Long-term locking oriented multi-loop cooperative control method and system for laser

Abstract

The application discloses a kind of long-term locking-oriented laser multi-loop cooperative control method and system, comprising: generating frequency discrimination signal related to laser frequency deviation;The frequency discrimination signal is input into the first controller, with the zero-crossing point of the frequency discrimination signal as the first set point, to generate the first control signal for driving the first actuator of the laser;The first control signal is input into the second controller, with the zero point of the first control signal as the second set point, to generate the second control signal for driving the second actuator of the laser;The second control signal is input into the third controller, with the deviation amount of the second control signal deviating from the center value of its linear operating region as input, with the center value of the linear operating region as the third set point, to generate the third control signal for driving the third actuator of the laser;Wherein, the response bandwidth of the first actuator, the second actuator and the third actuator decreases in turn, and the dynamic range increases in turn.

Description

Technical Field

[0001] This invention relates to the field of laser frequency stabilization technology, and more specifically, to a multi-loop collaborative control method and system for long-term locked lasers. Background Technology

[0002] Long-term frequency stability of lasers is a crucial indicator in fields such as precision measurement, quantum sensing, and optical communication. In existing technologies, frequency stabilization systems often employ multiple actuators (such as current-driven or piezoelectric ceramic PZTs) that independently respond to the same frequency discrimination error signal. However, due to the significant differences in the response bandwidth of each actuator, instantaneous adjustments in the fast loop can directly perturb the slow loop through the shared error signal path, leading to inter-loop coupling and limiting long-term locking performance.

[0003] To address the multi-loop coupling problem in laser frequency stabilization systems, existing technology (patent CN118825767A) proposes integrating the control unit onto a system-on-a-chip (SoC) and employing a two-stage cascaded PID controller: the first-stage PID calculates and outputs a PZT feedback control signal for rapid modulation; the second-stage PID uses the output of the first-stage PID as input to output a temperature feedback control signal for slow modulation. This structure achieves loop decoupling to a certain extent. However, during long-term operation, influenced by factors such as ambient temperature drift, the temperature crystal, acting as an outer-loop actuator, continuously increases its control input towards the boundary, eventually exceeding the controllable range and causing lockout. The patent's solution is to add an additional output channel with a larger dynamic range. When the temperature crystal's control signal is detected approaching the saturation boundary, overflow compensation is performed through this channel to pull the temperature crystal back from the saturation edge.

[0004] However, this overflow compensation architecture has inherent limitations: First, it is an event-driven, passive remediation mechanism that only triggers compensation when the temperature crystal approaches saturation, rather than real-time collaborative control; second, the compensation process is essentially the intervention and withdrawal of the control channel, which may introduce transient disturbances and affect locking continuity. Especially for lasers such as Faraday lasers that are extremely sensitive to PZT tuning, frequent compensation actions of the outer loop control quantity will be transmitted to the inner loop through cascade links, severely affecting the long-term locking robustness of the system.

[0005] Therefore, there is an urgent need for a multi-loop control method that can alleviate actuator saturation, eliminate the need for control channel switching, and achieve full-time coordination in order to address the above challenges. Summary of the Invention

[0006] This invention addresses the problem of long-term lock-up in existing laser frequency stabilization technologies, which is susceptible to actuator saturation and loop coupling. It proposes a multi-loop cooperative control method and system for lasers designed to address this issue. This method redefines the cooperative logic between control loops, shifting the outer loop's control objective from "directly correcting frequency deviation" to "maintaining the inner loop control output at the optimal operating point in the best possible way." Furthermore, it utilizes a heterogeneous hardware architecture to implement differentiated deployment of three-level PID controllers, thereby resolving the long-term lock-up challenge.

[0007] According to a first aspect of the embodiments of this application, a method for multi-loop cooperative control of a laser with long-term lock-in is provided, comprising: Generate a frequency discrimination signal related to the laser frequency deviation; The frequency discrimination signal is input into the first controller, and a first control signal for driving the first actuator of the laser is generated using the zero-crossing point of the frequency discrimination signal as the first set point. The first control signal is input into the second controller, and a second control signal for driving the second actuator of the laser is generated with the zero point of the first control signal as the second set point. The second control signal is input to the third controller, and the deviation of the second control signal from the center value of its linear working area is used as input. The center value of the linear working area is used as the third set point to generate a third control signal for driving the third actuator of the laser. The response bandwidth of the first actuator, the second actuator, and the third actuator decreases sequentially, while their dynamic range increases sequentially.

[0008] Furthermore, the first actuator is a laser injection current control unit, the second actuator is a piezoelectric ceramic (PZT) drive unit, and the third actuator is a temperature control unit (TEC).

[0009] Furthermore, the control bandwidth of the first controller is in the MHz range, the control bandwidth of the second controller is in the kHz range, and the control bandwidth of the third controller is in the Hz range or below.

[0010] Furthermore, the first controller employs an incremental PID algorithm with fixed or adjustable parameters, implemented on the FPGA side of the on-chip system in a purely hardware logic manner, and ensures real-time performance at the highest sampling frequency through a multi-stage pipeline architecture.

[0011] Furthermore, the second controller adopts a fuzzy PID controller, which is implemented on the FPGA side of the system-on-a-chip; the input of the fuzzy PID controller is the first control signal and its rate of change, and the output is the correction amount of the PID parameters; its membership function uses a Gaussian function to achieve fine adjustment in the region near the zero point, and a trigonometric function to achieve fast response in the region far from the zero point.

[0012] Furthermore, the third controller is implemented in software on the ARM processor side of the on-chip system; its parameters are configured to have low sensitivity to small fluctuations in the input signal and to respond to long-term drift in a slow and smooth manner.

[0013] Furthermore, the frequency discrimination signal is generated based on modulation transfer spectrum (MTS) technology and obtained by detecting the laser signal transmitted through the atomic gas cell and then digitally demodulating it.

[0014] Furthermore, the digital demodulation is performed on the FPGA side of the system-on-chip, using the CORDIC algorithm to generate a DDS reference signal, which is then mixed and low-pass filtered before the discriminant signal is extracted.

[0015] According to a second aspect of the embodiments of this application, a multi-loop cooperative control system for a laser with long-term locking is provided, comprising: The frequency discrimination signal generation module is used to generate a frequency discrimination signal related to the laser frequency deviation; The first control module, deployed on the FPGA side of the system-on-a-chip, is configured to receive the frequency discrimination signal and generate a first control signal with the zero-crossing point of the frequency discrimination signal as the first set point, so as to drive the first actuator of the laser. The second control module, deployed on the FPGA side of the system-on-a-chip, is configured to take the zero point of the first control signal as the second set point, receive the first control signal and generate the second control signal to drive the second actuator of the laser. The third control module, deployed on the ARM processor side of the on-chip system, is configured to take the deviation of the second control signal from its linear working area center value as input, and take the linear working area center value as a third set point, receive the second control signal and generate a third control signal to drive the third actuator of the laser. The response bandwidth of the first actuator, the second actuator, and the third actuator decreases sequentially, while their dynamic range increases sequentially.

[0016] The technical solutions provided by the embodiments of this application may include the following beneficial effects: First, the three-level zero-level control collaborative architecture mitigates the risk of actuator saturation. The input to each outer loop is the output of the inner loop, and the setpoint of each outer loop is the optimal operating point (zero point or center value of the linear operating range) for the inner loop control signal. The temperature loop ensures the PZT control quantity remains at its center value, the PZT loop ensures the current control quantity returns to zero, and the current loop focuses on high-frequency noise suppression. The three-level loop progressively distributes the load, fundamentally mitigating the saturation risk of any single actuator.

[0017] Second, the heterogeneous hardware architecture enables differentiated deployment. The first controller is implemented on the FPGA side in a pure hardware pipeline, ensuring deterministic real-time response at MHz-level control bandwidth; the second controller, also implemented on the FPGA side, utilizes a longer control cycle to introduce fuzzy PID for adaptive adjustment; the third controller is implemented in software on the ARM side, fully leveraging the flexibility and low-load characteristics of the ARM processor for easy online parameter adjustment. This heterogeneous division of labor maximizes the performance of the three-level loop within limited hardware resources.

[0018] Third, the fuzzy PID membership function is deeply bound to the physical control objective. The membership function of the second controller uses a Gaussian function near zero for fine-tuning, a trigonometric function for fast response in the region far from zero, and an asymmetric design in the transition range to balance the adjustment intensity and prevent overshoot. This design directly embeds the control objective of "rapid zeroing and smooth locking" into the membership function design.

[0019] Fourth, the temperature loop's inertial response design. The third controller is deliberately designed to be insensitive to small fluctuations, with a slower response speed, and to eliminate static deviations through long-term accumulation, taking into account the thermal inertia of the temperature actuator. This avoids introducing new frequency disturbances into the temperature regulation action itself, making it particularly suitable for laser systems such as Faraday lasers that are extremely sensitive to temperature changes.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0022] Figure 1 This is a flowchart illustrating a multi-loop cooperative control method for a long-term locked laser, according to an exemplary embodiment.

[0023] Figure 2 This is a block diagram illustrating a multi-loop cooperative control method for a long-term locked laser, according to an exemplary embodiment.

[0024] Figure 3This is an exemplary embodiment illustrating the overall structure of a laser frequency stabilization system based on modulation-transfer spectrum, wherein: 1. Faraday laser; 2. Digital laser frequency stabilization circuit; 3. Modulation-transfer spectrum frequency stabilization optical path; 4. Optical isolator; 5. First half-wave plate; 6. First polarizing beam splitter prism; 101. Light-emitting diode; 102. Semiconductor cooler (TEC); 103. First GlanTeller prism; 104. First atomic gas cell; 105. Second GlanTeller prism; 106. Resonant cavity mirror; 107. Piezoelectric ceramic (PZT); 108. Current source; 201. Bandpass filter. 202. Amplifier; 203. Multiplier; 204. DDS signal generator; 205. Low-pass filter; 206. First controller (current loop); 207. Third controller (temperature loop); 208. Second controller (PZT loop); 301. Second half-wave plate; 302. Second polarizing beam splitter; 303. Third half-wave plate; 304. Electro-optic modulator (EOM); 305. First mirror; 306. Second mirror; 307. Second atomic gas cell; 308. Third polarizing beam splitter; 309. Convex lens; 310. Photodetector.

[0025] Figure 4 This is a block diagram of a second controller loop according to an exemplary embodiment.

[0026] Figure 5 The second controller input variable is shown according to an exemplary embodiment. error The membership function.

[0027] Figure 6 This is a block diagram of a multi-loop cooperative control system for a laser with long-term locking, according to an exemplary embodiment. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of systems and methods consistent with some aspects of this application.

[0029] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0030] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0031] Figure 1 This is a flowchart illustrating a multi-loop cooperative control method for a long-term locked laser, according to an exemplary embodiment. Figure 2 This is a block diagram illustrating a multi-loop cooperative control method for long-term locked lasers according to an exemplary embodiment. The method generates a frequency discrimination signal based on modulation transfer spectrum technology and implements three-level PID cooperative control on a heterogeneous on-chip system, and may include the following steps: S1: Generate a frequency discrimination signal based on the modulation transfer spectrum; Specifically, Figure 3 The structure of the laser frequency stabilization system based on modulation transfer spectrum in this embodiment is shown. The Faraday laser 1 mainly consists of a light-emitting diode 101, a semiconductor cooler 102, a first GlanTeller prism 103, a first atomic gas cell 104, a second GlanTeller prism 105, a resonant cavity mirror 106, a piezoelectric ceramic 107, and a current source 108. The laser output from the Faraday laser 1 is split into two beams by a first polarizing beam splitter prism 6 after passing through an optical isolator 4 and a first half-wave plate 5: one beam serves as the output laser, and the other is used for frequency stabilization, entering the modulation transfer spectrum optical path 3. The frequency-stabilized laser is first split into two beams—a pump beam and a probe beam—by a second half-wave plate 301 and a second polarizing beam splitter prism 302. The stronger beam serves as the pump beam, and the weaker beam serves as the probe beam. The pump light enters the electro-optic modulator 304 after passing through the third half-wave plate 303, and undergoes phase modulation under the drive of the DDS signal generator 204. It is then guided into the second atomic gas chamber 307 by the first reflecting mirror 305 and the third polarizing beam splitter 308. The probe light is guided by the second reflecting mirror 306 and enters the second atomic gas chamber 307 in a counter-current propagation manner to the pump light. Inside the second atomic gas chamber 307, the pump light and probe light interact through a nonlinear four-wave mixing effect, transferring the modulation signal loaded on the pump light to the probe light. The probe light carrying the modulation signal is transmitted through the convex lens 309 and received by the photodetector 310, converted into an electrical signal, and used as the input to the subsequent digital laser frequency stabilization circuit 2.

[0032] Because light undergoes nonlinear modulation, unwanted DC and high-frequency harmonic components are generated. To filter out these signals, the electrical signal first needs to pass through a pre-stage bandpass filter 201 to remove these high-frequency harmonic components and DC components. The modulated signal to be acquired is then amplified by amplifier 202, and after being acquired by a high-speed ADC and entering the on-chip system, it undergoes mixing by multiplier 203 within the FPGA, followed by filtering by low-pass filter 205 to extract the frequency discrimination signal of the modulation transfer spectrum. The frequency discrimination signal enters the first controller 206, the second controller 207, and the third controller 208, which respectively control the current source 108, the semiconductor cooler 102, and the piezoelectric ceramic 107. In this embodiment, the FPGA uses the CORDIC algorithm to generate a DDS reference signal. The frequency discrimination signal provides a precise indication of the deviation of the laser frequency relative to the atomic spectral line center at the zero-crossing point, and the signal polarity on both sides of the zero-crossing point indicates the direction of the frequency deviation.

[0033] S2: Input the frequency discrimination signal into the first controller, and use the zero-crossing point of the frequency discrimination signal as the first set point to generate a first control signal for driving the first actuator of the laser. Specifically, the following combination Figure 2 Provide a detailed explanation of the signal flow and collaborative control logic of the three-level PID controller in the digital processing domain.

[0034] The frequency discrimination signal is first sent to the first controller 206 (current loop). The first controller 206 adopts an incremental PID algorithm with adjustable parameters. k Control increment for each control cycle Calculate using the following formula:

[0035] in, For the first Input deviation per control cycle For the first Input deviation per control cycle For the first Input deviation per control cycle This is the proportional coefficient in the first PID control loop. The integral coefficient in the first PID control loop. The derivative coefficients in the first PID control loop are denoted as .

[0036] In this embodiment, ,in, Indicates the first controller loop The frequency discrimination signal feedback quantity for each control cycle. In this embodiment, the system sampling frequency is 50MHz, i.e., the sampling period is 20ns. The first controller 206 is implemented in the FPGA using pure hardware logic, and its hardware logic adopts a multi-stage pipeline architecture design.

[0037] Specifically, the computational path of the incremental PID algorithm is divided into a three-stage pipeline: First-stage pipeline: Calculate the... Input deviation per control cycle That is, the deviation of the frequency discrimination signal from the zero-crossing point, and at the same time, the first value is read from the holding register. Input deviation per control cycle and the Input deviation per control cycle ; Second-stage pipeline: Parallel computation of the scaling term Integral terms and differential terms ; Third-level pipeline: Multiply the above three items by their corresponding coefficients. , , And sum them up to obtain the control increment. Then, compared with the control amount of the previous cycle. Add them together and output the current control quantity. .

[0038] The first controller 206 uses the zero-crossing point of the frequency discrimination signal as the first setpoint. Based on the magnitude and direction of the frequency discrimination signal deviating from the zero-crossing point, it generates a first control signal through the aforementioned pipeline PID controller. The first control signal is output to the laser current driver via a digital-to-analog converter, and the frequency deviation is quickly compensated by adjusting the laser injection current. The control bandwidth of the first controller 206 is on the order of MHz, with the fastest response speed. It is mainly used to suppress high-frequency noise and transient disturbances, narrowing the laser linewidth to its limit.

[0039] S3: Input the first control signal into the second controller, and use the zero point of the first control signal as the second set point to generate a second control signal for driving the second actuator of the laser; Specifically, refer to Figure 4The second controller 208 uses the zero point of the first control signal as its second setpoint, meaning its control objective is to bring the first control signal to zero. The control bandwidth of the second controller 208 (PZT loop) is on the order of tens of kHz, corresponding to a control cycle of tens of microseconds. This timescale provides ample computational window for fuzzy inference. The second controller employs fuzzy PID control. In this embodiment, the second controller 208 is also implemented on the FPGA side, but its control cycle is longer. After completing the PID calculations, the FPGA has sufficient logic resources and time margin to perform fuzzy inference calculations.

[0040] In one embodiment, the second controller 208 may be constructed by connecting a fuzzy inference unit and a conventional PID controller in parallel. The input to the fuzzy inference unit is the difference between the first control signal and the set value. and its rate of change The output is PID parameters (proportional coefficient of the second controller). Integral coefficient Differential coefficients Correction amount , and This enables online adaptive adjustment of PID coefficients.

[0041] Input and its rate of change They are divided into seven fuzzy sets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB).

[0042] Membership function design, such as Figure 5 As shown. This embodiment, based on the control characteristics of the PZT loop—"rapid zeroing and smooth locking"—uses a combination of Gaussian membership functions and trigonometric membership functions: In the region near zero (ZO, NS, PS fuzzy sets), a Gaussian membership function is used to achieve fine and smooth fine-tuning and prevent overshoot; In regions far from zero (NB, NM, PM, PB fuzzy sets), triangular membership functions are used to quickly respond to larger deviations and drive PZT to act rapidly. In the two transition fuzzy sets NM and PM, an asymmetric design is adopted: a larger slope is set near the boundary to ensure the adjustment strength, and a smaller slope is set near the zero point to prevent overshooting the zero point.

[0043] Rate of change of input The membership functions follow the same principle: Gaussian functions are used when changes are slow, and trigonometric functions are used when changes are drastic.

[0044] , and The membership function is uniformly adopted as a triangular function to reduce the computational complexity of FPGA defuzzification and save resources.

[0045] Error after blurring (corresponding to the first control signal) and error change rate The rate of change of the corresponding first control signal enters the fuzzy inference system and performs inference according to the preset fuzzy rule table. Based on the PID adjustment experience accumulated in engineering practice, this embodiment summarizes the fuzzy inference rules for the goal of "resetting the inner loop control signal to zero", and its fuzzy rule table is shown in Table 1.

[0046] Table 1

[0047] Specifically, Adjustment principle: When the input deviates significantly from zero, increase... To drive the PZT to move quickly and return rapidly; when the input is close to zero, reduce... To prevent overshoot.

[0048] Specifically, Adjustment principle: When the input deviates significantly from zero, reduce... To prevent integral saturation; when the input is near zero and changes gradually, appropriately increase the value. To eliminate steady-state error.

[0049] Specifically, The adjustment principle is: when the rate of change of the input is large (about to cross zero), increase the [input rate]. To provide damping and prevent overshoot; reduce when the input changes gradually. To reduce sensitivity to noise.

[0050] Calculate the first digit based on the fuzzy rule table. ( ) fuzzy rules corresponding to , , Fuzzy membership , , The formula for its calculation is: ; ; ; In the formula, For error The corresponding membership degree, Error change rate The corresponding membership degree.

[0051] The weighted average method is used to convert fuzzy membership degrees into actual output values ​​of the controller. The defuzzification formula is as follows:

[0052] In the formula, They represent the first The membership degree of the output quantity of the fuzzy rule.

[0053] The increment obtained from defuzzification , and By multiplying each parameter by the corresponding proportional factor, the three parameters of the PID controller are adjusted so that the controller can adaptively optimize the control effect according to changes in the system state. The adjusted PID parameter calculation formula is as follows:

[0054] In the formula, These are the proportional, integral, and differential coefficients after tuning, respectively; These are the proportional factors, integral factors, and differential factors, respectively. These are the initial values ​​for the proportional, integral, and differential coefficients, respectively.

[0055] The second control signal generated by the second controller 208 is output to the PZT driver via a DAC. Frequency deviation is compensated by adjusting the resonant cavity length through the piezoelectric ceramic 107. The control bandwidth of the second controller is in the kHz range, with a moderate response speed, primarily responsible for compensating for low-to-mid-frequency disturbances. By returning the first control signal to zero, the second controller 208 actively transfers the static and quasi-static frequency loads from the current loop to itself, ensuring that the current loop always operates near zero output and reserving the maximum bidirectional dynamic range to cope with transient disturbances.

[0056] S4: Input the second control signal into the third controller, take the deviation of the second control signal from the center value of its linear working area as input, and take the center value of the linear working area as the third set point to generate a third control signal for driving the third actuator of the laser. Specifically, the second control signal is output to the PZT driver and also serves as the input signal to the third controller 207. The third controller 207 uses the center value of the linear operating range of the second control signal as its third setpoint, that is, its control objective is to keep the second control signal near this center value, thereby preventing the PZT from deviating from its linear operating range due to long-term drift.

[0057] Unlike the second controller 208, the temperature actuator controlled by the third controller 207 has unique thermal response characteristics: temperature changes have a significant impact on the laser frequency, and the temperature actuator itself has extremely high thermal inertia. Therefore, the third controller 207 employs pure PI control without introducing a derivative element to avoid unnecessary responses to high-frequency noise during temperature regulation.

[0058] The temperature loop only addresses long-term, slow frequency drift and does not respond to short-term fluctuations. In PI parameter design, the proportional gain... The value is set to a small value, which makes the controller less sensitive to small instantaneous fluctuations in the second control signal, thus avoiding frequent temperature regulation triggered by small, instantaneous fluctuations in the PZT control quantity.

[0059] When the second control signal continuously deviates from the center value due to long-term temperature drift, the PI controller uses a smaller proportional coefficient. and smaller integral coefficients The control increment is generated slowly and smoothly. The response speed of the temperature loop is deliberately designed to be slow, not to quickly pull the PZT back to the center, but to gradually correct it through long-term accumulation, so as to avoid the temperature regulation action itself introducing new frequency disturbances.

[0060] Integral coefficient The value is set to a small value, allowing the temperature loop to gradually eliminate static deviations over a long period. This design prevents integral saturation and ensures that the temperature regulation process is extremely smooth, avoiding overshoot or oscillation caused by excessive integral action.

[0061] The third controller 207 is deployed on the ARM processor side of the system-on-a-chip and runs in software. The ARM processor reads the current value of the second control signal from the FPGA via the on-chip bus, executes the PI control algorithm, calculates the third control signal, and then writes the result back to the FPGA via the on-chip bus. The FPGA then outputs the result to the TEC driver via a DAC. The control cycle of the third controller 207 is on the order of seconds, giving the ARM processor ample time to complete the calculation task.

[0062] The third control signal generated by the third controller 207 is output to the laser temperature control unit (TEC driver) via a digital-to-analog converter, compensating for long-term frequency drift by adjusting the laser's operating temperature. The third controller 207 has a control bandwidth in the Hz range or below, the slowest response speed, but the largest dynamic range, undertaking the task of compensating for long-term temperature drift. The core advantage of this architecture is that the current loop focuses on suppressing high-frequency noise, always maintaining a symmetrical dynamic range near zero output; the PZT loop handles mid-to-low frequency disturbance compensation and transfers the static load of the current loop to itself; the temperature loop compensates for long-term temperature drift in an inertial and stable manner, preventing the PZT from deviating from its linear operating region. The three loops are naturally decoupled in the frequency domain, each performing its own function, mitigating the saturation risk of any single actuator and achieving true long-term continuous locking.

[0063] The response bandwidth of the first actuator, the second actuator, and the third actuator decreases sequentially, while their dynamic range increases sequentially.

[0064] In this invention, the "actuator" refers to a physical device or apparatus that can respond to control signals and change the operating state of the laser to adjust the output laser frequency.

[0065] In a preferred embodiment of the present invention, the first actuator is a laser injection current control unit, which changes the carrier density and temperature of the active region of the laser by adjusting the injection current, thereby rapidly tuning the laser frequency. Its response bandwidth is on the order of MHz or above, and its dynamic range is relatively small.

[0066] The second actuator is a piezoelectric ceramic (PZT) drive unit. By changing the voltage applied to the PZT, the length of the resonant cavity is changed, thereby tuning the laser frequency. Its response bandwidth is in the kHz range and its dynamic range is moderate.

[0067] The third actuator is a temperature control unit (TEC and drive circuit), which tunes the laser frequency by changing the laser's operating temperature. Its response bandwidth is on the order of Hz or below, and its dynamic range is the largest.

[0068] It should be understood that the specific actuator types described above are merely illustrative and not intended to limit the invention. In other embodiments, the first, second, and third actuators may also be other frequency tuning devices with corresponding response bandwidth and dynamic range characteristics.

[0069] Corresponding to the aforementioned embodiments of the laser multi-loop cooperative control method for long-term locked lasers, this application also provides embodiments of a laser multi-loop cooperative control system for long-term locked lasers.

[0070] Figure 6 This is a block diagram illustrating a multi-loop cooperative control system for a laser with long-term locking, according to an exemplary embodiment. (Refer to...) Figure 6The device includes: Frequency discrimination signal generation module 1 is used to generate a frequency discrimination signal related to the laser frequency deviation; The first control module 2 is deployed on the FPGA side of the system-on-a-chip and is configured to receive the frequency discrimination signal and generate a first control signal with the zero-crossing point of the frequency discrimination signal as the first set point, so as to drive the first actuator of the laser. The second control module 3 is deployed on the FPGA side of the system-on-a-chip and is configured to take the zero point of the first control signal as the second set point, receive the first control signal and generate the second control signal to drive the second actuator of the laser. The third control module 4 is deployed on the ARM processor side of the on-chip system and is configured to take the deviation of the second control signal from its linear working area center value as input, and take the linear working area center value as a third set point, receive the second control signal and generate a third control signal to drive the third actuator of the laser. The response bandwidth of the first actuator, the second actuator, and the third actuator decreases sequentially, while their dynamic range increases sequentially.

[0071] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0072] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.

[0073] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A multi-loop cooperative control method for long-term locked lasers, characterized in that, include: Generate a frequency discrimination signal related to the laser frequency deviation; The frequency discrimination signal is input into the first controller, and a first control signal for driving the first actuator of the laser is generated using the zero-crossing point of the frequency discrimination signal as the first set point. The first control signal is input into the second controller, and a second control signal for driving the second actuator of the laser is generated with the zero point of the first control signal as the second set point. The second control signal is input to the third controller, and the deviation of the second control signal from the center value of its linear working area is used as input. The center value of the linear working area is used as the third set point to generate a third control signal for driving the third actuator of the laser. The response bandwidth of the first actuator, the second actuator, and the third actuator decreases sequentially, while their dynamic range increases sequentially.

2. The method according to claim 1, characterized in that, The first actuator is a laser injection current control unit, the second actuator is a piezoelectric ceramic drive unit, and the third actuator is a temperature control unit.

3. The method according to claim 1, characterized in that, The control bandwidth of the first controller is in the MHz range, the control bandwidth of the second controller is in the kHz range, and the control bandwidth of the third controller is in the Hz range or below.

4. The method according to claim 1, characterized in that, The first controller uses an incremental PID algorithm with fixed or adjustable parameters, and is implemented on the FPGA side of the on-chip system in a pure hardware logic manner. The real-time performance at the highest sampling frequency is guaranteed through a multi-stage pipeline architecture.

5. The method according to claim 1, characterized in that, The second controller is a fuzzy PID controller, implemented on the FPGA side of the system-on-a-chip. The input of the fuzzy PID controller is the first control signal and its rate of change, and the output is the correction amount of the PID parameters. Its membership function uses a Gaussian function to achieve fine adjustment in the region near zero, and a trigonometric function to achieve fast response in the region far from zero.

6. The method according to claim 1, characterized in that, The third controller is implemented in software on the ARM processor side of the system-on-a-chip.

7. The method according to claim 1, characterized in that, The frequency discrimination signal is generated based on modulation transfer spectrum technology and is obtained by detecting the laser signal transmitted through the atomic gas cell and then digitally demodulating it.

8. The method according to claim 7, characterized in that, The digital demodulation is performed on the FPGA side of the system-on-chip. The CORDIC algorithm is used to generate the DDS reference signal, which is then mixed and low-pass filtered before the discriminant signal is extracted.

9. A multi-loop cooperative control system for a laser with long-term lock-in capability, characterized in that, include: The frequency discrimination signal generation module is used to generate a frequency discrimination signal related to the laser frequency deviation; The first control module, deployed on the FPGA side of the system-on-a-chip, is configured to receive the frequency discrimination signal and generate a first control signal with the zero-crossing point of the frequency discrimination signal as the first set point, so as to drive the first actuator of the laser. The second control module, deployed on the FPGA side of the system-on-a-chip, is configured to take the zero point of the first control signal as the second set point, receive the first control signal and generate the second control signal to drive the second actuator of the laser. The third control module, deployed on the ARM processor side of the on-chip system, is configured to take the deviation of the second control signal from its linear working area center value as input, and take the linear working area center value as a third set point, receive the second control signal and generate a third control signal to drive the third actuator of the laser. The response bandwidth of the first actuator, the second actuator, and the third actuator decreases sequentially, while their dynamic range increases sequentially.

Images