Ytterbium-doped tunable fiber laser data rapid processing system based on quantum algorithm

Abstract

The invention provides an ytterbium-doped tunable fiber laser data rapid processing system based on a quantum algorithm, and relates to the technical field of fiber lasers, and the system comprises a data collection module which is used for collecting various data generated in the operation process of an ytterbium-doped tunable fiber laser, and the data comprises the output power, the lasing wavelength and the spectral characteristics of the laser; the quantum processing module is used for executing a quantum algorithm program through quantum computing equipment and processing various data generated in the operation process of the ytterbium-doped tunable fiber laser so as to obtain a processing result; and the feedback control module is used for generating a feedback control signal through a reinforcement learning algorithm according to a processing result of the quantum processing module, and sending the feedback control signal to the ytterbium-doped tunable fiber laser so as to realize optimization and real-time control of the performance of the laser. The laser operation data can be efficiently processed, the performance is optimized in real time, faults are prevented, and the stability and the operation efficiency of the laser are improved.

Description

Technical Field

[0001] The invention relates to the technical field of optical fiber lasers, and in particular to a fast data processing system for ytterbium-doped tunable optical fiber lasers based on quantum algorithms. Background Art

[0002] Traditional data processing methods have slow computing speeds and limited data processing capabilities when processing massive, high-dimensional data. This may result in an inability to respond in time and accurately analyze data from ytterbium-doped tunable fiber lasers, which require high real-time performance.

[0003] For example, during the operation of an Yb-doped tunable fiber laser, its status data (such as output power, lasing wavelength, temperature, etc.) will change in real time. Traditional algorithms may not be able to process this large amount of data in a short period of time, resulting in the inability to obtain the accurate operating status of the laser in a timely manner, thus affecting the real-time performance prediction and control.

[0004] In addition, with the continuous development of laser technology, the dimensions of data that need to be monitored and processed are increasing. When processing these high-dimensional data, traditional algorithms may face the problems of high computational complexity and slow processing speed, and cannot meet the data processing speed requirements of high-performance fiber lasers. When processing nonlinear and complex changing data, the prediction accuracy may not be high enough and the adaptability is poor. This may lead to large deviations in predicting the changing trends of the performance indicators of ytterbium-doped tunable fiber lasers, thereby affecting the adjustment effect of the working parameters.

[0005] The performance of Yb-doped tunable fiber lasers is affected by many factors, and there may be complex nonlinear relationships between these factors. Traditional algorithms may not be able to accurately capture these nonlinear changes, resulting in inaccurate prediction results.

[0006] For example, when predicting the trend of laser output power changing with temperature, ignoring the nonlinear interaction between temperature and other factors (such as pump power fluctuations) may lead to deviations in the prediction results.

[0007] The working environment of the laser may change, such as temperature, humidity, vibration, etc. Traditional algorithms may not be able to adapt to these environmental changes in time, resulting in deviations in predicting laser performance. For example, when the ambient temperature changes suddenly, traditional algorithms may not be able to accurately predict the fluctuation of the laser output power, thus affecting the adjustment of the laser operating parameters. Summary of the invention

[0008] The technical problem to be solved by the present invention is to provide a fast data processing system for an ytterbium-doped tunable fiber laser based on a quantum algorithm, which can realize continuous tunability of the laser wavelength.

[0009] In order to solve the above technical problems, the technical solution of the present invention is as follows: First, the fast data processing system of ytterbium-doped tunable fiber laser based on quantum algorithm includes: Data acquisition module, used to collect various data generated by the Yb-doped tunable fiber laser during operation, including the output power, lasing wavelength, and spectral characteristics of the laser; A quantum processing module is used to execute a quantum algorithm program through a quantum computing device to process various data generated during the operation of the ytterbium-doped tunable fiber laser to obtain a processing result; A feedback control module is used to generate a feedback control signal through a reinforcement learning algorithm according to the processing result of the quantum processing module, and send the feedback control signal to the ytterbium-doped tunable fiber laser to achieve optimization and real-time control of the laser performance; The monitoring and analysis module is used to monitor and analyze laser data in real time, identify and warn of potential faults during laser operation, and generate warning information; The real-time performance optimization module is used to monitor the operating status of the laser in real time, predict the changing trend of the laser performance indicators, and adjust the laser operating parameters according to the prediction results to achieve real-time performance optimization.

[0010] Furthermore, the quantum algorithm program is executed by the quantum computing device to process various data generated during the operation of the ytterbium-doped tunable fiber laser to obtain processing results, including: Use the quantum computer as a quantum computing device, initialize the quantum computing device, and configure the operating environment, including the number of quantum bits, quantum gate set, and measurement method; The various data generated by the Yb-doped tunable fiber laser during operation, including output power, lasing wavelength, and spectral characteristics, are processed through the quantum approximate optimization algorithm, and the process of the quantum approximate optimization algorithm is set up, including the data encoding method, the operation sequence of the quantum gate, and the final measurement plan; On the quantum computing device, various data generated by the Yb-doped tunable fiber laser during operation are used as input data to debug and test the quantum approximate optimization algorithm program, and the program parameters are adjusted by comparing the program output with the expected results; When the quantum approximate optimization algorithm program is running, various data generated during the operation of the ytterbium-doped tunable fiber laser are encoded and converted into the state of quantum bits, and quantum transformations are performed on the quantum bits according to a predefined sequence of quantum gate operations; At the end of the quantum approximate optimization algorithm program, a measurement operation is performed to obtain the final state of the quantum bit, and the final state of the quantum bit is decoded into classical data to obtain the processing results, including the optimized output power value, lasing wavelength value, and spectral characteristic parameters.

[0011] Furthermore, according to the processing results of the quantum processing module, a feedback control signal is generated through a reinforcement learning algorithm, and the feedback control signal is sent to the ytterbium-doped tunable fiber laser to achieve optimization and real-time control of the laser performance, including: Initialize the reinforcement learning algorithm and define the state space, action space, reward function and learning rate parameters; the state space includes the current output power, lasing wavelength and spectral characteristic parameters of the Yb-doped tunable fiber laser; the action space includes the actions to adjust the laser parameters, including changing the pump power and adjusting the resonant cavity length; For each time step, the reinforcement learning algorithm selects a current state from the state space, uses the current state as the input data of the quantum approximate optimization algorithm, runs the quantum approximate optimization algorithm program, processes the ytterbium-doped tunable fiber laser data, and obtains the optimized output power value, lasing wavelength value, and spectral characteristic parameter processing results; According to the processing results and reward function, calculate the reward value of the current state, and update the strategy function according to the current state, action and reward value; According to the updated policy function, the action to be performed in the next time step is determined, and a feedback control signal is generated according to the determined action; The feedback control signal is sent to the ytterbium-doped tunable fiber laser, and the iterative process of state acquisition, quantum approximation optimization, reward calculation, strategy update, action selection, feedback signal generation and sending is continued until the preset performance indicators are reached to achieve optimization and real-time control of the laser performance.

[0012] Furthermore, the calculation formula of the reward function is: ; in, Represents the reward function value; Represents the weight coefficient of the output power error term; Indicates the current output power of the laser; Indicates the target output power of the laser; , represents a positive number; represents the weight coefficient of the lasing wavelength error term; Indicates the current lasing wavelength of the laser; Indicates the target lasing wavelength of the laser; Represents the weight coefficient of the spectral characteristic similarity term; A vector representing the current spectral characteristics of the laser; Represents the target spectral characteristic vector of the laser.

[0013] Furthermore, the calculation formula of the strategy function is: ; in, represents the parameter vector; represents the learning rate; Indicates the maximum number of time steps in a round; represents the time step index; Represents the discount factor of Power; Represents the time step The instant reward value obtained; Indicates in status Take action The eigenvector of , Indicates action; Represents the parameter vector The transpose of Indicates in status Take action The eigenvector of Indicates in status Take action The feature vector of .

[0014] Furthermore, the laser data is monitored and analyzed in real time to identify and warn of potential failures during laser operation and generate warning information, including: Collect real-time data from ytterbium-doped tunable fiber lasers, including laser output power, wavelength, temperature, and spectral information; Extract key features from real-time data, including the rate of change of laser output power, wavelength stability, temperature variation trend, and abnormal spectral fluctuations; Using the normal operation data in the historical data set to train the preset anomaly detection model to obtain a trained anomaly detection model; After training, the anomaly detection model processes and analyzes the key features, calculates the degree of deviation between the current operating state of the laser and the normal state, and obtains the deviation degree index value to evaluate the degree of deviation between the laser and the normal state; According to the indicators obtained by the trained anomaly detection model, a threshold is set. When the indicator exceeds the threshold, the fault warning mechanism is triggered and warning information is generated, including the fault type, fault severity, and recommended maintenance measures.

[0015] Furthermore, the calculation formula for the degree of deviation between the current operating state of the laser and the normal state is: ; in, Indicates the degree of deviation between the current operating state of the laser and the normal state; The weight coefficient representing the rate of change of laser output power; Indicates the current output power of the laser; Indicates that the laser is Output power at the moment; The weight coefficient representing the wavelength stability; represents the length of the time window used to calculate wavelength stability; Indicates the current moment; Represents a time index; Indicates at time The wavelength value of represents the average value of wavelength; Indicates the current time Temperature value; express Temperature value at the moment; Indicates the length of the time window used to calculate the temperature trend; The weight coefficient representing the temperature change trend; The weight coefficient representing the abnormal spectral fluctuation; represents the total number of spectral features; The index representing the spectral feature value; Indicates the current time The spectrum eigenvalues; express The moment spectrum feature values.

[0016] Furthermore, by real-time monitoring of the laser operating status, the change trend of the laser performance indicators is predicted, and the laser operating parameters are adjusted according to the prediction results to achieve real-time performance optimization, including; Obtain historical operating status data of ytterbium-doped tunable fiber lasers, including output power, lasing wavelength, temperature, and spectral characteristics; Using historical operating status data to train a preset prediction model, and adjusting prediction model parameters to obtain a trained prediction model; Monitor the operating status data of the ytterbium-doped tunable fiber laser in real time, and input the real-time monitored operating status data into the trained prediction model to calculate the change trend of the laser performance indicators in the future to obtain the prediction results, including the rising, falling or stable trend information of the performance indicators; According to the changing trend of laser performance indicators in the future, combined with the preset optimization strategy, including maintaining output power stability and controlling wavelength drift, corresponding adjustment instructions are generated, including adjusting pump power, changing resonant cavity length, and controlling temperature operation; Adjustment instructions are sent to the Ytterbium-doped tunable fiber laser to dynamically adjust the operating parameters to achieve real-time performance optimization.

[0017] In a second aspect, a computing device includes: one or more processors; The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0018] In a third aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described above is implemented.

[0019] The above solution of the present invention includes at least the following beneficial effects: The data acquisition module can collect various data of the ytterbium-doped tunable fiber laser in real time and accurately during operation. The quantum processing module uses the powerful parallel processing capabilities of quantum computing devices to execute quantum algorithm programs and quickly process data. Compared with traditional algorithms, it improves the data processing speed and efficiency, allowing the system to process massive, high-dimensional laser data in a timely manner. Quantum algorithms have unique advantages in processing complex and nonlinear data, and can more accurately capture the changing trends of laser performance indicators and improve prediction accuracy. The feedback control module combines the reinforcement learning algorithm to generate feedback control signals based on the processing results of the quantum processing module, enabling the system to make more accurate decisions, optimize laser performance, and achieve real-time control.

[0020] The monitoring and analysis module monitors and analyzes the laser data in real time, and can promptly detect and identify potential faults during the operation of the laser. By generating early warning information, the system can take measures in advance to prevent the occurrence of faults, reduce downtime and maintenance costs caused by faults, and improve the reliability and stability of the laser. The real-time performance optimization module dynamically adjusts the laser operating parameters by monitoring the operating status of the laser in real time and combining the prediction results of the quantum processing module. This real-time optimization mechanism enables the laser to always maintain the best performance state, meet the ever-changing application requirements, and improve the efficiency and application effect of the laser. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of a fast data processing system for an ytterbium-doped tunable fiber laser based on a quantum algorithm provided in an embodiment of the present invention.

[0022] Figure 2 The invention provides a flow chart of a quantum algorithm-based ytterbium-doped tunable fiber laser data rapid processing system for real-time monitoring and analysis of laser data, identification and early warning of potential faults during laser operation, and generation of early warning information. DETAILED DESCRIPTION

[0023] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0024] like Figure 1 As shown, an embodiment of the present invention proposes a fast data processing system for an ytterbium-doped tunable fiber laser based on a quantum algorithm, comprising: The data acquisition module 11 is used to collect various data generated by the ytterbium-doped tunable fiber laser during operation, including the output power, lasing wavelength, and spectral characteristics of the laser; The quantum processing module 12 is used to execute a quantum algorithm program through a quantum computing device to process various data generated during the operation of the ytterbium-doped tunable fiber laser to obtain a processing result; A feedback control module 13 is used to generate a feedback control signal through a reinforcement learning algorithm according to the processing result of the quantum processing module, and send the feedback control signal to the ytterbium-doped tunable fiber laser to achieve optimization and real-time control of the laser performance; The monitoring and analysis module 14 is used to monitor and analyze the laser data in real time, identify and warn of potential faults during the operation of the laser, and generate warning information; The real-time performance optimization module 15 is used to monitor the operating status of the laser in real time, predict the change trend of the laser performance indicators, and adjust the laser operating parameters according to the prediction results to achieve real-time performance optimization.

[0025] In the embodiment of the present invention, the data acquisition module 11 collects key data (such as output power, lasing wavelength, spectral characteristics, etc.) of the ytterbium-doped tunable fiber laser during operation in real time and comprehensively, providing an accurate and reliable data basis for subsequent processing and analysis of the system, ensuring the integrity and accuracy of the data, and helping to improve the system's ability to monitor and evaluate the performance status of the laser.

[0026] The quantum processing module 12 uses the powerful computing power of quantum computing devices to execute efficient quantum algorithm programs to quickly process the collected massive, high-dimensional data, thereby improving the speed and efficiency of data processing. Quantum algorithms have unique advantages in processing complex, nonlinear data, and can more accurately extract useful information from the data, providing strong support for system decision-making and optimization.

[0027] The feedback control module 13 generates an optimized feedback control signal through a reinforcement learning algorithm according to the processing results of the quantum processing module, thereby realizing intelligent control and optimization of the laser performance. The reinforcement learning algorithm can continuously learn and adapt to the operating state of the laser, making the feedback control signal more accurate and effective, and improving the control performance and stability of the system.

[0028] The monitoring and analysis module 14 monitors and analyzes the laser data in real time, and can timely discover and identify potential faults in the operation of the laser, providing a powerful means for system fault warning and prevention. By generating early warning information, the system can take measures in advance to prevent the occurrence of faults, reduce downtime and maintenance costs caused by faults, and improve the reliability and availability of the laser.

[0029] The real-time performance optimization module 15 monitors the operating status of the laser in real time and dynamically adjusts the operating parameters of the laser in combination with the prediction results of the quantum processing module, thereby achieving real-time optimization of the laser performance.

[0030] This real-time optimization mechanism enables the laser to always maintain optimal performance, meet ever-changing application requirements, and improve the efficiency and application effect of the laser.

[0031] In a preferred embodiment of the present invention, various data generated by the Yb-doped tunable fiber laser during operation are collected, including the output power, lasing wavelength, and spectral characteristics of the laser, which may include: In an embodiment of the present invention, In a preferred embodiment of the present invention, a quantum algorithm program is executed by a quantum computing device to process various data generated during the operation of the ytterbium-doped tunable fiber laser to obtain processing results, which may include: Use the quantum computer as a quantum computing device, initialize the quantum computing device, and configure the operating environment, including the number of quantum bits, quantum gate set, and measurement method; The various data generated by the Yb-doped tunable fiber laser during operation, including output power, lasing wavelength, and spectral characteristics, are processed through the quantum approximate optimization algorithm, and the process of the quantum approximate optimization algorithm is set up, including the data encoding method, the operation sequence of the quantum gate, and the final measurement plan; On the quantum computing device, various data generated by the Yb-doped tunable fiber laser during operation are used as input data to debug and test the quantum approximate optimization algorithm program, and the program parameters are adjusted by comparing the program output with the expected results; When the quantum approximate optimization algorithm program is running, various data generated during the operation of the ytterbium-doped tunable fiber laser are encoded and converted into the state of quantum bits, and quantum transformations are performed on the quantum bits according to a predefined sequence of quantum gate operations; At the end of the quantum approximate optimization algorithm program, a measurement operation is performed to obtain the final state of the quantum bit, and the final state of the quantum bit is decoded into classical data to obtain the processing results, including the optimized output power value, lasing wavelength value, and spectral characteristic parameters.

[0032] In an embodiment of the present invention, according to the characteristics of the ytterbium-doped tunable fiber laser, a sensor that can accurately measure the output power, lasing wavelength and spectral characteristics is selected. For example, a power meter is used to measure the output power, and a spectrometer is used to measure the lasing wavelength and spectral characteristics. Before using the sensor, calibration is performed to ensure the accuracy of the measurement. The calibration process should follow the instructions for use of the sensor, including comparison with a known standard or adjustment of internal parameters. According to the interface type and protocol, the connection between the data acquisition device and the laser is configured. This involves setting parameters such as the IP address, port number, and baud rate.

[0033] Use programming languages ​​(such as Python, C++, LabVIEW, etc.) to write a data acquisition program to achieve communication with the laser and data acquisition. The program should be able to read the sensor data and convert it into an easy-to-process format (such as CSV, JSON, etc.). Integrate basic data processing functions such as filtering, denoising, data format conversion, etc. in the data acquisition program, which will help improve the accuracy and usability of the data.

[0034] Start the laser according to the laser's instructions and ensure that it is in a stable operating state. Start the data acquisition program and begin collecting the laser's output power, lasing wavelength, and spectral characteristics data. The program continuously collects data. During the data acquisition process, monitor the program's operating status and data acquisition quality. If any abnormal situation is found (such as data acquisition interruption, data abnormality, etc.), it should be handled in a timely manner. Store the collected data in a local hard disk, network storage device, or cloud storage. Ensure that the storage medium has sufficient capacity and reliability. Establish a data management system to classify, archive, and retrieve the collected data. This will facilitate subsequent data analysis and processing.

[0035] In a preferred embodiment of the present invention, according to the processing result of the quantum processing module, a feedback control signal is generated by a reinforcement learning algorithm, and the feedback control signal is sent to the ytterbium-doped tunable fiber laser to achieve optimization and real-time control of the laser performance, which may include: Initialize the reinforcement learning algorithm and define the state space, action space, reward function and learning rate parameters; the state space includes the current output power, lasing wavelength and spectral characteristic parameters of the Yb-doped tunable fiber laser; the action space includes the actions to adjust the laser parameters, including changing the pump power and adjusting the resonant cavity length; For each time step, the reinforcement learning algorithm selects a current state from the state space, uses the current state as the input data of the quantum approximate optimization algorithm, runs the quantum approximate optimization algorithm program, processes the ytterbium-doped tunable fiber laser data, and obtains the optimized output power value, lasing wavelength value, and spectral characteristic parameter processing results; According to the processing results and reward function, calculate the reward value of the current state, and update the strategy function according to the current state, action and reward value; According to the updated policy function, the action to be performed in the next time step is determined, and a feedback control signal is generated according to the determined action; The feedback control signal is sent to the ytterbium-doped tunable fiber laser, and the iterative process of state acquisition, quantum approximation optimization, reward calculation, strategy update, action selection, feedback signal generation and sending is continued until the preset performance indicators are reached to achieve optimization and real-time control of the laser performance.

[0036] In an embodiment of the present invention, the state space includes the current output power, lasing wavelength, and spectral characteristic parameters of the ytterbium-doped tunable fiber laser. These parameters are acquired in real time through the data acquisition module and used as the input of the reinforcement learning algorithm. The action space includes actions to adjust the laser parameters, such as changing the pump power, adjusting the resonant cavity length, etc. These actions are selected by the reinforcement learning algorithm based on the current state and are used to optimize the laser performance. The reward function is used to evaluate the quality of the current state and is usually related to the performance indicators of the laser (such as output power stability, wavelength accuracy, etc.). The design of the reward function needs to be adjusted according to the specific application requirements. The learning rate parameter controls the speed at which the strategy function is updated, affecting the convergence speed and stability of the reinforcement learning algorithm.

[0037] The reinforcement learning algorithm selects a current state from the state space, that is, obtains the output power, lasing wavelength, and spectral characteristic parameters of the current laser. The current state is used as the input data of the quantum approximate optimization algorithm to run the quantum approximate optimization algorithm program. The program uses the powerful computing power of the quantum computing device to quickly process the data of the ytterbium-doped tunable fiber laser to obtain the optimized output power value, lasing wavelength value, and spectral characteristic parameter processing results. According to the processing results and the reward function, the reward value of the current state is calculated. The reward value reflects the contribution of the current state to the optimization of laser performance. According to the current state, action and reward value, the policy function is updated. The policy function determines the probability distribution of the reinforcement learning algorithm to select an action in a given state. According to the updated policy function, the action to be performed in the next time step is determined. These actions will be used to adjust the parameters of the laser to optimize its performance. According to the determined action, a feedback control signal is generated and sent to the ytterbium-doped tunable fiber laser. The iterative process of state acquisition, quantum approximation optimization, reward calculation, strategy update, action selection, feedback signal generation and sending is continuously carried out until the preset performance indicators are reached (such as the output power stability reaches a certain threshold, the wavelength accuracy meets the requirements, etc.), and the iteration is stopped to achieve the optimization and real-time control of the laser performance.

[0038] Suppose there is an Yb-doped tunable fiber laser whose goal is to maintain the stability of the output power. When the reinforcement learning algorithm is initialized, the state space (output power, lasing wavelength, spectral characteristics), action space (changing pump power, adjusting resonant cavity length), reward function (output power stability) and learning rate parameters are defined. At each time step, the reinforcement learning algorithm selects the current state from the state space, that is, obtains the output power, lasing wavelength and spectral characteristic parameters of the current laser. Then, these parameters are used as input data of the quantum approximate optimization algorithm, and the quantum approximate optimization algorithm program is run to obtain the optimized output power value and lasing wavelength value. According to the optimized output results and reward function, the reward value of the current state is calculated. If the output power stability is high, the reward value is large; otherwise, the reward value is small. Then, according to the current state, action and reward value, the policy function is updated so that the reinforcement learning algorithm is more inclined to choose actions that can optimize the performance of the laser in subsequent time steps.

[0039] Assume that the current policy function selects increasing the pump power as the action for the next time step. Therefore, the reinforcement learning algorithm generates the corresponding feedback control signal and sends it to the Ytterbium-doped tunable fiber laser. After receiving the feedback control signal, the laser adjusts the pump power to optimize its performance. The reinforcement learning algorithm continues to acquire new state information and repeats the above process, continuously iterating state acquisition, quantum approximation optimization, reward calculation, strategy update, action selection, feedback signal generation and sending. Finally, when the output power stability reaches the preset performance index, the iteration is stopped to achieve the optimization and real-time control of the laser performance.

[0040] By combining the reinforcement learning algorithm with the quantum approximate optimization algorithm, the performance of the ytterbium-doped tunable fiber laser can be quickly optimized. This optimization method is more efficient and accurate than the traditional trial-and-error method or rule-based method. The reinforcement learning algorithm can dynamically adjust the laser parameters according to the current state to cope with changes in the external environment or fluctuations inside the laser. This real-time adjustment mechanism helps to maintain the stability of the laser performance and improve its reliability and availability. Through continuous iteration and optimization, the reinforcement learning algorithm can continuously learn and adapt to the operating state of the laser and find the final parameter combination to maximize the optimization of the laser performance. The automated and intelligent optimization process reduces the need for manual intervention and reduces labor and time costs. At the same time, it also avoids errors and uncertainties caused by human factors.

[0041] In a preferred embodiment of the present invention, the calculation formula of the reward function is: ; in, Represents the reward function value; Represents the weight coefficient of the output power error term; Indicates the current output power of the laser; Indicates the target output power of the laser; , represents a positive number; represents the weight coefficient of the lasing wavelength error term; Indicates the current lasing wavelength of the laser; Indicates the target lasing wavelength of the laser; Represents the weight coefficient of the spectral characteristic similarity term; A vector representing the current spectral characteristics of the laser; Represents the target spectral characteristic vector of the laser.

[0042] In the embodiment of the present invention, the weight coefficient is set , , , these coefficients determine the relative importance of output power error, lasing wavelength error, and spectral characteristic similarity in the reward function. Set a positive number and , which is used to avoid the situation where the denominator of the error term is zero when the output power or laser wavelength is close to zero, thereby ensuring the stability of the reward function. Get the current output power of the laser from the data acquisition module , Current lasing wavelength and the current spectral characteristic vector . Set the target output power of the laser , target lasing wavelength and the target spectral characteristic vector .

[0043] Using the formula Calculate the output power error term, which reflects the deviation between the current output power and the target output power. The smaller the deviation, the higher the reward. Use the formula Calculate the lasing wavelength error term, which reflects the deviation between the current lasing wavelength and the target lasing wavelength. The smaller the deviation, the higher the reward. Use the cosine similarity formula Calculate the spectral characteristic similarity item, which reflects the similarity between the current spectral characteristic and the target spectral characteristic. The higher the similarity, the higher the reward.

[0044] The above three items are divided into , and Perform weighted summation to obtain the reward function value . Reward function value It reflects the overall matching degree between the current laser state and the target state. In the reinforcement learning algorithm, the reward function value is used To update the policy function.

[0045] The reward function comprehensively considers the output power error, laser wavelength error and spectral characteristic similarity, and can more comprehensively evaluate the effect of laser performance optimization. This comprehensive evaluation method helps to find the optimal parameter combination and improve the accuracy of laser performance optimization. Through continuous iteration and optimization, the reward function can guide the reinforcement learning algorithm to continuously learn and adapt to the operating state of the laser. This real-time adjustment mechanism helps to maintain the stability of laser performance and reduce the impact of external environment or internal fluctuations on laser performance. The various weight coefficients in the reward function can be adjusted according to specific application requirements to achieve comprehensive optimization of laser performance. For example, in applications that require higher output power stability, the weight coefficient of the output power error term can be increased. By reasonably designing the reward function and weight coefficient, the reinforcement learning algorithm can be guided to find the final solution faster. This optimization method is more efficient and faster than the traditional trial and error method or rule-based method.

[0046] In a preferred embodiment of the present invention, the calculation formula of the strategy function is: ; in, represents the parameter vector; represents the learning rate; Indicates the maximum number of time steps in a round; represents the time step index; Represents the discount factor of Power; Represents the time step The instant reward value obtained; Indicates in status Take action The eigenvector of , Indicates action; Represents the parameter vector The transpose of Indicates in status Take action The eigenvector of Indicates in status Take action The feature vector of .

[0047] In the embodiment of the present invention, the parameter vector is set , which is the core of the policy function and is used to decide which actions to choose in a given state. Set to a zero vector or a random vector, depending on the application requirements and algorithm implementation. Learning rate Controls the speed at which the strategy function is updated, affecting the speed and stability of the algorithm convergence. Used to weigh the importance of current rewards and future rewards, usually set between 0 and 1. In one round, record each time step Status ,action , Instant Rewards and the corresponding eigenvector At the end of the round, a complete set of data is obtained for subsequent policy function updates.

[0048] For each time step , calculate all possible actions The eigenvector of The weighted sum of .

[0049] The weight is , which means that in the state Next select action The probability of . Using the formula Calculate the policy gradient, which reflects the gap between the current policy function and the optimal policy and is used to guide the update of the policy function. Use policy gradient and learning rate Update parameter vector , the updated parameter vector will be used for strategy selection in the next round.

[0050] By updating the parameter vector through policy gradient, the relationship between state and action can be more accurately reflected, thereby improving the accuracy of policy selection. This gradient-based optimization method is more direct and efficient than the traditional value function-based optimization method. By reasonably setting the learning rate and discount factor, the algorithm can be guided to find the final solution faster while maintaining the stability of the algorithm. This optimization method is more efficient and stable than the traditional trial and error method or random search method. The policy gradient algorithm allows optimization in a continuous action space, thereby achieving fine adjustment of laser parameters. This continuous optimization method is more flexible and accurate than discrete optimization methods. By continuously iterating and optimizing the policy function, comprehensive optimization of laser performance can be achieved, including output power, laser wavelength, and spectral characteristics. This comprehensive optimization method helps to improve the overall performance and reliability of the laser.

[0051] In a preferred embodiment of the present invention, real-time monitoring and analysis of laser data, identification and early warning of potential faults during laser operation, and generation of early warning information may include: Collect real-time data from ytterbium-doped tunable fiber lasers, including laser output power, wavelength, temperature, and spectral information; Extract key features from real-time data, including the rate of change of laser output power, wavelength stability, temperature variation trend, and abnormal spectral fluctuations; Using the normal operation data in the historical data set to train the preset anomaly detection model to obtain a trained anomaly detection model; After training, the anomaly detection model processes and analyzes the key features, calculates the degree of deviation between the current operating state of the laser and the normal state, and obtains the deviation degree index value to evaluate the degree of deviation between the laser and the normal state; According to the indicators obtained by the trained anomaly detection model, a threshold is set. When the indicator exceeds the threshold, the fault warning mechanism is triggered and warning information is generated, including the fault type, fault severity, and recommended maintenance measures.

[0052] In an embodiment of the present invention, data is collected in real time by sensors integrated on the laser (such as photodetectors, spectrometers, temperature sensors, etc.), and the data includes laser output power (measured by photodetectors), wavelength (spectral peak position analyzed by spectrometers), temperature (read directly by temperature sensors), and spectral information (complete spectrum obtained by spectrometers). The collected data is sampled at a certain frequency (such as every second or every few minutes) and stored in a data cache or database. The collected real-time data is preprocessed, such as denoising, normalization, etc., to improve data quality. The rate of change of the laser output power is extracted by calculating the ratio of the power difference between adjacent sampling points to the sampling interval. The stability of the calculated wavelength can be evaluated by statistically analyzing the standard deviation or range of variation of the wavelength over a period of time. The trend of temperature change can be analyzed by fitting a linear or nonlinear model of the temperature data to the time axis.

[0053] Detecting abnormal fluctuations in the spectrum can be achieved by comparing the difference between the current spectrum and the normal spectrum template, or by applying spectral anomaly detection algorithms (such as principal component analysis, cluster analysis, etc.).

[0054] Select data samples from the historical data set during normal operation to ensure that the samples are representative and diverse. Select a suitable anomaly detection model. Use the selected normal operation data to train the model and adjust the model parameters to optimize the detection performance. After training, verify and test the model to ensure its detection accuracy on normal data. Input the extracted key features into the trained anomaly detection model.

[0055] The model processes and analyzes the features and calculates the degree of deviation between the current operating state of the laser and the normal state.

[0056] The degree of deviation can be quantified by metrics such as anomaly score, probability, or distance output by the model.

[0057] Based on the output of the anomaly detection model, calculate the deviation index value. The index value is used to evaluate the degree of deviation between the current operating state of the laser and the normal state. Set a threshold based on historical data and actual needs.

[0058] When the deviation index value exceeds the threshold, the fault warning mechanism is triggered. The warning information includes the fault type (such as power drop, wavelength drift, overtemperature, etc.), fault severity (such as minor, serious, emergency, etc.) and recommended maintenance measures (such as checking fiber connections, adjusting laser parameters, contacting professional maintenance, etc.).

[0059] Warning information can be conveyed to users or maintenance personnel through interface display, email notification, SMS sending, etc.

[0060] Assume that an Ytterbium-doped tunable fiber laser is being monitored: Real-time data collection: The machine collects laser output power, wavelength, temperature and spectral information once per second.

[0061] Calculate the power difference between two adjacent seconds to obtain the rate of change sequence. Fit the temperature data in the past five minutes to the linear model of the time axis. Compare the difference between the current spectrum and the normal spectrum template and calculate the difference value. Use the normal operation data in the past month to train an anomaly detection model, input the extracted key features into the trained model, and obtain the deviation index value. Set the threshold to 0.8 (assuming the index value ranges from 0 to 1). When the deviation index value exceeds 0.8, the fault warning mechanism is triggered and the warning information is generated: "Fault type: power drop; Fault severity: severe; Recommended maintenance measures: check the fiber connection and laser power supply." By monitoring and analyzing laser data in real time, potential faults can be discovered in time to prevent the faults from further deteriorating and causing the laser to shut down or be damaged. This helps improve the operational reliability of the laser and ensures that it can work properly when needed. With early warning information, maintenance personnel can take maintenance measures before a fault occurs to avoid unnecessary downtime and repair costs, which helps reduce the maintenance cost of the laser and improve its efficiency. By continuously monitoring and analyzing laser data, its operating parameters can be optimized and performance indicators such as output power, wavelength stability, and spectral quality can be improved, which helps improve the overall performance of the laser and meet a wider range of application needs. With early warning information and recommended maintenance measures, users can manage lasers more easily and reduce troubles and losses caused by faults, which helps enhance the user experience and improve their satisfaction and trust in the laser.

[0062] In a preferred embodiment of the present invention, the calculation formula for the degree of deviation between the current operating state of the laser and the normal state is: ; in, Indicates the degree of deviation between the current operating state of the laser and the normal state; The weight coefficient representing the rate of change of laser output power; Indicates the current output power of the laser; Indicates that the laser is Output power at the moment; The weight coefficient representing the wavelength stability; represents the length of the time window used to calculate wavelength stability; Indicates the current moment; Represents a time index; Indicates at time The wavelength value of represents the average value of wavelength; Indicates the current time Temperature value; express Temperature value at the moment; Indicates the length of the time window used to calculate the temperature trend; The weight coefficient representing the temperature change trend; The weight coefficient representing the abnormal spectral fluctuation; represents the total number of spectral features; The index representing the spectral feature value; Indicates the current time The spectrum eigenvalues; express The moment spectrum feature values.

[0063] In the embodiment of the present invention, the current output power is collected in real time by the built-in power sensor of the laser. And the output power at the previous moment . Use the spectrometer to measure and record the current moment and the past moment (from arrive ) wavelength value . Calculate the average wavelength . Get the current time through the temperature sensor Temperature value and the past moment (from arrive ) is used to calculate the temperature change trend. Use the spectrometer to collect the current time and the previous moment The spectral information of Spectral eigenvalues and .

[0064] Calculate the difference between the output power at the current moment and the previous moment, that is, , and multiply by the weight coefficient , and the laser output power change rate is obtained. Calculate the past The average value of the sum of squares of the deviations of the wavelength value at each moment from the average wavelength, that is, , and multiply by the weight coefficient , get the wavelength stability. Calculate the current time and the past The average rate of change of temperature value at each moment is , and multiply by the weight coefficient , and get the temperature change trend. Calculate the average value of the absolute difference between the current moment and the previous moment's spectral characteristic value, that is, , and multiply by the weight coefficient , and get the abnormal spectrum fluctuation. Add the above deviations to get the deviation between the current operating state of the laser and the normal state. .

[0065] By integrating multiple factors such as the rate of change of laser output power, wavelength stability, temperature change trend, and abnormal spectral fluctuations, the formula can more accurately reflect the degree of deviation of the current operating state of the laser from the normal state, thereby improving the accuracy of fault detection. Since the formula can calculate the degree of deviation in real time, it can trigger the early warning mechanism at the early stage of the fault, providing maintenance personnel with sufficient response time to avoid further deterioration of the fault. By analyzing the components of the degree of deviation, the main causes of the fault can be identified, so as to formulate a more targeted maintenance strategy. For example, if the degree of deviation is mainly caused by the rate of change of laser output power, it may be necessary to check components such as fiber connections or laser power supplies. Through early warning and targeted maintenance, unnecessary downtime and maintenance costs can be reduced. At the same time, since the formula can accurately reflect the operating status of the laser, it can also avoid excessive or insufficient maintenance. By continuously monitoring and analyzing the degree of deviation, the operating parameters of the laser can be optimized, and its performance indicators such as output power, wavelength stability, and spectral quality can be improved. This helps to improve the overall performance of the laser and meet a wider range of application needs. The application of this formula not only helps to improve the reliability and performance of the laser, but also provides strong support for the research and development and improvement of laser technology. By accumulating a large amount of operating data and failure cases, we can further study the operating mechanism and failure mode of the laser and promote the continuous development of laser technology.

[0066] In a preferred embodiment of the present invention, by real-time monitoring of the laser operating state, predicting the change trend of the laser performance index, and adjusting the laser operating parameters according to the prediction results, the performance is optimized in real time, which may include: Obtain historical operating status data of ytterbium-doped tunable fiber lasers, including output power, lasing wavelength, temperature, and spectral characteristics; Using historical operating status data to train a preset prediction model, and adjusting prediction model parameters to obtain a trained prediction model; Monitor the operating status data of the ytterbium-doped tunable fiber laser in real time, and input the real-time monitored operating status data into the trained prediction model to calculate the change trend of the laser performance indicators in the future to obtain the prediction results, including the rising, falling or stable trend information of the performance indicators; According to the changing trend of laser performance indicators in the future, combined with the preset optimization strategy, including maintaining output power stability and controlling wavelength drift, corresponding adjustment instructions are generated, including adjusting pump power, changing resonant cavity length, and controlling temperature operation; Adjustment instructions are sent to the Ytterbium-doped tunable fiber laser to dynamically adjust the operating parameters to achieve real-time performance optimization.

[0067] In an embodiment of the present invention, the operating status data of the laser is continuously collected by connecting to a sensor and a data acquisition system of an ytterbium-doped tunable fiber laser. The data includes key performance indicators such as output power, lasing wavelength, temperature, and spectral characteristics. The collected data is stored in a database. A prediction model is selected, such as a time series analysis model. The historical operating status data is used as input to train the prediction model. During the training process, the parameters of the model, such as the learning rate, the number of hidden layers, the number of neurons, etc., are adjusted to optimize the prediction performance of the model. The prediction accuracy of the model is evaluated using methods such as cross-validation or leave-out validation, and the model parameters are further adjusted based on the evaluation results.

[0068] The operating status data of the laser is continuously monitored in real time through sensors and data acquisition systems. The real-time monitoring data includes key performance indicators such as output power, laser wavelength, temperature, and spectral characteristics. The real-time monitoring data is transmitted to the processing system in real time. The real-time monitored operating status data is used as input to the trained prediction model. The prediction model calculates the predicted values ​​of laser performance indicators (such as output power and laser wavelength) in the future based on the input data and the internal logic of the model. Assume that in the future The output power prediction value at the moment is The predicted lasing wavelength is ;in, It’s the future Output power prediction value at a moment; It is at the moment The predicted value of the laser output power when ; It is at the moment The predicted value of the laser output power when ; It is at the moment The predicted value of the laser output power when ; Indicates the future The predicted value of the lasing wavelength at a certain moment.

[0069] By comparing the predicted value with the current value or historical value, the change trend of the performance indicator is determined, including rising, falling or remaining stable. The performance indicator change trend information calculated by the prediction model is output as the prediction result. The prediction result includes the rising, falling or remaining stable trend information of the performance indicator, as well as the corresponding predicted value and time range.

[0070] Assume that an Ytterbium-doped tunable fiber laser is being monitored: The sensor monitors the laser output power of 100W, the laser wavelength of 1064nm, the temperature of 25℃, and the spectral characteristics are normal. The real-time monitoring data is input into the trained prediction model. The prediction model calculates that in the next hour, the laser output power will increase by 5W to 105W, the laser wavelength will remain stable at around 1064nm, and the temperature will rise by 1℃ to 26℃. Based on the prediction results, the machine determines that the increase in output power may cause system instability, so it generates an adjustment instruction to reduce the pump power by 5% to control the stability of the output power.

[0071] By real-time monitoring and predicting the changing trends of laser performance indicators, factors that may lead to unstable performance can be discovered and adjusted in a timely manner, thereby improving the performance stability of the laser. By avoiding the operation of the laser under extreme or unfavorable conditions, the service life of the laser can be extended and the frequency of maintenance and replacement can be reduced.

[0072] By dynamically adjusting the laser's operating parameters, such as pump power, resonant cavity length, and temperature, based on the prediction results, the laser's performance can be optimized, and key performance indicators such as output power and wavelength stability can be improved. By optimizing the laser's performance in real time, production efficiency can be improved, and production interruptions or scrap rates caused by unstable laser performance can be reduced. By extending the laser's service life, reducing the frequency of repairs and replacements, and optimizing operating parameters, the laser's operating costs can be reduced, and overall economic benefits can be improved.

[0073] The embodiment of the present invention further provides a computing device, including: a processor, a memory storing a computer program, and when the computer program is executed by the processor, the system described above is executed. All implementation methods in the above system embodiment are applicable to this embodiment and can achieve the same technical effect.

[0074] The embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the system described above. All implementations in the above system embodiment are applicable to this embodiment and can achieve the same technical effect.

[0075] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A fast data processing system for ytterbium-doped tunable fiber laser based on quantum algorithm, characterized by: include: Data acquisition module, used to collect various data generated by the Yb-doped tunable fiber laser during operation, including the output power, lasing wavelength, and spectral characteristics of the laser; A quantum processing module is used to execute a quantum algorithm program through a quantum computing device to process various data generated during the operation of the ytterbium-doped tunable fiber laser to obtain a processing result; A feedback control module is used to generate a feedback control signal through a reinforcement learning algorithm according to the processing result of the quantum processing module, and send the feedback control signal to the ytterbium-doped tunable fiber laser to achieve optimization and real-time control of the laser performance; The monitoring and analysis module is used to monitor and analyze laser data in real time, identify and warn of potential faults during laser operation, and generate warning information; The real-time performance optimization module is used to monitor the operating status of the laser in real time, predict the changing trend of the laser performance indicators, and adjust the laser operating parameters according to the prediction results to achieve real-time performance optimization.

2. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 1, characterized in that: The quantum algorithm program is executed by the quantum computing device to process various data generated by the Yb-doped tunable fiber laser during operation to obtain processing results, including: Use the quantum computer as a quantum computing device, initialize the quantum computing device, and configure the operating environment, including the number of quantum bits, quantum gate set, and measurement method; The various data generated by the Yb-doped tunable fiber laser during operation, including output power, lasing wavelength, and spectral characteristics, are processed through the quantum approximate optimization algorithm, and the process of the quantum approximate optimization algorithm is set up, including the data encoding method, the operation sequence of the quantum gate, and the final measurement plan; On the quantum computing device, various data generated by the Yb-doped tunable fiber laser during operation are used as input data to debug and test the quantum approximate optimization algorithm program, and the program parameters are adjusted by comparing the program output with the expected results; When the quantum approximate optimization algorithm program is running, various data generated during the operation of the ytterbium-doped tunable fiber laser are encoded and converted into the state of quantum bits, and quantum transformations are performed on the quantum bits according to a predefined sequence of quantum gate operations; At the end of the quantum approximate optimization algorithm program, a measurement operation is performed to obtain the final state of the quantum bit, and the final state of the quantum bit is decoded into classical data to obtain the processing results, including the optimized output power value, lasing wavelength value, and spectral characteristic parameters.

3. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 2, characterized in that: Based on the processing results of the quantum processing module, a feedback control signal is generated through a reinforcement learning algorithm and sent to the Ytterbium-doped tunable fiber laser to achieve optimization and real-time control of the laser performance, including: Initialize the reinforcement learning algorithm and define the state space, action space, reward function and learning rate parameters; the state space includes the current output power, lasing wavelength and spectral characteristic parameters of the Yb-doped tunable fiber laser; the action space includes the actions to adjust the laser parameters, including changing the pump power and adjusting the resonant cavity length; For each time step, the reinforcement learning algorithm selects a current state from the state space, uses the current state as the input data of the quantum approximate optimization algorithm, runs the quantum approximate optimization algorithm program, processes the ytterbium-doped tunable fiber laser data, and obtains the optimized output power value, lasing wavelength value, and spectral characteristic parameter processing results; According to the processing results and reward function, calculate the reward value of the current state, and update the strategy function according to the current state, action and reward value; According to the updated policy function, the action to be performed in the next time step is determined, and a feedback control signal is generated according to the determined action; The feedback control signal is sent to the ytterbium-doped tunable fiber laser, and the iterative process of state acquisition, quantum approximation optimization, reward calculation, strategy update, action selection, feedback signal generation and sending is continued until the preset performance indicators are reached to achieve optimization and real-time control of the laser performance.

4. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 3, characterized in that: The reward function is calculated as: ； in, Represents the reward function value; Represents the weight coefficient of the output power error term; Indicates the current output power of the laser; Indicates the target output power of the laser; , represents a positive number; represents the weight coefficient of the lasing wavelength error term; Indicates the current lasing wavelength of the laser; Indicates the target lasing wavelength of the laser; Represents the weight coefficient of the spectral characteristic similarity term; A vector representing the current spectral characteristics of the laser; Represents the target spectral characteristic vector of the laser.

5. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 4, characterized in that: The calculation formula of the strategy function is: ； in, represents the parameter vector; represents the learning rate; Indicates the maximum number of time steps in a round; represents the time step index; Represents the discount factor of Power; Represents the time step The instant reward value obtained; Indicates in status Take action The eigenvector of , Indicates action; Represents the parameter vector The transpose of Indicates in status Take action The eigenvector of Indicates in status Take action The feature vector of .

6. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 5, characterized in that: Monitor and analyze laser data in real time, identify and warn of potential faults during laser operation, and generate warning information, including: Collect real-time data from ytterbium-doped tunable fiber lasers, including laser output power, wavelength, temperature, and spectral information; Extract key features from real-time data, including the rate of change of laser output power, wavelength stability, temperature variation trend, and abnormal spectral fluctuations; Using the normal operation data in the historical data set to train the preset anomaly detection model to obtain a trained anomaly detection model; After training, the anomaly detection model processes and analyzes the key features, calculates the degree of deviation between the current operating state of the laser and the normal state, and obtains the deviation degree index value to evaluate the degree of deviation between the laser and the normal state; According to the indicators obtained by the trained anomaly detection model, a threshold is set. When the indicator exceeds the threshold, the fault warning mechanism is triggered and warning information is generated, including the fault type, fault severity, and recommended maintenance measures.

7. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 6, characterized in that: The calculation formula for the degree of deviation between the current operating state of the laser and the normal state is: ； in, Indicates the degree of deviation between the current operating state of the laser and the normal state; The weight coefficient representing the rate of change of laser output power; Indicates the current output power of the laser; Indicates that the laser is Output power at the moment; The weight coefficient representing the wavelength stability; represents the length of the time window used to calculate wavelength stability; Indicates the current moment; Represents a time index; Indicates at time The wavelength value of represents the average value of wavelength; Indicates the current time Temperature value; express Temperature value at the moment; Indicates the length of the time window used to calculate the temperature trend; The weight coefficient representing the temperature change trend; The weight coefficient representing the abnormal spectral fluctuation; represents the total number of spectral features; The index representing the spectral feature value; Indicates the current time The spectrum eigenvalues; express The moment spectrum feature values.

8. The quantum algorithm-based Ytterbium-doped tunable fiber laser data rapid processing system according to claim 7, characterized in that: By real-time monitoring of the laser operating status, predicting the change trend of laser performance indicators, and adjusting the laser operating parameters according to the prediction results, the performance can be optimized in real time, including; Obtain historical operating status data of ytterbium-doped tunable fiber lasers, including output power, lasing wavelength, temperature, and spectral characteristics; Using historical operating status data to train a preset prediction model, and adjusting prediction model parameters to obtain a trained prediction model; Monitor the operating status data of the ytterbium-doped tunable fiber laser in real time, and input the real-time monitored operating status data into the trained prediction model to calculate the change trend of the laser performance indicators in the future to obtain the prediction results, including the rising, falling or stable trend information of the performance indicators; According to the changing trend of laser performance indicators in the future, combined with the preset optimization strategy, including maintaining output power stability and controlling wavelength drift, corresponding adjustment instructions are generated, including adjusting pump power, changing resonant cavity length, and controlling temperature operation; Adjustment instructions are sent to the Ytterbium-doped tunable fiber laser to dynamically adjust the operating parameters to achieve real-time performance optimization.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the system according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, and when the program is executed by a processor, the system according to any one of claims 1 to 8 is implemented.

Images