallel, the mutual interference between channels will cause light intensity fluctuations and spectral distortion.

[0003] As optical test systems develop towards high precision and multi-functions, the performance requirements for analog light source controllers continue to increase. Existing control methods often only consider a single control target, such as light intensity stability or color temperature control, and lack comprehensive consideration of multi-objective optimization, making it difficult to simultaneously meet the requirements of multiple performance indicators such as light intensity stability, spectral fidelity, energy efficiency, and temperature balance. In addition, in practical applications, analog light source controllers face a complex and changeable working environment, including the influence of factors such as input voltage fluctuations, load changes, and temperature drift, and there are complex coupling relationships between these factors. Existing control methods lack an effective parameter adaptive adjustment mechanism and cannot dynamically adjust control parameters according to environmental changes, resulting in problems such as delayed system response and poor anti-interference ability. Summary of the invention

[0004] The present application provides a method and device for adaptively adjusting parameters of an analog light source controller, thereby effectively suppressing crosstalk between channels of the analog light source controller and solving the mutual interference problem when multiple channels work in parallel.

[0005] In a first aspect, the present application provides a method for adaptively adjusting parameters of a simulated light source controller, the method comprising:

[0006] Collect the input voltage signal, output current signal, ambient temperature signal and channel power signal of the analog light source controller, and input them into the neural network model for parameter mapping to obtain the dynamic response matrix of the light source parameters;

[0007] Using principal component analysis algorithm to perform feature dimension reduction on the dynamic response matrix of the light source parameters, and constructing a target optimization model;

[0008] Gridding the parameter space based on the target optimization model, generating the steepest descent direction vector and the vertical search direction vector at the grid points after the division, and obtaining a parameter optimization direction set;

[0009] According to the parameter optimization direction set, each control parameter is iteratively updated using a dynamic step size to obtain a parameter iteration sequence set;

[0010] For each group of control parameters in the parameter iteration sequence set, a voltage stabilization output test is performed through a constant voltage controller, and a comprehensive score of a voltage stability index, an overcurrent protection index, a trigger response index, and a temperature balance index is calculated using a multi-objective evaluation function to obtain a plurality of candidate control parameter combinations;

[0011] The multiple candidate control parameter combinations are respectively input into the multi-channel collaborative controller for parallel verification testing, and the optimal control parameter combination is determined according to the degree of interference suppression between the channels.

[0012] A second aspect of the present application provides a device for adaptively adjusting parameters of a simulated light source controller, the device comprising:

[0013] The acquisition module is used to collect the input voltage signal, output current signal, ambient temperature signal and channel power signal of the analog light source controller, and input them into the neural network model for parameter mapping to obtain the dynamic response matrix of the light source parameters;

[0014] A dimension reduction module, used to perform feature dimension reduction on the dynamic response matrix of the light source parameters by using a principal component analysis algorithm, and to construct a target optimization model;

[0015] A partitioning module, used to perform grid partitioning on the parameter space based on the target optimization model, and to generate a steepest descent direction vector and a vertical search direction vector on the divided grid points to obtain a parameter optimization direction set;

[0016] An updating module, configured to iteratively update each control parameter using a dynamic step size according to the parameter optimization direction set to obtain a parameter iteration sequence set;

[0017] A calculation module, for performing a voltage stabilization output test on each group of control parameters in the parameter iteration sequence set through a constant voltage controller, and calculating a comprehensive score of a voltage stability index, an overcurrent protection index, a trigger response index, and a temperature balance index using a multi-objective evaluation function to obtain a plurality of candidate control parameter combinations;

[0018] The test module is used to input the multiple candidate control parameter combinations into the multi-channel collaborative controller for parallel verification testing, and determine the optimal control parameter combination according to the interference suppression degree between the channels.

[0019] The third aspect of the present application provides an electronic device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the electronic device executes the above-mentioned method for adaptively adjusting parameters of a simulation light source controller.

[0020] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned method for adaptively adjusting parameters of a simulation light source controller.

[0021] Compared with the prior art, the present application has the following beneficial effects: by performing parameter mapping and dimensionality reduction through principal component analysis through a neural network model, accurate modeling of the complex dynamic characteristics of the analog light source controller is achieved, providing a reliable mathematical basis for parameter optimization; by adopting multi-objective optimization and grid search strategies, multiple control objectives such as light intensity stability, spectral fidelity, energy efficiency and temperature balance are taken into account at the same time, significantly improving the overall performance of the system; by introducing a dynamic step size adaptive adjustment mechanism, the control parameters can be automatically adjusted according to changes in the working environment, thereby enhancing the environmental adaptability of the system; by using a multi-channel collaborative control strategy, crosstalk between channels is effectively suppressed, and the mutual interference problem when multiple channels work in parallel is solved; based on a hierarchical optimization structure, the parameter optimization process is divided into multiple links such as feature extraction, target construction, parameter search and performance evaluation, thereby improving the convergence efficiency of the optimization algorithm; a complete performance evaluation system is designed, and the control parameters are comprehensively evaluated through a multi-objective evaluation function, thereby ensuring the reliability and practicality of the optimization results. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0023] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0024] Figure 1 It is a schematic flow chart of a method for adaptively adjusting parameters of a simulated light source controller provided by an embodiment of the present invention;

[0025] Figure 2 is a schematic block diagram of the structure of a device for adaptively adjusting parameters of a simulated light source controller provided by an embodiment of the present invention;

[0026] Figure 3 It is a schematic block diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0027] 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.

[0028] 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 according to actual conditions.

[0029] 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.

[0030] It should be further understood that the term "and / or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. Figure 1 , an embodiment of the method for adaptively adjusting parameters of a simulated light source controller in an embodiment of the present application includes:

[0031] Step 100: collecting input voltage signals, output current signals, ambient temperature signals and channel power signals of the analog light source controller, and inputting them into a neural network model for parameter mapping to obtain a dynamic response matrix of light source parameters;

[0032] It is understandable that the execution subject of the present application may be a simulated light source controller parameter adaptive adjustment device, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0033] Specifically, the input voltage signal collected by the SMP-03V-BC port is converted into digital quantity and phase compensated by a 16-bit high-precision analog-to-digital converter to obtain a voltage input sequence. Digital quantity conversion is to convert analog signals into digital signals that can be processed by computers, and phase compensation is used to correct the phase offset caused by sampling delay and system response to ensure the accuracy of voltage signal acquisition. When processing the output current signal, the output current signal is collected by a constant voltage drive circuit, and a 2.5A range Hall current sensor is used to perform current detection and baseline calibration to obtain a current input sequence. The use of the Hall current sensor can accurately detect the current signal and provide a higher response speed and lower energy loss. The baseline calibration process ensures that the current signal will not be affected by baseline drift during the detection process, thereby improving the accuracy of current detection. At the same time, the ambient temperature signal is collected by a forced cooling controller, and the temperature sampling and temperature compensation are performed using an NTC thermistor to obtain a temperature input sequence. The NTC thermistor is a temperature sensor whose resistance changes with temperature and can sensitively sample the ambient temperature. The temperature compensation process is to ensure that the temperature signal is not interfered by the external environment during the measurement process and obtain accurate ambient temperature data. For the channel power signal, a four-channel power acquisition method is adopted. The power signal of each channel is detected by the channel isolation amplifier, and the load of the channel is calculated to obtain the power input sequence. The use of channel isolation amplifiers can avoid signal interference between channels and improve the accuracy of signal acquisition and the stability of the system. The voltage input sequence, current input sequence, temperature input sequence and power input sequence are sampled and aligned and feature matrices are constructed to obtain the input feature sequence matrix. The sampling timing alignment ensures the time synchronization of each signal so that the neural network can receive input data with the same timestamp at the same time, and the feature matrix construction organizes various input signals in an orderly manner for subsequent neural network processing. The input feature sequence matrix is ​​input into the first hidden layer of the neural network model. The first hidden layer contains 256 neuron nodes with a dropout mechanism. The nonlinear feature transformation is performed through the ReLU activation function. The introduction of the dropout mechanism can effectively prevent the model from overfitting and improve the generalization ability of the model. The ReLU activation function performs nonlinear mapping on the input features, so that the network has the ability to process complex features. The batch normalization method is used to dynamically adjust the feature distribution. Batch normalization can accelerate model training and improve the stability of the model. By standardizing the data of each batch, the feature distribution is ensured to remain stable during the training process. The first layer of feature maps is input into the second hidden layer of the neural network model, which contains 128 neuron nodes with residual connections. Through the residual connection method, the gradient vanishing problem in the deep network is effectively alleviated, and the training efficiency of the network is improved.In the second hidden layer, the Sigmoid activation function is used to compress the features. The Sigmoid activation function can compress the feature values ​​to the range of 0 to 1, enhancing the model's ability to distinguish features. At the same time, the attention mechanism is introduced in this layer to enhance the key features. The introduction of the attention mechanism enables the model to pay more attention to the features that have an important impact on the final output, improving the model's ability to understand the input data and the accuracy of the output. The second layer feature map is input into the output layer of the neural network model. The output layer contains 64 fully connected neuron nodes. The light source parameters are mapped through the linear activation function to obtain the dynamic response matrix of the light source parameters. The role of the linear activation function is to maintain the linear relationship between the input and output so that the features of the previous layer can be mapped to the final light source parameter output.

[0034] Step 200: Use the principal component analysis algorithm to perform feature dimension reduction on the dynamic response matrix of the light source parameters, and construct a target optimization model;

[0035] Specifically, the dynamic response matrix of the light source parameters is subjected to mean centering and standard deviation normalization operations to obtain a standardized parameter matrix. Mean centering can eliminate the bias in the data, make the mean of the data of each parameter zero, and improve the stability of subsequent operations, while standard deviation normalization keeps the dimension of each parameter consistent and eliminates the scale differences between various input features. The standardized parameter matrix is ​​subjected to covariance calculation and eigenvector decomposition to obtain a set of characteristic components that characterize the characteristics of the light source. These characteristic components include light intensity response function, color temperature response function, temperature response function, and power consumption response function, etc. Each characteristic component reflects the change characteristics of a certain aspect of the light source parameters. The purpose of covariance calculation is to obtain the correlation between each feature, so as to retain important feature information in the subsequent dimensionality reduction, while eigenvector decomposition is to map these features to a new space, so that their changes are easier to explain and process in the new space. Based on the set of characteristic components, the eigenvalue of each characteristic component is calculated, and the principal component features are screened according to the threshold of 95% contribution rate to obtain the dimensionality reduction mapping matrix. The eigenvalue reflects the explanatory power of the corresponding characteristic component for the original data set, and the contribution rate threshold of 95% is to retain as much important information as possible while reducing the data dimension to achieve the purpose of dimensionality reduction. The standardized parameter matrix and the dimension reduction mapping matrix are orthogonally transformed to obtain a low-dimensional feature vector containing light intensity stability parameters, spectral fidelity parameters, energy efficiency parameters and temperature balance parameters. The orthogonal transformation operation maps the features in the high-dimensional space to a lower-dimensional space through the dimension reduction mapping matrix. In the new low-dimensional space, different features are independent of each other, which is conducive to the subsequent parameter optimization process. When constructing the target optimization model, the light intensity optimization target item is constructed based on the deviation between the light intensity stability parameter calculation and the set threshold in the low-dimensional feature vector. The goal of the light intensity stability parameter is to ensure that the light intensity output by the light source can remain stable and will not fluctuate significantly due to changes in external conditions. By setting a threshold, the deviation is controlled within an acceptable range to ensure the stability of the light source. At the same time, the spectral fidelity parameter calculation in the low-dimensional feature vector matches the standard spectrum to construct the spectrum optimization target item. The spectral fidelity parameter reflects the difference between the output spectrum of the light source and the ideal standard spectrum. The purpose of constructing the spectrum optimization target is to ensure that the output spectrum of the light source can meet the expected light quality requirements to ensure the effect of the light source in practical applications. The energy efficiency parameter in the low-dimensional feature vector is used to calculate the ratio to the input power to construct the energy optimization target item. The energy efficiency parameter reflects the efficiency of the light source in the energy conversion process. The purpose of constructing the energy optimization target is to maximize the energy utilization efficiency of the light source, reduce unnecessary energy consumption, and improve the economy and environmental performance of the system. For the temperature balance parameter in the low-dimensional feature vector, the temperature balance target item is constructed based on the difference between it and the maximum allowable temperature.The construction of the temperature balance target item is to ensure that the temperature of the light source does not exceed the preset safety threshold during operation, thereby avoiding equipment damage or safety hazards caused by excessive temperature. The light intensity optimization target item, spectrum optimization target item, energy consumption optimization target item and temperature balance target item are combined together to form a comprehensive target optimization model. Through this target optimization model, a trade-off is made between different optimization targets to obtain a set of control parameters that perform well in terms of light intensity stability, spectrum fidelity, energy consumption efficiency and temperature balance.

[0036] Step 300: gridding the parameter space based on the target optimization model, generating the steepest descent direction vector and the vertical search direction vector at the grid points after the division, and obtaining a parameter optimization direction set;

[0037] It should be noted that the target optimization model is input into the multi-parameter grid generator for parameter interval division, and a search boundary matrix containing the light intensity parameter interval, the color temperature parameter interval, the power parameter interval and the temperature parameter interval is obtained, so as to optimize the parameters in a constrained manner to prevent exceeding the physically allowed range. The search boundary matrix is ​​divided into equal intervals, and the minimum sampling interval is determined according to the physical constraints of each parameter to obtain the initial grid sampling point matrix. A uniform sampling point grid is constructed in the parameter space so that the search process can cover the entire parameter space, and the minimum sampling interval is set according to the physical characteristics of each parameter to ensure the balance between sampling accuracy and computational overhead. On this basis, the first-order derivative of the objective function at each sampling point is calculated based on the initial grid sampling point matrix to obtain a gradient vector group in the parameter space. The calculation of the first-order derivative can describe the changing trend of the objective function at different parameter points and is the basis for determining the optimal direction. The gradient vector group is orthogonalized and decomposed to construct the basis vectors of the local coordinate system. The purpose of orthogonal decomposition is to eliminate the correlation between gradient vectors, so that the basis vectors of the local coordinate system are independent of each other, which is convenient for subsequent search direction generation. At the same time, the constructed basis vectors are normalized to obtain an orthogonalized basis vector set. The length of the basis vectors is normalized so that optimization operations with equal trade-offs can be performed in different directions to ensure that the search in each direction has the same importance. The orthogonalized basis vector set is input into the steepest descent optimizer to calculate the negative gradient direction and obtain the steepest descent direction vector. The steepest descent method is an optimization method that searches along the negative gradient direction of the objective function. This direction is usually the direction in which the objective function value decreases the fastest. Therefore, the calculation of the steepest descent direction vector can effectively guide the parameters to converge quickly toward the optimization target. The steepest descent direction vector is projected into the orthogonal complement space, and a search direction orthogonal to the steepest descent direction is generated through Schmidt orthogonalization to obtain a vertical search direction vector. The purpose of the projection operation in the orthogonal complement space and Schmidt orthogonalization is to construct other search directions that are unrelated to the steepest descent direction, thereby avoiding falling into local optimality and increasing the diversity and flexibility of parameter optimization. In this way, the vertical search direction vector can provide a search method that is independent of the steepest descent direction, so that the entire parameter optimization process has a higher exploration capability. The steepest descent direction vector and the vertical search direction vector are input into the linear searcher to determine the optimal combination weight and obtain the hybrid search direction. In this process, the linear searcher finds the direction combination that makes the objective function drop the fastest by performing a weighted combination between the steepest descent direction and the vertical search direction. The introduction of the hybrid search direction helps to overcome some defects in the steepest descent method, such as being prone to falling into a zigzag path, thereby improving the efficiency of the optimization process. The hybrid search direction is corrected and updated by the conjugate gradient optimizer to obtain the final set of parameter optimization directions.The conjugate gradient optimizer is an improved optimization method that can avoid the common repeated calculation path problem in gradient methods by introducing conjugate directions and accelerate convergence. Under the action of the conjugate gradient optimizer, the hybrid search direction is corrected, making the parameter optimization direction more global and efficient.

[0038] Step 400: According to the parameter optimization direction set, each control parameter is iteratively updated using a dynamic step size to obtain a parameter iteration sequence set;

[0039] Specifically, based on the parameter optimization direction set, the change trend of the light intensity parameter, color temperature parameter, power parameter and temperature parameter is analyzed, the initial step size of each control parameter in the current optimization direction is determined, and the initial step size control matrix is ​​obtained. The objective function gradient is calculated for each parameter in the initial step size control matrix, the sensitivity of the parameter to the target optimization model under the current step size is evaluated, and the dynamic step size factor is obtained. Through this step, the dynamic step size factor adapts to the change requirements of different parameters and realizes flexible step size adjustment, so that it can converge quickly and avoid missing the global optimal solution during the optimization process. Based on the dynamic step size factor, a damping attenuation coefficient is applied to each control parameter to impose a certain degree of suppression on the change of the parameter to prevent the parameter from fluctuating too much and causing system instability, while ensuring that the step size is gradually reduced during the iteration process to achieve accurate convergence. The step size convergence speed is adjusted by the feedback control algorithm, so that the step size can be adaptively reduced when approaching the optimal solution, thereby improving the final convergence accuracy. After adjustment by the feedback control algorithm, a step size update vector is obtained, which describes the step size adjustment amplitude of each parameter in the current optimization direction. The step update vector is inner-producted with the parameter optimization direction set to obtain the candidate parameter update amount. The step update amount is projected onto the optimization direction to determine the actual update amplitude of each control parameter. In this way, the candidate parameter update amount can accurately reflect the change of each control parameter in the current iteration process and provide specific operation values ​​for subsequent parameter updates. The candidate parameter update amount is validated and the parameters are truncated based on the hardware restrictions of the input voltage range AC100-240V and the output voltage DC24V to obtain a valid parameter increment. The purpose of the validity verification is to check whether the candidate parameter update amount meets the physical and hardware constraints of the system and ensure that the generated parameters will not exceed the working range of the device in actual applications. The parameter truncation based on hardware restrictions ensures that the update amount of each control parameter works within the power supply and output restrictions to avoid system failure or hardware damage caused by improper parameter settings. The valid parameter increment is linearly combined with the current iteration parameter through a dynamic weight allocator with memory items to obtain a parameter iteration sequence set. The dynamic weight allocator with memory items retains some historical information during the parameter update process to increase the stability of parameter adjustment. The linear combination process makes the current parameters not only depend on the new effective parameter increments, but also considers the historical parameters of the previous iterations, so as to achieve a smooth transition in the entire optimization process and avoid system oscillations caused by excessive parameter changes. Through this process, the final parameter iteration sequence set contains the update status of each control parameter in each iteration. These parameter iteration sequences can gradually approach the optimal solution of the target optimization model, thereby realizing the adaptive adjustment of the parameters of the simulated light source controller.

[0040] Step 500: for each group of control parameters in the parameter iteration sequence set, a constant voltage controller is used to perform a voltage stabilization output test, and a multi-objective evaluation function is used to calculate the comprehensive scores of the voltage stability index, the overcurrent protection index, the trigger response index and the temperature balance index to obtain multiple candidate control parameter combinations;

[0041] Specifically, the parameter iteration sequence set is input into the constant voltage controller, and the constant voltage controller stabilizes the input voltage AC100-240V through the PWM rectifier conversion circuit to obtain a regulated output sequence. The PWM rectifier conversion circuit can rectify and filter the AC voltage and convert it into a stable DC output. The voltage ripple detection and overcurrent threshold detection are performed on the regulated output sequence through a 2.5A range sampling circuit to obtain the single-channel operation characteristic data. The voltage ripple detection evaluates the smoothness of the output voltage and the quality of the power supply. Low ripple means good voltage regulation performance, which helps to reduce interference with the load. The overcurrent threshold detection is used to monitor whether the output current exceeds the set safety threshold to ensure that the system can operate within a safe current range to avoid damage to the hardware. The single-channel operation characteristic data is input into the performance evaluation module. In the performance evaluation module, the four indicators of voltage stability, overcurrent protection, trigger response and temperature balance are quantitatively calculated to obtain a performance indicator matrix. The voltage stability index is used to quantify the stability of the output voltage under different working conditions and reflect the ability of the voltage regulator controller to suppress input voltage fluctuations; the overcurrent protection index is used to measure whether the controller can respond in time and take protective measures when encountering overcurrent conditions; the trigger response index is used to evaluate the system's response speed and accuracy to input signal changes; the temperature balance index is used to evaluate the temperature changes during system operation to ensure that the temperature fluctuates within a reasonable range to prevent overheating from causing system failure or performance degradation. Through the quantitative calculation of these indicators, the performance indicator matrix can fully reflect the performance of the control parameters in all aspects. The various indicators in the performance indicator matrix are weighted and summed by the weight distributor to obtain a comprehensive performance score. The weight distributor assigns different weights to different indicators according to their contribution to the overall performance of the system to ensure that the comprehensive performance score can objectively reflect the overall quality of the control parameters. The comprehensive performance scores are arranged in descending order and the screening threshold is set to obtain candidate control parameter combinations. The descending order is to give priority to control parameters with higher comprehensive scores to ensure that the candidate combinations have better performance in all performance indicators. The setting of the screening threshold is to eliminate parameter combinations with substandard comprehensive scores to narrow the range of the candidate set.

[0042] Step 600: Input a plurality of candidate control parameter combinations into a multi-channel cooperative controller for parallel verification testing, and determine an optimal control parameter combination according to the degree of interference suppression between the channels.

[0043] Specifically, the candidate control parameter combinations are input into the SMP-03V-BC ports of the four-channel 60W power controller, and the four channels are tested for parallel operation at the same time to simulate the actual operating environment of the light source controller when working in multi-channel collaboration, so as to ensure that the candidate control parameter combinations can still achieve stable control under complex multi-channel conditions. By collecting the output voltage, current, power and temperature parameters of each channel, a multi-channel operation data matrix is ​​formed to reflect the operating characteristics of the four channels when working in parallel. The multi-channel operation data matrix is ​​crosstalk detected by the channel isolation amplifier to evaluate the signal interference between the channels, that is, to determine whether the signal of one channel will have an adverse effect on other channels. In order to analyze the mutual influence between channels, the load ratio of each channel is calculated based on the dynamic power allocation algorithm to obtain the interference characteristic data between channels. The dynamic power allocation algorithm analyzes the load characteristics of each channel and dynamically adjusts the power allocation ratio of each channel so that the load of each channel can be considered in the interference characteristic analysis to provide a more accurate interference evaluation result. The interference characteristic data between channels are input into the compensation controller, and the interference between channels is suppressed by the zero drift compensation circuit and the temperature compensation network. The function of the zero drift compensation circuit is to eliminate the zero drift error in the signal and ensure that the reference voltage of the circuit remains stable during the measurement and control process, thereby improving the accuracy of interference suppression. The temperature compensation network is to compensate for the performance fluctuations caused by changes in ambient temperature during the operation of the system and ensure the stability of the controller under different temperature conditions. Through the dual compensation of zero drift and temperature, the compensation controller can effectively reduce the mutual interference between channels and improve the anti-interference ability of the system. The compensation control parameters are corrected by the segmented linearization circuit, and the nonlinear compensation parameters are transformed into a linear form to facilitate subsequent control calculation and adjustment. After the linearization correction is completed, the compensation control parameters are delayed controlled by the trigger delay compensation network to obtain a multi-channel cooperative control sequence. The purpose of trigger delay compensation is to adjust the signal delay in the compensation control, ensure that the control signals of all channels can arrive synchronously, realize coordinated control between channels, and avoid the aggravation of mutual interference caused by asynchronous signals. The multi-channel cooperative control sequence is feature mapped with the candidate control parameter combination, and the parameter combination with the strongest anti-interference ability is selected by the minimum interference criterion. The compensated control parameters are compared and analyzed with the candidate parameters to find the optimal solution based on specific criteria. The minimum interference criterion is used to evaluate the performance of each parameter combination in interference suppression, and the goal is to find a parameter combination that can minimize the interference between channels. Through this process, the optimal control parameters that best meet the requirements are screened out from multiple candidate control parameter combinations to obtain the optimal control parameter combination.

[0044] In the embodiments of the present application, parameter mapping and principal component analysis dimensionality reduction are performed through a neural network model, thereby achieving accurate modeling of the complex dynamic characteristics of the analog light source controller and providing a reliable mathematical basis for parameter optimization; multi-objective optimization and grid search strategies are adopted, while taking into account multiple control objectives such as light intensity stability, spectral fidelity, energy efficiency and temperature balance, thereby significantly improving the overall performance of the system; a dynamic step size adaptive adjustment mechanism is introduced, so that the control parameters can be automatically adjusted according to changes in the working environment, thereby enhancing the environmental adaptability of the system; through a multi-channel collaborative control strategy, crosstalk between channels is effectively suppressed, and the problem of mutual interference when multiple channels work in parallel is solved; based on a hierarchical optimization structure, the parameter optimization process is divided into multiple links such as feature extraction, target construction, parameter search and performance evaluation, thereby improving the convergence efficiency of the optimization algorithm; a complete performance evaluation system is designed, and the control parameters are comprehensively evaluated through a multi-objective evaluation function, thereby ensuring the reliability and practicality of the optimization results.

[0045] In a specific embodiment, the process of executing step 100 may specifically include the following steps:

[0046] The input voltage signal collected by the SMP-03V-BC port is converted into digital quantity and phase compensated by a 16-bit high-precision analog-to-digital converter to obtain a voltage input sequence; the output current signal collected by the constant voltage drive circuit is detected and baseline calibrated by a 2.5A range Hall current sensor to obtain a current input sequence; the ambient temperature signal collected by the forced cooling controller is sampled and temperature compensated by an NTC thermistor to obtain a temperature input sequence; the channel power signals of the four channels are detected and channel load calculated by a channel isolation amplifier to obtain a power input sequence;

[0047] Perform sampling timing alignment and feature matrix construction on the voltage input sequence, current input sequence, temperature input sequence and power input sequence to obtain an input feature sequence matrix;

[0048] The input feature sequence matrix is ​​input into the first hidden layer of the neural network model. The first hidden layer contains 256 neuron nodes with a dropout mechanism. The nonlinear feature transformation is performed through the ReLU activation function, and the feature distribution is dynamically adjusted using the batch normalization method to obtain the first layer feature map.

[0049] The first layer feature map is input into the second hidden layer of the neural network model. The second hidden layer contains 128 neuron nodes with residual connections. The feature is compressed by the Sigmoid activation function, and the key features are enhanced by the attention mechanism to obtain the second layer feature map.

[0050] The second layer feature map is input into the output layer of the neural network model. The output layer contains 64 fully connected neuron nodes. The light source parameters are mapped through a linear activation function to obtain the dynamic response matrix of the light source parameters.

[0051] Specifically, the input voltage signal collected by the SMP-03V-BC port is converted into digital quantity and phase compensated by a 16-bit high-precision analog-to-digital converter to obtain a voltage input sequence. The input voltage signal is a continuous analog signal, and the 16-bit high-precision analog-to-digital converter converts the analog signal into a corresponding digital signal. The high-precision analog-to-digital conversion ensures that the details of the sampled signal are retained, and its resolution can express subtle voltage changes, so that the sampling accuracy meets the requirements of the control system. At the same time, phase compensation is used to compensate for the phase lag or advance caused by signal acquisition and system transmission. The compensation process is described as:

[0052] ;

[0053] in, represents the compensated voltage input signal, is the uncompensated voltage signal, is the correction amount for phase compensation. The output current signal collected by the constant voltage drive circuit is processed, and this process relies on a 2.5A range Hall current sensor. The Hall current sensor is used to realize current detection, and baseline calibration is performed to eliminate baseline drift during the measurement process. The current signal detection is expressed as:

[0054] ;

[0055] in, is the calibrated output current signal, is the current initially measured by the Hall sensor, The offset is caused by sensor characteristics or environmental conditions, which is eliminated through the calibration process to improve measurement accuracy. At the same time, the ambient temperature signal is collected and processed, the ambient temperature signal is collected through the forced cooling controller, and the temperature is sampled and compensated through the NTC thermistor to obtain the temperature input sequence. The characteristic that the resistance value of the NTC thermistor changes with temperature is used to obtain the temperature information of the surrounding environment. The purpose of temperature compensation is to ensure that the measured temperature signal truly reflects the monitored ambient temperature when the external temperature fluctuates. It is described as:

[0056] ;

[0057] in, is the temperature after compensation, is the temperature signal measured directly, is the temperature compensation coefficient. and are the ambient temperature and the reference temperature respectively. For the four-channel channel power signals, the power is detected by the channel isolation amplifier to avoid mutual interference between channels. The power of each channel is expressed as:

[0058] ;

[0059] in, is the detected single channel power, and are channel voltage and current respectively. By calculating the load of each channel, the load distribution is obtained for further power balancing and management. The voltage input sequence, current input sequence, temperature input sequence and power input sequence are sampled and aligned and the feature matrix is ​​constructed to obtain the input feature sequence matrix. Sampling timing alignment ensures that all input signals have consistent timestamps on the same time axis, so that each column in the input feature matrix represents the features at the same time point. Input to the first hidden layer of the neural network model, the first hidden layer contains 256 neuron nodes with dropout mechanism, and nonlinear feature transformation is performed through ReLU activation function. The function of ReLU activation function is to introduce nonlinear characteristics so that the network can process more complex input features. Its expression is:

[0060] ;

[0061] The dropout mechanism prevents network overfitting and improves the generalization ability of the model by randomly setting the output of some neuron nodes to 0. At the same time, the batch normalization method is used to dynamically adjust the feature distribution. The expression of batch normalization is:

[0062] ;

[0063] in, and Represent the mean and variance of the mini-batch input, respectively. is a very small value used to avoid division by 0. The introduction of batch normalization helps speed up training and improve the stability of the network. The first layer of features is input into the second hidden layer of the neural network. The second hidden layer contains 128 neuron nodes with residual connections, and features are compressed by the Sigmoid activation function. The Sigmoid activation function compresses the input value between 0 and 1, and its expression is:

[0064] ;

[0065] This makes the output feature value fall within a fixed range, which helps maintain the numerical stability of the model. The introduction of residual connection allows the original input to bypass some hidden layers and pass directly to the next layer, alleviating the gradient disappearance problem in deep networks. The residual connection is expressed as:

[0066] ;

[0067] in, is the input, is the output of the hidden layer. At the same time, the attention mechanism is used to enhance the key features. By calculating the importance of different features and weighting them, the model can pay more attention to the features that have a significant impact on the final output. The second layer feature map is input to the output layer of the neural network model. The output layer contains 64 fully connected neuron nodes. The light source parameters are mapped through the linear activation function to obtain the dynamic response matrix of the light source parameters. The linear activation function is used to maintain the linear relationship between input and output. Its expression is:

[0068] ;

[0069] The light source parameter dynamic response matrix contains the response characteristics of the light source under input conditions such as input voltage, current, temperature and power.

[0070] In a specific embodiment, the process of executing step 200 may specifically include the following steps:

[0071] Perform mean centering and standard deviation normalization operations on the light source parameter dynamic response matrix to obtain a standardized parameter matrix;

[0072] Perform covariance calculation and eigenvector decomposition on the standardized parameter matrix to obtain a set of characteristic components that characterize the light intensity response function, color temperature response function, temperature response function, and power consumption response function;

[0073] The eigenvalues ​​of each eigencomponent are calculated based on the eigencomponent set, and the principal component features are screened according to the contribution rate threshold of 95% to obtain the dimension reduction mapping matrix;

[0074] The standardized parameter matrix and the dimension reduction mapping matrix are orthogonally transformed to obtain a low-dimensional feature vector including light intensity stability parameters, spectral fidelity parameters, energy efficiency parameters and temperature balance parameters;

[0075] Based on the deviation between the light intensity stability parameter calculation in the low-dimensional feature vector and the set threshold, the light intensity optimization target item is constructed; based on the spectral fidelity parameter calculation in the low-dimensional feature vector and the matching degree with the standard spectrum, the spectrum optimization target item is constructed; based on the energy efficiency parameter calculation in the low-dimensional feature vector and the ratio with the input power, the energy consumption optimization target item is constructed; based on the temperature balance parameter calculation in the low-dimensional feature vector and the difference with the maximum allowable temperature, the temperature balance target item is constructed;

[0076] The light intensity optimization target item, spectrum optimization target item, energy consumption optimization target item, and temperature balance target item are combined into a target optimization model.

[0077] Specifically, the light source parameter dynamic response matrix is ​​subjected to mean centering and standard deviation normalization to obtain a standardized parameter matrix. Mean centering eliminates the offset of each parameter in the data, making the average value of each feature zero, making subsequent calculations more stable. The formula is:

[0078] ;

[0079] in, is the light source parameter dynamic response matrix, is the mean vector of each column in the matrix, is the matrix after centering. Then the standard deviation normalization operation is performed to make the scale of the data consistent to eliminate the influence of different dimensions between features. The formula is:

[0080] ;

[0081] in, is the standard deviation vector, is the normalized parameter matrix. Through preprocessing, the standardized data of each feature at the same scale is obtained. The covariance calculation and eigenvector decomposition of the standardized parameter matrix are performed to obtain the characteristic component set that characterizes the light intensity response function, color temperature response function, temperature response function and power consumption response function. Covariance matrix The calculation formula is:

[0082] ;

[0083] in, is the sample size, is the transposed matrix of the normalized parameter matrix. The covariance matrix Describes the linear relationship between features and is used to determine the correlation between features. Perform eigenvector decomposition on the covariance matrix:

[0084] ;

[0085] in, are the eigenvectors of the covariance matrix, is the corresponding eigenvalue. The eigenvector represents the main direction of the covariance matrix, and the eigenvalue represents the variance of these directions. By combining the eigenvector and the eigenvalue, a set of characteristic components describing the response of light intensity, color temperature, temperature and power consumption is obtained. Based on the set of characteristic components, the eigenvalues ​​of each characteristic component are calculated, and the main component features are filtered according to the threshold of 95% contribution rate to obtain the dimension reduction mapping matrix. The contribution rate is the proportion of the eigenvalue in all eigenvalues, which is expressed by the formula:

[0086] ;

[0087] in, The former eigenvalues, is the total number of eigenvalues. When the cumulative contribution rate reaches 95%, the corresponding eigenvector is selected as the principal component feature to construct the dimensionality reduction mapping matrix , where each column is a principal component eigenvector. Perform orthogonal transformation operations to obtain a low-dimensional feature vector containing light intensity stability parameters, spectral fidelity parameters, energy efficiency parameters, and temperature balance parameters :

[0088] ;

[0089] in, It is a low-dimensional feature vector, each element of which represents different key characteristics of light source control, including light intensity stability parameters, spectral fidelity parameters, energy efficiency parameters and temperature balance parameters. The orthogonal transformation maps the original high-dimensional data to a lower-dimensional space through a dimensionality reduction mapping matrix, while retaining the most important features to simplify subsequent optimization calculations. The target optimization model is constructed based on each parameter in the low-dimensional feature vector. Based on the light intensity stability parameters in the low-dimensional feature vector Calculate its deviation from the set threshold and construct the light intensity optimization target item:

[0090] ;

[0091] in, is the target value of light intensity, is the optimization target item of light intensity. This target item measures the difference between the actual light intensity stability and the set target. By minimizing this difference, the light intensity output by the light source is ensured to remain stable within the set range. Based on the spectral fidelity parameters in the low-dimensional feature vector Calculate the degree of match with the standard spectrum to construct the spectrum optimization objective:

[0092] ;

[0093] in, The standard spectrum Quantity, is the number of components in the spectrum, Represents the spectrum optimization objective. By minimizing the error between the spectrum fidelity parameter and the standard spectrum, the spectral output of the light source is ensured to match the expected standard spectrum, thereby improving the light quality of the light source. Based on the energy efficiency parameter in the low-dimensional feature vector , calculate its value with input power The ratio of , constructs the energy consumption optimization target item:

[0094] ;

[0095] By minimizing the negative energy consumption optimization target, maximizing the energy conversion efficiency, and ensuring that the light source controller can work at the highest possible energy efficiency, energy consumption can be reduced and the economic and environmental performance of the overall system can be improved. Calculate its maximum allowable temperature The difference between and is used to construct the temperature balance objective:

[0096] ;

[0097] By minimizing the deviation between the temperature parameters and the maximum allowable temperature, the light source controller is ensured to work within a reasonable temperature range to avoid equipment damage or performance degradation caused by overheating. Through the above steps, the light intensity optimization target item, spectrum optimization target item, energy consumption optimization target item and temperature balance target item are combined to finally obtain the target optimization model:

[0098] ;

[0099] in, are weight coefficients of each target item, which are used to balance the relative importance of different optimization targets. By adjusting these weight coefficients, specific performance indicators can be optimized according to different application requirements. For example, if the energy efficiency requirement is high in a certain application scenario, increasing value to be more inclined to improve energy efficiency.

[0100] In a specific embodiment, the process of executing step 300 may specifically include the following steps:

[0101] The target optimization model is input into a multi-parameter grid generator to divide the parameter intervals, and a search boundary matrix including a light intensity parameter interval, a color temperature parameter interval, a power parameter interval, and a temperature parameter interval is obtained;

[0102] The search boundary matrix is ​​divided into equal intervals, and the minimum sampling interval is determined according to the physical constraints of each parameter to obtain the initial grid sampling point matrix. The first-order derivative of the objective function at each sampling point is calculated based on the initial grid sampling point matrix to obtain the gradient vector group in the parameter space.

[0103] Perform an orthogonal decomposition operation on the gradient vector group, construct the basis vector of the local coordinate system, and perform unitization processing to obtain an orthogonalized basis vector set;

[0104] The orthogonalized basis vector set is input into the steepest descent optimizer to calculate the negative gradient direction and obtain the steepest descent direction vector;

[0105] Perform orthogonal complementary space projection operation on the steepest descent direction vector, and generate a search direction orthogonal to it through Schmidt orthogonalization processing to obtain a vertical search direction vector;

[0106] The steepest descent direction vector and the vertical search direction vector are input into the linear searcher to determine the optimal combination weight, and a hybrid search direction is obtained. The hybrid search direction is corrected and updated by a conjugate gradient optimizer to obtain a parameter optimization direction set.

[0107] Specifically, the target optimization model is input into the multi-parameter grid generator to reasonably divide the parameter intervals. The target optimization model contains the optimization requirements of light intensity parameters, color temperature parameters, power parameters and temperature parameters, and the role of the multi-parameter grid generator is to divide these parameters into different intervals to form a search boundary matrix containing the ranges of each parameter. Assume that the intervals of light intensity parameters, color temperature parameters, power parameters and temperature parameters are , the search boundary matrix is ​​expressed as:

[0108] ;

[0109] Among them, the matrix Describe the upper and lower limits of light intensity, color temperature, power and temperature, and Represent the lower and upper limits of each parameter respectively. Define the range of the parameter to ensure that the subsequent search process is carried out within a reasonable boundary and avoid exceeding the physical limitations of the device. Divide the search boundary matrix into equal intervals to construct the initial grid sampling point matrix. When performing equal interval division, determine the minimum sampling interval based on the physical constraints of each parameter. Assume that for a certain parameter, its interval is , the minimum sampling interval is , then the expression of the sampling point is expressed as:

[0110] ;

[0111] in, is the number of sampling points, Indicates After all parameters are divided equally in this way, the initial grid sampling point matrix is ​​obtained. , which contains the possible values ​​of each parameter in the parameter space. Based on the initial grid sampling point matrix, the first-order derivative of the objective function at each sampling point is calculated to obtain the gradient vector group in the parameter space. The first-order derivative of the objective function describes the changing trend of the function at the current point. Assuming that the objective function is ,in represents the parameter vector, and its gradient vector is expressed as:

[0112] ;

[0113] in, Represent different parameters respectively. It means that the objective function is at point The gradient vector at . By calculating the first-order derivative of each sampling point in the initial grid sampling point matrix, the gradient vector group is obtained , each gradient vector describes the direction of function change at a specific point in the parameter space. The gradient vector group is orthogonalized to construct the basis vectors of the local coordinate system, and these basis vectors are normalized to obtain the orthogonalized basis vector set. Assume that the original gradient vector group is , through orthogonal decomposition, we get the orthogonal basis vector set , where each basis vector is orthogonal to each other. The orthogonalization process is implemented by the Gram-Schmidt method, and its recursive relationship is:

[0114] ;

[0115] After obtaining the orthogonal basis vectors, normalize them to ensure that the length of all basis vectors is 1:

[0116] ;

[0117] in, are normalized orthogonal basis vectors, Representation vector The modulus length of the target function. Through this process, the orthogonalized basis vector set is obtained, which is used to accurately describe the direction of parameter change in the local coordinate system. The orthogonalized basis vector set is input into the steepest descent optimizer to calculate the negative gradient direction and obtain the steepest descent direction vector. The steepest descent method searches along the negative gradient direction of the objective function to reduce the value of the objective function as quickly as possible. The steepest descent direction is expressed as:

[0118] ;

[0119] in, Represents the steepest descent direction vector. Following this direction, the value of the objective function can be quickly reduced, guiding the parameters to move toward the optimal point. Perform orthogonal complementary space projection on the steepest descent direction vector to obtain other search directions. By performing orthogonal projection on the steepest descent direction vector, directions that are not associated with it can be obtained, avoiding the optimization process from falling into a local optimum. In order to ensure that the newly generated direction is orthogonal to the steepest descent direction, Schmidt orthogonalization is used to generate a search direction orthogonal to the steepest descent direction. Assume that this direction is , whose expression is:

[0120] ;

[0121] After this process, we get The direction is orthogonal to the steepest descent direction, providing more exploration dimensions for parameter optimization. The steepest descent direction vector and the vertical search direction vector are input into the linear searcher to determine the optimal combination weight and obtain the hybrid search direction. The hybrid search direction is a weighted combination of the steepest descent direction and the vertical search direction, and the formula is:

[0122] ;

[0123] in, and are the combined weights of the steepest descent direction and the vertical direction, respectively. The optimal combined weight that can minimize the objective function value is found through linear search. The hybrid search direction is corrected and updated by the conjugate gradient optimizer to obtain the final set of parameter optimization directions. Conjugate gradient optimization is an efficient iterative optimization method. By making the new direction conjugate with the previous search direction, repeated searches in the same direction are avoided, thereby improving the convergence speed and optimization efficiency. Assume that the current search direction is , the updated search direction is , then the recursive relation of the conjugate gradient is:

[0124] ;

[0125] in, is the conjugate coefficient, usually calculated as:

[0126] ;

[0127] After the hybrid search direction is corrected by the conjugate gradient optimizer, the parameter optimization direction set is finally obtained. , these directions are used to guide each control parameter to update towards the optimal value to achieve global optimization of the target optimization model.

[0128] In a specific embodiment, the process of executing step 400 may specifically include the following steps:

[0129] Based on the parameter optimization direction set, the change trend of the light intensity parameter, color temperature parameter, power parameter and temperature parameter is analyzed to obtain the initial step size control matrix;

[0130] Calculate the objective function gradient for each parameter in the initial step size control matrix to obtain the dynamic step size factor;

[0131] Based on the dynamic step size factor, a damping attenuation coefficient is applied to each control parameter, and the step size convergence speed is adjusted through the feedback control algorithm to obtain the step size update vector;

[0132] Perform an inner product operation on the step size update vector and the parameter optimization direction set to obtain the candidate parameter update amount;

[0133] The validity of the candidate parameter update amount is verified, and the parameter is truncated based on the hardware limitations of the input voltage range AC100-240V and the output voltage DC24V to obtain the valid parameter increment;

[0134] The effective parameter increment is linearly combined with the current iteration parameter through a dynamic weight allocator with memory item to obtain a set of parameter iteration sequences.

[0135] Specifically, based on the parameter optimization direction set, the change trend of the light intensity parameter, color temperature parameter, power parameter and temperature parameter is analyzed to obtain the initial step size control matrix. middle, They represent the optimization directions of light intensity, color temperature, power and temperature parameters respectively. The change trend of each parameter is analyzed in these directions to determine the initial step size control matrix . Let the change of each parameter in the direction of the initial step be , the initial step size control matrix is ​​expressed as:

[0136] ;

[0137] in Representative The sample The step size of each parameter is determined by analyzing the change trend. The initial step size of each parameter in different optimization directions is obtained by calculating the gradient of the objective function for each parameter in the initial step size control matrix to obtain the dynamic step size factor. Assume that the objective function is ,in Represents the light intensity, color temperature, power and temperature parameters, and the objective function gradient at each step is expressed as:

[0138] ;

[0139] The dynamic step size factor is obtained by calculating the gradient of the objective function value of each parameter in different step sizes. , whose expression is:

[0140] ;

[0141] It indicates the step size adjustment required for each control parameter in different optimization directions, reflecting the change trend of the objective function in the current direction and its sensitivity to the step size. Based on the dynamic step size factor, a damping attenuation coefficient is applied to each control parameter to reduce the instability caused by excessive step size, and the step size convergence speed is adjusted through the feedback control algorithm. Damping attenuation coefficient The role of is to appropriately decay the step size at each iteration, so that the step size is gradually reduced to ensure that it converges to the optimal point in the end. Assume that the current step size is updated to , the step size update after damping is expressed as:

[0142] ;

[0143] in , which is usually a positive number less than 1. The feedback control algorithm is used to adjust the convergence speed of the step size. The step size is dynamically adjusted by monitoring the change of the objective function value. If the change of the objective function slows down, the step size is gradually reduced, otherwise the step size is increased to improve the efficiency of the optimization process. Through these processes, the final step size update vector is obtained. . Perform the inner product operation on the step size update vector and the parameter optimization direction set to obtain the candidate parameter update amount. Suppose the step size update vector is , the parameter optimization direction set is , then the candidate parameter update amount is expressed as:

[0144] ;

[0145] in, Indicates the parameter update amplitude in the current optimization direction. The step size and optimization direction are combined through the inner product operation to determine the update amount of each parameter. The validity of the candidate parameter update amount is verified. Considering the range of input voltage and output voltage Due to the hardware limitation, the candidate parameter update amount is truncated to obtain a valid parameter increment The purpose of validity verification is to ensure that all parameter updates do not exceed the safe operating range of the hardware, thereby ensuring the stability and security of the system. Parameter truncation is expressed as:

[0146] ;

[0147] in, and The parameters are The lower and upper limits of Represents the effective parameter increment after truncation. The effective parameter increment is linearly combined with the current iteration parameter through a dynamic weight allocator with memory to obtain a parameter iteration sequence set. The function of the dynamic weight allocator with memory is to retain some historical information to improve the stability of parameter updates and prevent excessive fluctuations in parameters during the update process. Assume that the parameters of the current iteration are , the effective parameter increment is , then the parameters of the next iteration It is expressed as:

[0148] ;

[0149] in, is the memory coefficient, and its value range is , which is used to control the weight ratio between historical parameters and new parameter increments. When it is larger, it means that it is more inclined to retain historical information, and when When it is smaller, it means that the parameter increment of the new update is more acceptable.

[0150] In a specific embodiment, the process of executing step 500 may specifically include the following steps:

[0151] The parameter iteration sequence set is input into the constant voltage controller, and the input voltage AC100-240V is stabilized by the PWM rectifier conversion circuit to obtain a stabilized output sequence;

[0152] The voltage ripple detection and over-current threshold detection are performed on the regulated output sequence through a 2.5A range sampling circuit to obtain single-channel operation characteristic data;

[0153] The single-channel operation characteristic data is input into the performance evaluation module, and the four indicators of voltage stability, overcurrent protection, trigger response and temperature balance are quantitatively calculated to obtain the performance indicator matrix;

[0154] The various indicators in the performance indicator matrix are weighted and summed through the weight distributor to obtain a comprehensive performance score;

[0155] The comprehensive performance scores are arranged in descending order and a screening threshold is set to obtain candidate control parameter combinations.

[0156] Specifically, the parameter iteration sequence set is input into the constant voltage controller, and the input voltage is rectified by the PWM rectifier conversion circuit. The input voltage range is AC100-240V, and the function of the PWM rectifier conversion circuit is to convert the AC input voltage into a stable DC output voltage. PWM (Pulse Width Modulation) technology controls the average value of the output voltage by quickly switching the input voltage on and off to keep it constant. The PWM rectification process is expressed as:

[0157] ;

[0158] in, is the proportionality coefficient of the rectification transformation, is the duty cycle. By adjusting the duty cycle Regulating the output voltage To achieve a stable target value. In the case where the input voltage may fluctuate, the PWM circuit can effectively compensate for these fluctuations and obtain a regulated output sequence. The voltage ripple detection and overcurrent threshold detection are performed on the regulated output sequence through a 2.5A range sampling circuit to obtain the operating characteristic data of a single channel. The voltage ripple detection monitors whether there are irregular small amplitude oscillations in the output voltage. Such oscillations will interfere with the system load and thus affect the stability of the system. Assuming the voltage ripple size is , and its quantitative calculation formula is:

[0159] ;

[0160] in, and Respectively represent the maximum and minimum values ​​of the output voltage, Indicates the ripple amplitude of the output voltage. , to determine whether the voltage regulation effect is good and ensure the output stability of the system. Overcurrent threshold detection can determine whether the output current exceeds the safe operating range to prevent the system from being damaged due to excessive current. Assuming the output current is , the maximum safe current threshold is , the conditions for overcurrent detection are:

[0161] ;

[0162] If the output current Exceed , the overcurrent protection mechanism is triggered to limit the current. The output current is monitored in real time through the 2.5A range sampling circuit to effectively ensure that the output current is within a safe range, thereby avoiding damage to the control system. The single-channel operation characteristic data is input into the performance evaluation module to quantify the four indicators of voltage stability, overcurrent protection, trigger response and temperature balance to obtain a performance indicator matrix. Voltage stability indicator Reflects the variation of output voltage under different working conditions. The calculation formula is:

[0163] ;

[0164] in, To evaluate the time interval, is the target voltage, For time The output voltage at the moment. The voltage deviation in the entire time interval is calculated by integration to obtain the quantitative result of the voltage stabilization effect. Overcurrent protection index It is used to quantify the effectiveness of the overcurrent protection function. Its value is 0 or 1, indicating no overcurrent and overcurrent respectively. Indicates the system's response speed to the input signal. Assuming that the input signal changes in time The time required for the system output to reach a steady state is , then the trigger response index is calculated as:

[0165] ;

[0166] This indicator reflects the system's ability to respond quickly. The larger the value, the faster the system responds. Temperature balance indicator It is used to quantify the stability of system temperature during operation. Assuming the system temperature is , the target temperature is , the temperature balance index is calculated as:

[0167] ;

[0168] By calculating these four indicators, we can get the performance indicator matrix , which is expressed as:

[0169] ;

[0170] Performance Indicator Matrix Reflects the specific performance of the voltage stabilization output sequence in various performance indicators. The various indicators in the performance indicator matrix are weighted and summed through the weight distributor to obtain a comprehensive performance score. Suppose the weight coefficient is , then the comprehensive performance score Calculated as:

[0171] ;

[0172] Among them, the weight coefficient It is used to measure the relative importance of each indicator in the overall performance. The weights are different in different application scenarios. For example, in some cases, the voltage stability of the system is more important than the response speed. By weighted summation, the quantitative results of each indicator are combined to obtain a comprehensive performance evaluation value. , to fully reflect the overall operating performance of the system. Arrange the comprehensive performance scores in descending order and set a screening threshold to obtain candidate control parameter combinations. Assume that the set of comprehensive performance scores is , sorting these scores in descending order:

[0173] ;

[0174] Set a screening threshold , filter out all the comprehensive performance scores greater than or equal to The parameter combination is expressed as:

[0175] ;

[0176] Through this screening process, a set of candidate control parameter combinations are obtained, which perform well in various performance indicators and can meet the optimization requirements of the system.

[0177] In a specific embodiment, the process of executing step 600 may specifically include the following steps:

[0178] The candidate control parameter combinations are input into the SMP-03V-BC ports of the four-channel 60W power controller, and the four channels are tested in parallel to obtain a multi-channel operation data matrix;

[0179] The multi-channel operation data matrix is ​​tested for crosstalk through channel isolation amplifiers, and the load ratio of each channel is calculated based on the dynamic power allocation algorithm to obtain the interference characteristic data between channels;

[0180] The interference characteristic data between channels is input into the compensation controller, and interference suppression processing is performed through the zero drift compensation circuit and the temperature compensation network to obtain the compensation control parameters;

[0181] The compensation control parameters are corrected by a piecewise linearization circuit, and the delay is controlled by a trigger delay compensation network to obtain a multi-channel coordinated control sequence.

[0182] The multi-channel cooperative control sequence and the candidate control parameter combination are feature mapped, and the parameter combination with the strongest anti-interference ability is screened out through the minimum interference criterion to obtain the optimal control parameter combination.

[0183] Specifically, the candidate control parameter combinations are input into the SMP-03V-BC port of the four-channel 60W power controller to perform parallel operation tests on the four channels at the same time, and obtain a multi-channel operation data matrix. Assume that the power control parameter combination of each channel is , these parameters are input into the respective channel power controllers through the SMP-03V-BC port to achieve independent regulation of each channel. The purpose of the four-channel parallel operation test is to simulate the multi-channel collaborative operation environment of the system in actual operation, and to ensure that the candidate control parameter combination can provide ideal control effects under complex working conditions. After the test, the data of the four channels during operation are recorded to form a multi-channel operation data matrix , which is expressed as:

[0184] ;

[0185] in, Indicates Time point channel The multi-channel operation data matrix is ​​tested for crosstalk through channel isolation operational amplifiers (op amps) to analyze whether the signal of one channel interferes with other channels, especially in the case of high power and multi-channel parallel operation. This interference will significantly affect the performance of the system. The output is , for channel The crosstalk generated It is measured by detecting the interference component in its output:

[0186] ;

[0187] in, Is the channel without interference The output, Representative channel For Channel After crosstalk detection, the load ratio of each channel is calculated based on the dynamic power allocation algorithm to obtain the interference characteristic data between channels. The dynamic power allocation algorithm dynamically adjusts the power allocation according to the load of each channel to minimize the mutual interference between channels and balance the power load of each channel. The power output of each channel is The total system power is ,aisle Load ratio It is expressed as:

[0188] ;

[0189] Calculate the load ratio of all channels , and obtain characteristic data reflecting the load conditions of each channel. The interference characteristic data between channels is input into the compensation controller, and interference suppression is processed through the zero drift compensation circuit and the temperature compensation network. The zero drift compensation circuit can eliminate the DC offset generated by the amplifier or sensor during operation, which will affect the control accuracy and signal authenticity. Zero drift compensation is expressed as:

[0190] ;

[0191] in, is the uncompensated output, is the zero drift offset, which is obtained by subtracting the offset , and obtain a more accurate output signal. The temperature compensation network is used to compensate for performance fluctuations caused by changes in ambient temperature. The temperature compensation is expressed by the following formula:

[0192] ;

[0193] in, is the temperature compensation coefficient, and Represent the ambient temperature and reference temperature respectively. Through temperature compensation, the system can maintain a relatively stable output under different temperature conditions. After zero drift compensation and temperature compensation, the compensation control parameters are obtained. The compensation control parameters are corrected by the segmented linearization circuit and delayed by the trigger delay compensation network to obtain a multi-channel collaborative control sequence. The role of the segmented linearization circuit is to approximate the nonlinear compensation parameters to multi-segment linearity for subsequent control calculations. Suppose the compensated parameters are , then it can be expressed by piecewise linearization as:

[0194] ;

[0195] in, and Respectively The slope and intercept of the segment, Represents the range of linearization processing. By approximating the compensated parameters to a linear form, the control process can be made simpler and more predictable. After piecewise linearization, the compensation control parameters are delayed by the trigger delay compensation network. Trigger delay compensation ensures synchronization of control signals between multiple channels to avoid increased mutual interference caused by asynchronous delays. Assume that the delay time is , the output after delay is expressed as:

[0196] ;

[0197] Through delay compensation, the signals of each channel arrive synchronously, forming a multi-channel cooperative control sequence to improve the coordination ability and control accuracy of the system. The multi-channel cooperative control sequence is combined with the candidate control parameters for feature mapping, and the parameter combination with the strongest anti-interference ability is selected through the minimum interference criterion. Feature mapping is to analyze the relationship between the compensation control sequence and the original candidate parameters to determine which parameter combination performs best in suppressing inter-channel interference. Let the interference feature be , the anti-interference index obtained after feature mapping is , the goal of the minimum interference criterion is to find The minimum parameter combination is expressed as:

[0198] ;

[0199] in, Represents a candidate control parameter combination. By minimizing the interference factor, the parameter combination with the minimum interference characteristic is screened out to obtain the optimal control parameter combination.

[0200] The above describes the method for adaptively adjusting the parameters of the analog light source controller in the embodiment of the present application. The following describes the device 10 for adaptively adjusting the parameters of the analog light source controller in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the simulated light source controller parameter adaptive adjustment device 10 includes:

[0201] The acquisition module 11 is used to acquire the input voltage signal, output current signal, ambient temperature signal and channel power signal of the analog light source controller, and input them into the neural network model for parameter mapping to obtain the dynamic response matrix of the light source parameters;

[0202] A dimension reduction module 12 is used to perform feature dimension reduction on the dynamic response matrix of light source parameters by using a principal component analysis algorithm, and to construct a target optimization model;

[0203] A partitioning module 13 is used to partition the parameter space into grids based on the target optimization model, and to generate the steepest descent direction vector and the vertical search direction vector at the partitioned grid points to obtain a parameter optimization direction set;

[0204] An updating module 14 is used to iteratively update each control parameter using a dynamic step size according to a parameter optimization direction set to obtain a parameter iteration sequence set;

[0205] The calculation module 15 is used to perform a voltage stabilization output test on each group of control parameters in the parameter iteration sequence set through a constant voltage controller, and calculate the comprehensive scores of the voltage stability index, the overcurrent protection index, the trigger response index and the temperature balance index using a multi-objective evaluation function to obtain multiple candidate control parameter combinations;

[0206] The test module 16 is used to input multiple candidate control parameter combinations into the multi-channel cooperative controller for parallel verification testing, and determine the optimal control parameter combination according to the interference suppression degree between each channel.

[0207] Through the coordinated cooperation of the above-mentioned components, parameter mapping and principal component analysis dimensionality reduction are carried out through the neural network model, which realizes the accurate modeling of the complex dynamic characteristics of the analog light source controller and provides a reliable mathematical basis for parameter optimization; multi-objective optimization and grid search strategies are adopted to take into account multiple control objectives such as light intensity stability, spectral fidelity, energy efficiency and temperature balance, which significantly improves the overall performance of the system; the dynamic step adaptive adjustment mechanism is introduced to enable the control parameters to be automatically adjusted according to changes in the working environment, thereby enhancing the environmental adaptability of the system; through the multi-channel collaborative control strategy, the crosstalk between channels is effectively suppressed, and the mutual interference problem when multiple channels work in parallel is solved; based on the hierarchical optimization structure, the parameter optimization process is divided into multiple links such as feature extraction, target construction, parameter search and performance evaluation, which improves the convergence efficiency of the optimization algorithm; a complete performance evaluation system is designed, and the control parameters are comprehensively evaluated through a multi-objective evaluation function to ensure the reliability and practicality of the optimization results.

[0208] See also Figure 3 , Figure 3 This is a schematic block diagram of the structure of an electronic device 300 provided in an embodiment of the present application. The electronic device 300 includes a processor 301 and a memory 302. The processor 301 and the memory 302 are connected via a device bus 303, wherein the memory 302 may include a non-volatile storage medium and an internal memory.

[0209] The non-volatile storage medium can store a computer program. The computer program includes program instructions, and when the program instructions are executed by the processor 301, the processor 301 can execute any of the above-mentioned methods for adaptively adjusting parameters of a simulation light source controller.

[0210] The processor 301 is used to provide computing and control capabilities to support the operation of the entire electronic device 300 .

[0211] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor 301, the processor 301 can execute any of the above-mentioned methods for adaptively adjusting parameters of the analog light source controller.

[0212] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a partial structure related to the present application scheme, and does not constitute a limitation on the electronic device 300 involved in the present application scheme. The specific electronic device 300 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0213] It should be understood that the processor 301 may be a central processing unit (CPU), and the processor 301 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0214] It should be noted that technicians in the relevant field can clearly understand that for the convenience and simplicity of description, the specific working process of the electronic device 300 described above can refer to the corresponding process of the aforementioned analog light source controller parameter adaptive adjustment method, which will not be repeated here.

[0215] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by one or more processors, the one or more processors implement the method for adaptively adjusting parameters of a simulation light source controller as provided in the embodiment of the present application.

[0216] The computer-readable storage medium may be an internal storage unit of the electronic device 300 in the aforementioned embodiment, such as a hard disk or memory of the electronic device 300. The computer-readable storage medium may also be an external storage device of the electronic device 300, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped with the electronic device 300.

[0217] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0218] 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 an electronic device (which can be a personal computer, server, or network device, etc.) to perform 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.

[0219] 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 adaptively adjusting parameters of a simulated light source controller, characterized in that: The method comprises: Collect the input voltage signal, output current signal, ambient temperature signal and channel power signal of the analog light source controller, and input them into the neural network model for parameter mapping to obtain the dynamic response matrix of the light source parameters; The principal component analysis algorithm is used to perform feature dimensionality reduction on the dynamic response matrix of the light source parameters, and a target optimization model is constructed; specifically, the method comprises: performing mean centering operation and standard deviation normalization operation on the dynamic response matrix of the light source parameters to obtain a standardized parameter matrix; performing covariance calculation and eigenvector decomposition on the standardized parameter matrix to obtain a set of characteristic components that characterize the light intensity response function, the color temperature response function, the temperature response function, and the power consumption response function; calculating the eigenvalues ​​of each characteristic component based on the set of characteristic components, and filtering the principal component features according to a threshold value of 95% of the contribution rate to obtain a dimensionality reduction mapping matrix; performing an orthogonal transformation operation on the standardized parameter matrix and the dimensionality reduction mapping matrix to obtain a matrix containing light intensity stability parameters. , a low-dimensional feature vector of spectral fidelity parameters, energy efficiency parameters and temperature balance parameters; based on the deviation between the light intensity stability parameter calculation in the low-dimensional feature vector and the set threshold, construct a light intensity optimization target item; based on the matching degree between the spectral fidelity parameter calculation in the low-dimensional feature vector and the standard spectrum, construct a spectrum optimization target item; based on the ratio of the energy efficiency parameter calculation in the low-dimensional feature vector to the input power, construct an energy consumption optimization target item; based on the difference between the temperature balance parameter calculation in the low-dimensional feature vector and the maximum allowable temperature, construct a temperature balance target item; combine the light intensity optimization target item, the spectrum optimization target item, the energy consumption optimization target item and the temperature balance target item into a target optimization model; Gridding the parameter space based on the target optimization model, generating the steepest descent direction vector and the vertical search direction vector at the grid points after the division, and obtaining a parameter optimization direction set; According to the parameter optimization direction set, each control parameter is iteratively updated using a dynamic step size to obtain a parameter iteration sequence set; For each group of control parameters in the parameter iteration sequence set, a voltage stabilization output test is performed through a constant voltage controller, and a comprehensive score of a voltage stability index, an overcurrent protection index, a trigger response index, and a temperature balance index is calculated using a multi-objective evaluation function to obtain a plurality of candidate control parameter combinations; The multiple candidate control parameter combinations are respectively input into the multi-channel collaborative controller for parallel verification testing, and the optimal control parameter combination is determined according to the degree of interference suppression between the channels.

2. The method for adaptively adjusting parameters of a simulated light source controller according to claim 1, characterized in that: The input voltage signal, output current signal, ambient temperature signal and channel power signal of the analog light source controller are collected and input into the neural network model for parameter mapping to obtain a dynamic response matrix of light source parameters, including: The input voltage signal collected by the SMP-03V-BC port is converted into digital quantity and phase compensated by a 16-bit high-precision analog-to-digital converter to obtain a voltage input sequence; the output current signal collected by the constant voltage drive circuit is detected and baseline calibrated by a 2.5A range Hall current sensor to obtain a current input sequence; the ambient temperature signal collected by the forced cooling controller is sampled and temperature compensated by an NTC thermistor to obtain a temperature input sequence; the channel power signals of the four channels are detected and channel load calculated by a channel isolation amplifier to obtain a power input sequence; Performing sampling timing alignment and characteristic matrix construction on the voltage input sequence, the current input sequence, the temperature input sequence, and the power input sequence to obtain an input characteristic sequence matrix; Inputting the input feature sequence matrix into the first hidden layer of the neural network model, the first hidden layer includes 256 neuron nodes with a dropout mechanism, performing nonlinear feature transformation through a ReLU activation function, and dynamically adjusting the feature distribution using a batch normalization method to obtain a first-layer feature map; Inputting the first layer feature map into the second hidden layer of the neural network model, the second hidden layer includes 128 neuron nodes with residual connections, performing feature compression through a Sigmoid activation function, and enhancing key features using an attention mechanism to obtain a second layer feature map; The second layer feature map is input into the output layer of the neural network model, the output layer includes 64 fully connected neuron nodes, and the light source parameter mapping is performed through a linear activation function to obtain a light source parameter dynamic response matrix.

3. The method for adaptively adjusting parameters of a simulated light source controller according to claim 1, characterized in that: The parameter space is gridded based on the target optimization model, and the steepest descent direction vector and the vertical search direction vector are respectively generated on the grid points after the division to obtain a parameter optimization direction set, including: Inputting the target optimization model into a multi-parameter grid generator to divide the parameter intervals, and obtaining a search boundary matrix including a light intensity parameter interval, a color temperature parameter interval, a power parameter interval, and a temperature parameter interval; The search boundary matrix is ​​divided into equal intervals, and the minimum sampling interval is determined according to the physical constraints of each parameter to obtain an initial grid sampling point matrix, and the first-order derivative of the objective function at each sampling point is calculated based on the initial grid sampling point matrix to obtain a gradient vector group in the parameter space; Performing an orthogonal decomposition operation on the gradient vector group, constructing a basis vector of a local coordinate system, and performing unitization processing to obtain an orthogonalized basis vector set; Inputting the orthogonalized basis vector set into the steepest descent optimizer to calculate the negative gradient direction to obtain the steepest descent direction vector; Performing an orthogonal complementary space projection operation on the steepest descent direction vector, and generating a search direction orthogonal to the steepest descent direction vector through Schmidt orthogonalization processing, thereby obtaining a vertical search direction vector; The steepest descent direction vector and the vertical search direction vector are input into a linear searcher to determine the optimal combination weight to obtain a hybrid search direction, and the hybrid search direction is corrected and updated by a conjugate gradient optimizer to obtain a parameter optimization direction set.

4. The method for adaptively adjusting parameters of a simulated light source controller according to claim 3, characterized in that: The step of iteratively updating each control parameter using a dynamic step size according to the parameter optimization direction set to obtain a parameter iteration sequence set includes: Based on the parameter optimization direction set, a change trend analysis is performed on the light intensity parameter, the color temperature parameter, the power parameter and the temperature parameter to obtain an initial step length control matrix; Perform objective function gradient calculation on each parameter in the initial step size control matrix to obtain a dynamic step size factor; Applying a damping attenuation coefficient to each control parameter based on the dynamic step size factor, and adjusting the step size convergence speed through a feedback control algorithm to obtain a step size update vector; Perform an inner product operation on the step size update vector and the parameter optimization direction set to obtain a candidate parameter update amount; The validity of the candidate parameter update amount is verified, and the parameter is truncated based on the hardware limitations of the input voltage range AC100-240V and the output voltage DC24V to obtain a valid parameter increment; The effective parameter increment is linearly combined with the current iteration parameter through a dynamic weight allocator with a memory item to obtain a parameter iteration sequence set.

5. The method for adaptively adjusting parameters of a simulated light source controller according to claim 4, characterized in that: For each group of control parameters in the parameter iteration sequence set, a voltage stabilization output test is performed through a constant voltage controller, and a comprehensive score of a voltage stability index, an overcurrent protection index, a trigger response index, and a temperature balance index is calculated using a multi-objective evaluation function to obtain a plurality of candidate control parameter combinations, including: The parameter iteration sequence set is input into the constant voltage controller, and the input voltage AC100-240V is stabilized by the PWM rectification conversion circuit to obtain a stabilized output sequence; The voltage ripple detection and overcurrent threshold detection are performed on the regulated output sequence through a sampling circuit with a 2.5A range to obtain single-channel operation characteristic data; The single-channel operation characteristic data is input into a performance evaluation module, and four indicators, namely, voltage stability, overcurrent protection, trigger response and temperature balance, are quantitatively calculated to obtain a performance indicator matrix; Performing weighted summation on each indicator in the performance indicator matrix through a weight allocator to obtain a comprehensive performance score; The comprehensive performance scores are arranged in descending order and a screening threshold is set to obtain candidate control parameter combinations.

6. The method for adaptively adjusting parameters of a simulated light source controller according to claim 5, characterized in that: The step of inputting the plurality of candidate control parameter combinations into the multi-channel collaborative controller for parallel verification testing, and determining the optimal control parameter combination according to the interference suppression degree between the channels, comprises: The candidate control parameter combinations are respectively input into the SMP-03V-BC ports of the four-channel 60W power controller, and the four channels are tested in parallel to obtain a multi-channel operation data matrix; Performing crosstalk detection on the multi-channel operation data matrix through a channel isolation amplifier, and calculating the load ratio of each channel based on a dynamic power allocation algorithm to obtain inter-channel interference characteristic data; Inputting the inter-channel interference characteristic data into a compensation controller, performing interference suppression processing through a zero drift compensation circuit and a temperature compensation network, and obtaining compensation control parameters; The compensation control parameters are subjected to curve correction through a piecewise linearization circuit, and delay control is performed through a trigger delay compensation network to obtain a multi-channel coordinated control sequence; The multi-channel cooperative control sequence and the candidate control parameter combination are feature mapped, and the parameter combination with the strongest anti-interference ability is screened out through the minimum interference criterion to obtain the optimal control parameter combination.

7. A device for adaptively adjusting parameters of a simulated light source controller, characterized in that: Used to execute the method for adaptively adjusting parameters of a simulated light source controller according to any one of claims 1 to 6, the device for adaptively adjusting parameters of a simulated light source controller comprises: The acquisition module is used to collect the input voltage signal, output current signal, ambient temperature signal and channel power signal of the analog light source controller, and input them into the neural network model for parameter mapping to obtain the dynamic response matrix of the light source parameters; A dimension reduction module, used to perform feature dimension reduction on the dynamic response matrix of the light source parameters by using a principal component analysis algorithm, and to construct a target optimization model; A partitioning module, used to perform grid partitioning on the parameter space based on the target optimization model, and to generate a steepest descent direction vector and a vertical search direction vector on the divided grid points to obtain a parameter optimization direction set; An updating module, configured to iteratively update each control parameter using a dynamic step size according to the parameter optimization direction set to obtain a parameter iteration sequence set; A calculation module, for performing a voltage stabilization output test on each group of control parameters in the parameter iteration sequence set through a constant voltage controller, and calculating a comprehensive score of a voltage stability index, an overcurrent protection index, a trigger response index, and a temperature balance index using a multi-objective evaluation function to obtain a plurality of candidate control parameter combinations; The test module is used to input the multiple candidate control parameter combinations into the multi-channel collaborative controller for parallel verification testing, and determine the optimal control parameter combination according to the interference suppression degree between the channels.

8. An electronic device, characterized in that: The electronic device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the electronic device executes the method for adaptively adjusting parameters of a simulation light source controller as described in any one of claims 1-6.

9. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the method for adaptively adjusting parameters of a simulation light source controller according to any one of claims 1 to 6 is implemented.

Images