An led driving circuit topology and control method

Abstract

The application discloses an LED driving circuit topology and a control method, and relates to the field of circuit topology control.The application realizes dynamic mapping of a use scene characteristic parameter and a driving power parameter through a model, can automatically adapt to different application scene requirements such as vehicle-mounted lighting and dimming control, and enables an LED system to still maintain a current stable output of ±10% when a power voltage fluctuates by ±15%. Mapping results of double models are compared, and whether the load characteristics of a current use scene match the driving circuit currently used is confirmed from two angles, so that it is judged whether the topology structure needs to be adjusted through a parameter set to ensure that the topology structure always matches the current load characteristics, and the system efficiency is improved. Through construction of a final LED driving circuit parameter mapping model, a mapping tool is provided for subsequent precise matching between a use scene and a corresponding LED driving circuit.

Description

Technical Field

[0001] This invention belongs to the field of circuit topology control, and more specifically, relates to an LED driver circuit topology and control method. Background Technology

[0002] When installing LED driver circuits in different application scenarios (e.g., mobile device backlighting, home / commercial lighting, industrial high-power lighting, outdoor advertising screens, vehicle lighting systems), the existing LED driver circuits are often designed based on existing experience. However, due to differences in the understanding of different circuit designers, the designed LED driver circuits cannot be fully adapted to the current application scenarios, thus requiring design changes and increasing the design cycle. Summary of the Invention

[0003] In view of the problems in the related technologies, the present invention proposes an LED driving circuit topology and control method to overcome the above-mentioned technical problems existing in the existing related technologies.

[0004] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0005] This invention relates to an LED driver circuit topology and control method, comprising the following steps:

[0006] S1. Define several usage scenarios for the LED driver circuit to obtain a set of usage scenarios for the LED driver circuit;

[0007] S2. Based on the set of LED driver circuit usage scenarios, collect multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit usage scenario, and train and test the final usage scenario mapping model.

[0008] S3. Collect data from various usage scenarios and corresponding LED driver power supply parameter data to train and test the final LED driver circuit parameter mapping model.

[0009] S4. Based on the final application scenario mapping model and the final LED driver circuit parameter mapping model, map the collected current scenario feature parameter dataset and the current driver power parameter dataset respectively; then compare the mapping results of the two datasets; when the two mapping results are different, adjust the current driver power parameter dataset to obtain the current final driver power parameter dataset.

[0010] S5. Adjust the topology of the LED driver power supply used in the current usage scenario based on the current final driver power supply parameter dataset.

[0011] Preferably, step S1 includes the following steps:

[0012] S11. Define several usage scenarios for the LED driver circuit to obtain a set of LED driver circuit usage scenarios; then define several characteristic parameter types for the LED driver circuit usage scenarios to obtain a set of circuit usage scenario characteristic parameter types.

[0013] S12. Set several types of LED driver circuit scene parameters to obtain a set of LED driver circuit scene parameter types;

[0014] By setting the set of usage scenarios for LED driver circuits and the set of characteristic parameter types for each usage scenario, a basis is provided for determining the current scenario type based on the characteristic parameter types of different scenarios. In addition, by setting the set of scenario parameter types for LED driver circuits, a quantifiable adjustment object is provided for adjusting the parameters of the LED driver circuits corresponding to the respective usage scenarios, thereby making the adjusted LED driver circuit scenarios more compatible with the corresponding usage scenarios.

[0015] Preferably, step S2 includes the following steps:

[0016] S21. Based on the set of LED driver circuit usage scenarios and the set of circuit usage scenario feature parameter types, collect multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit usage scenario to obtain a set of circuit usage scenario feature parameter data matrix.

[0017] S22. Encode each type of LED driver circuit usage scenario data in the LED driver circuit usage scenario set into binary values ​​to obtain the encoded LED driver circuit usage scenario set; then normalize each feature parameter data in the circuit usage scenario feature parameter data matrix set to obtain the processed circuit usage scenario feature parameter data matrix set.

[0018] S23. Construct an initial usage scenario mapping model; use the processed circuit usage scenario feature parameter data matrix set and the encoded LED driver circuit usage scenario set to train and test the initial usage scenario mapping model. After training and testing are completed, the final usage scenario mapping model is obtained.

[0019] By normalizing the feature parameters, such as scaling the voltage value to the [0,1] range, the convergence speed of the model is improved; the hidden layer uses the ReLU function, which can alleviate the gradient vanishing problem and is computationally efficient; through training and testing, the final scene mapping model has a good mapping ability between scene feature parameters and scene type numerical encoding.

[0020] Preferably, step S3 includes the following steps:

[0021] S31. Based on the set of LED driver circuit scenario parameter types and the set of LED driver circuit usage scenarios, collect parameter data of multiple existing usage scenarios and corresponding LED driver circuits to obtain an LED driver circuit parameter data matrix and an existing usage scenario dataset.

[0022] The existing use scenario data in the existing use scenario dataset is binary numerically encoded using the method in S22 to obtain the encoded existing use scenario dataset.

[0023] S32. Construct an initial LED driver circuit parameter mapping model; train and test the initial LED driver circuit parameter mapping model using the encoded existing application scenario dataset and the LED driver circuit parameter data matrix; after training and testing, obtain the final LED driver circuit parameter mapping model.

[0024] By constructing the final LED driver circuit parameter mapping model, a mapping tool is provided for the subsequent accurate matching between the application scenarios and the corresponding LED driver circuits. This ensures that each LED driver circuit with different parameters can be matched to its most suitable application scenario, thereby enabling the LED driver circuits to play their maximum role.

[0025] Preferably, the initial usage scenario mapping model constructed in S23 and the initial LED driving circuit parameter mapping model constructed in S32 both adopt a multilayer perceptron model.

[0026] Preferably, step S4 includes the following steps:

[0027] S41. Set the feature parameter data of the current usage scenario and the parameter dataset of the LED driver power supply corresponding to the current usage scenario data to obtain the current scenario feature parameter dataset and the current driver power supply parameter dataset.

[0028] S42. Normalize the feature parameter data in the current scene feature parameter dataset using the method in S22 to obtain the current scene processed feature parameter dataset; input the current scene processed feature parameter dataset into the final use scene mapping model for mapping to obtain the first current use scene data.

[0029] The current driving power parameter dataset is then input into the final LED driving circuit parameter mapping model for mapping to obtain the second current usage scenario data;

[0030] S43. When the second current usage scenario data is different from the first current usage scenario data, the current drive power parameter dataset is adjusted until the second current usage scenario data is the same as the first current usage scenario data, and the current final drive power parameter dataset is obtained; otherwise, no adjustment is required.

[0031] Since people's judgment of the current usage scenario type is subjective, this solution collects characteristic parameter data of the current usage scenario instead of directly collecting the current usage scenario data. A constructed mapping model objectively maps the collected characteristic parameter data of the current usage scenario to the corresponding current usage scenario type data, ensuring data accuracy. By comparing the current usage scenario data obtained from these two mappings, it is determined whether the current LED driver is suitable for the current usage scenario. This determines whether adjustments to the current driver parameter data are needed to achieve a high degree of matching between the current LED driver and the current usage scenario.

[0032] Preferably, adjusting the current drive power supply parameter dataset in S43 includes the following steps:

[0033] S431. Set the value range of each current drive power parameter data in the current drive power parameter data set to obtain the current drive power parameter value range set;

[0034] S432. Adjust the current drive power supply parameter dataset according to the current drive power supply parameter value range set and using the Black-and-White tern optimization algorithm.

[0035] Preferably, a population of terns for adjusting driving power parameters is constructed; the maximum number of iterations for the population is set as b1 and the current number of iterations is set as b2, which are respectively denoted as the maximum number of iterations for power parameters and the current number of iterations for power parameters; the search space dimension of the population for adjusting driving power parameters is... same;

[0036] S432. Based on the current range of driving power parameters, generate the initial position of each tern in the tern population to adjust the driving power parameters, and obtain the initial position matrix.

[0037] S433. Construct the fitness function for adjusting the driving power supply parameters of the Black-crowned Tern population;

[0038] S434. Start the iteration. Before the iteration, set the current iteration count of the power supply parameter to 1. During the first iteration, use the fitness function of the driving power supply parameter to adjust the fitness of the tern population to calculate the fitness value of the initial position of each tern in the initial position matrix, and obtain the first fitness value set. Take the fitness of the largest fitness value set and the corresponding initial position of the tern as the first global best fitness and the first global best position, respectively. Update the initial position of each tern in the initial position matrix according to the first global best fitness and the first global best position. After the update is completed, increment the current iteration count of the power supply parameter by 1 and enter the next iteration.

[0039] In each iteration, the fitness function of the driving power parameters is used to adjust the population of terns. The fitness value of each tern in the population is calculated based on the position obtained in the previous iteration, resulting in a second fitness value set. The maximum fitness value and the corresponding tern position are taken as the second global optimal fitness and the second global optimal position, respectively. The position of each tern in the population is updated based on the second global optimal fitness and the second global optimal position. After the update, the current iteration number of the power parameters is incremented by 1, and the next iteration begins.

[0040] S435. When b2 ≥ b1, stop the iteration and obtain the final global optimal position and the final global optimal fitness; otherwise, continue the iteration until b2 ≥ b1; when the value of the final global optimal fitness is... If the final global optimal position is selected, it becomes the current final drive power parameter dataset; otherwise, it returns to S434 to continue iteration until the final global optimal fitness value is [value missing]. And it tends to stabilize; α1 is an integer representing the protection parameter;

[0041] For strongly coupled parameters such as circuit input / output voltage, LED series-parallel structure, and the number of Buck converter units, ISTOA achieves multi-objective balanced optimization through Pareto front solution set, reducing THD to below 2% while maintaining efficiency >90%. Its adaptive weighting mechanism can dynamically adjust the optimization priority of different parameters. The algorithm's built-in chaotic mapping strategy can effectively handle the nonlinear relationship between voltage and current in LED driving, improving the constraint satisfaction rate by 15% compared to traditional algorithms. By randomly generating discretization schemes, it directly outputs the optimal configuration of integer parameters such as the number of LEDs in series and the number of driving circuits, avoiding double rounding errors.

[0042] Preferably, step S5 includes the following steps:

[0043] S51. Adjust the topology of the LED driver power supply corresponding to the current usage scenario data according to the current final driver power supply parameter dataset.

[0044] An LED driver circuit topology and control system includes an LED driver circuit usage scenario setting module, an LED driver circuit scenario parameter setting module, a usage scenario mapping model construction module, an LED driver power supply parameter mapping model, a current driver power supply parameter adjustment module, and a current LED driver power supply topology adjustment module.

[0045] The LED driver circuit usage scenario setting module sets several usage scenarios for the LED driver circuit to obtain an LED driver circuit usage scenario set.

[0046] The LED driver circuit scenario parameter setting module sets several characteristic parameter types for LED driver circuit usage scenarios, thereby obtaining a set of characteristic parameter types for circuit usage scenarios.

[0047] The usage scenario mapping model construction module collects multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit based on the LED driver circuit usage scenario set, and then trains and tests the collected data to obtain the final usage scenario mapping model.

[0048] The LED driver power supply parameter mapping model collects various usage scenario data and corresponding LED driver power supply parameter data based on the circuit usage scenario characteristic parameter type set, and then trains and tests based on the collected data to obtain the final LED driver circuit parameter mapping model.

[0049] The current drive power parameter adjustment module maps the collected current scene feature parameter dataset and the current drive power parameter dataset according to the final use scenario mapping model and the final LED drive circuit parameter mapping model, respectively; then compares the mapping results of the two datasets; when the two mapping results are different, the current drive power parameter dataset is adjusted to obtain the current final drive power parameter dataset.

[0050] The current LED driver power supply topology adjustment module adjusts the topology of the LED driver power supply used in the current usage scenario based on the current final driver power supply parameter dataset.

[0051] The present invention has the following beneficial effects:

[0052] 1. In this invention, a model is used to dynamically map the characteristic parameters of the usage scenario to the parameters of the driving power supply. This model can automatically adapt to the needs of different application scenarios such as vehicle lighting and dimming control, so that the LED system can maintain a stable current output of ±10% even when the power supply voltage fluctuates by ±15%. By comparing the mapping results of the two models, it is confirmed from two perspectives whether the load characteristics of the current usage scenario match the current driving circuit. This determines whether it is necessary to adjust the parameter set to ensure that the topology always matches the current load characteristics, thereby improving system efficiency.

[0053] 2. In this invention, by constructing a final LED driver circuit parameter mapping model, a mapping tool is provided for the subsequent accurate matching between the application scenario and the corresponding LED driver circuit. This ensures that each LED driver circuit with different parameters can be matched to its most suitable application scenario, thereby enabling the LED driver circuit to play its maximum role.

[0054] 3. In this invention, a multi-objective balanced optimization is achieved using the U-Tern optimization algorithm for strongly coupled parameters such as circuit input / output voltage, LED series-parallel structure, and the number of Buck converter units. This reduces THD to below 2% while maintaining an efficiency greater than 90%. Its adaptive weighting mechanism can dynamically adjust the optimization priority of different parameters. The algorithm's built-in chaotic mapping strategy can effectively handle the nonlinear relationship between voltage and current in LED driving, improving the constraint satisfaction rate by 15% compared to traditional algorithms. By randomly generating discretization schemes, the optimal configuration of integer parameters such as the number of LEDs in series and the number of driving circuits is directly output.

[0055] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0056] To more clearly illustrate the technical solutions of the embodiments of the invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a flowchart illustrating an LED driving circuit topology and control method according to the present invention.

[0058] Figure 2 This is a flowchart illustrating the process of constructing the final application scenario mapping model and the final LED driver circuit parameter mapping model for this invention.

[0059] Figure 3 This is a schematic diagram illustrating the process of adjusting the current drive power supply parameter dataset according to the present invention;

[0060] Figure 4 This is a schematic diagram of an LED driving circuit topology and control system according to the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the invention, and not all embodiments. Based on the embodiments of the invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the invention.

[0062] Example 1

[0063] Please see Figure 1-3 This embodiment is an LED driver circuit topology and control method, including the following steps:

[0064] S1. Define several usage scenarios for the LED driver circuit to obtain a set of usage scenarios for the LED driver circuit;

[0065] S1 includes the following steps:

[0066] S11. Define several application scenarios for the LED driver circuit to obtain a set of application scenarios for the LED driver circuit. The set of application scenarios for the LED driver circuit includes mobile device backlighting, home / commercial lighting, industrial high-power lighting, outdoor advertising screens, vehicle lighting systems (such as headlights / taillights, ambient lighting, and automotive displays), and solar photovoltaic systems. Then define several characteristic parameter types for the application scenarios of the LED driver circuit to obtain a set of characteristic parameter types for the application scenarios of the circuit. The set of characteristic parameter types for the application scenarios of the circuit includes input voltage range (such as narrow voltage range (180V-260VAC) and wide voltage range (85V-265VAC)) and output characteristics (such as constant current output, constant voltage output, efficiency and power factor).

[0067] S12. Set several types of LED driving circuit scene parameters to obtain an LED driving circuit scene parameter type set; the LED driving circuit scene parameter type set includes circuit input voltage, circuit output voltage, number of LEDs in series, number of LED driving circuits, and number of Buck conversion units.

[0068] S2. Based on the set of LED driver circuit usage scenarios, collect multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit usage scenario, and train and test the final usage scenario mapping model.

[0069] S2 includes the following steps:

[0070] S21. Based on the set of LED driver circuit usage scenarios and the set of circuit usage scenario feature parameter types, collect multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit usage scenario to obtain a set of circuit usage scenario feature parameter data matrix a1={a 11 ,...,a 1i ,...,a 1a′}, a i This represents the circuit usage scenario feature parameter data matrix corresponding to the i-th type of LED driver circuit usage scenario, and a′ represents the total number of set LED driver circuit usage scenario types;

[0071] S22. Encode each type of LED driver circuit usage scenario data in the LED driver circuit usage scenario set into binary values ​​to obtain the encoded LED driver circuit usage scenario set; then normalize each feature parameter data in the circuit usage scenario feature parameter data matrix set to obtain the processed circuit usage scenario feature parameter data matrix set; for example, mobile device backlight, home / commercial lighting, industrial high-power lighting, outdoor advertising screen, and vehicle lighting system are represented by 000, 001, 010, 011, and 100 respectively.

[0072] S23. Construct an initial usage scenario mapping model; the number of input layer nodes of the initial usage scenario mapping model is equal to the feature parameter dimension. For example, if the circuit scenario contains 10 feature indicators such as voltage, frequency, and temperature, then the number of input layer nodes is 10; if the data contains time-series features (such as waveform sampling points), the number of input nodes corresponds to the time-series window length × the number of features, 3-5 hidden layers, and the number of nodes in each hidden layer is usually 1-2 times the feature dimension.

[0073] The hidden layer uses the ReLU function; the number of nodes in the output layer is the same as the number of set scene types; the initial usage scene mapping model is trained and tested using the processed circuit usage scene feature parameter data matrix set and the encoded LED driver circuit usage scene set; after training and testing, the final usage scene mapping model is obtained.

[0074] In S23, the initial use case mapping model is trained and tested. After training and testing are completed, the final use case mapping model is obtained, including the following steps:

[0075] S231. Set a first training data ratio; use the first training data ratio to divide the processed circuit usage scenario feature parameter data matrix set and the encoded LED driver circuit usage scenario set into data, to obtain the processed circuit usage scenario feature parameter training data matrix set, the encoded LED driver circuit usage scenario training set, the processed circuit usage scenario feature parameter test data matrix set, and the encoded LED driver circuit usage scenario test set; the first training data ratio can be 0.7 or 0.8;

[0076] S232. Set a first training error threshold; input the processed circuit usage scenario feature parameter training data matrix set as training data and the encoded LED driver circuit usage scenario training set as training labels into the initial usage scenario mapping model for training operation; during the training operation, when the training error is less than the first training error threshold, stop training and obtain the trained usage scenario mapping model; otherwise, continue training until the training error is less than the first training error threshold.

[0077] S233. Set a first test accuracy threshold; input the processed circuit usage scenario feature parameter test data matrix set as test data and the encoded LED driver circuit usage scenario test set as test labels into the trained usage scenario mapping model for testing; after the test is completed, obtain the first test accuracy data; when the first test accuracy data is greater than or equal to the first test accuracy threshold, use the trained usage scenario mapping model as the final usage scenario mapping model; otherwise, return to S232 to continue training until the first test accuracy data is greater than or equal to the first test accuracy threshold.

[0078] S3. Collect data from various usage scenarios and corresponding LED driver power supply parameter data to train and test the final LED driver circuit parameter mapping model.

[0079] S3 includes the following steps:

[0080] S31. Based on the set of LED driver circuit scenario parameter types and the set of LED driver circuit usage scenarios, collect parameter data of multiple existing usage scenarios and corresponding LED driver circuits to obtain an LED driver circuit parameter data matrix and an existing usage scenario dataset.

[0081] The existing use scenario data in the existing use scenario dataset is binary numerically encoded using the method in S22 to obtain the encoded existing use scenario dataset.

[0082] S32. Construct an initial LED driver circuit parameter mapping model; train and test the initial LED driver circuit parameter mapping model using the encoded existing application scenario dataset and the LED driver circuit parameter data matrix; after training and testing, obtain the final LED driver circuit parameter mapping model.

[0083] Preferably, in step S32, the initial LED driver circuit parameter mapping model is trained and tested; after training and testing are completed, the final LED driver circuit parameter mapping model is obtained by the following steps:

[0084] S321. Set a second training data ratio; use the second training data ratio to divide the encoded existing usage scenario dataset and the LED driver circuit parameter data matrix into data, respectively, to obtain the encoded existing usage scenario training dataset, the LED driver circuit parameter training data matrix, the encoded existing usage scenario test dataset, and the LED driver circuit parameter test data matrix.

[0085] S322. Set a second training error threshold; input the LED driver circuit parameter training data matrix as training data and the encoded existing usage scenario training dataset as training labels into the initial LED driver circuit parameter mapping model for training operation; during the training process, when the training error is less than the second training error threshold, stop training and obtain the trained LED driver circuit parameter mapping model; otherwise, continue training until the training error is less than the second training error threshold.

[0086] S323. Set a second test accuracy threshold; input the LED driver circuit parameter test data matrix as test data and the encoded existing usage scenario test dataset as test labels into the trained LED driver circuit parameter mapping model for testing; after the test is completed, obtain the second test accuracy data; when the second test accuracy data is greater than or equal to the second test accuracy threshold, use the trained LED driver circuit parameter mapping model as the final LED driver circuit parameter mapping model; otherwise, return to S322 to continue training until the second test accuracy data is greater than or equal to the second test accuracy threshold;

[0087] S4. Based on the final application scenario mapping model and the final LED driver circuit parameter mapping model, map the collected current scenario feature parameter dataset and the current driver power parameter dataset respectively; then compare the mapping results of the two datasets; when the two mapping results are different, adjust the current driver power parameter dataset to obtain the current final driver power parameter dataset.

[0088] S4 includes the following steps:

[0089] S41. Set the feature parameter data of the current usage scenario and the parameter dataset of the LED driver power supply corresponding to the current usage scenario data to obtain the current scenario feature parameter dataset and the current driver power supply parameter dataset.

[0090] S42. Normalize the feature parameter data in the current scene feature parameter dataset using the method in S22 to obtain the current scene processed feature parameter dataset; input the current scene processed feature parameter dataset into the final use scene mapping model for mapping to obtain the first current use scene data.

[0091] The current driving power parameter dataset is then input into the final LED driving circuit parameter mapping model for mapping to obtain the second current usage scenario data;

[0092] S43. When the second current usage scenario data is different from the first current usage scenario data, the current drive power parameter dataset is adjusted until the second current usage scenario data is the same as the first current usage scenario data, and the current final drive power parameter dataset is obtained; otherwise, no adjustment is required.

[0093] Adjusting the current drive power supply parameter dataset in S43 includes the following steps:

[0094] S431. Set the value range for each current drive power parameter in the current drive power parameter dataset to obtain the current drive power parameter value range set. as follows,

[0095]

[0096] in, These represent the lower limit and upper limit of the value of the i-th current drive power parameter in the current drive power parameter dataset, respectively. Indicates the total number of drive power parameter types set;

[0097] S432. Adjust the current drive power supply parameter dataset according to the current drive power supply parameter value range set and using the Black-and-White tern optimization algorithm.

[0098] Preferably, a population of terns for adjusting driving power parameters is constructed; the maximum number of iterations for the population is set as b1 and the current number of iterations is set as b2, which are respectively denoted as the maximum number of iterations for power parameters and the current number of iterations for power parameters; the search space dimension of the population for adjusting driving power parameters is... same;

[0099] S432. Based on the current set of driving power parameter values, generate the initial position of each tern in the tern population to adjust the driving power parameters, thus obtaining the initial position matrix; the generation formula is as follows.

[0100]

[0101] In the formula, b′ ji This represents the position component of the j-th tern in the ith current driving power parameter data dimension, indicating the initial position of the j-th tern in the tern population adjusted by the driving power parameters; rand ji Indicates that for b′ ji Generates random numbers between 0 and 1; ceil represents the floor function;

[0102] S433. Construct the fitness function for adjusting the driving power supply parameters of the Black-crowned Tern population. as follows,

[0103]

[0104] In the formula, These represent the first and second current usage scenario data corresponding to a set of drive power supply parameters obtained in each iteration, respectively. α1 is an integer representing the protection parameter used to avoid... The case where the denominator is 0 occurs;

[0105] S434. Begin iteration. Before iteration, set the current iteration number of the power supply parameter to 1. During the first iteration, adjust the fitness function of the Black-crowned Tern population using the driving power supply parameter. Calculate the fitness value of the initial position of each tern in the initial position matrix to obtain the first fitness value set; take the fitness value of the largest fitness value set and the corresponding initial position of the tern as the first global best fitness and the first global best position, respectively; update the initial position of each tern in the initial position matrix according to the first global best fitness and the first global best position; after the update is completed, increment the current iteration number of the power parameters by 1 and enter the next iteration.

[0106] In each subsequent iteration, the fitness function of the Black-crowned Tern population is adjusted using the aforementioned driving power parameters. Calculate the fitness value of each Black-eared Terns in the population adjusted by the driving power parameters updated in the previous iteration, resulting in a second fitness value set. The maximum fitness value and the corresponding Black-eared Tern position in the second fitness value set are respectively taken as the second global optimal fitness and the second global optimal position. Update the position of each Black-eared Terns in the population based on the driving power parameters updated in the previous iteration using the second global optimal fitness and the second global optimal position. After the update is completed, increment the current iteration number of the power parameters by 1 and proceed to the next iteration.

[0107] S435. When b2 ≥ b1, stop the iteration and obtain the final global optimal position and the final global optimal fitness; otherwise, continue the iteration until b2 ≥ b1; when the value of the final global optimal fitness is... If the final global optimal position is selected, it becomes the current final drive power parameter dataset; otherwise, it returns to S434 to continue iteration until the final global optimal fitness value is [value missing]. And until it tends to stabilize;

[0108] S5. Adjust the topology of the LED driver power supply used in the current application scenario based on the current final driver power supply parameter dataset;

[0109] S5 includes the following steps:

[0110] S51. Adjust the topology of the LED driver power supply corresponding to the current usage scenario data according to the current final driver power supply parameter dataset.

[0111] Example 2

[0112] Please see Figure 4 This embodiment discloses an LED driving circuit topology and control system, which can implement the method of the above embodiment, including...

[0113] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0114] The preferred embodiments of the invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. An LED driver circuit topology and control method, characterized in that, Includes the following steps: S1. Define several usage scenarios for the LED driver circuit to obtain a set of usage scenarios for the LED driver circuit; S2. Based on the set of LED driver circuit usage scenarios, collect multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit usage scenario, and train and test the final usage scenario mapping model. S3. Collect data from various usage scenarios and corresponding LED driver power supply parameter data to train and test the final LED driver circuit parameter mapping model. S4. Based on the final application scenario mapping model and the final LED driver circuit parameter mapping model, map the collected current scenario feature parameter dataset and the current driver power supply parameter dataset respectively; then compare the mapping results of the two datasets. When the two mapping results are different, the current drive power parameter dataset is adjusted to obtain the current final drive power parameter dataset. S4 specifically includes: S41, setting the feature parameter data of the current usage scenario and the parameter dataset of the LED driver power supply corresponding to the current usage scenario data, to obtain the current scenario feature parameter dataset and the current driver power supply parameter dataset; S42. Normalize the feature parameter data in the current scene feature parameter dataset to obtain the current scene processed feature parameter dataset; input the current scene processed feature parameter dataset into the final use scenario mapping model for mapping to obtain the first current use scenario data; then input the current driving power supply parameter dataset into the final LED driving circuit parameter mapping model for mapping to obtain the second current use scenario data. S43. When the second current usage scenario data is different from the first current usage scenario data, the current drive power parameter dataset is adjusted until the second current usage scenario data is the same as the first current usage scenario data, and the current final drive power parameter dataset is obtained; otherwise, no adjustment is required. S43 specifically includes: S431, setting the value range of each current drive power parameter data in the current drive power parameter data set to obtain the current drive power parameter value range set; S432, adjusting the current drive power parameter data set according to the current drive power parameter value range set and using the Black-and-White tern optimization algorithm; S5. Adjust the topology of the LED driver power supply used in the current usage scenario based on the current final driver power supply parameter dataset.

2. The LED driving circuit topology and control method according to claim 1, characterized in that, S1 includes the following steps: S11. Define several usage scenarios for the LED driver circuit to obtain a set of LED driver circuit usage scenarios; then define several characteristic parameter types for the LED driver circuit usage scenarios to obtain a set of circuit usage scenario characteristic parameter types. S12. Set several types of LED driver circuit scene parameters to obtain a set of LED driver circuit scene parameter types.

3. The LED driving circuit topology and control method according to claim 2, characterized in that, S2 includes the following steps: S21. Based on the set of LED driver circuit usage scenarios and the set of circuit usage scenario feature parameter types, collect multiple usage scenarios and corresponding feature parameter data for each type of LED driver circuit usage scenario to obtain a set of circuit usage scenario feature parameter data matrix. S22. Encode each type of LED driver circuit usage scenario data in the LED driver circuit usage scenario set into binary values ​​to obtain the encoded LED driver circuit usage scenario set; then normalize each feature parameter data in the circuit usage scenario feature parameter data matrix set to obtain the processed circuit usage scenario feature parameter data matrix set. S23. Construct an initial usage scenario mapping model; use the processed circuit usage scenario feature parameter data matrix set and the encoded LED driver circuit usage scenario set to train and test the initial usage scenario mapping model. After training and testing are completed, the final usage scenario mapping model is obtained.

4. The LED driving circuit topology and control method according to claim 3, characterized in that, S3 includes the following steps: S31. Based on the set of LED driver circuit scenario parameter types and the set of LED driver circuit usage scenarios, collect parameter data of multiple existing usage scenarios and corresponding LED driver circuits to obtain an LED driver circuit parameter data matrix and an existing usage scenario dataset. The existing use scenario data in the existing use scenario dataset is binary numerically encoded using the method in S22 to obtain the encoded existing use scenario dataset. S32. Construct an initial LED driver circuit parameter mapping model; train and test the initial LED driver circuit parameter mapping model using the encoded existing usage scenario dataset and the LED driver circuit parameter data matrix; after training and testing, obtain the final LED driver circuit parameter mapping model.

5. The LED driving circuit topology and control method according to claim 4, characterized in that: The initial usage scenario mapping model constructed in S23 and the initial LED driver circuit parameter mapping model constructed in S32 both adopt the multilayer perceptron model.

6. The LED driving circuit topology and control method according to claim 1, characterized in that, S432 includes the following steps: S4321. Construct a population of terns for adjusting drive power parameters; set the maximum number of iterations for the tern population for adjusting drive power parameters to be... b 1 and the current iteration number is b 2, denoted as the maximum number of iterations for the power supply parameters and the current number of iterations for the power supply parameters, respectively; S4322. Based on the current range of driving power parameters, generate the initial position of each tern in the tern population to adjust the driving power parameters, and obtain the initial position matrix. S4323. Construct the fitness function for adjusting the driving power supply parameters of the Black-crowned Tern population; S4324. Start the iteration. Before the iteration, set the current iteration count of the power supply parameter to 1. In each iteration, use the fitness function of the driving power supply parameter to adjust the tern population to calculate the fitness value of the position of each tern in the driving power supply parameter adjusted tern population updated in the previous iteration, and obtain the second fitness value set. Take the fitness of the largest fitness value set and the corresponding tern position as the second global best fitness and the second global best position, respectively. Update the position of each tern in the driving power supply parameter adjusted tern population updated in the previous iteration according to the second global best fitness and the second global best position. After the update is completed, increment the current iteration count of the power supply parameter by 1 and enter the next iteration. S4325, when If the condition is met, stop iterating to obtain the final global optimal position and the final global optimal fitness; otherwise, continue iterating until... Up to this point; when the final global optimal fitness value is... If the final global optimal position is obtained, it is used as the current final drive power parameter dataset; otherwise, the iteration continues in S4324 until the final global optimal fitness value is obtained. And until it tends to stabilize, An integer representing a protection parameter.

7. The LED driving circuit topology and control method according to claim 6, characterized in that, S5 includes the following steps: S51. Adjust the topology of the LED driver power supply corresponding to the current usage scenario data according to the current final driver power supply parameter dataset.

8. A system for implementing an LED driving circuit topology and control method as described in any one of claims 1-7.

Images