Closed-loop control system for signal stabilization and noise suppression of semiconductor light source

Abstract

The invention discloses a closed-loop control system for signal stabilization and noise suppression of a semiconductor light source, and relates to the field of optical detection.The closed-loop control system comprises the steps that linear polarization laser and an input electric signal are obtained, the linear polarization laser is filtered, and filtered laser is obtained; acquiring a target electric signal and a reference signal according to the laser; generating a comparison signal according to the reference signal, performing noise reduction on the target electric signal according to the comparison signal, and obtaining an effective signal after noise reduction; extracting a signal standard preset in advance, and calculating a control quantity according to the effective signal; an input electric signal is adjusted according to a control quantity, a brand new design scheme of'modulation-phase locking demodulation-PID feedback adjustment 'is provided, a target signal is effectively modulated to a specific frequency, selective extraction is performed in a frequency domain by means of a phase locking technology, and active control and fine adjustment of laser intensity, phase or optical power are realized in combination with a PID feedback mode. Therefore, strict requirements for ultrahigh-sensitivity measurement and control and high-stability output under the background of strong noise are met.

Description

Technical Field

[0001] This invention relates to the field of optical detection, and more specifically to a closed-loop control system for stabilizing semiconductor light source signals and suppressing noise. Background Technology

[0002] Traditional bias-type photodetectors typically use transimpedance amplifier circuits to convert photocurrent into voltage signals for measurement. However, this method has a series of key drawbacks that severely limit its performance improvement and application expansion.

[0003] From a noise suppression perspective, while the broadband design of transimpedance amplifiers aims to achieve a fast signal response, it also makes the system highly susceptible to a wide range of noise, including ambient light, thermal noise, and power frequency interference. This noise is superimposed on the measurement signal without obstruction, often overwhelming weak signals and making accurate capture and analysis difficult. Particularly in DC or low-frequency measurement scenarios, 1 / f noise becomes the dominant noise source. Due to its characteristics, simple bandpass filtering or other conventional filtering methods are insufficient to adequately reduce its interference, posing a significant challenge to improving the signal-to-noise ratio (SNR).

[0004] In terms of signal recognition capability, traditional transimpedance circuits perform poorly when detecting extremely weak signals. When measuring laser intensity fluctuations below 0.1%, the inherent noise layer of the system often fails to effectively distinguish the signal, leading to detection failure. For example, a typical photodiode operating at a bandwidth of approximately 10Hz can experience background noise in the picowatt range (approximately 10 Hz). -12 This noise level (W) directly limits the maximum signal strength that the system can detect.

[0005] Dynamic range and feedback control are also shortcomings of traditional bias-type photodetectors. Using high feedback resistors to improve gain leads to problems such as limited output voltage swing and poor low-frequency stability. Furthermore, adjusting the transimpedance range always faces a trade-off between response speed and sensitivity, making it difficult to achieve high-precision measurement while ensuring rapid control, thus failing to meet the application requirements for accurate measurement and control of rapidly changing signals.

[0006] While transimpedance amplification can provide significant gain, it cannot perform frequency-selective filtering of noise. Bandwidth can only be limited by physical or hardware filters, but this approach does not significantly improve the signal-to-noise ratio (SNR). In contrast, lock-in gradient signal detection, without feedback, can indeed significantly improve the SNR by several orders of magnitude using a lock-in amplifier alone. However, due to the lack of active control and compensation mechanisms, it lacks robustness to light source drift and system dynamic disturbances, making it difficult to operate stably in complex and variable real-world environments. Therefore, this invention proposes a closed-loop control system for semiconductor light source signal stabilization and noise suppression. Summary of the Invention

[0007] To address the aforementioned technical issues, a closed-loop control system for stabilizing semiconductor light source signals and suppressing noise is provided. This system solves the problem that, due to the lack of active control and compensation mechanisms, it lacks robustness against light source drift and system dynamic disturbances, making it difficult to operate stably in complex and ever-changing real-world environments.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows: Closed-loop control methods for semiconductor light source signal stabilization and noise suppression include: Acquire linearly polarized laser light and input electrical signal, filter the linearly polarized laser light, and obtain the filtered laser light. The target electrical signal and reference signal are obtained based on the laser. A comparison signal is generated based on a reference signal, and the target electrical signal is denoised based on the comparison signal to obtain the denoised effective signal. Extract the pre-set signal standard and calculate the control quantity based on the valid signal; Adjust the input electrical signal according to the control quantity.

[0009] Preferably, the step of acquiring linearly polarized laser light and an input electrical signal, filtering the linearly polarized laser light, and acquiring the filtered laser light includes the following steps: Acquire linearly polarized laser light and input electrical signal, and then transmit the linearly polarized laser light to the liquid crystal delay unit; The input electrical signal and the liquid crystal delay device generate polarized light; By filtering polarized light using a linear polarizer, laser light parallel to the axis of the linear polarizer can be obtained.

[0010] Preferably, the step of acquiring the target electrical signal and reference signal based on the laser includes the following steps: The laser is input into the beam splitter, which splits the laser into a detection beam and a reference beam. The detection beam is converted into a current signal by a photodiode. The current signal is converted into the target electrical signal by a transimpedance amplifier; The reference beam is amplified to obtain the amplified signal through lock-in amplification; The 1f frequency in the amplified signal is extracted and marked as the reference signal.

[0011] Preferably, the step of generating a comparison signal based on a reference signal, and denoising the target electrical signal based on the comparison signal to obtain a denoised effective signal includes the following steps: A comparison signal of the same frequency is generated based on the reference signal; The mixed signal is obtained by multiplying the comparison signal and the target electrical signal; The effective signal is obtained by filtering the mixed signal using low-pass filtering; The formula for obtaining the mixing signal is as follows: ; In the formula, The phase difference between the target electrical signal and the reference signal. The amplitude of the target electrical signal. It is the angular frequency of the signal. It is a time variable. Indicates the noise term. It is a reference signal.

[0012] Preferably, the step of extracting a pre-set signal standard and calculating the control quantity based on the valid signal includes the following steps: Extract pre-set signal standards; Signal error is obtained based on signal standards and valid signals; Adjust the proportional gain, integral gain, and derivative gain; The control quantity is calculated based on the signal error, proportional gain, integral gain, and derivative gain.

[0013] Preferably, adjusting the proportional gain, integral gain, and derivative gain includes the following steps: Set the proportional gain, integral gain, and derivative gain to 0; Gradually increase the proportional gain and obtain the error convergence rate and polarization state; If the error convergence rate is low, the proportional gain is low, so continue to increase the value of the proportional gain; If the error convergence speed is fast and the polarized light does not oscillate, then the proportional gain is appropriate. Determine the value of the proportional gain and adjust the integral gain and derivative gain. If the error convergence speed is fast but the polarized light oscillates, the proportional gain is high, and the value of the proportional gain needs to be reduced.

[0014] Preferably, determining the value of the proportional gain and adjusting the integral and derivative gains includes the following steps: Gradually increase the integral gain and obtain the laser power and laser state; If the laser power is lower than the set value, the integral gain is low, so the value of the integral gain should be increased further. If the laser power is equal to the set value and there is no fluctuation in the laser power, then the integral gain is appropriate. Determine the value of the integral gain and adjust the differential gain. If the laser power is higher than the set value and the laser power fluctuates, the integral gain is too high and the value of the integral gain needs to be reduced.

[0015] Preferably, determining the value of the integral gain and adjusting the differential gain includes the following steps: Gradually increase the differential gain and obtain the drop-off amplitude and signal fluctuation after the laser power exceeds the set value; If the laser power drops significantly after exceeding the set value, the differential gain is low, and the value of the differential gain should be increased further. If the laser power drops slightly after exceeding the set value and the signal does not fluctuate, then the differential gain is appropriate. If the laser power drops slightly after exceeding the set value and the signal fluctuates, then the integral-derivative gain needs to be reduced, and the derivative gain value needs to be decreased.

[0016] Preferably, a closed-loop control system for semiconductor light source signal stabilization and noise suppression is proposed to implement the above-mentioned closed-loop control method for semiconductor light source signal stabilization and noise suppression, including: Control module: The control module is used to control data transmission within the system; Liquid crystal delay unit: The liquid crystal delay unit adjusts the state of laser polarization according to the input AC signal; Linear polarizer: The linear polarizer filters out laser light that does not conform to the specified polarization direction; Beam splitter: The laser signal from the beam splitter is precisely distributed to the photodiode and the reference section; Photodiode: The photodiode converts the laser signal into a current signal; Transimpedance amplifier: The transimpedance amplifier is used to convert the current signal output by the photodiode into a voltage signal; RedPitaya Module: The RedPitaya module is responsible for digital signal processing and includes lock-in amplification, PID control, and oscilloscope functions. Interaction module: The interaction module facilitates staff to read signals and adjust corresponding parameters.

[0017] Compared with the prior art, the advantages of the present invention are as follows: The present invention proposes a brand-new design scheme of "modulation → phase-locked demodulation → PID feedback adjustment", which effectively modulates the target signal to a specific frequency, selects and extracts it in the frequency domain with the help of phase-locked technology, and combines it with PID feedback to realize active control and fine adjustment of laser intensity, phase or optical power, thereby meeting the strict requirements of ultra-high sensitivity measurement and control and high stability output in the context of strong noise. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating steps S10-S50 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 2This is a flowchart illustrating steps S11-S13 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 3 This is a flowchart illustrating steps S21-S25 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 4 This is a flowchart illustrating steps S31-S33 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 5 This is a flowchart illustrating steps S41-S45 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 6 This is a flowchart illustrating steps S431-S435 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 7 This is a flowchart illustrating steps S4341-S4344 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 8 This is a flowchart illustrating steps S43431-S43434 in the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Figure 9 This is a system structure diagram of the closed-loop control system for semiconductor light source signal stabilization and noise suppression proposed in this invention. Detailed Implementation

[0019] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art.

[0020] Reference Figures 1-9 As shown, a closed-loop control method for stabilizing semiconductor light source signals and suppressing noise includes: S10. Acquire the linearly polarized laser and the input electrical signal, filter the linearly polarized laser, and acquire the filtered laser. S20. Obtain the target electrical signal and reference signal based on the laser; S30. Generate a comparison signal based on the reference signal, and perform noise reduction on the target electrical signal based on the comparison signal to obtain the effective signal after noise reduction. S40. Extract the pre-set signal standard and calculate the control quantity based on the valid signal; S50. Adjust the input electrical signal according to the control quantity; Those skilled in the art will understand that by combining a liquid crystal retarder with a linear polarizer, nonlinear polarization components in the laser can be eliminated, ensuring the stability of the beam splitting ratio of the subsequent beam splitter. The liquid crystal retarder can dynamically adjust the polarization direction to compensate for the polarization state degradation caused by environmental disturbances (such as temperature drift and mechanical vibration). The linear polarizer, as a spatial filter, can block stray light (such as reflected light and scattered light) and reduce background noise in the optical path. Its extinction ratio directly affects the system's signal-to-noise ratio. The beam splitter divides the laser into a detection beam (90%) and a reference beam (1%) to achieve balanced power distribution. The detection beam is used to monitor the output characteristics in real time, while the reference beam provides a phase reference. The photodiode converts the detection beam into a current signal, and the transimpedance amplifier converts it into a voltage signal. The reference signal generates a comparison signal at the same frequency, which is mixed with the target electrical signal and then filtered by a low-pass filter to extract the DC component. The high-frequency components in the mixed signal are filtered out by the low-pass filter, and the remaining effective signal only contains the amplitude and phase information of the target signal, thus separating the noise band from the signal band. The effective signal is compared with a preset standard to generate an error signal. This signal reflects the deviation between the current state and the desired state of the system. The control quantity is converted into the driving voltage of the liquid crystal delay unit to adjust the polarization direction in real time, so that the output laser characteristics track the preset standard. Through rapid iteration, the laser parameter drift caused by environmental disturbances (such as temperature fluctuations and airflow vibrations) is suppressed, achieving long-term stability.

[0021] like Figure 2 As shown, acquiring linearly polarized laser light and an input electrical signal, and filtering the linearly polarized laser light to obtain the filtered laser light includes the following steps: S11. Acquire linearly polarized laser and input electrical signal, and send the linearly polarized laser to the liquid crystal delay unit; S12, Input electrical signal and liquid crystal delay device to obtain polarized light; S13. Filter the polarized light using a linear polarizer to obtain laser light that is parallel to the axis of the linear polarizer; As will be understood by those skilled in the art, a liquid crystal retarder is used to adjust the polarization state of an input laser. By inputting an alternating current signal, this element can change the phase or polarization state of the light, thereby creating favorable conditions for subsequent signal analysis and noise suppression. The liquid crystal retarder exhibits birefringence under the influence of an electric field, and its equivalent phase retardation... It has a non-linear relationship with the input voltage V. , For the difference in birefringence, (The thickness of the liquid crystal layer) By precisely controlling V, the polarization state of the emitted light can be dynamically adjusted. The linear polarizer, acting as a spatial filter, allows only the light component parallel to its transmission axis to pass through, suppressing the perpendicular component. When polarized light passes through the linear polarizer, it effectively filters out laser light that does not conform to the specified polarization direction, thus providing a stable and satisfactory laser signal. like Figure 3 As shown, acquiring the target electrical signal and reference signal based on laser light includes the following steps: S21. Input the laser into the beam splitter, which splits the laser into a detection beam and a reference beam. S22. The detection beam is converted into a current signal by a photodiode; S23. The current signal is converted into the target electrical signal through a transimpedance amplifier; S24. The reference beam obtains the amplified signal through lock-in amplification; S25. Extract the 1f frequency from the amplified signal and mark it as the reference signal; Those skilled in the art will understand that the beam splitter divides the incident laser into two paths at a preset ratio (90:1). The detection beam (main path) is used to monitor the output characteristics in real time, while the reference beam (auxiliary path) provides a phase and amplitude reference. The typical splitting ratio needs to be optimized based on the photodiode responsivity (RD) to ensure that the dynamic range of the two signals is matched. The photodiode (such as InGaAs PIN type) converts the photon energy of the detection beam into a current signal. Through the conversion of the transimpedance amplifier, the signal enters the subsequent signal processing stage in the form of a voltage, further enhancing the system's signal processing capability. The lock-in amplifier extracts the signal component with the same frequency as the reference signal from the reference beam, suppressing incoherent noise. The 1f signal serves as the system clock synchronization source, ensuring the phase correlation between the detection beam and the reference beam.

[0022] like Figure 4 As shown, the process of generating a comparison signal based on a reference signal, and then denoising the target electrical signal based on the comparison signal to obtain the denoised effective signal includes the following steps: S31. Generate a comparison signal of the same frequency based on the reference signal; S32. Multiply the comparison signal and the target electrical signal to obtain the mixed signal; S33. Filter the mixing signal using low-pass filtering to obtain the effective signal; The formula for obtaining the mixing signal is as follows: ; In the formula, The phase difference between the target electrical signal and the reference signal. The amplitude of the target electrical signal. It is the angular frequency of the signal. It is a time variable. Indicates the noise term. It is a reference signal; Those skilled in the art will understand that the comparison signal and the reference signal (1f frequency) are strictly in phase and frequency, providing a coherent reference for subsequent mixing. The amplitude of the comparison signal is dynamically adjusted by a digital attenuator (range -60dB to 0dB) to match the amplitude of the target electrical signal (typical matching error <0.5dB), preventing mixer saturation or insufficient sensitivity. The mixer multiplies the target electrical signal (including the 1f modulation component) with the comparison signal, realizing the shift of the spectrum from the carrier frequency band (e.g., MHz level) to the baseband (DC to kHz level). The mixed signal contains the phase difference information between the target signal and the comparison signal. The mixing process expands the spectrum of incoherent noise (e.g., broadband white noise) to the 2f frequency band, while the baseband signal energy is concentrated, achieving preliminary noise suppression. The cutoff frequency of the low-pass filter (LPF) is set to 2f / 1 (e.g., when f=1kHz, the cutoff frequency is 2kHz), completely suppressing the 2f component generated by mixing. The low-pass filter retains the baseband component in the mixed signal, whose amplitude is proportional to the amplitude of the target signal (the proportionality coefficient is determined by the amplitude of the comparison signal), realizing signal demodulation.

[0023] like Figure 5 As shown, the process of extracting a pre-set signal standard and calculating the control quantity based on the valid signal includes the following steps: S41. Extract the pre-set signal standard; S42. Obtain signal error based on signal standards and valid signals; S43. Adjust the proportional gain, integral gain, and derivative gain; S44. Calculate the control quantity based on the signal error, proportional gain, integral gain, and derivative gain; S45. The signal standard, as the target value for system control, is usually a preset amplitude; Those skilled in the art will understand that the signal standard, as the target value for system control, is usually a preset amplitude (e.g., 1V), phase (e.g., 0°), or frequency (e.g., 1f=1kHz). Its accuracy directly affects the closed-loop control performance. It supports dynamic switching of the standard signal (e.g., through register configuration or host computer instructions) to adapt to the needs of different operating modes (e.g., high-precision mode, high-speed response mode). The difference between the effective signal (the demodulated baseband signal) and the standard signal is calculated using a subtractor (or digital algorithm). The error signal needs to be limited to a reasonable range by a limiting circuit (or software threshold). Proportional gain (Kp): adjusts the system response speed (the larger Kp, the faster the response, but prone to overshoot), typical adjustment range 0.1~10, needs optimization based on system bandwidth (e.g., 1kHz). Integral gain (Ki): eliminates steady-state error (the larger Ki, the higher the steady-state accuracy, but prone to integral saturation), typical adjustment range 0.01~1, needs to be combined with anti-saturation algorithms (e.g., integral separation). Differential gain (Kd): suppresses overshoot and oscillation (the larger Kd, the stronger the damping, but prone to amplifying high-frequency noise), typical adjustment range 0.001~1, needs to be smoothed by low-pass filtering (e.g., cutoff frequency 10Hz) and by PID algorithm (…). ) Calculate the control quantity to achieve accurate tracking of the standard signal by the system.

[0024] like Figure 6 As shown, adjusting the proportional gain, integral gain, and derivative gain includes the following steps: S431. Set the proportional gain, integral gain, and derivative gain to 0; S432. Gradually increase the proportional gain and obtain the error convergence rate and polarization state; S433. If the error convergence speed is low, the proportional gain is low, so continue to increase the value of the proportional gain. S434. If the error convergence speed is fast and the polarized light does not oscillate, then the proportional gain is appropriate. Determine the value of the proportional gain and adjust the integral gain and differential gain. S435. If the error convergence speed is fast but the polarized light oscillates, the proportional gain is high and the value of the proportional gain needs to be reduced. Those skilled in the art will understand that by resetting the PID parameters (Kp=Ki=Kd=0), switching the closed-loop control system to the open-loop state eliminates the influence of historical control quantities on the system (such as saturation caused by the accumulation of integral terms). Without control action, the initial values ​​of the polarized light state (such as polarization extinction ratio and phase delay) and the steady-state values ​​of the error signal (such as the demodulated DC offset) are recorded to provide a benchmark for subsequent parameter adjustments. By increasing Kp in steps (e.g., increasing by 1% each time, from 0.1→10), the convergence speed of the error signal (defined as the time required for the error to decrease from the initial value to 1%) and the changes in the polarized light state (such as extinction ratio fluctuations and phase jumps) are observed. The critical point (i.e., Kp,crit) at which the system transitions from stability to oscillation is identified. Typically, the error convergence speed first increases and then decreases, and the polarized light exhibits periodic oscillations (amplitude > 5% of the initial value). When the error convergence speed falls below a threshold (e.g., t90% > 10ms), a judgment is made... If the given Kp is insufficient, it needs to be increased further to improve the system response speed (theoretical convergence time is inversely proportional to Kp). By gradually increasing Kp, the upper limit of the system's linear operating region (such as the maximum output force / displacement of the actuator) is explored to ensure that the control quantity covers the target error range (typical coverage multiple ≥ 2). When the error convergence speed meets the standard (such as t90% < 1ms) and the polarization light stability meets the requirements (such as extinction ratio fluctuation < 1%), the current Kp is locked as the basic parameter. When the polarization light oscillates (such as periodic fluctuation of extinction ratio > 5%), it is determined that Kp exceeds the critical value and needs to be gradually reduced (such as reducing by 1% each time) until the oscillation disappears. By reducing Kp, the system phase lag is reduced (typical phase margin improvement > 1°), ensuring that the poles of the closed-loop transfer function are located in the left half-plane (avoiding divergence caused by poles in the right half-plane). Appropriately reducing Kp can improve the system's tolerance to parameter perturbations (such as actuator nonlinearity and sensor noise) (typical robustness index improvement of 15%).

[0025] like Figure 7 As shown, determining the value of the proportional gain and adjusting the integral and derivative gains includes the following steps: S4341. Gradually increase the integral gain and obtain the laser power and laser state; S4342. If the laser power is lower than the set value, the integral gain is low, so continue to increase the value of the integral gain. S4343. If the laser power is equal to the set value and the laser power does not fluctuate, then the integral gain is appropriate. Determine the value of the integral gain and adjust the differential gain. S4344. If the laser power is higher than the set value and the laser power fluctuates, the integral gain is too high and the value of the integral gain needs to be reduced. Those skilled in the art will understand that by gradually increasing the integral gain (Ki, for example, starting from 0.001 and increasing by 1% each time), an integral control term can be introduced. To eliminate steady-state errors under proportional control (typical target: reducing laser power steady-state error from ±5% to ±0.1%), real-time monitoring of laser power (sampled via photodetector, resolution ≥16 bits) and laser state (e.g., mode jump, polarization degradation, thermal lensing effect) is implemented. A mapping relationship between Ki and system dynamic characteristics (e.g., power convergence time, overshoot, steady-state fluctuation) is established. Ki adjustment requires a slow incremental strategy (step size ≤ 1% of the current value) to avoid rapid accumulation of the integral term leading to actuator (e.g., laser diode drive current) saturation (typical limit value ±1A). When the laser power remains below the set value (e.g., Pout < 0.95Pset) and shows no upward trend, the integral action is deemed insufficient, and Ki needs to be further increased to accelerate error accumulation and elimination (theoretically, the contribution of the integral term is proportional to Ki). By increasing Ki, the system's ability to compensate for long-term errors (e.g., efficiency decline due to laser aging) is enhanced (typical steady-state error improvement rate ≥80%). When the laser power stabilizes at the set value (e.g., ... And the fluctuation range is extremely small (e.g.) When Ki is determined, the current Ki is locked as the optimal parameter. After Ki is determined, the differential gain (Kd, for example, starting from 0.0001 and increasing by 1% each time) is gradually adjusted to suppress overshoot (typical target: overshoot <1%) and improve the smoothness of dynamic response (e.g., oscillation count <0.5 times). When the laser power exceeds the set value (e.g., ... If periodic fluctuations occur (e.g., amplitude > 1% of initial value), it is determined that the integral effect is too strong. Ki should be gradually reduced (e.g., by 1% each time) until the fluctuations disappear. By reducing Ki, the accumulation rate of the integral term is reduced (typically the integral time constant Ti = 1 / Ki increases), thus avoiding excessive phase lag in the system (typically the phase margin is increased by > 15°).

[0026] like Figure 8 As shown, determining the value of the integral gain and adjusting the differential gain includes the following steps: S43431. Gradually increase the differential gain and obtain the drop-off amplitude and signal fluctuation after the laser power exceeds the set value; S43432. If the laser power drops significantly after exceeding the set value, the differential gain is low, and the value of the differential gain should be increased further. S43433. If the laser power drops slightly after exceeding the set value and the signal does not fluctuate, then the differential gain is appropriate. S43434 If the laser power drops slightly after exceeding the set value and the signal fluctuates, then the integral-derivative gain needs to be reduced, and the value of the derivative gain needs to be decreased. Those skilled in the art will understand that by gradually increasing the differential gain (Kd, e.g., starting from 0.0001 and increasing by 1% each time), a differential control term can be introduced. To suppress overshoot under proportional-integral control (typical target: reduce overshoot from 15% to <2%), the system monitors in real time the drop-off amplitude of laser power in the step response after exceeding the set value (defined as the percentage difference between the overshoot peak and the set value relative to the set value) and signal fluctuations (calculated by the power standard deviation σ, sampling rate ≥1kHz). It establishes a quantitative relationship between Kd and the system's dynamic characteristics, identifying the critical point (i.e., Kd,crit) where the differential gain effectively suppresses overshoot but triggers oscillation. Typically, this is manifested as a decrease followed by an increase in the drop-off amplitude, accompanied by high-frequency oscillations (amplitude > 1% of the set value, frequency > 10Hz). When the drop-off amplitude exceeds the threshold (drop-off amplitude > 5%), the differential effect is considered weak, and Kd needs to be further increased to enhance the suppression of the error rate of change (theoretically, the contribution of the differential term is proportional to Kd and the error rate of change). By increasing Kd, the system's dynamic characteristics are shortened. The adjustment time is reduced (typically from 50ms to 20ms), while the overshoot peak value is reduced (typical overshoot improvement rate ≥60%). When the fallback amplitude is below the threshold (fallback amplitude <2%) and the signal fluctuation is minimal (e.g., σ < 0.1% of the set value), the current Kd is locked as the optimal parameter. Through time-domain waveform analysis (e.g., no periodic oscillation, no decaying oscillation) and frequency-domain analysis (e.g., Nyquist plot showing amplitude margin > 6dB, phase margin > 45°), the system is confirmed to be in a stable state. When the signal exhibits high-frequency oscillation (e.g., amplitude > 1% of the set value, frequency > 10Hz), it is determined that the derivative effect is too strong, and Kd needs to be gradually reduced (e.g., reduced by 1% each time) until the fluctuation disappears. By reducing Kd, the amplification effect of the derivative term on sensor noise (e.g., dark current noise of photodetectors, power supply ripple noise) is reduced (typical noise power attenuation > 1dB@1kHz).

[0027] like Figure 9 As shown, a closed-loop control system for semiconductor light source signal stabilization and noise suppression is proposed to implement the aforementioned closed-loop control method for semiconductor light source signal stabilization and noise suppression, including: Control module: The control module is used to control data transmission within the system; Liquid crystal delay unit: The liquid crystal delay unit adjusts the state of laser polarization according to the input AC signal; Linear polarizer: The linear polarizer filters out laser light that does not conform to the specified polarization direction; Beam splitter: The laser signal from the beam splitter is precisely distributed to the photodiode and the reference section; Photodiode: The photodiode converts the laser signal into a current signal; Transimpedance amplifier: The transimpedance amplifier is used to convert the current signal output by the photodiode into a voltage signal; RedPitaya Module: The RedPitaya module is responsible for digital signal processing and includes lock-in amplification, PID control, and oscilloscope functions. Interaction module: The interaction module facilitates staff to read signals and adjust corresponding parameters.

[0028] In summary, the advantages of this invention are as follows: it proposes a novel design scheme of "modulation → phase-locked demodulation → PID feedback adjustment", which effectively modulates the target signal to a specific frequency, selects and extracts it in the frequency domain with the help of phase-locked technology, and combines it with PID feedback to achieve active control and fine adjustment of laser intensity, phase or optical power, thereby meeting the stringent requirements for ultra-high sensitivity measurement and control and high stability output under strong noise background.

[0029] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A closed-loop control method for signal stabilization and noise suppression of semiconductor light sources, characterized in that, include: Acquire linearly polarized laser light and input electrical signal, filter the linearly polarized laser light, and obtain the filtered laser light. The target electrical signal and reference signal are obtained based on the laser. A comparison signal is generated based on a reference signal, and the target electrical signal is denoised based on the comparison signal to obtain the denoised effective signal. Extract the pre-set signal standard and calculate the control quantity based on the valid signal; Adjust the input electrical signal according to the control quantity.

2. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 1, characterized in that: The process of acquiring linearly polarized laser light and an input electrical signal, filtering the linearly polarized laser light, and obtaining the filtered laser light includes the following steps: Acquire linearly polarized laser light and input electrical signal, and then transmit the linearly polarized laser light to the liquid crystal delay unit; The input electrical signal and the liquid crystal delay device generate polarized light; By filtering polarized light using a linear polarizer, laser light parallel to the axis of the linear polarizer can be obtained.

3. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 2, characterized in that: The process of acquiring the target electrical signal and reference signal based on laser light includes the following steps: The laser is input into the beam splitter, which splits the laser into a detection beam and a reference beam. The detection beam is converted into a current signal by a photodiode. The current signal is converted into the target electrical signal by a transimpedance amplifier; The reference beam is amplified to obtain the amplified signal through lock-in amplification; The 1f frequency in the amplified signal is extracted and marked as the reference signal.

4. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 3, characterized in that: The steps of generating a comparison signal based on a reference signal, denoising the target electrical signal based on the comparison signal, and obtaining the denoised effective signal include the following: A comparison signal of the same frequency is generated based on the reference signal; The mixed signal is obtained by multiplying the comparison signal and the target electrical signal; The effective signal is obtained by filtering the mixed signal using low-pass filtering; The formula for obtaining the mixing signal is as follows: ； In the formula, The phase difference between the target electrical signal and the reference signal. The amplitude of the target electrical signal. It is the angular frequency of the signal. It is a time variable. Indicates the noise term. It is a reference signal.

5. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 4, characterized in that: The process of extracting a pre-set signal standard and calculating the control quantity based on the valid signal includes the following steps: Extract pre-set signal standards; Signal error is obtained based on signal standards and valid signals; Adjust the proportional gain, integral gain, and derivative gain; The control quantity is calculated based on the signal error, proportional gain, integral gain, and derivative gain.

6. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 5, characterized in that: The adjustment of the proportional gain, integral gain, and derivative gain includes the following steps: Set the proportional gain, integral gain, and derivative gain to 0; Gradually increase the proportional gain and obtain the error convergence rate and polarization state; If the error convergence rate is low, the proportional gain is low, so continue to increase the value of the proportional gain; If the error convergence speed is fast and the polarized light does not oscillate, then the proportional gain is appropriate. Determine the value of the proportional gain and adjust the integral gain and derivative gain. If the error convergence speed is fast but the polarized light oscillates, the proportional gain is high, and the value of the proportional gain needs to be reduced.

7. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 6, characterized in that: Determining the value of the proportional gain and adjusting the integral and derivative gains includes the following steps: Gradually increase the integral gain and obtain the laser power and laser state; If the laser power is lower than the set value, the integral gain is low, so the value of the integral gain should be increased further. If the laser power is equal to the set value and there is no fluctuation in the laser power, then the integral gain is appropriate. Determine the value of the integral gain and adjust the differential gain. If the laser power is higher than the set value and the laser power fluctuates, the integral gain is too high and the value of the integral gain needs to be reduced.

8. The closed-loop control method for semiconductor light source signal stabilization and noise suppression according to claim 6, characterized in that: Determining the value of the integral gain and adjusting the differential gain includes the following steps: Gradually increase the differential gain and obtain the drop-off amplitude and signal fluctuation after the laser power exceeds the set value; If the laser power drops significantly after exceeding the set value, the differential gain is low, and the value of the differential gain should be increased further. If the laser power drops slightly after exceeding the set value and the signal does not fluctuate, then the differential gain is appropriate. If the laser power drops slightly after exceeding the set value and the signal fluctuates, then the integral-derivative gain needs to be reduced, and the derivative gain value needs to be decreased.

9. A DC insulator degradation diagnosis system based on multi-modal data fusion, used to implement the DC insulator degradation diagnosis method based on multi-modal data fusion as described in claims 1-8, characterized in that, include: Control module: The control module is used to control data transmission within the system; Liquid crystal delay unit: The liquid crystal delay unit adjusts the state of laser polarization according to the input AC signal; Linear polarizer: The linear polarizer filters out laser light that does not conform to the specified polarization direction; Beam splitter: The laser signal from the beam splitter is precisely distributed to the photodiode and the reference section; Photodiode: The photodiode converts the laser signal into a current signal; Transimpedance amplifier: The transimpedance amplifier is used to convert the current signal output by the photodiode into a voltage signal; RedPitaya Module: The RedPitaya module is responsible for digital signal processing and includes lock-in amplification, PID control, and oscilloscope functions. Interaction module: The interaction module facilitates staff to read signals and adjust corresponding parameters.

Images