Semiconductor laser temperature and current dual control method and system

Abstract

The application discloses a kind of semiconductor laser temperature and current double control method and system, it is related to temperature and current double control technical field.The semiconductor laser temperature and current double control method, by in setting sliding period, obtain the device temperature time series data of setting semiconductor laser, and combining pre-trained temperature dynamic regulation and control model, analyze the temperature regulation and control characteristic value of setting semiconductor laser;Obtain current temperature effect time series data, and combining temperature regulation and control characteristic value, analyze initial current regulation and control characteristic value;Obtain current effect time series data, and combining initial current regulation and control characteristic value, analyze current comprehensive regulation and control characteristic value, the application is based on temperature regulation and control characteristic value, current comprehensive regulation and control characteristic value carries out double collaborative control processing to setting semiconductor laser, to ensure that laser can maintain stable output under complex dynamic change environment, to realize the efficient stable operation of semiconductor laser.

Description

Technical Field

[0001] This invention relates to the field of temperature and current dual control technology, specifically to a method and system for temperature and current dual control of a semiconductor laser. Background Technology

[0002] Semiconductor lasers, as a new type of light source device, have advantages such as small size, high efficiency, long lifespan, and flexible control. They have been widely used in fields such as optical communication, optical storage, lidar, medical detection, and industrial processing. In these applications, the output performance of semiconductor lasers is highly dependent on the stability of their operating current and operating temperature.

[0003] Typically, semiconductor lasers use constant current drive circuits as their power supply. Constant current drive can provide a stable operating current within a certain range, thereby ensuring that the device emits light normally above the threshold current. However, since semiconductor materials themselves have obvious temperature sensitivity, changes in device temperature will cause the threshold voltage and threshold current to drift, which in turn will affect the stability of the output current. At the same time, fluctuations in the operating current will cause the device's own temperature to rise, creating a coupling effect between temperature and current.

[0004] Existing technology, such as the semiconductor temperature control device and its control method disclosed in patent application CN119806246B, includes: a temperature detection module for multi-point detection of the temperature field of a semiconductor laser to obtain multiple temperature measurements, thereby performing thermal field simulation calculations to obtain a temperature field distribution matrix; a heat flux analysis module for calculating the heat flux density of the semiconductor laser based on the temperature field distribution matrix, thereby obtaining a temperature regulation drive signal by performing multi-channel pulse width modulation processing on a thermoelectric cooling component; and a closed-loop control module for adaptively adjusting the thermoelectric cooling component based on the temperature regulation drive signal to generate a temperature regulation feedback signal, thereby performing closed-loop adjustment of the drive current of the thermoelectric cooling component until the thermoelectric cooling component maintains the semiconductor laser temperature within the target temperature range based on the adjusted drive current. This device ensures dynamic balance in temperature control.

[0005] Based on the above findings, the limitations of existing technologies include at least the following problems: Existing technologies neglect the synergistic relationship between temperature and current. During the operation of semiconductor lasers, temperature and current are closely related. For example, an increase in device temperature will cause a decrease in threshold voltage, which will lead to a further increase in driving current. The increase in current will bring additional heat, causing the temperature to continue to rise, which can easily lead to the semiconductor laser falling into an unstable thermoelectric cycle. When the current overshoots under transient disturbances, the local temperature rise intensifies, making it difficult for temperature control to respond in a timely manner, resulting in a misalignment between temperature and current regulation. In addition, the dynamic delay in the temperature regulation process may also be superimposed on the current feedback loop, leading to a decrease in control accuracy. For example, when the current is lowered, the temperature is still accumulating, or the current remains at a high level after the temperature is lowered, resulting in an imbalance in the regulation rhythm. Existing technologies lack synergistic control of temperature and current, making it difficult to ensure the efficient and stable output of semiconductor lasers under complex operating conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for dual control of temperature and current in semiconductor lasers, which solves the problem that existing technologies neglect the coordinated regulation of temperature and current, making it difficult for semiconductor lasers to achieve efficient and stable output.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for dual control of temperature and current in a semiconductor laser, comprising the following steps: within a set sliding period, acquiring the device temperature timing data of the set semiconductor laser, and combining it with a pre-trained temperature dynamic control model to analyze the temperature control characteristic value of the set semiconductor laser; acquiring the current temperature effect timing data of the set semiconductor laser, and combining it with the temperature control characteristic value to analyze the initial current control characteristic value of the set semiconductor laser; acquiring the current effect timing data of the set semiconductor laser, and combining it with the initial current control characteristic value to analyze the comprehensive current control characteristic value of the set semiconductor laser; and performing dual collaborative control processing on the set semiconductor laser based on the temperature control characteristic value and the comprehensive current control characteristic value.

[0008] Furthermore, the device temperature timing data includes the temperature signal data of the regulating tube and the temperature signal data of the sampling resistor. The specific steps for analyzing and setting the temperature control characteristic value of the semiconductor laser are as follows: input the device temperature timing data of the semiconductor laser into the pre-trained temperature dynamic control model, analyze the temperature control characteristic set of the semiconductor laser, including thermal mismatch characteristic value, thermal conduction deviation characteristic value, and thermal synergy characteristic value; based on the temperature control characteristic set of the semiconductor laser, analyze the temperature control characteristic value of the semiconductor laser.

[0009] Furthermore, the temperature signal data of the regulating tube includes the temperature value of the regulating tube at each time point, the temperature signal data of the sampling resistor includes the temperature value of the sampling resistor at each time point, and the temperature dynamic control model includes an input layer, an LSTM temperature control extraction layer, and an output layer.

[0010] Furthermore, the specific steps for analyzing and setting the temperature control feature set of the semiconductor laser are as follows: In the input layer of the temperature dynamic control model, the device temperature time series data of the semiconductor laser is received and preprocessed; in the LSTM temperature control extraction layer of the temperature dynamic control model, the temperature mapping feature vector of the semiconductor laser is extracted from the preprocessed device temperature time series data; in the output layer of the temperature dynamic control model, the temperature control feature set of the semiconductor laser is output based on the temperature mapping feature vector of the semiconductor laser.

[0011] Furthermore, the specific steps for analyzing and setting the initial current control characteristic value of the semiconductor laser are as follows: obtain the reference parameter set of the semiconductor laser and, in combination with the current-temperature effect time series data, analyze the equivalent current-temperature response characteristic value of the semiconductor laser; based on the current-temperature coefficient and temperature control characteristic value of the semiconductor laser, analyze the initial current control characteristic value of the semiconductor laser.

[0012] Furthermore, the current effect timing data includes voltage conversion efficiency timing data and feedback voltage timing data. The specific steps for analyzing and setting the comprehensive current control characteristic value of the semiconductor laser are as follows: Based on the current effect timing data of the semiconductor laser, analyze and set the electro-efficiency synergistic control characteristic value of the semiconductor laser; based on the electro-efficiency synergistic control characteristic value and the initial current control characteristic value of the semiconductor laser, analyze and set the comprehensive current control characteristic value of the semiconductor laser.

[0013] Furthermore, the specific steps for analyzing and setting the synergistic control characteristic value of the semiconductor laser are as follows: Based on the voltage conversion efficiency time series data of the semiconductor laser, analyze and set the energy efficiency evolution characteristic value of the semiconductor laser; based on the feedback voltage time series data of the semiconductor laser, analyze and set the feedback evolution characteristic value of the semiconductor laser, and combine the energy efficiency evolution characteristic value to analyze and set the synergistic control characteristic value of the semiconductor laser.

[0014] Furthermore, the specific formula for calculating the overall current control characteristic value of the semiconductor laser is as follows: ;in, , , The parameters are, in order, the current comprehensive control characteristic value, the power efficiency synergistic control characteristic value, and the initial current control characteristic value of the semiconductor laser. , These are, in order, the coupling regulation coefficient and the inhibition regulation coefficient stored in the database.

[0015] Furthermore, the specific steps for performing dual coordinated control processing on the set semiconductor laser based on temperature regulation characteristic values ​​and current comprehensive regulation characteristic values ​​are as follows: normalize the temperature regulation characteristic values ​​and current comprehensive regulation characteristic values ​​of the set semiconductor laser; compare the normalized temperature regulation characteristic values ​​and current comprehensive regulation characteristic values ​​of the set semiconductor laser with several preset dual adjustment intervals for judgment and analysis; and take corresponding dual coordinated control measures based on the judgment and analysis results.

[0016] A dual temperature and current control system for a semiconductor laser includes: a temperature acquisition and analysis module for acquiring time-series data of the device temperature of a set semiconductor laser within a set sliding period, and analyzing the temperature regulation characteristic value of the set semiconductor laser in conjunction with a pre-trained temperature dynamic regulation model; a current temperature effect analysis module for acquiring time-series data of the current temperature effect of the set semiconductor laser, and analyzing the initial current regulation characteristic value of the set semiconductor laser in conjunction with the temperature regulation characteristic value; a current regulation analysis module for acquiring time-series data of the current effect of the set semiconductor laser, and analyzing the comprehensive current regulation characteristic value of the set semiconductor laser in conjunction with the initial current regulation characteristic value; and a dual collaborative control module for performing dual collaborative control processing on the set semiconductor laser based on the temperature regulation characteristic value and the comprehensive current regulation characteristic value.

[0017] The present invention has the following beneficial effects:

[0018] (1) The dual temperature and current control method of the semiconductor laser combines temperature regulation characteristic value and current regulation characteristic value and adopts dual collaborative control method to realize the collaborative regulation of temperature and current. This fully considers and adjusts the mutual influence between temperature and current, and avoids the misalignment of temperature and current regulation. Temperature change of semiconductor laser will affect threshold voltage, which will cause current fluctuation. In turn, the increase of current may cause temperature rise. Based on the dual collaborative control of temperature and current, the delay and overshoot phenomenon between temperature and current regulation can be effectively reduced, thereby ensuring that the laser can maintain stable output in complex dynamic change environment. This can effectively reduce the laser performance degradation caused by excessive temperature or excessive current, and thus realize the efficient and stable operation of semiconductor laser.

[0019] (2) The dual temperature and current control method for semiconductor lasers introduces a pre-trained temperature dynamic control model and uses LSTM to extract features from temperature time series data to achieve accurate extraction of temperature control features. This model can not only capture the temperature change trend, but also effectively capture the influence of temperature on semiconductor lasers under different dynamic changes, thereby improving the temperature control accuracy. Based on the accurate extraction of temperature control features, it can ensure that the laser maintains a stable operating temperature during dynamic changes, thereby improving the response speed of temperature control and enhancing the high efficiency and stability of semiconductor lasers under complex application conditions.

[0020] (3) The dual control method of temperature and current of semiconductor laser combines the timing data of current temperature effect with the temperature regulation characteristic value, thereby establishing a more refined synergistic relationship between current regulation and temperature regulation. It also considers the influence of temperature on current regulation and the feedback effect of current fluctuation on temperature, realizing the effective interaction between current and temperature in dual control, thereby accurately adjusting the current to deal with problems such as excessive current that may occur in the laser under variable working conditions, and avoiding the loss of laser efficiency caused by excessive current. This enhances the adaptability of semiconductor laser under complex load conditions and ensures the long-term stable output of the laser.

[0021] (4) The dual temperature and current control system of the semiconductor laser significantly improves the accuracy of temperature and current regulation through the collaborative work between modules. Through the collaborative work of the temperature acquisition and analysis module, the current temperature effect analysis module and the current regulation analysis module, the system can accurately acquire and analyze the temperature and current related data of the semiconductor laser, thereby ensuring that the laser can dynamically adapt to environmental changes during operation and achieve optimal temperature and current regulation. This effectively solves the problem of slow response or large deviation in complex working environments. Furthermore, through the feedback control of the dual collaborative control module, not only is the matching relationship between current and temperature optimized, but the system's adaptability to transient changes is also improved, thereby improving the reliability of the laser and reducing the performance degradation caused by power consumption and temperature fluctuations.

[0022] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0023] Figure 1 This is a flowchart of a method for dual control of temperature and current in a semiconductor laser according to the present invention.

[0024] Figure 2 This is a schematic diagram illustrating the temperature control timing feature set data of a semiconductor laser in a dual temperature and current control method for a semiconductor laser according to the present invention.

[0025] Figure 3 This is a flowchart illustrating the specific steps involved in analyzing and setting the comprehensive current regulation characteristic value of a semiconductor laser in a dual temperature and current control method for a semiconductor laser according to the present invention.

[0026] Figure 4 This is a block diagram of a dual control system for temperature and current of a semiconductor laser according to the present invention. Detailed Implementation

[0027] Please see Figure 1 This invention provides a technical solution: a method for dual control of temperature and current in a semiconductor laser, comprising the following steps: setting a sliding period (the current time point, i.e., the last time point of the sliding period, and the period can be 0.5s to 1s). It should be noted that when enough time points have not been collected during the sliding period, any of the following methods can be used: filling in missing data by default based on set reference temperature and reference current values; expanding the existing time point data using interpolation to meet the data requirements of the sliding period; delaying the execution of the control steps, and performing analysis and dual control only when the number of collected data points meets the sliding period requirements. Within the dual-coordinated control mechanism, the device temperature timing data (in the constant current drive circuit) of the set semiconductor laser is acquired, and combined with a pre-trained temperature dynamic control model, the temperature control characteristic value of the set semiconductor laser is analyzed; the current temperature effect timing data of the set semiconductor laser is acquired, and combined with the temperature control characteristic value, the initial current control characteristic value of the set semiconductor laser is analyzed; the current effect timing data of the set semiconductor laser is acquired, and combined with the initial current control characteristic value, the comprehensive current control characteristic value of the set semiconductor laser is analyzed; based on the temperature control characteristic value and the comprehensive current control characteristic value, the set semiconductor laser is subjected to dual-coordinated control processing.

[0028] The specific steps for dual coordinated control of a semiconductor laser based on temperature regulation characteristic values ​​and current comprehensive regulation characteristic values ​​are as follows: The temperature regulation characteristic values ​​and current comprehensive regulation characteristic values ​​of the semiconductor laser are normalized (i.e., their values ​​are mapped to a range of 0-1); the normalized temperature regulation characteristic values ​​and current comprehensive regulation characteristic values ​​of the semiconductor laser are compared with several preset dual adjustment intervals for judgment and analysis. Each dual adjustment interval includes a temperature regulation interval and a current regulation interval, and each dual adjustment interval corresponds to a dual control strategy.

[0029] Based on the judgment and analysis results, corresponding dual-coordinated control measures are adopted. Specifically, based on the normalized set temperature regulation characteristic value and current comprehensive regulation characteristic value of the semiconductor laser being within the preset star adjustment range, dual-coordinated control measures are implemented on the set semiconductor laser, including but not limited to the following examples:

[0030] Interval group 1 (temperature too high, current too low):

[0031] Temperature control range: 0.8-1.0 (indicates that the device temperature is too high and the temperature control accuracy decreases);

[0032] Current regulation range: 0.0-0.2 (indicates that the drive current is too low, possibly indicating insufficient power);

[0033] Dual-coordinated control strategy: Prioritize temperature control, increase the cooling current of the TEC (thermal cooler) drive circuit, and strengthen the cooling of key components such as the regulating tube and sampling resistor. The goal is to quickly reduce the device temperature to the ideal operating point of 25°C to prevent performance degradation due to overheating. Under the premise of effective temperature control, gradually increase the reference voltage setting of the linear constant current source circuit to slowly increase the output current to the rated value. At the same time, monitor the optical power output through a photodetector to avoid the introduction of unstable factors due to the simultaneous changes in current and temperature.

[0034] Interval group 2 (temperature too low, current too high):

[0035] Temperature control range: 0.0-0.2 (indicates that if the device temperature is too low, it may affect laser efficiency).

[0036] Current regulation range: 0.8-1.0 (indicates that the drive current is too high and there is a risk of overshoot);

[0037] Dual collaborative control strategy: Reduce the reference voltage setting of the linear constant current source circuit or reduce its feedback loop gain to bring the output current back to a safe range, ensuring that the laser is not impacted by large current. After the current stabilizes, switch the TEC operating mode from strong cooling to weak cooling or micro heating, and provide reverse current through the H-bridge circuit to raise the temperature of key components from an excessively low state to around 25°C, thereby optimizing the electro-optical conversion efficiency of the laser.

[0038] Interval group 3 (temperature and current are both in ideal conditions):

[0039] Temperature control range: 0.4-0.6 (indicating that the temperature is stable within the optimal range);

[0040] Current regulation range: 0.4-0.6 (indicating that the current is stable within the optimal range);

[0041] Dual collaborative control strategy: Maintain the current parameters of the power supply voltage adaptive circuit and the TEC temperature control platform so that the system is at its optimal operating point. At this point, efforts should be made to optimize energy efficiency, such as finely adjusting the switching frequency of the DC / DC converter to reduce its own power consumption.

[0042] Interval group 4 (both temperature and current deviate significantly):

[0043] Temperature control range: 0.8-1.0 (too high temperature);

[0044] Current adjustment range: 0.8-1.0 (current is too high);

[0045] Dual-coordinated control strategy: Immediately and significantly reduce the output current setpoint to reduce the power consumption and heat generation of the regulating tube, while simultaneously maximizing the TEC cooling power to force cooling of critical components.

[0046] Specifically, the device temperature timing data includes the temperature signal data of the regulating tube and the temperature signal data of the sampling resistor. The specific steps for analyzing the temperature control characteristic value of the set semiconductor laser are as follows: Input the device temperature timing data of the set semiconductor laser into the pre-trained temperature dynamic control model, analyze the temperature control characteristic set of the set semiconductor laser, including thermal mismatch characteristic value, thermal conduction deviation characteristic value, and thermal synergy characteristic value; Based on the temperature control characteristic set of the set semiconductor laser, analyze the temperature control characteristic value of the set semiconductor laser (used to characterize the overall temperature dynamic control capability of the set semiconductor laser; when the temperature control characteristic value is large, it indicates that the device temperature is prone to being too high and the temperature control accuracy is reduced), the specific formula is as follows: ;in, To set the temperature control characteristic value of the semiconductor laser, To set the thermal mismatch characteristic value of the semiconductor laser, To set the thermal synergistic characteristic value of the semiconductor laser, This is the thermal mismatch adjustment coefficient in the database. To define the thermal conduction deviation characteristics of semiconductor lasers, This refers to the heat conduction adjustment coefficient in the database. The thermal mismatch adjustment coefficient is the thermal coordination adjustment coefficient stored in the database, and in this embodiment, the thermal mismatch adjustment coefficient is the thermal coordination adjustment coefficient stored in the database. Thermal conductivity adjustment coefficient Thermal Coordination Coefficient The values ​​were 0.400, 0.350, and 0.250, respectively.

[0047] The following is a specific implementation example for calculating the temperature control characteristic values ​​of a semiconductor laser. The available data includes thermal mismatch characteristic values, thermal conduction deviation characteristic values, and thermal synergy characteristic values ​​within five randomly selected sliding cycles, as detailed in Table 1 and... Figure 2 As shown:

[0048] Table 1. Example of setting the temperature control timing feature set data for a semiconductor laser.

[0049] Thermal mismatch characteristic value thermal conduction deviation characteristic value Thermal synergistic eigenvalues Sliding cycle 1 0.478 0.543 0.436 Sliding cycle 2 0.512 0.568 0.429 Sliding cycle 3 0.506 0.554 0.468 Sliding cycle 4 0.483 0.519 0.492 Sliding cycle 5 0.526 0.524 0.504

[0050] Thermal mismatch adjustment coefficients stored in the database The value is: 0.400;

[0051] The thermal conductivity regulation coefficient stored in the database The value is: 0.350;

[0052] Thermal Coordination Coefficients Stored in Database The value is: 0.250;

[0053] Substituting the data from Table 1 and the aforementioned coefficients into the specific formula for calculating the temperature control characteristic value of the semiconductor laser, we obtain:

[0054] The temperature control characteristic value of the first set sliding cycle semiconductor laser = 0.400×0.478+0.350×√0.543+0.250×0.436≈0.558;

[0055] The temperature control characteristic value of the semiconductor laser for the second set sliding cycle is approximately 0.400 × 0.512 + 0.350 × √0.568 + 0.250 × 0.429 ≈ 0.576.

[0056] The third setting for the sliding cycle of the temperature control characteristic value of the semiconductor laser is: 0.400×0.512+0.350×√0.554+0.250×0.468≈0.580;

[0057] The temperature control characteristic value of the semiconductor laser for the fourth set sliding cycle is approximately 0.400 × 0.483 + 0.350 × √0.519 + 0.250 × 0.492 ≈ 0.568.

[0058] The temperature control characteristic value of the semiconductor laser for the fifth set sliding cycle is approximately 0.400 × 0.526 + 0.350 × √0.524 + 0.250 × 0.502 ≈ 0.590.

[0059] The regulating transistor temperature signal data includes the temperature value of the regulating transistor (i.e., MOS transistor) at each time point, the sampling resistor temperature signal data includes the temperature value of the sampling resistor at each time point, and the temperature dynamic control model includes the input layer, the LSTM temperature control extraction layer, and the output layer.

[0060] The specific steps for analyzing and setting the temperature control feature set of the semiconductor laser are as follows: In the input layer of the temperature dynamic control model, the device temperature time series data of the semiconductor laser is received and preprocessed. Specifically, outlier detection and removal are performed on the device temperature time series data, eliminating abnormal time points caused by sensor jitter, environmental electromagnetic interference, or sampling errors. Interpolation or moving average methods are used to compensate for missing data. The compensated device temperature time series data is then normalized, mapping the results to a unified numerical range [0, 1]. In the LSTM temperature control extraction layer of the temperature dynamic control model, the temperature mapping feature vector of the semiconductor laser is extracted from the preprocessed device temperature time series data. Specifically, the following steps are taken:

[0061] The device temperature time series data (including the temperature sequence of the regulating tube and the temperature sequence of the sampling resistor) after preprocessing by the input layer is input into the LSTM temperature control extraction layer. The LSTM network utilizes the structural characteristics of a recurrent neural network (RNN), updating the hidden state and memory state at each time step through its memory cells and gating mechanisms (including input gates, forget gates, and output gates). This allows it to capture long-term dependencies and short-term dynamic trends in the time series data. During this process, the LSTM layer progressively learns the temperature change patterns at each time point, memorizing and accumulating historical information in the temperature sequence while dynamically updating the response to future time steps. Through this process, the LSTM layer generates a temperature mapping feature vector that reflects the thermal dynamic behavior of the semiconductor laser device under different operating conditions, such as:

[0062] The temperature difference between the regulating tube temperature value and the sampling resistor temperature value at each time point is processed to obtain the temperature difference value at each time point. Then, root mean square (RMS) processing is performed to obtain the temperature MMS value. The first derivative of the temperature difference value at each time point is processed, and the result is averaged to obtain the temperature difference change value. Furthermore, a cross-correlation analysis method is used to progressively slide the sampling resistor temperature signal along the time axis, multiplying and accumulating it point-by-point with the regulating tube temperature signal. (That is, the entire sampling resistor temperature signal is shifted backward along the time axis by a set time offset, and its overlapping time interval with the regulating tube temperature signal is determined. Each shift forms a new overlapping interval containing the synchronous time points of the two signals at that offset. Within the overlapping interval, the sample values ​​of the two signals at the same moment are multiplied point-by-point, and all products are summed.) The process involves summing the sums and dividing the sum by the product of the number of overlapping sample points and their standard deviations. This step is repeated with different time offsets to obtain similarity metrics at different time offsets. When the similarity metric reaches its maximum, the time offset at this point is recorded as the optimal time delay between the two signals. The mean square value of temperature, the temperature difference change value, and the time offset are normalized. Based on the normalization results, a weighted average is applied to extract thermal mismatch features (used to characterize the difference between the temperature of the regulating tube and the temperature of the sampling resistor. It can reflect the degree of mismatch between the two temperature signals in the dynamic process. This feature can ensure that the constant current drive system maintains high accuracy of current detection under temperature disturbances and supports high stability. The larger the feature value, the more serious the difference between the two temperature signals, indicating that the device as a whole is in an overheating risk state).

[0063] For each time point, the temperature difference of the regulating tube at adjacent time points is analyzed sequentially. If multiple consecutive temperature differences (e.g., four time points) are greater than zero, the interval formed by these time points is determined to be the heating stage. If multiple consecutive temperature differences (e.g., four time points) are less than zero, the interval formed by these time points is determined to be the cooling stage. Within each heating stage, a linear fit is performed using the least squares method with the time point as the x-axis and the regulating tube temperature as the y-axis. The slope of the fitted line is extracted, and the average value is taken as the heating rate. Similarly, the cooling rate can be obtained. The heating rate and cooling rate are then combined, i.e., (heating rate - cooling rate) / (cooling rate + The heating rate is used to obtain the temperature response rate of the regulating tube. Similarly, the temperature response rate of the sampling resistor can be obtained. After weighted averaging, the thermal conduction deviation feature is extracted (used to characterize the difference in conduction efficiency of semiconductor lasers during the heating and cooling stages. It can reveal the inconsistency of the dynamic response rate of the device in different thermal processes and serve as an important parameter for judging the adjustment range of heating and cooling processes in temperature control. This feature can improve its energy efficiency control capability in dynamic processes and ensure the high efficiency of constant current drive. The larger the feature value, the more significant the inconsistency between heating and cooling processes. Generally, the heating rate is higher than the cooling rate, and the device is more likely to accumulate heat and the temperature is higher).

[0064] The temperature values ​​of the regulating tube and the sampling resistor at each time point are read. The average temperature values ​​of the regulating tube and the sampling resistor are extracted and their differences are processed to obtain the temperature difference value. The correlation between the regulating tube temperature value and the sampling resistor temperature value is analyzed using the Pearson correlation coefficient. Rank transformation is performed on the regulating tube temperature value and the sampling resistor temperature value at each time point, replacing each value with its sorted position in the signal sequence. The Spearman rank correlation coefficient is then used to analyze the rank correlation value, which is standardized with the temperature difference value and the correlation value. Based on the results, a weighted average is applied to extract thermal synergy feature values ​​(used to characterize the overall synergy between the regulating tube temperature signal and the sampling resistor temperature signal; it can characterize the similarity of the two temperature signals in terms of time sequence and monotonicity. The larger the feature value, the weaker the overall synergy between the two temperature signals, and the more obvious the temperature imbalance, indicating that the regulating tube temperature is generally higher than the sampling resistor temperature, and there is a risk of excessive heat generation). The thermal mismatch feature, thermal conduction deviation feature, and thermal synergy feature are concatenated into a temperature mapping feature vector.

[0065] In the output layer of the temperature dynamic control model, based on the temperature mapping feature vector of the set semiconductor laser, the temperature control feature set of the set semiconductor laser is output. Specifically, the thermal mismatch feature, thermal conduction deviation feature, and thermal synergy feature in the temperature mapping feature vector of the set semiconductor laser are activated by the Sigmoid function to obtain thermal mismatch feature value, thermal conduction deviation feature value, and thermal synergy feature value between 0 and 1.

[0066] The pre-training steps of the temperature dynamic control model are as follows: In the pre-training data preparation stage of the temperature dynamic control model, the temperature time series data of key components in the semiconductor laser constant current drive system are first divided and structured. Specifically, the temperature time series data of the regulating tube and sampling resistor after outlier removal, missing value compensation and normalization are divided into training set, validation set and test set according to time order. The division ratio follows the principle of 80% for model training, 10% for hyperparameter validation and 10% for final performance testing. Each sample data is a continuous temperature sequence within a certain time window, and its corresponding real temperature control feature set (including the true values ​​of thermal mismatch feature, thermal conduction deviation feature and thermal coordination feature) is labeled. The sliding window method is used to convert the long time series data into input samples of fixed length (such as 128 time steps) and organize them into a three-dimensional tensor format of (number of samples × time step length × feature dimension) to adapt to the input requirements of the LSTM network.

[0067] In the pre-training stage of the LSTM temperature extraction layer of the temperature dynamic regulation model, the time-series temperature dependency is modeled based on the training set. Specifically, the structured temperature time-series training data is input into the LSTM layer, which is trained using the backpropagation time time (BPTT) algorithm. The network weights are optimized to minimize the loss function (such as mean squared error MSE or smoothing L1 loss) between the predicted features and the true features. The memory units and gating mechanisms (input gate, forget gate, output gate) of the LSTM dynamically learn the long-term and short-term patterns in the temperature sequence. The hidden state is updated iteratively with time steps, thereby capturing the dynamic coupling relationship, changing trend and delay effect between the temperature of the regulating tube and the sampling resistor. During the training process, a gradient clipping strategy is used to prevent gradient explosion, and Dropout technology is used to suppress overfitting and improve the model's generalization ability.

[0068] In the pre-training stage of the output layer of the temperature dynamic regulation model, the high-level features extracted by the LSTM layer are mapped and integrated. Specifically, the final hidden state of the L layer output is nonlinearly transformed by a fully connected layer and mapped to a representation that matches the dimension of the target features. Then, through three independent Sigmoid activation branches, thermal mismatch feature values, thermal conduction deviation feature values, and thermal synergy feature values ​​are calculated respectively to ensure that the output range is within the [0, 1] interval. The model training adopts the adaptive moment estimation (Adam) optimizer to dynamically adjust the learning rate to accelerate convergence. The loss function and feature prediction error are monitored on the validation set. Overfitting is prevented by an early stopping strategy. The hyperparameters (such as the number of LSTM hidden units, learning rate, and batch size) are tuned by grid search or Bayesian optimization methods until the model achieves stable and excellent performance on the training and validation sets.

[0069] During the model evaluation and deployment phase, the pre-trained temperature dynamic control model is evaluated using a test set. Specifically, the mean absolute error (MAE), root mean square error (RMSE), and correlation coefficient (R²) between the predicted temperature control feature set and the true feature set on the test set are calculated to comprehensively evaluate the accuracy of the model's feature extraction. Once the model performance meets the requirements, it is solidified into an inference model and deployed in the embedded control unit of the semiconductor laser constant current drive system. This model is used to process temperature time series data in real time and output the temperature control feature set, providing forward-looking parameter basis for subsequent current regulation.

[0070] In this implementation scheme, a dynamic temperature control model is introduced, combined with LSTM analysis of temperature time-series data, to extract temperature control feature sets. This allows for in-depth analysis of the thermal dynamic behavior of the semiconductor laser. Secondly, by performing difference processing and root mean square calculation on the temperature signals of the control tube and the sampling resistor at each time point, the amplitude and trend of temperature fluctuations can be accurately extracted, providing a precise reference for temperature control adjustment. Furthermore, by combining cross-correlation analysis and Spearman rank correlation coefficient methods, the synergy and differences between the control tube temperature and the sampling resistor temperature are analyzed, effectively revealing the degree of temperature distribution imbalance. Finally, through comprehensive analysis of thermal mismatch, thermal conduction deviation, and thermal synergy characteristics, temperature control feature values ​​are generated. This ensures timely and accurate adjustments when the temperature is too high or too low, avoiding temperature control imbalance and device performance degradation, and preventing temperature control instability caused by temperature fluctuations. This enables the semiconductor laser to maintain a stable temperature in a dynamically changing working environment, reducing performance degradation caused by overheating or underheating, and ultimately ensuring the efficient and stable operation of the laser.

[0071] Specifically, the current-temperature effect timing data includes the MOSFET threshold voltage and MOSFET output current at each time point. The specific steps for analyzing and setting the initial current modulation characteristic value of the semiconductor laser are as follows:

[0072] A set of reference parameters for the semiconductor laser is obtained and, combined with current-temperature effect timing data, the equivalent current-temperature response characteristic value of the semiconductor laser is analyzed. Specifically, the reference parameter set includes the MOSFET temperature reference value (obtained from the MOSFET factory calibration stored in the database, and taken as 25℃ in this example), the MOSFET threshold voltage temperature coefficient (obtained from the MOSFET manufacturer's datasheet stored in the database, and taken as 2mV / ℃ in this example), the reference voltage value (set by the constant current drive circuit and stored in the database), and the reference current value (obtained by the sampling resistor measurement during the initial calibration stage and stored in the database). Based on the output current at each time point, the relative rate of change of the output current is extracted and read. The temperature value of the regulating tube at each time point is taken and the difference is processed with the temperature reference value of the MOSFET to obtain the temperature difference value of the MOSFET at each time point. The output current value of the MOSFET at each time point is compared with the reference current value, i.e., (reference current value - MOSFET output current value) / reference voltage value, to obtain the relative rate of change of current at each time point. Based on the reference voltage value of the set semiconductor laser, the temperature coefficient of the MOSFET threshold voltage, and the temperature difference value, relative rate of change of current value, and threshold voltage value of the MOSFET at each time point, the current temperature change coefficient of the set semiconductor laser at each time point is analyzed, and the average value is processed. The structure is then standardized to obtain the equivalent current temperature response characteristic value of the set semiconductor laser.

[0073] Based on the current temperature coefficient and temperature control characteristic value of the set semiconductor laser, the initial current control characteristic value of the set semiconductor laser is analyzed (used to characterize the current deviation sensitivity of the set semiconductor laser under temperature effects; its value reflects the degree of deviation of the current relative to the reference current when it is not controlled and its temperature dependence; when this characteristic value is large, it indicates that the temperature drift of the threshold voltage of the MOS transistor has a significant impact on the output current, resulting in a larger current relative to the reference current, and a stronger demand for current control). Specifically:

[0074] The equivalent current-temperature response characteristic value and temperature control characteristic value of the set semiconductor laser are standardized, and then the standardized equivalent current-temperature response characteristic value and temperature control characteristic value of the set semiconductor laser are fused. In this embodiment, a multiplicative fusion method is adopted, that is, the standardized current-temperature response characteristic value and the temperature control characteristic value are multiplied, and then the result is square rooted to obtain the initial current control characteristic value of the set semiconductor laser.

[0075] The threshold voltage of a MOSFET is the minimum gate voltage required for the MOSFET to turn on. It can read the MOSFET temperature value at that time and perform difference processing with the MOSFET temperature reference value stored in the database to obtain the MOSFET temperature difference value. It can also read the MOSFET reference threshold voltage stored in the database under the MOSFET temperature reference value and perform comprehensive processing with the MOSFET temperature difference value and the MOSFET threshold voltage temperature coefficient stored in the database. That is, MOSFET threshold voltage = MOSFET reference threshold voltage - MOSFET threshold voltage temperature coefficient × MOSFET temperature difference value. The result is taken as the MOSFET threshold voltage value.

[0076] The output current of the MOSFET is the operating current provided by the constant current drive circuit of the semiconductor laser. It can be obtained by acquiring the MOSFET electron mobility, MOSFET width, MOSFET length, gate oxide capacitance per unit area, and reference voltage stored in the database, and reading the MOSFET threshold voltage at the MOSFET temperature at that time point. The results are then processed as follows: MOSFET output current = 0.5 × MOSFET electron mobility × gate oxide capacitance per unit area × (MOSFET width / MOSFET length) × (reference voltage - MOSFET threshold voltage)^2. The result is taken as the MOSFET output current value.

[0077] The specific formula for calculating the current temperature coefficient of a semiconductor laser at a given time point is as follows: ;in, To set the current temperature coefficient of a semiconductor laser at a specific point in time. To set the relative rate of change of current in a semiconductor laser at a specific point in time. To set the temperature coefficient of the threshold voltage of the MOS transistor in a semiconductor laser, To set the reference voltage value for the semiconductor laser, To set the threshold voltage value of the MOS transistor in a semiconductor laser at a specific time point, To set the relative rate of change of current in a semiconductor laser at a specific point in time. To set the temperature difference value of the MOS tube at a certain time point in the semiconductor laser.

[0078] In this implementation scheme, the current control accuracy of the semiconductor laser is effectively improved by deeply analyzing the relationship between temperature and current. Secondly, based on the threshold voltage and output current of the MOS transistor at each time point, combined with the temperature control characteristic value, the temperature response characteristics of the current can be accurately extracted. By analyzing the temperature difference change of the MOS transistor and its impact on the current, the dynamic changes of temperature fluctuations on the output current can be revealed. Especially when temperature drift significantly affects the current output, the current deviation from the reference value can be effectively prevented by adjusting the temperature control strategy. Finally, by adopting standardization processing and multiplicative fusion method, the temperature control characteristic value and the current temperature response characteristic value are combined to optimize the current control characteristic value, thereby improving the accuracy of current control. This makes the current control of the semiconductor laser more precise in dynamic environments, thereby improving the working stability of the laser under complex temperature changes.

[0079] Specifically, such as Figure 3 As shown, the current effect time-series data includes voltage conversion efficiency time-series data (i.e., voltage conversion efficiency value at each time point) and feedback voltage time-series data (i.e., feedback voltage value at each time point). The specific steps for analyzing and setting the comprehensive current control characteristic value of the semiconductor laser are as follows: Based on the current effect time-series data of the semiconductor laser, analyze and set the electro-efficiency synergistic control characteristic value of the semiconductor laser; based on the electro-efficiency synergistic control characteristic value and the initial current control characteristic value of the semiconductor laser, analyze and set the comprehensive current control characteristic value of the semiconductor laser (used to characterize the overall current deviation of the semiconductor laser; when the comprehensive current control characteristic value is large, it indicates that the overall current is too high, which is prone to overshoot risk).

[0080] The voltage conversion efficiency value is the energy utilization rate of the constant current drive circuit. It can be obtained in real time by the voltage divider sampling circuit and ADC conversion to obtain the power supply voltage value of the constant current drive system; the sampling resistor voltage value across the sampling resistor can be obtained in real time by the differential amplifier circuit and ADC conversion; and the drain-source voltage value of the MOSFET can be obtained in real time by the differential voltage detection circuit and ADC conversion. The power supply voltage value, sampling resistor voltage value, and MOSFET drain-source voltage value are then combined and processed into a formula, i.e., 1 - [(sampling resistor voltage value + MOSFET drain-source voltage value) / power supply voltage value], and the result is used as the voltage conversion efficiency value.

[0081] The feedback voltage value is the voltage drop across the sampling resistor. It can be obtained by bringing out a detection port across the sampling resistor, inputting the feedback voltage signal to the inverting input of the operational amplifier for comparison and control, and simultaneously sampling it in real time through an analog-to-digital converter.

[0082] The specific steps for analyzing and setting the energy efficiency synergistic control characteristic value of the semiconductor laser are as follows: Based on the time series data of the voltage conversion efficiency of the semiconductor laser, analyze the energy efficiency evolution characteristic value of the semiconductor laser. Specifically, read the voltage conversion efficiency value at each time point, and process the change of the voltage conversion efficiency value at adjacent time points in sequence, such as |the difference between the voltage conversion efficiency values ​​at the first time point and the second time point| / the voltage conversion efficiency value at the second time point. Then, perform a moving average on the change processing result to obtain the energy efficiency response characteristic value of the semiconductor laser (used to characterize the sensitivity of the semiconductor laser's energy efficiency to external disturbances or transient changes during operation. The larger the characteristic value, the more severe the energy efficiency fluctuation under current-driven conditions, reflecting a higher demand for current regulation to adapt to dynamic changes).

[0083] The voltage conversion efficiency value at each time point is divided into several time intervals (i.e., the voltage conversion efficiency values ​​at adjacent time points are read and processed, and compared with a preset threshold. If the value is higher than the preset threshold, the first time point in the adjacent time points is taken as the end time point of the time interval, and the second time point is taken as the start time point of the next time interval). The minimum voltage conversion efficiency value in each time interval is calculated and the difference is processed with the voltage conversion efficiency reference value stored in the database, i.e., voltage conversion efficiency reference value - minimum voltage conversion efficiency value, to obtain the voltage conversion efficiency difference for each time interval. The minimum voltage conversion efficiency difference and the average voltage conversion efficiency difference are extracted and ratio processed to extract the energy efficiency margin robustness feature (used to characterize the energy efficiency maintenance capability of semiconductor lasers in different time intervals. The larger the feature value, the easier it is for the energy efficiency to drop below the reference value under natural operating conditions, reflecting that the current regulation needs to be more robust to ensure long-term energy efficiency stability).

[0084] The voltage conversion efficiency value at each time point is differentially processed between adjacent points to obtain a voltage conversion efficiency difference sequence (i.e., the result of calculating the difference between the voltage conversion efficiency values ​​at adjacent time points). A fluctuation threshold is set based on the statistical mean and standard deviation of the difference sequence (e.g., the mean + standard deviation of the difference sequence results is used as the fluctuation threshold). If the voltage conversion efficiency difference result at adjacent time points exceeds the set fluctuation threshold, it is determined that there is an abnormal energy efficiency change at that adjacent time point. The absolute value of the difference value is used as the abnormal change amplitude, and the mean of the abnormal change amplitude is extracted. At the same time, the number of abnormal changes is accumulated and counted, and the frequency is calculated (i.e., the cumulative count result is compared with the total number of time points). This is then weighted with the mean of the constant change amplitude to obtain the energy efficiency abnormal change characteristic value (the non-smoothness and sudden abnormality of the energy efficiency curve of the semiconductor laser during operation; the larger the characteristic value, the more sudden energy efficiency anomalies exist in the uncontrolled state of the system).

[0085] The energy efficiency response characteristic value, energy efficiency margin robustness characteristic value, and energy efficiency abnormal change characteristic value are weighted to obtain the energy efficiency evolution characteristic value of the set semiconductor laser (used to characterize the comprehensive evolution characteristics of voltage conversion efficiency during operation. The larger the characteristic value, the more severe the energy efficiency fluctuation or the more obvious the drop. At this time, the current utilization is insufficient and the current needs to be increased).

[0086] Based on the timing data of the feedback voltage of the set semiconductor laser, the feedback evolution characteristic value of the set semiconductor laser is analyzed, and combined with the energy efficiency evolution characteristic value, the electro-efficiency coordinated control characteristic value of the set semiconductor laser is analyzed. Specifically, the feedback evolution characteristic value and the energy efficiency evolution characteristic value of the set semiconductor laser are weighted, and in this weighting process, the energy efficiency evolution characteristic value is inverted, i.e., 1 / (1+energy efficiency evolution characteristic value), so as to obtain the electro-efficiency coordinated control characteristic value of the set semiconductor laser (used to characterize the current regulation requirement of the set semiconductor laser under the combined effect of energy efficiency characteristics and feedback characteristics. The larger the characteristic value, the higher the degree of coordinated imbalance between energy efficiency utilization and current feedback deviation, reflecting that the overall current is too large).

[0087] The specific steps for extracting the feedback evolution characteristic value of the set semiconductor laser are as follows: Obtain the feedback voltage reference value stored in the database (which can be obtained by measuring the output current through the sampling resistor during the initial calibration stage and converting it into a reference feedback voltage value, and storing it in the database, i.e., the product of the set target output current and the resistance value of the sampling resistor), and perform difference processing on the feedback voltage value at each time point and the feedback voltage reference value to obtain the feedback voltage deviation amplitude (absolute value) at each time point, and determine whether it is higher than the preset feedback voltage deviation amplitude threshold. If it is higher than the preset feedback voltage deviation amplitude threshold, mark the time point as the abnormal start time point, and the time point below the preset feedback voltage deviation amplitude threshold is the abnormal end time point, and use this as an abnormal interval. Statistically calculate the duration value and the average feedback voltage deviation amplitude of all abnormal intervals, and perform standardization processing. Based on the results, perform weighted processing to obtain the feedback deviation margin evolution characteristic value (used to characterize the degree of deviation of the feedback voltage from the reference voltage. The larger this characteristic value, the larger it is, indicating that the overall current is too large and the current needs to be reduced).

[0088] The feedback voltage time series data is differentially processed between adjacent time points to obtain a feedback voltage change rate sequence (i.e., the result of calculating the difference between feedback voltage values ​​at adjacent time points). The portion with a change rate greater than zero is defined as the rising change rate sequence, and the portion with a change rate less than zero is defined as the falling change rate sequence. The maximum change rate is extracted from the rising change rate sequence as the maximum rising change rate, and the maximum absolute value of the change rate is extracted from the falling change rate sequence as the maximum falling change rate. The ratio of the maximum rising change rate to the maximum falling change rate is processed to obtain the feedback response asymmetry characteristic value (used to characterize the dynamic imbalance of the feedback voltage during the rising and falling processes; the larger the characteristic value, the more significant the current change trend in the rising direction, indicating a risk of excessive current that needs to be reduced).

[0089] Based on the feedback voltage value at each time point, a least squares method is used for linear fitting to extract the slope of the fitted line, which is then used as the feedback trend feature value (used to characterize the overall trend of feedback voltage change over time; when it is large, it indicates that the feedback voltage is on an upward trend, the current is gradually increasing, and the current needs to be reduced). The feedback deviation margin evolution feature value, the feedback response asymmetry feature value, and the feedback trend feature value are weighted to obtain the feedback evolution feature value of the set semiconductor laser (used to characterize the comprehensive evolution characteristics of the feedback voltage during operation; the larger this feature value, the larger the current is, and the current needs to be reduced).

[0090] The specific formula for calculating the overall current control characteristic value of a semiconductor laser is as follows: ;in, To set the comprehensive current control characteristic value of the semiconductor laser, To set the characteristic values ​​for the synergistic control of the electrical efficiency of semiconductor lasers, To set the initial current modulation characteristic value of the semiconductor laser, These are the coupling adjustment coefficients stored in the database. The inhibition regulation coefficients are stored in the database, and in this embodiment, the coupling regulation coefficients are stored in the database. Suppression regulation coefficient The values ​​were 0.650 and 0.350, respectively.

[0091] In this implementation scheme, through in-depth analysis of voltage conversion efficiency time-series data and feedback voltage time-series data, corresponding features are extracted. Based on this, synergistic control feature values ​​of electrical efficiency are generated, thereby enabling precise control of the semiconductor laser current. Secondly, by weighting the energy efficiency response feature values ​​and feedback evolution feature values, the current output can be precisely adjusted, avoiding excessive current fluctuations or overshoot, ensuring that the semiconductor laser can maintain stable output power. Finally, the synergistic control feature values ​​of electrical efficiency are synergistically processed with the initial current regulation feature values ​​to generate comprehensive current regulation feature values. This comprehensively optimizes the current regulation strategy of the semiconductor laser, thereby improving the response speed of current regulation and avoiding performance fluctuations caused by excessive or insufficient current, thus achieving efficient and stable laser output to meet the high-performance requirements of complex application environments.

[0092] Please see Figure 4 This invention provides a technical solution: a dual temperature and current control system for a semiconductor laser, comprising: a temperature acquisition and analysis module, used to acquire time-series data of the device temperature of a set semiconductor laser within a set sliding period, and analyze the temperature control characteristic value of the set semiconductor laser in conjunction with a pre-trained temperature dynamic control model; a current temperature effect analysis module, used to acquire time-series data of the current temperature effect of the set semiconductor laser, and analyze the initial current control characteristic value of the set semiconductor laser in conjunction with the temperature control characteristic value; a current control analysis module, used to acquire time-series data of the current effect of the set semiconductor laser, and analyze the comprehensive current control characteristic value of the set semiconductor laser in conjunction with the initial current control characteristic value; and a dual collaborative control module, used to perform dual collaborative control processing on the set semiconductor laser based on the temperature control characteristic value and the comprehensive current control characteristic value.

[0093] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0094] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dual control of temperature and current in a semiconductor laser, characterized in that, Includes the following steps: Within a set sliding period, the device temperature timing data of the semiconductor laser is acquired, including the temperature signal data of the control tube and the temperature signal data of the sampling resistor. The temperature signal data of the control tube includes the temperature value of the control tube at each time point, and the temperature signal data of the sampling resistor includes the temperature value of the sampling resistor at each time point. Combined with a pre-trained temperature dynamic control model, which includes an input layer, an LSTM temperature control extraction layer, and an output layer, the temperature control characteristic values ​​of the semiconductor laser are analyzed. The specific steps are as follows: The device temperature timing data of the semiconductor laser is set and input into the pre-trained temperature dynamic control model to analyze the temperature control feature set of the semiconductor laser, including thermal mismatch feature value, thermal conduction deviation feature value, and thermal synergy feature value. Based on the set temperature control characteristic set of the semiconductor laser, the temperature regulation characteristic value of the semiconductor laser is analyzed, and the specific formula is as follows: ; in, , , , The parameters are set sequentially as follows: temperature control characteristic value, thermal mismatch characteristic value, thermal compatibility characteristic value, and thermal conduction deviation characteristic value of the semiconductor laser. , , The coefficients in order are the thermal mismatch adjustment coefficient, thermal conduction adjustment coefficient, and thermal synergy adjustment coefficient in the database; The specific steps for analyzing and setting the temperature control characteristic set of a semiconductor laser are as follows: In the input layer of the temperature dynamic control model, the device temperature timing data of the semiconductor laser is received and preprocessed. In the LSTM temperature control extraction layer of the temperature dynamic control model, the temperature mapping feature vector of the set semiconductor laser is extracted from the preprocessed device temperature time series data of the set semiconductor laser. In the output layer of the temperature dynamic control model, the temperature control feature set of the semiconductor laser is output based on the temperature mapping feature vector of the semiconductor laser. Acquire the timing data of the current-temperature effect of the set semiconductor laser, and combine it with the temperature control characteristic value to analyze the initial current control characteristic value of the set semiconductor laser; The following steps are taken to obtain the timing data of the current effect of the set semiconductor laser, including the timing data of voltage conversion efficiency and feedback voltage, and combine it with the initial current regulation characteristic value to analyze the comprehensive current regulation characteristic value of the set semiconductor laser: Based on the timing data of the current effect of the semiconductor laser, the characteristic values ​​of the electro-efficiency coordinated control of the semiconductor laser are analyzed, specifically as follows: Based on the voltage conversion efficiency time series data of a set semiconductor laser, the energy efficiency evolution characteristic value of the set semiconductor laser is analyzed; Based on the timing data of the feedback voltage of the set semiconductor laser, the feedback evolution characteristic value of the set semiconductor laser is analyzed, and combined with the energy efficiency evolution characteristic value, the energy efficiency coordinated control characteristic value of the set semiconductor laser is analyzed. Based on the set characteristic values ​​of the synergistic control of the electro-efficiency and the initial current regulation of the semiconductor laser, the comprehensive current regulation characteristic value of the semiconductor laser is analyzed, and its specific formula is as follows: ;in, , , The parameters are, in order, the current comprehensive control characteristic value, the power efficiency synergistic control characteristic value, and the initial current control characteristic value of the semiconductor laser. , The coupling adjustment coefficient and the inhibition adjustment coefficient are stored in the database in the following order; The semiconductor laser is subjected to dual coordinated control based on temperature regulation characteristic value and current comprehensive regulation characteristic value.

2. The method for dual temperature and current control of a semiconductor laser according to claim 1, characterized in that, The specific steps for analyzing and setting the initial current modulation characteristic value of a semiconductor laser are as follows: Obtain the reference parameter set of the set semiconductor laser, and combine it with the current-temperature effect time series data to analyze the equivalent current-temperature response characteristic value of the set semiconductor laser; Based on the current temperature coefficient and temperature control characteristic value of the semiconductor laser, the initial current control characteristic value of the semiconductor laser is analyzed.

3. The method for dual temperature and current control of a semiconductor laser according to claim 1, characterized in that, The specific steps for dual coordinated control of a semiconductor laser based on temperature-controlled characteristic values ​​and current-controlled characteristic values ​​are as follows: The temperature control characteristic value and the current comprehensive control characteristic value of the semiconductor laser are normalized. The normalized temperature control characteristic value and current comprehensive control characteristic value of the set semiconductor laser are compared with several preset dual adjustment intervals for judgment and analysis. Based on the judgment and analysis results, corresponding dual-coordinated control measures are adopted.

4. A dual temperature and current control system for a semiconductor laser, employing the dual temperature and current control method for a semiconductor laser as described in any one of claims 1-3, characterized in that, include: The temperature acquisition and analysis module is used to acquire the device temperature time series data of the set semiconductor laser within a set sliding period, and to analyze the temperature control characteristic value of the set semiconductor laser in combination with the pre-trained temperature dynamic control model. The current-temperature effect analysis module is used to acquire the timing data of the current-temperature effect of the set semiconductor laser, and combine it with the temperature control characteristic value to analyze the initial current control characteristic value of the set semiconductor laser. The current regulation analysis module is used to acquire the timing data of the current effect of the set semiconductor laser, and analyze the comprehensive current regulation characteristic value of the set semiconductor laser by combining the initial current regulation characteristic value. The dual collaborative control module is used to perform dual collaborative control processing on the set semiconductor laser based on temperature regulation characteristic value and current comprehensive regulation characteristic value.

Images