Temperature control method, device and equipment for multi-wavelength LED detection light source

Abstract

The present invention relates to the field of temperature control technology of light sources, and discloses a temperature control method, device and equipment for a multi-wavelength LED detection light source, wherein the method comprises: modeling each wavelength LED in the multi-wavelength LED detection light source to obtain a transient temperature rise balance model; performing noise reduction preprocessing and temperature coupling spectrum analysis to obtain a multi-wavelength temperature signal feature vector; performing combined power resampling and envelope spectrum analysis to obtain a multi-wavelength temperature anomaly feature library; constructing a temperature control strategy; performing zero temperature drift distribution calculation considering thermal coupling, establishing a multi-channel first-order inertia compensation model, and obtaining multi-channel PWM modulation parameters; collecting multi-wavelength real-time temperature data and matching calculation with the multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and collaboratively updating multi-channel PWM modulation parameters based on temperature balance. The method effectively solves the problems of temperature balance and spectral stability of the multi-wavelength LED detection light source during use.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature control of light sources, and in particular to a temperature control method, device and equipment for detecting a multi-wavelength LED light source. Background Art

[0002] As an important device in the field of precision measurement and detection, the luminous stability and spectral characteristics of multi-wavelength LED detection light sources directly affect the detection accuracy. During use, the temperature change of LED chips with different wavelengths will cause their spectra to drift, affecting the accuracy of the detection results. The current LED temperature control method is mainly designed for single-wavelength LEDs and cannot effectively deal with the temperature coupling effect and spectral interaction between multi-wavelength LEDs.

[0003] Traditional LED temperature control methods usually use simple PID control or fitting models based on steady-state characteristics. These methods have problems such as response lag and insufficient precision when dealing with dynamic changes in temperature of multi-wavelength LEDs. Especially in high-precision detection applications, the thermal coupling effect between multiple LED chips can cause unstable phenomena such as oscillation and overshoot in the temperature control system, seriously affecting the performance of the detection light source. In addition, existing temperature control strategies often ignore the impact of temperature changes on the spectral characteristics of LEDs, and lack in-depth analysis and modeling of spectral-temperature coupling characteristics. At the same time, in practical applications, due to the lack of an effective temperature anomaly diagnosis mechanism, it is difficult to promptly detect and deal with problems such as heat dissipation anomalies and chip overheating, which increases the risk of equipment failure and life loss. Summary of the invention

[0004] The present invention provides a temperature control method, device and equipment for a multi-wavelength LED detection light source, which effectively solves the problems of temperature balance and spectral stability of the multi-wavelength LED detection light source during use.

[0005] In a first aspect, the present invention provides a temperature control method for a multi-wavelength LED detection light source, the temperature control method for a multi-wavelength LED detection light source comprising:

[0006] The ambient temperature data, heat dissipation substrate temperature data and chip temperature data of each wavelength LED in the multi-wavelength LED detection light source are collected and the spectrum-temperature coupling characteristics are modeled to obtain a transient temperature rise balance model including spectrum-temperature characteristics;

[0007] Performing noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector;

[0008] Performing combined power resampling and envelope spectrum analysis on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library;

[0009] According to the multi-wavelength temperature anomaly feature library, a temperature control strategy including spectral stability and temperature balance is constructed;

[0010] According to the temperature control strategy, the zero temperature drift distribution of the PWM wave of each wavelength LED is calculated considering thermal coupling, a multi-channel first-order inertia compensation model is established, and multi-channel PWM modulation parameters are obtained;

[0011] The multi-channel PWM modulation parameters are respectively applied to the corresponding wavelength LED chips, and the multi-wavelength real-time temperature data is collected and matched with the multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and the multi-channel PWM modulation parameters are collaboratively updated based on temperature balance.

[0012] In a second aspect, the present invention provides a temperature control device for a multi-wavelength LED detection light source, the temperature control device for a multi-wavelength LED detection light source comprising:

[0013] The acquisition module is used to collect the ambient temperature data, heat dissipation substrate temperature data and chip temperature data of each wavelength LED in the multi-wavelength LED detection light source and model the spectrum-temperature coupling characteristics to obtain a transient temperature rise balance model including the spectrum-temperature characteristics;

[0014] A preprocessing module, used for performing noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector;

[0015] A resampling module, used for performing combined power resampling and envelope spectrum analysis on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library;

[0016] A construction module, used to construct a temperature control strategy including spectral stability and temperature balance according to the multi-wavelength temperature anomaly feature library;

[0017] A calculation module, for calculating the zero temperature drift distribution of the PWM wave of each wavelength LED considering thermal coupling according to the temperature control strategy, establishing a multi-channel first-order inertia compensation model, and obtaining multi-channel PWM modulation parameters;

[0018] The updating module is used to apply the multi-channel PWM modulation parameters to the corresponding wavelength LED chips respectively, collect multi-wavelength real-time temperature data and match and calculate with the multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and collaboratively update the multi-channel PWM modulation parameters based on temperature balance.

[0019] A third aspect of the present invention provides a temperature control device for a multi-wavelength LED detection light source, comprising: 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 temperature control device for the multi-wavelength LED detection light source executes the above-mentioned temperature control method for the multi-wavelength LED detection light source.

[0020] A fourth aspect of the present invention 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 temperature control method of the multi-wavelength LED detection light source described above.

[0021] In the technical solution provided by the present invention, a transient temperature rise balance model including spectral-temperature characteristics is established, and temperature-coupled spectrum analysis and multi-dimensional feature extraction technology are combined to achieve an accurate description of the dynamic temperature characteristics of multi-wavelength LEDs. A temperature control strategy based on PSO-GWO hybrid optimization and a PWM wave zero temperature drift distribution technology considering thermal coupling are adopted to significantly improve the accuracy and stability of temperature control. A complete temperature anomaly diagnosis and compensation mechanism is established through real-time monitoring of multi-channel temperature data and abnormal feature matching, which effectively solves the temperature balance and spectral stability problems of multi-wavelength LED detection light sources during use, and provides reliable technical support for high-precision optical detection.

[0022] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of an embodiment of a temperature control method of a multi-wavelength LED detection light source in an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of an embodiment of a temperature control device for a multi-wavelength LED detection light source in an embodiment of the present invention;

[0026] Figure 3 It is a schematic diagram of an embodiment of a temperature control device for a multi-wavelength LED detection light source in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. 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 terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0029] To facilitate understanding of this embodiment, a temperature control method for a multi-wavelength LED detection light source disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps:

[0030] 101. Collect the ambient temperature data, heat dissipation substrate temperature data and chip temperature data of each wavelength LED in the multi-wavelength LED detection light source and model the spectrum-temperature coupling characteristics to obtain a transient temperature rise balance model including the spectrum-temperature characteristics;

[0031] It is understandable that the execution subject of the present invention may be a temperature control device of a multi-wavelength LED detection light source, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0032] Specifically, in the multi-wavelength LED detection light source, the ambient temperature, heat dissipation substrate temperature and chip temperature of each wavelength LED are sampled synchronously at multiple points to obtain a high-resolution temperature original sampling data matrix. The temperature original sampling data matrix is ​​subjected to wavelet transform filtering to remove the noise therein, significantly improve the quality and reliability of the data, and obtain the denoised temperature data matrix. The corresponding radial temperature rise rate RITRR value and temperature coupling coefficient of each wavelength LED are calculated to obtain the temperature dynamic characteristic parameters, which reflect the thermal response characteristics of the LED under different environmental conditions. According to the temperature dynamic characteristic parameters, the transfer function from the ambient temperature to the heat dissipation substrate temperature and the transfer function from the heat dissipation substrate temperature to the chip temperature are established considering the thermal coupling effect. These two transfer functions together constitute a multi-level temperature transfer coupling model, which describes the temperature transfer process between different levels and takes into account the influence of thermal coupling. The multi-level temperature transfer coupling model is subjected to parameter identification calculation, and the temperature transfer coefficient, thermal resistance coefficient and thermal coupling coefficient are extracted to help the system understand the contribution of different factors in the temperature transfer process, and these parameters are substituted into the multi-wavelength LED thermodynamic equilibrium equation to construct a transient temperature rise basic model including the temperature response time constant. Multi-band spectrum tests are performed on LEDs of various wavelengths to obtain spectrum drift and interaction data under different temperature combinations. On this basis, a temperature-spectrum coupling mapping relationship is established to obtain a spectrum-temperature response model that reflects how temperature changes affect the spectrum characteristics of LEDs. The constructed transient temperature rise basic model and the spectrum-temperature response model are subjected to multi-dimensional data fusion operations to construct a temperature and spectrum coupling mathematical model for multi-wavelength LEDs. The model includes spectrum-temperature characteristics and monitors and adjusts the working status of LEDs in real time in practical applications, thereby achieving efficient temperature control of multi-wavelength LED detection light sources.

[0033] 102. Perform noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector;

[0034] Specifically, wavelet multi-resolution analysis is performed on the multi-channel temperature sensor signals output by the transient temperature rise balance model, and the components of the temperature signals are extracted from different scales to reveal their details and characteristics. By analyzing the energy distribution characteristics of the temperature signal components, the noise components are identified, and adaptive threshold processing is applied to obtain the multi-wavelength temperature noise reduction threshold parameters. The multi-wavelength temperature noise reduction threshold parameters are substituted into the wavelet inverse transform formula to perform synchronous reconstruction of the multi-channel signals to obtain the filtered temperature sensor signal, which more realistically reflects the temperature changes of each wavelength LED. Multi-dimensional empirical mode decomposition is performed on the filtered temperature sensor signal to obtain a series of intrinsic mode components (IMFs) to effectively characterize the temperature coupling characteristics. Based on the frequency characteristics and phase characteristics of the extracted intrinsic mode components IMFs, the mutual correlation coefficients and phase delays between the temperature signals of each wavelength LED are calculated to reveal the temperature coupling relationship between different wavelength LEDs, and then a multi-dimensional temperature coupling feature matrix is ​​constructed. The singular value decomposition operation is performed on the feature matrix to extract the main characteristic direction of temperature coupling and obtain the multi-wavelength temperature feature space. Based on the multi-wavelength temperature feature space, the analytical basis function of the multi-dimensional Hilbert-Huang transform is constructed. Using these basis functions, the instantaneous frequency spectrum considering temperature coupling is calculated to obtain a multi-dimensional time-frequency characteristic distribution map. The multi-dimensional time-frequency characteristic distribution map is subjected to principal component extraction and feature fusion to extract the most important feature information and obtain a multi-wavelength temperature signal feature vector.

[0035] 103. Perform combined power resampling and envelope spectrum analysis on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library;

[0036] Specifically, the driving power of each wavelength LED is calculated by combining the ratios of temperature coupling, and a multi-dimensional power combination matrix is ​​generated, which can reflect the driving power distribution of different wavelength LEDs under specific temperature conditions. In this way, the relationship between the temperature signal characteristics under different driving powers is accurately captured. According to the constructed multi-dimensional power combination matrix, the multi-wavelength temperature signal feature vector is subjected to a multi-dimensional power domain mapping transformation. Through this mapping transformation, the power-temperature coupling correspondence is obtained, and the mutual influence mechanism between temperature and power is provided. The power-temperature coupling correspondence is substituted into the cubic spline interpolation function considering temperature coupling to perform multi-dimensional resampling point calculation, and an equally spaced multi-dimensional power sampling sequence is obtained, so that the data for subsequent analysis is more uniform and easy to process. The equally spaced multi-dimensional power sampling sequence is subjected to a multi-dimensional Hilbert transform, and the coupled instantaneous envelope of the multi-wavelength temperature signal is extracted. The instantaneous envelope provides the amplitude information of the temperature signal, which can intuitively reflect the trend of temperature change over time. Based on this, the variance, peak coefficient and coupling coefficient of temperature fluctuation under different power combinations are calculated to obtain multidimensional temperature anomaly evaluation indicators, which effectively evaluate the stability and abnormality of temperature signals and help the system monitor the performance of LEDs under different operating conditions. According to the multidimensional temperature anomaly evaluation indicators, a temperature anomaly classification criterion considering heat dissipation anomalies and chip overheating is established. This classification criterion not only considers the performance of LEDs of different wavelengths under different temperature coupling conditions, but also effectively distinguishes between normal working conditions and potential overheating or heat dissipation anomalies. The temperature anomaly classification criterion is optimized by the maximum entropy principle to ensure that the classification model retains information as much as possible while dealing with uncertainty and complexity, improves the accuracy and reliability of classification, and obtains a multidimensional anomaly feature classification model. The multidimensional anomaly feature classification model is feature mapped with the formed multidimensional power combination matrix to obtain a multi-wavelength temperature anomaly feature library.

[0037] 104. Based on the multi-wavelength temperature anomaly feature library, a temperature control strategy including spectral stability and temperature balance is constructed;

[0038] Specifically, the data in the multi-wavelength temperature anomaly feature library are normalized considering temperature coupling to eliminate the dimensional influence between different features, so that each feature can be compared and optimized on the same scale. After normalization, a multi-objective optimization space is constructed to form a multi-dimensional feature optimization matrix, which can reflect the performance of temperature anomaly features in terms of spectral stability and temperature balance. Based on the multi-dimensional feature optimization matrix, a multi-objective function including temperature control accuracy, spectral stability and temperature balance is established. This function provides quantitative indicators for subsequent optimization, so that effective trade-offs and selections can be made between different objectives. After determining the multi-objective function, the multi-dimensional optimization weight coefficient is extracted from it to indicate the importance of each objective, helping the system to better allocate resources and attention during the optimization process. The multi-dimensional optimization weight coefficient is mapped to a cosine function considering temperature coupling. This mapping can convert the weight coefficient into a more adaptable form and construct a multi-dimensional convergence factor that changes dynamically with the number of iterations. The adaptive convergence factor can effectively improve the convergence speed and accuracy of the optimization algorithm, making the final optimization result more reliable. On this basis, based on the multi-dimensional adaptive convergence parameters, the optimal positions of the multi-wavelength LED individuals and the optimal positions of the group in the particle swarm optimization (PSO) algorithm are collaboratively updated and calculated. This collaborative update process obtains the multi-dimensional PSO search trajectory, which provides direction guidance for subsequent optimization. After completing the calculation of the PSO search trajectory, these trajectories are input into the prey encirclement model of the gray wolf optimization (GWO) algorithm considering temperature coupling. Through this model, the position update of the α, β, and δ wolves of the multi-wavelength LED temperature control is calculated to obtain the multi-dimensional GWO search direction. According to the multi-dimensional GWO search direction and multi-dimensional adaptive convergence parameters, a hybrid iterative update strategy considering temperature coupling is constructed. Combining the advantages of PSO and GWO, and ensuring that the influence of temperature coupling is fully considered in the search process, a better multi-dimensional PSO-GWO hybrid optimization parameter is obtained. The multi-dimensional PSO-GWO hybrid optimization parameters are balanced by the local search and global search considering temperature equilibrium. Through this calculation, the temperature control parameters in different search spaces are effectively explored, and the multi-wavelength temperature control parameter set is obtained. Substitute the multi-wavelength temperature control parameter set into the temperature-spectrum coupling mapping model to perform multi-wavelength stability verification calculations. Through this verification process, the effectiveness and stability of the constructed temperature control strategy in practical applications are evaluated. The resulting temperature control strategy will include spectral stability and temperature balance, thereby ensuring that the multi-wavelength LED detection light source can maintain stable output performance and good temperature management effects under various working conditions.

[0039] 105. According to the temperature control strategy, the zero temperature drift distribution of the PWM wave of each wavelength LED is calculated considering thermal coupling, a multi-channel first-order inertia compensation model is established, and multi-channel PWM modulation parameters are obtained;

[0040] Specifically, the multi-wavelength temperature control parameters in the temperature control strategy are decomposed considering temperature coupling, and the target temperature control amount of each wavelength LED and its thermal coupling influence amount are extracted from them to obtain the multi-channel temperature control reference value. Based on the multi-channel temperature control reference value, a multi-dimensional PWM reference modulation matrix of the PWM duty cycle of each wavelength LED considering the thermal coupling effect is constructed. The multi-dimensional PWM reference modulation matrix is ​​subjected to frequency domain analysis to extract the main frequency components of multi-channel temperature fluctuations and their interactive effects. Through this analysis, a multi-dimensional temperature drift characteristic spectrum is obtained, which helps to understand the law and characteristics of temperature fluctuations. Based on the multi-dimensional temperature drift characteristic spectrum, a zero temperature drift PWM wave reordering rule considering thermal coupling is constructed. This reordering rule will optimize the distribution of PWM waves, making it more suitable for the thermal coupling effect of each wavelength LED, and forming a multi-dimensional PWM wave distribution optimization model. The multi-dimensional PWM wave distribution optimization model is substituted into the multi-channel first-order inertia transfer function considering temperature coupling, and the temperature response characteristics of the multi-wavelength LED are calculated to obtain a multi-dimensional dynamic temperature response curve. In order to ensure the precise control of the system, the multi-dimensional dynamic temperature response curve is subjected to pure delay compensation and phase correction of thermal coupling to ensure that the PWM signal maintains good synchronization with the target temperature control amount when it is output, thereby constructing the first-order inertia compensation parameters of the multi-channel PWM signal. Based on the first-order inertia compensation parameters, the PWM waves of each wavelength LED are subjected to phase synchronization and waveform reconstruction considering temperature coupling. Through this waveform reconstruction, the compensated multi-dimensional PWM wave sequence is obtained. In order to improve the effect of temperature control, the compensated multi-dimensional PWM wave sequence is subjected to multi-channel collaborative optimization operation considering temperature balance, and multi-channel PWM modulation parameters considering thermal coupling effects are constructed to provide precise PWM waveform modulation for actual temperature control, ensuring that each wavelength LED can maintain a stable temperature and good spectral output under different working conditions.

[0041] 106. Apply multiple PWM modulation parameters to corresponding wavelength LED chips respectively, collect multi-wavelength real-time temperature data and perform matching calculation with a multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and collaboratively update multi-channel PWM modulation parameters based on temperature balance.

[0042] Specifically, the multi-channel PWM modulation parameters are synchronously applied to each wavelength LED chip through a multi-channel driving circuit, and multi-dimensional real-time temperature data is collected to form a temperature real-time monitoring sequence considering temperature coupling. The temperature real-time monitoring sequence is subjected to wavelet filtering and normalization operations to eliminate noise and nonlinear effects, and a multi-dimensional temperature feature monitoring matrix is ​​obtained. The multi-dimensional temperature feature monitoring matrix and the multi-wavelength temperature anomaly feature library are subjected to multi-dimensional distance measurement calculation. By comparing the real-time monitoring data with the abnormal data in the feature library, the multi-dimensional temperature feature matching value is obtained. This matching value helps to evaluate the deviation between the current temperature state and the normal working conditions, and then extracts the deviation between the real-time temperature of each wavelength LED and its target temperature and its interactive influence, forming a multi-dimensional temperature deviation vector, which can effectively reflect the temperature performance of each wavelength LED under the current operating conditions. The multi-dimensional temperature deviation vector is subjected to multi-dimensional Kalman filtering considering temperature coupling. Kalman filtering is an effective state estimation method that can dynamically adjust the estimated value of temperature deviation according to the dynamic model of the system and the measurement data. After this processing, multi-channel temperature deviation compensation values ​​are obtained, and these compensation values ​​are directly used to correct the previously calculated PWM modulation parameters. Based on the temperature deviation compensation value, a recursive correction calculation considering temperature balance is performed to obtain the multi-dimensional PWM compensation update amount. This step ensures that when adjusting the PWM waveform, the temperature balance of each wavelength LED can be considered at the same time to prevent spectral drift caused by temperature differences. According to the multi-dimensional PWM compensation update amount, a multi-wavelength LED temperature collaborative control optimization function considering temperature coupling is constructed. Through this optimization function, the temperature deviation of each wavelength LED and its corresponding PWM modulation parameters are integrated to form an overall temperature control strategy to achieve collaborative adjustment of LED light sources. The obtained multi-dimensional PWM collaborative adjustment strategy will be converted into specific PWM wave frequency and duty cycle adjustment instructions, which will be applied to the update of multi-channel PWM modulation parameters.

[0043] In the embodiment of the present invention, by establishing a transient temperature rise balance model including spectral-temperature characteristics, combined with temperature coupling spectrum analysis and multi-dimensional feature extraction technology, an accurate description of the dynamic temperature characteristics of multi-wavelength LEDs is achieved; a temperature control strategy based on PSO-GWO hybrid optimization and a PWM wave zero temperature drift distribution technology considering thermal coupling are adopted to significantly improve the accuracy and stability of temperature control; through real-time monitoring of multi-channel temperature data and abnormal feature matching, a complete temperature anomaly diagnosis and compensation mechanism is established, which effectively solves the temperature balance and spectral stability problems of multi-wavelength LED detection light sources during use, and provides reliable technical support for high-precision optical detection.

[0044] In a specific embodiment, the process of executing step 101 may specifically include the following steps:

[0045] (1) Perform multi-point synchronous sampling of the ambient temperature, heat dissipation substrate temperature, and chip temperature of each wavelength LED in the multi-wavelength LED detection light source to obtain the temperature original sampling data matrix;

[0046] (2) Perform wavelet transform filtering on the original temperature sampling data matrix to obtain a denoised temperature data matrix. Based on the denoised temperature data matrix, calculate the corresponding radial temperature rise rate RITRR value and temperature coupling coefficient of each wavelength LED to obtain the temperature dynamic characteristic parameters.

[0047] (3) Based on the temperature dynamic characteristic parameters, the transfer function from the ambient temperature to the heat sink temperature and the transfer function from the heat sink temperature to the chip temperature are established considering the thermal coupling effect, and a multi-level temperature transfer coupling model is obtained;

[0048] (4) Perform parameter identification calculation on the multi-stage temperature transfer coupling model, extract the temperature transfer coefficient, thermal resistance coefficient and thermal coupling coefficient, obtain the temperature transfer characteristic parameters, and substitute the temperature transfer characteristic parameters into the multi-wavelength LED thermodynamic equilibrium equation to construct a transient temperature rise basic model including the temperature response time constant;

[0049] (5) Multi-band spectral tests are performed on LEDs of various wavelengths to obtain spectral drift and interaction data under different temperature combinations, and a temperature-spectrum coupling mapping relationship is established to obtain a spectral temperature response model. The transient temperature rise basic model and the spectral temperature response model are then subjected to multi-dimensional data fusion operations to construct a mathematical model of temperature and spectrum coupling for multi-wavelength LEDs, thereby obtaining a transient temperature rise balance model that includes spectral-temperature characteristics.

[0050] Specifically, the ambient temperature, heat dissipation substrate temperature and chip temperature of each wavelength LED are sampled at multiple points synchronously to obtain the temperature original sampling data matrix, which is expressed as ,in is the ambient temperature, is the heat sink substrate temperature, is the LED chip temperature. This matrix contains the temperature information of all LEDs under different conditions. The original temperature sampling data matrix is ​​processed by wavelet transform filtering to remove noise and obtain the denoised temperature data matrix , expressed as ,in is a wavelet transform function. Through wavelet transform, the temperature signal is decomposed and high-frequency noise components are removed to improve the quality of the data. Based on the denoised temperature data matrix, the corresponding radial heating rate of each wavelength LED is calculated. and temperature coupling coefficient The radial heating rate is defined as the rate of temperature change per unit time and is expressed as ,in It is The temperature coupling coefficient reflects the mutual influence between temperatures and is expressed as Based on the calculated temperature dynamic characteristic parameters, a transfer function from ambient temperature to heat dissipation substrate temperature is established considering the thermal coupling effect. and the transfer function from the heat sink temperature to the chip temperature . These transfer functions are expressed as first-order inertial models, for example:

[0051] ;

[0052] ;

[0053] in and are the transfer coefficients, and is the time constant. Through these transfer functions, a multi-stage temperature transfer coupling model is obtained, which can describe the dynamic response characteristics from ambient temperature to chip temperature. The multi-stage temperature transfer coupling model is subjected to parameter identification calculation, and the temperature transfer coefficient, thermal resistance coefficient and thermal coupling coefficient are extracted, which are expressed as , and The extraction of these parameters depends on experimental data fitting and optimization algorithms. After extracting these temperature transfer characteristic parameters, they are substituted into the multi-wavelength LED thermodynamic equilibrium equation to construct a transient temperature rise basic model including the temperature response time constant, which is expressed as:

[0054] ;

[0055] This equation describes the relationship between the transient change of chip temperature and the temperature of the environment and heat sink. Multi-band spectrum tests are performed on LEDs of various wavelengths to obtain spectral drift and interaction data under different temperature combinations. A coupled mapping relationship between temperature and spectrum is established to obtain a spectral temperature response model. For example, the following relationship is constructed using spectral measurement results:

[0056] ;

[0057] in, is the wavelength The spectral intensity under is the spectral intensity at the reference temperature, is the change in spectral drift. The transient temperature rise basic model and the spectral temperature response model are fused with multi-dimensional data to construct a temperature and spectrum coupling mathematical model for multi-wavelength LEDs, and a transient temperature rise balance model including spectral-temperature characteristics is obtained, which is formalized as:

[0058] ;

[0059] The model integrates the effect of temperature changes on spectral output, allowing the system to accurately control the output characteristics of the LED. In this way, the system can not only monitor and adjust the temperature in real time, but also maintain the stability of the spectral output to ensure the best performance under various working conditions.

[0060] In a specific embodiment, the process of executing step 102 may specifically include the following steps:

[0061] (1) Wavelet multi-resolution analysis is performed on the multi-channel temperature sensor signals output by the transient temperature rise balance model to obtain temperature signal components of different scales. Based on the energy distribution characteristics of the temperature signal components, adaptive threshold processing is performed on the noise components to obtain the multi-wavelength temperature noise reduction threshold parameters.

[0062] (2) Substitute the multi-wavelength temperature noise reduction threshold parameter into the wavelet inverse transform formula to perform multi-channel signal synchronous reconstruction to obtain the filtered temperature sensor signal, and perform multi-dimensional empirical mode decomposition on the filtered temperature sensor signal to obtain multiple intrinsic mode components (IMFs) that characterize the temperature coupling characteristics;

[0063] (3) According to the frequency characteristics and phase characteristics of the intrinsic mode component IMF, the mutual correlation coefficient and phase delay between the temperature signals of the LEDs at each wavelength are calculated to obtain a multi-dimensional temperature coupling characteristic matrix. The multi-dimensional temperature coupling characteristic matrix is ​​subjected to singular value decomposition to extract the main characteristic direction of temperature coupling and obtain a multi-wavelength temperature feature space.

[0064] (4) Based on the multi-wavelength temperature feature space, the analytical basis functions of the multi-dimensional Hilbert-Huang transform are constructed, and the instantaneous frequency spectrum considering temperature coupling is calculated to obtain a multi-dimensional time-frequency feature distribution map. The principal components of the multi-dimensional time-frequency feature distribution map are extracted and features are fused to obtain the multi-wavelength temperature signal feature vector.

[0065] Specifically, wavelet multi-resolution analysis is performed on the multi-channel temperature sensor signals output by the transient temperature rise balance model, and wavelet transform is performed on these signals to extract temperature signal components of different scales. This is achieved by using discrete wavelet transform, which is expressed as:

[0066] ;

[0067] in It is The transform coefficients of the layer, is the input signal, is the wavelet basis function. Through transformation, signal components in different frequency ranges are obtained, allowing the system to analyze the characteristics of temperature signals at various scales. The energy distribution characteristics of the obtained temperature signal components are realized by calculating the energy of each component, which is defined as:

[0068] ;

[0069] On this basis, the noise component is processed by adaptive threshold. The adaptive threshold is calculated in the following form:

[0070] ;

[0071] in It is The threshold of the layer, is a constant, usually taken as 1.5. It is The standard deviation of the layer. Use this threshold to identify and remove noise, and obtain the multi-wavelength temperature noise reduction threshold parameter. Substitute the multi-wavelength temperature noise reduction threshold parameter into the wavelet inverse transform formula to perform multi-channel signal synchronous reconstruction, which is expressed as:

[0072] ;

[0073] After inverse transformation, the filtered temperature sensor signal is obtained, which is recorded as The filtered signal is subjected to multidimensional empirical mode decomposition to obtain multiple intrinsic mode components (IMFs), which are expressed as ,in Represents different modal components. For each natural modal component , analyze its frequency characteristics and phase characteristics. The frequency characteristics are realized by fast Fourier transform and expressed as:

[0074] ;

[0075] The phase characteristic is obtained by calculating the phase angle:

[0076] ;

[0077] By calculating the mutual correlation coefficient and phase delay between the temperature signals of LEDs of different wavelengths, a multi-dimensional temperature coupling characteristic matrix is ​​obtained. , the matrix is ​​defined as:

[0078] ;

[0079] Where Cov is the covariance, is the standard deviation. This matrix can characterize the correlation of temperature changes between different LEDs. Perform singular value decomposition on the multidimensional temperature coupling feature matrix to extract the main characteristic direction of temperature coupling. The singular value decomposition is expressed as:

[0080] ;

[0081] in is the left singular vector matrix, is a diagonal matrix, is the right singular vector matrix. The principal eigendirections are determined by choosing Through these steps, a multi-wavelength temperature feature space is constructed, and the analysis basis function of the multi-dimensional Hilbert-Huang transform is constructed using this space. This process involves constructing the instantaneous frequency spectrum of the analysis basis function, which is achieved through the following formula:

[0082] ;

[0083] The instantaneous frequency corresponding to each IMF is obtained. These instantaneous frequencies are combined together to form a multi-dimensional time-frequency feature distribution map. The multi-dimensional time-frequency feature distribution map is subjected to principal component extraction and feature fusion to obtain the multi-wavelength temperature signal feature vector, which is expressed as:

[0084] ;

[0085] in is the weight coefficient, is the number of intrinsic modes. Through the fusion process, the final feature vector Provides comprehensive characterization of multi-wavelength LED temperature signals.

[0086] In a specific embodiment, the process of executing step 103 may specifically include the following steps:

[0087] (1) The driving power of each wavelength LED is calculated by combining the ratios of the driving power considering the temperature coupling to obtain a multi-dimensional power combination matrix. Based on the multi-dimensional power combination matrix, the multi-wavelength temperature signal feature vector is transformed into a multi-dimensional power domain to obtain the power-temperature coupling correspondence.

[0088] (2) Substituting the power-temperature coupling correspondence into the cubic spline interpolation function considering temperature coupling, multidimensional resampling point calculation is performed to obtain an equally spaced multidimensional power sampling sequence, and a multidimensional Hilbert transform is performed on the equally spaced multidimensional power sampling sequence to obtain the coupled instantaneous envelope of the multi-wavelength temperature signal;

[0089] (3) Based on the coupled instantaneous envelope of multi-wavelength temperature signals, the temperature fluctuation variance, peak coefficient and coupling coefficient under different power combinations are calculated to obtain the multi-dimensional temperature anomaly evaluation index. Based on the multi-dimensional temperature anomaly evaluation index, the temperature anomaly classification criteria for heat dissipation anomaly and chip overheating considering temperature coupling are established;

[0090] (4) The temperature anomaly classification criterion is optimized by the maximum entropy principle considering temperature coupling to obtain a multi-dimensional anomaly feature classification model. The multi-dimensional anomaly feature classification model is then feature mapped with the multi-dimensional power combination matrix to obtain a multi-wavelength temperature anomaly feature library.

[0091] Specifically, the temperature coupling influence model is defined, taking into account the temperature influence between LEDs of different wavelengths. By constructing a power combination matrix, it is expressed as:

[0092] ;

[0093] in, Representative wavelength The LED at a given temperature and And power The driving power under is a function that represents the relationship between power and temperature. Through this combination, a multi-dimensional power combination matrix is ​​obtained , whose elements reflect the power configuration of LEDs with different wavelengths under different temperature combinations. According to the multi-dimensional power combination matrix, the multi-wavelength temperature signal feature vector is transformed into a multi-dimensional power domain mapping. The feature vector is defined as , by combining the power combination matrix with the eigenvector, the power-temperature coupling correspondence is obtained

[0094] ;

[0095] in, Represents the temperature signal characteristics corresponding to each power combination. This mapping helps the system understand the impact of different power configurations on LED temperature, thereby more effectively controlling temperature. Substitute the power-temperature coupling correspondence into the cubic spline interpolation function that considers temperature coupling to perform multi-dimensional resampling point calculations. The spline interpolation function is expressed as:

[0096] ;

[0097] in are the coefficients determined by the spline interpolation algorithm, is the horizontal coordinate of the known point. By interpolation, the power sampling points under specific temperature conditions are calculated to form a multi-dimensional power sampling sequence with equal intervals. . Perform multidimensional Hilbert transform on the equally spaced multidimensional power sampling sequence to extract its instantaneous envelope. The mathematical form of Hilbert transform is:

[0098] ;

[0099] By calculating, the instantaneous envelope of the signal is obtained , this envelope can represent the coupling characteristics of LED temperature and help the system further analyze temperature fluctuations. On this basis, based on the coupled instantaneous envelope of the multi-wavelength temperature signal, the temperature fluctuation variance, peak coefficient and coupling coefficient under different power combinations are calculated to establish a multi-dimensional temperature anomaly evaluation index. Through the above calculations, the multi-dimensional temperature anomaly evaluation index is obtained, and then the temperature anomaly classification criteria for heat dissipation abnormalities and chip overheating are established. For example, if the temperature fluctuation variance exceeds the preset threshold, or the peak coefficient exceeds a specific range, it can be determined as a temperature anomaly. For the temperature anomaly classification criteria, the maximum entropy principle considering temperature coupling is applied for optimization. This process aims to improve the accuracy of the classification model. The maximum entropy principle is to select the probability distribution with the largest entropy under all constraints, which is defined as:

[0100] ;

[0101] in is the probability of each abnormal type in the classification model. By maximizing entropy, a multidimensional abnormal feature classification model is obtained, which can effectively distinguish normal state from abnormal state. The multidimensional abnormal feature classification model is mapped with the multidimensional power combination matrix to construct a multi-wavelength temperature anomaly feature library. :

[0102] ;

[0103] in It is an abnormal feature classification model, which contains specific classification information of temperature anomalies. This feature library provides an important basis for subsequent temperature monitoring and control strategies, ensuring the stability and reliability of LEDs under different working conditions.

[0104] In a specific embodiment, the process of executing step 104 may specifically include the following steps:

[0105] (1) The data in the multi-wavelength temperature anomaly feature library are normalized and multi-objective optimization space is constructed considering temperature coupling to obtain a multi-dimensional feature optimization matrix. Based on the multi-dimensional feature optimization matrix, a multi-objective function including temperature control accuracy, spectral stability and temperature balance is established to obtain the multi-dimensional optimization weight coefficient;

[0106] (2) The multidimensional optimization weight coefficient is mapped to a cosine function that takes temperature coupling into consideration, and a multidimensional convergence factor that changes dynamically with the number of iterations is constructed to obtain a multidimensional adaptive convergence parameter. Based on the multidimensional adaptive convergence parameter, the optimal position of the multi-wavelength LED individual and the optimal position of the group of the PSO algorithm are collaboratively updated to obtain a multidimensional PSO search trajectory.

[0107] (3) The multidimensional PSO search trajectory is input into the prey encirclement model of the GWO algorithm considering temperature coupling, and the position update of the α, β, and δ wolves controlled by the multi-wavelength LED temperature is calculated to obtain the multidimensional GWO search direction. Based on the multidimensional GWO search direction and the multidimensional adaptive convergence parameters, a hybrid iterative update strategy considering temperature coupling is constructed to obtain the multidimensional PSO-GWO hybrid optimization parameters.

[0108] (4) A balanced calculation of local search and global search considering temperature balance is performed on the multi-dimensional PSO-GWO hybrid optimization parameters to obtain a multi-wavelength temperature control parameter set. The multi-wavelength temperature control parameter set is substituted into the temperature-spectrum coupling mapping model to perform multi-wavelength stability verification calculations to obtain a temperature control strategy that includes spectral stability and temperature balance.

[0109] Specifically, the data in the multi-wavelength temperature anomaly feature library are normalized considering temperature coupling and a multi-objective optimization space is constructed. The goal of normalization is to standardize the feature data for better subsequent analysis. After normalization, all feature values ​​will be compressed to the range of [0, 1] to avoid the dominant influence of certain features on the optimization process. A multi-objective optimization space is constructed, and a multi-objective function including temperature control accuracy, spectral stability, and temperature balance is defined. It is expressed as:

[0110] ;

[0111] in, is the objective function, is the weight coefficient of each target, is a function related to the temperature control accuracy, is a function related to spectral stability, is a function related to temperature balance. By analyzing the feature optimization matrix, these weight coefficients are obtained using methods such as non-dominated sorting and Pareto optimization to ensure the balance between different objectives. The multi-dimensional optimization weight coefficients are mapped to a cosine function that takes temperature coupling into account to achieve a dynamically changing convergence factor. A common form of the cosine function is:

[0112] ;

[0113] in, is the dynamic convergence factor, and is the parameter of the adjustment factor, is the iteration cycle, is the current iteration number. This mapping enables the convergence factor to change dynamically in each iteration, helping the optimization process to converge faster. Based on the multi-dimensional adaptive convergence parameter, it is applied to the particle swarm optimization (PSO) algorithm. The core of PSO is to update the optimal position of the individual and the optimal position of the group , and its update formula is as follows:

[0114] ;

[0115] ;

[0116] in, It is a particle speed, is the position of the particle, and is the acceleration constant, and is a random number, is the inertia weight. On this basis, the multidimensional adaptive convergence parameter is introduced into the particle update process to obtain the multidimensional PSO search trajectory. The multidimensional PSO search trajectory is input into the gray wolf optimization (GWO) algorithm considering temperature coupling. The prey encirclement model of the GWO algorithm updates the position by simulating the behavior of the wolf pack. The specific update formula is as follows:

[0117] ;

[0118] in, is the location of the prey, It's a pack of wolves that hunt. They represent the best, second-best, and third-best wolves in the wolf pack, respectively, and are responsible for updating different weights of the positions. Combined with these update quantities, the multidimensional GWO search direction is calculated. According to the multidimensional GWO search direction and multidimensional adaptive convergence parameters, a hybrid iterative update strategy considering temperature coupling is constructed. This strategy combines the PSO and GWO algorithms to effectively balance the capabilities of local search and global search to obtain multidimensional PSO-GWO hybrid optimization parameters. The update form of the optimization parameters is:

[0119] ;

[0120] in, is the updated parameter, is the current parameter, and is a coefficient that is dynamically adjusted during the update process. A balanced calculation of local search and global search is performed on the multi-dimensional PSO-GWO hybrid optimization parameters to obtain the final multi-wavelength temperature control parameter set. This process uses technologies such as adaptive fuzzy control to achieve local search to reduce the impact of temperature imbalance. By substituting the final multi-wavelength temperature control parameter set into the temperature-spectrum coupling mapping model, a multi-wavelength stability verification calculation is performed to ensure the spectral stability and temperature balance of the system under different conditions, and a temperature control strategy that includes spectral stability and temperature balance is obtained.

[0121] In a specific embodiment, the process of executing step 105 may specifically include the following steps:

[0122] (1) The multi-wavelength temperature control parameters in the temperature control strategy are decomposed considering temperature coupling, the target temperature control amount of each wavelength LED and its thermal coupling influence amount are extracted, and the multi-channel temperature control reference value is obtained. Based on the multi-channel temperature control reference value, a multi-dimensional PWM reference modulation matrix of the PWM duty cycle of each wavelength LED considering the thermal coupling effect is constructed;

[0123] (2) Perform frequency domain analysis on the multidimensional PWM reference modulation matrix considering temperature coupling, extract the main frequency components of multi-channel temperature fluctuations and their interactive effects, and obtain the multidimensional temperature drift characteristic spectrum. Based on the multidimensional temperature drift characteristic spectrum, a zero temperature drift PWM wave reordering rule considering thermal coupling is constructed to obtain a multidimensional PWM wave distribution optimization model.

[0124] (3) Substitute the multi-dimensional PWM wave distribution optimization model into the multi-channel first-order inertia transfer function considering temperature coupling, calculate the temperature response characteristics of the multi-wavelength LED, obtain the multi-dimensional dynamic temperature response curve, and perform pure delay compensation and phase correction considering thermal coupling on the multi-dimensional dynamic temperature response curve to construct the first-order inertia compensation parameters of the multi-channel PWM signal;

[0125] (4) Based on the first-order inertia compensation parameters, the phase synchronization and waveform reconstruction of the PWM waves of each wavelength LED are performed considering temperature coupling to obtain a compensated multi-dimensional PWM wave sequence. The compensated multi-dimensional PWM wave sequence is then subjected to multi-path collaborative optimization calculation considering temperature balance to construct multi-path PWM modulation parameters considering thermal coupling effects.

[0126] Specifically, the multi-wavelength temperature control parameters in the temperature control strategy are decomposed considering temperature coupling, and the target temperature control amount of each wavelength LED and its thermal coupling influence amount are identified and extracted. In this process, the target temperature control amount is expressed as , and the thermal coupling influence is expressed as , the temperature control reference value is described by the following formula:

[0127] ;

[0128] in, is the obtained multi-channel temperature control reference value. The calculation of this reference value is based on an in-depth analysis of the temperature coupling effect, which usually involves data collection and processing of multiple sensors to ensure that the temperature changes when LEDs with different wavelengths are working can be accurately captured. Based on these reference values, a multi-dimensional PWM reference modulation matrix of the PWM (pulse width modulation) duty cycle of each wavelength LED considering the thermal coupling effect is constructed. , expressed as:

[0129] ;

[0130] in, It is The wavelength LED is The duty cycle in a PWM cycle, is the number of LEDs, is the number of PWM cycles. This matrix provides dynamic adjustment capability for the temperature control of each LED. The multi-dimensional PWM reference modulation matrix is ​​analyzed in the frequency domain to extract the main frequency components of multi-channel temperature fluctuations and their interactive effects. The frequency spectrum representation of the temperature signal is obtained using fast Fourier transform:

[0131] ;

[0132] in, is the spectrum representation, is a time domain signal, is the frequency, is the duration of the signal. Through spectrum analysis, the main frequency components that affect temperature fluctuations are identified, and the multi-dimensional temperature drift characteristic spectrum is obtained. Based on the extracted spectrum, a zero temperature drift PWM wave reordering rule considering thermal coupling is constructed. This rule is used to optimize the distribution of PWM signals to minimize the impact of temperature changes on spectral output. It is expressed as

[0133] ;

[0134] in, is the reordering rule for each PWM duty cycle, is the variance of temperature fluctuation, It is The application of this rule makes the optimization model The following relationship is formed:

[0135] ;

[0136] Substitute the multi-dimensional PWM wave distribution optimization model into the multi-channel first-order inertia transfer function considering temperature coupling to calculate the temperature response characteristics of the multi-wavelength LED. The form of the multi-channel first-order inertia transfer function is expressed as:

[0137] ;

[0138] in, is the gain, is the time constant, is a complex frequency domain variable. Through this model, a multi-dimensional dynamic temperature response curve is obtained , and use this curve for pure delay compensation and phase correction. The compensation formula for dynamic temperature response is expressed as:

[0139] ;

[0140] in, is the delay time. The first-order inertia compensation parameters of multiple PWM signals are constructed through the corrected dynamic temperature response. , in order to achieve more accurate temperature control. Based on the first-order inertia compensation parameters obtained, the PWM waves of each wavelength LED are phase synchronized and waveform reconstructed. Phase synchronization is achieved using phase compensation technology, and waveform reconstruction is performed through interpolation to ensure that the temperature remains stable in different cycles of the PWM signal. The compensated multi-dimensional PWM wave sequence is expressed as:

[0141] ;

[0142] Multi-channel collaborative optimization operation considering temperature balance. Substitute the compensated multi-dimensional PWM wave sequence into the construction process of multi-channel PWM modulation parameters to ensure that the operating temperature of all wavelength LEDs is balanced under corresponding conditions. The optimization model is expressed as:

[0143] ;

[0144] in, It is The actual temperature of the LED, is the desired temperature. The final multi-channel PWM modulation parameters will take into account the specific needs of each LED to achieve an optimized effect.

[0145] In a specific embodiment, the process of executing step 106 may specifically include the following steps:

[0146] (1) The multi-channel PWM modulation parameters are synchronously applied to the LED chips of each wavelength through the multi-channel driving circuit, and the multi-dimensional real-time temperature data is collected to obtain the real-time temperature monitoring sequence considering the temperature coupling. The real-time temperature monitoring sequence is subjected to wavelet filtering and normalization operation considering the thermal coupling effect to obtain the multi-dimensional temperature feature monitoring matrix;

[0147] (2) The multidimensional temperature feature monitoring matrix and the multi-wavelength temperature anomaly feature library are subjected to multi-dimensional distance measurement calculation considering temperature coupling to obtain the multi-dimensional temperature feature matching value. Based on the multi-dimensional temperature feature matching value, the deviation between the real-time temperature of each wavelength LED and the target temperature and their interaction influence are extracted to obtain the multi-dimensional temperature deviation vector.

[0148] (3) Performing multi-dimensional Kalman filtering processing on the multi-dimensional temperature deviation vector taking into account temperature coupling to obtain multi-channel temperature deviation compensation values, and based on the multi-channel temperature deviation compensation values, performing recursive correction calculation on the multi-channel PWM modulation parameters taking into account temperature balance to obtain the multi-dimensional PWM compensation update amount;

[0149] (4) According to the multi-dimensional PWM compensation update amount, a multi-wavelength LED temperature collaborative control optimization function considering temperature coupling is constructed to obtain a multi-dimensional PWM collaborative adjustment strategy. The multi-dimensional PWM collaborative adjustment strategy is then converted into frequency and duty cycle adjustment instructions of the PWM wave considering temperature balance to obtain updated multi-channel PWM modulation parameters.

[0150] Specifically, the multiple PWM modulation parameters are synchronously applied to each wavelength LED chip through a multiple drive circuit. Transmitted to each LED driving circuit, expressed as:

[0151] ;

[0152] in, It is The PWM modulation parameters received by the LED is the driving gain coefficient of the corresponding LED. After acting on each wavelength LED synchronously, multi-dimensional real-time temperature data is collected, and the obtained temperature real-time monitoring sequence is expressed as:

[0153] ;

[0154] in, Indicates Real-time temperature data of LEDs, is the number of LEDs. In this process, the temperature coupling effect is taken into account to more accurately monitor and control the temperature of each wavelength LED. Through wavelet transform, the real-time monitoring sequence is subjected to wavelet filtering and normalization, and the multi-dimensional temperature feature monitoring matrix is ​​obtained. It is expressed as:

[0155] ;

[0156] in, represents the filtering operation of wavelet transform, Is a normalization matrix, which is used to standardize temperature data to eliminate the influence of noise. With pre-built multi-wavelength temperature anomaly signature library Perform multi-dimensional distance measurement calculation. Using the Euclidean distance measurement method, the calculation method is as follows:

[0157] ;

[0158] Through this calculation, the multi-dimensional temperature characteristic matching value is obtained It can be used to evaluate the deviation between the actual monitored temperature and the abnormal characteristics. Based on this matching value, the real-time temperature of each wavelength LED and the target temperature are extracted. The deviation and its interaction influence constitute a multidimensional temperature deviation vector

[0159] ;

[0160] For multidimensional temperature deviation vector Perform multidimensional Kalman filtering considering temperature coupling. The state update formula of Kalman filtering is:

[0161] ;

[0162] in, is the current state estimate, is the Kalman gain, is the observed value, is the observation matrix. After Kalman filtering, the multi-channel temperature deviation compensation value is obtained It is expressed as:

[0163] ;

[0164] Based on temperature deviation compensation value , the multi-channel PWM modulation parameters are recursively corrected considering the temperature balance, and the multi-dimensional PWM compensation update amount is obtained It is expressed as:

[0165] ;

[0166] in, is the adjustment coefficient used to control the compensation amplitude. Through this process, the PWM signal of each LED is effectively adjusted to ensure that the temperature fluctuates within the set range. , construct a multi-wavelength LED temperature coordinated control optimization function considering temperature coupling , the optimization function is expressed as:

[0167] ;

[0168] in, is the weight coefficient, which adjusts the priority of temperature control. By solving this optimization function, we can get the multi-dimensional PWM coordinated regulation strategy , whose expression is:

[0169] ;

[0170] The multi-dimensional PWM coordinated regulation strategy is converted into the frequency and duty cycle regulation instructions of the PWM wave considering temperature balance to obtain the updated multi-channel PWM modulation parameters. , which is of the form:

[0171] ;

[0172] For example, in a high-power LED lighting system, through this process, the temperature of each wavelength LED is monitored and adjusted in real time to keep it within the optimal operating range, effectively preventing damage caused by overheating.

[0173] The temperature control method of the multi-wavelength LED detection light source in the embodiment of the present invention is described above. The temperature control device of the multi-wavelength LED detection light source in the embodiment of the present invention is described below. Figure 2 , an embodiment of the temperature control device of the multi-wavelength LED detection light source in the embodiment of the present invention includes:

[0174] The acquisition module 201 is used to collect the ambient temperature data, heat dissipation substrate temperature data and chip temperature data of each wavelength LED in the multi-wavelength LED detection light source and model the spectrum-temperature coupling characteristics to obtain a transient temperature rise balance model including the spectrum-temperature characteristics;

[0175] A preprocessing module 202 is used to perform noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector;

[0176] The resampling module 203 is used to perform combined power resampling and envelope spectrum analysis on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library;

[0177] A construction module 204 is used to construct a temperature control strategy including spectral stability and temperature balance according to a multi-wavelength temperature anomaly feature library;

[0178] The calculation module 205 is used to calculate the zero temperature drift distribution of the PWM wave of each wavelength LED considering thermal coupling according to the temperature control strategy, establish a multi-channel first-order inertia compensation model, and obtain multi-channel PWM modulation parameters;

[0179] The updating module 206 is used to apply the multi-channel PWM modulation parameters to the corresponding wavelength LED chips respectively, collect the multi-wavelength real-time temperature data and match and calculate the multi-wavelength temperature anomaly feature library, obtain the multi-channel temperature deviation compensation value, and collaboratively update the multi-channel PWM modulation parameters based on temperature balance.

[0180] Through the collaborative cooperation of the above-mentioned components, by establishing a transient temperature rise balance model including spectral-temperature characteristics, combined with temperature-coupled spectrum analysis and multi-dimensional feature extraction technology, an accurate description of the dynamic temperature characteristics of multi-wavelength LEDs is achieved; a temperature control strategy based on PSO-GWO hybrid optimization and PWM wave zero temperature drift distribution technology considering thermal coupling are adopted to significantly improve the accuracy and stability of temperature control; through real-time monitoring of multi-channel temperature data and abnormal feature matching, a complete temperature anomaly diagnosis and compensation mechanism is established, which effectively solves the temperature balance and spectral stability problems of multi-wavelength LED detection light sources during use, providing reliable technical support for high-precision optical detection.

[0181] above Figure 2 The temperature control device of the multi-wavelength LED detection light source in the embodiment of the present invention is described in detail from the perspective of modular functional entity. The temperature control device of the multi-wavelength LED detection light source in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0182] Figure 33 is a schematic diagram of the structure of a temperature control device for a multi-wavelength LED detection light source provided by an embodiment of the present invention. The temperature control device 300 for the multi-wavelength LED detection light source may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 can be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the temperature control device 300 for the multi-wavelength LED detection light source. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the temperature control device 300 for the multi-wavelength LED detection light source to implement the steps of the temperature control method for the multi-wavelength LED detection light source.

[0183] The temperature control device 300 of the multi-wavelength LED detection light source may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It can be understood by those skilled in the art that Figure 3 The temperature control device structure of the multi-wavelength LED detection light source shown does not constitute a limitation on the temperature control device of the multi-wavelength LED detection light source provided by the present invention, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0184] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the temperature control method of the multi-wavelength LED detection light source.

[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0186] 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 invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0187] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. 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 invention.

Claims

1. A temperature control method for a multi-wavelength LED detection light source, characterized in that: The method comprises: The ambient temperature data, heat dissipation substrate temperature data and chip temperature data of each wavelength LED in the multi-wavelength LED detection light source are collected and the spectrum-temperature coupling characteristics are modeled to obtain a transient temperature rise balance model containing spectrum-temperature characteristics; specifically, the model includes: multi-point synchronous sampling of the ambient temperature, heat dissipation substrate temperature and chip temperature of each wavelength LED in the multi-wavelength LED detection light source to obtain a temperature original sampling data matrix; wavelet transform filtering is performed on the original temperature sampling data matrix to obtain a denoised temperature data matrix, and based on the denoised temperature data matrix, the corresponding radial temperature rise rate RITRR value and temperature coupling coefficient of each wavelength LED are calculated to obtain temperature dynamic characteristic parameters; according to the temperature dynamic characteristic parameters, a transfer function from ambient temperature to heat dissipation substrate temperature considering the thermal coupling effect is established. The temperature transfer coefficient and the transfer function from the heat dissipation substrate temperature to the chip temperature are used to obtain a multi-stage temperature transfer coupling model; parameter identification calculation is performed on the multi-stage temperature transfer coupling model, and the temperature transfer coefficient, thermal resistance coefficient and thermal coupling coefficient are extracted to obtain temperature transfer characteristic parameters, and the temperature transfer characteristic parameters are substituted into the multi-wavelength LED thermodynamic equilibrium equation to construct a transient temperature rise basic model including a temperature response time constant; multi-band spectrum tests are performed on each wavelength LED to obtain spectrum drift and interaction influence data under different temperature combinations, and a temperature-spectrum coupling mapping relationship is established to obtain a spectrum temperature response model, and a multi-dimensional data fusion operation is performed on the transient temperature rise basic model and the spectrum temperature response model to construct a temperature and spectrum coupling mathematical model for multi-wavelength LEDs, and a transient temperature rise balance model including spectrum-temperature characteristics is obtained; Performing noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector; Performing combined power resampling and envelope spectrum analysis on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library; According to the multi-wavelength temperature anomaly feature library, a temperature control strategy including spectral stability and temperature balance is constructed; According to the temperature control strategy, the zero temperature drift distribution of the PWM wave of each wavelength LED is calculated considering thermal coupling, a multi-channel first-order inertia compensation model is established, and multi-channel PWM modulation parameters are obtained; The multi-channel PWM modulation parameters are respectively applied to the corresponding wavelength LED chips, and the multi-wavelength real-time temperature data is collected and matched with the multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and the multi-channel PWM modulation parameters are collaboratively updated based on temperature balance.

2. The temperature control method of the multi-wavelength LED detection light source according to claim 1, characterized in that: The method of performing noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector includes: Performing wavelet multi-resolution analysis on the multi-channel temperature sensor signals output by the transient temperature rise balance model to obtain temperature signal components of different scales, and performing adaptive threshold processing on the noise components based on the energy distribution characteristics of the temperature signal components to obtain multi-wavelength temperature noise reduction threshold parameters; Substituting the multi-wavelength temperature noise reduction threshold parameter into the wavelet inverse transform formula, performing multi-channel signal synchronous reconstruction to obtain a filtered temperature sensor signal, and performing multi-dimensional empirical mode decomposition on the filtered temperature sensor signal to obtain a plurality of intrinsic mode components IMFs characterizing temperature coupling characteristics; According to the frequency characteristics and phase characteristics of the intrinsic mode component IMF, the mutual correlation coefficient and phase delay between the temperature signals of the LEDs of each wavelength are calculated to obtain a multi-dimensional temperature coupling characteristic matrix, and a singular value decomposition operation is performed on the multi-dimensional temperature coupling characteristic matrix to extract the main characteristic direction of the temperature coupling to obtain a multi-wavelength temperature characteristic space; Based on the multi-wavelength temperature feature space, an analysis basis function of a multi-dimensional Hilbert-Huang transform is constructed, and an instantaneous frequency spectrum considering temperature coupling is calculated to obtain a multi-dimensional time-frequency feature distribution diagram. The multi-dimensional time-frequency feature distribution diagram is subjected to principal component extraction and feature fusion to obtain a multi-wavelength temperature signal feature vector.

3. The temperature control method of the multi-wavelength LED detection light source according to claim 2, characterized in that: The combining power resampling and envelope spectrum analysis are performed on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library, including: The driving power of each wavelength LED is calculated by combining and matching the driving power with temperature coupling in consideration to obtain a multi-dimensional power combination matrix, and the multi-wavelength temperature signal feature vector is transformed into a multi-dimensional power domain according to the multi-dimensional power combination matrix to obtain a power-temperature coupling correspondence relationship; Substituting the power-temperature coupling correspondence into a cubic spline interpolation function considering temperature coupling, performing multidimensional resampling point calculation to obtain an equally spaced multidimensional power sampling sequence, and performing a multidimensional Hilbert transform on the equally spaced multidimensional power sampling sequence to obtain a coupled instantaneous envelope of a multi-wavelength temperature signal; Based on the coupled instantaneous envelope of the multi-wavelength temperature signal, the temperature fluctuation variance, peak coefficient and coupling coefficient under different power combinations are calculated to obtain a multi-dimensional temperature anomaly evaluation index, and based on the multi-dimensional temperature anomaly evaluation index, a temperature anomaly classification criterion for heat dissipation anomaly and chip overheating considering temperature coupling is established; The temperature anomaly classification criterion is optimized by the maximum entropy principle considering temperature coupling to obtain a multi-dimensional anomaly feature classification model, and the multi-dimensional anomaly feature classification model is feature mapped with the multi-dimensional power combination matrix to obtain a multi-wavelength temperature anomaly feature library.

4. The temperature control method of the multi-wavelength LED detection light source according to claim 3, characterized in that: The method of constructing a temperature control strategy including spectral stability and temperature balance according to the multi-wavelength temperature anomaly feature library comprises: The data in the multi-wavelength temperature anomaly feature library are normalized and multi-objective optimization space is constructed considering temperature coupling to obtain a multi-dimensional feature optimization matrix, and a multi-objective function including temperature control accuracy, spectral stability and temperature balance is established according to the multi-dimensional feature optimization matrix to obtain a multi-dimensional optimization weight coefficient; The multidimensional optimization weight coefficient is mapped to a cosine function taking into account temperature coupling, a multidimensional convergence factor that changes dynamically with the number of iterations is constructed, a multidimensional adaptive convergence parameter is obtained, and based on the multidimensional adaptive convergence parameter, the optimal position of the multi-wavelength LED individual and the optimal position of the group of the PSO algorithm are collaboratively updated and calculated to obtain a multidimensional PSO search trajectory; The multidimensional PSO search trajectory is input into the prey encirclement model of the GWO algorithm considering temperature coupling, the position update amount of the α, β, and δ wolves controlled by the multi-wavelength LED temperature is calculated, and the multidimensional GWO search direction is obtained. According to the multidimensional GWO search direction and the multidimensional adaptive convergence parameter, a hybrid iterative update strategy considering temperature coupling is constructed to obtain the multidimensional PSO-GWO hybrid optimization parameters; A balance calculation of local search and global search considering temperature balance is performed on the multi-dimensional PSO-GWO hybrid optimization parameters to obtain a multi-wavelength temperature control parameter set, and the multi-wavelength temperature control parameter set is substituted into the spectral temperature response model to perform a multi-wavelength stability verification calculation to obtain a temperature control strategy that includes spectral stability and temperature balance.

5. The temperature control method of the multi-wavelength LED detection light source according to claim 4, characterized in that: According to the temperature control strategy, the zero temperature drift distribution of the PWM wave of each wavelength LED is calculated considering thermal coupling, a multi-channel first-order inertia compensation model is established, and multi-channel PWM modulation parameters are obtained, including: Decomposition operation is performed on the multi-wavelength temperature control parameters in the temperature control strategy considering temperature coupling, the target temperature control amount of each wavelength LED and its thermal coupling influence amount are extracted, and a multi-channel temperature control reference value is obtained. Based on the multi-channel temperature control reference value, a multi-dimensional PWM reference modulation matrix of the PWM duty ratio of each wavelength LED considering the thermal coupling effect is constructed; Performing frequency domain analysis on the multi-dimensional PWM reference modulation matrix considering temperature coupling, extracting the main frequency components of multi-channel temperature fluctuations and their interactive effects, obtaining a multi-dimensional temperature drift characteristic spectrum, and constructing a zero temperature drift PWM wave reordering rule considering thermal coupling based on the multi-dimensional temperature drift characteristic spectrum, to obtain a multi-dimensional PWM wave distribution optimization model; Substituting the multi-dimensional PWM wave distribution optimization model into the multi-channel first-order inertia transfer function considering temperature coupling, calculating the temperature response characteristics of the multi-wavelength LED, obtaining a multi-dimensional dynamic temperature response curve, and performing pure delay compensation and phase correction considering thermal coupling on the multi-dimensional dynamic temperature response curve to construct the first-order inertia compensation parameters of the multi-channel PWM signal; Based on the first-order inertia compensation parameters, the PWM waves of the LEDs of each wavelength are subjected to phase synchronization and waveform reconstruction considering temperature coupling to obtain a compensated multi-dimensional PWM wave sequence, and the compensated multi-dimensional PWM wave sequence is subjected to multi-path collaborative optimization operation considering temperature balance to construct multi-path PWM modulation parameters considering thermal coupling effects.

6. The temperature control method of the multi-wavelength LED detection light source according to claim 5, characterized in that: The method of applying the multi-channel PWM modulation parameters to the corresponding wavelength LED chips respectively, collecting multi-wavelength real-time temperature data and matching and calculating with the multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and collaboratively updating the multi-channel PWM modulation parameters based on temperature balance includes: The multi-channel PWM modulation parameters are synchronously applied to the LED chips of each wavelength through a multi-channel driving circuit, and multi-dimensional real-time temperature data is collected to obtain a temperature real-time monitoring sequence considering temperature coupling, and the temperature real-time monitoring sequence is subjected to wavelet filtering and normalization operation considering thermal coupling effect to obtain a multi-dimensional temperature feature monitoring matrix; The multidimensional temperature feature monitoring matrix and the multi-wavelength temperature anomaly feature library are subjected to a multidimensional distance measurement calculation considering temperature coupling to obtain a multidimensional temperature feature matching value, and based on the multidimensional temperature feature matching value, the deviation between the real-time temperature of each wavelength LED and the target temperature and the amount of their interaction influence are extracted to obtain a multidimensional temperature deviation vector; Performing multidimensional Kalman filtering processing considering temperature coupling on the multidimensional temperature deviation vector to obtain multi-channel temperature deviation compensation values, and based on the multi-channel temperature deviation compensation values, performing recursive correction calculation considering temperature balance on the multi-channel PWM modulation parameters to obtain a multi-dimensional PWM compensation update amount; According to the multi-dimensional PWM compensation update amount, a multi-wavelength LED temperature collaborative control optimization function considering temperature coupling is constructed to obtain a multi-dimensional PWM collaborative adjustment strategy, and the multi-dimensional PWM collaborative adjustment strategy is converted into frequency and duty cycle adjustment instructions of the PWM wave considering temperature balance to obtain updated multi-channel PWM modulation parameters.

7. A temperature control device for a multi-wavelength LED detection light source, characterized in that: The device is used to perform the temperature control method of the multi-wavelength LED detection light source according to any one of claims 1 to 6, comprising: The acquisition module is used to collect the ambient temperature data, heat dissipation substrate temperature data and chip temperature data of each wavelength LED in the multi-wavelength LED detection light source and model the spectrum-temperature coupling characteristics to obtain a transient temperature rise balance model containing spectrum-temperature characteristics; specifically comprising: performing multi-point synchronous sampling on the ambient temperature, heat dissipation substrate temperature and chip temperature of each wavelength LED in the multi-wavelength LED detection light source to obtain a temperature original sampling data matrix; performing wavelet transform filtering on the temperature original sampling data matrix to obtain a denoised temperature data matrix, and based on the denoised temperature data matrix, calculating the corresponding radial temperature rise rate RITRR value and temperature coupling coefficient of each wavelength LED to obtain temperature dynamic characteristic parameters; according to the temperature dynamic characteristic parameters, establishing a temperature-to-heat dissipation substrate temperature relationship considering the thermal coupling effect; The temperature transfer function and the transfer function from the heat dissipation substrate temperature to the chip temperature are used to obtain a multi-stage temperature transfer coupling model; parameter identification calculation is performed on the multi-stage temperature transfer coupling model, and the temperature transfer coefficient, thermal resistance coefficient and thermal coupling coefficient are extracted to obtain temperature transfer characteristic parameters, and the temperature transfer characteristic parameters are substituted into the multi-wavelength LED thermodynamic equilibrium equation to construct a transient temperature rise basic model including a temperature response time constant; multi-band spectrum tests are performed on each wavelength LED to obtain spectrum drift and interaction influence data under different temperature combinations, and a temperature-spectrum coupling mapping relationship is established to obtain a spectrum temperature response model, and a multi-dimensional data fusion operation is performed on the transient temperature rise basic model and the spectrum temperature response model to construct a temperature and spectrum coupling mathematical model of the multi-wavelength LED to obtain a transient temperature rise balance model including spectrum-temperature characteristics; A preprocessing module, used for performing noise reduction preprocessing and temperature coupling spectrum analysis on the LED temperature sensor signals of each wavelength in the transient temperature rise balance model to obtain a multi-wavelength temperature signal feature vector; A resampling module, used for performing combined power resampling and envelope spectrum analysis on the multi-wavelength temperature signal feature vector to obtain a multi-wavelength temperature anomaly feature library; A construction module, used to construct a temperature control strategy including spectral stability and temperature balance according to the multi-wavelength temperature anomaly feature library; A calculation module, for calculating the zero temperature drift distribution of the PWM wave of each wavelength LED considering thermal coupling according to the temperature control strategy, establishing a multi-channel first-order inertia compensation model, and obtaining multi-channel PWM modulation parameters; The updating module is used to apply the multi-channel PWM modulation parameters to the corresponding wavelength LED chips respectively, collect multi-wavelength real-time temperature data and match and calculate with the multi-wavelength temperature anomaly feature library to obtain multi-channel temperature deviation compensation values, and collaboratively update the multi-channel PWM modulation parameters based on temperature balance.

8. A temperature control device for a multi-wavelength LED detection light source, characterized in that: The temperature control device of the multi-wavelength LED detection light source comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instruction in the memory to enable the temperature control device of the multi-wavelength LED detection light source to execute the temperature control method of the multi-wavelength LED detection light source according to 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 temperature control method of the multi-wavelength LED detection light source according to any one of claims 1 to 6 is implemented.

Images