A laser control system for coherent population trapping magnetometer

Abstract

The present invention discloses a laser control system for a coherent population trapping magnetometer. The system comprises a core control unit, a drive circuit system, a microwave source signal modulation system, and an H-bridge bidirectionally connected to the core control unit. The drive circuit system includes a bandgap reference voltage circuit, which includes an input low-power bias startup circuit and a folding operational amplifier. The core control unit includes an FPGA and a temperature control circuit system, which uses an anti-integration windup algorithm combined with fuzzy PID control to dynamically adjust PID parameters to achieve temperature regulation. The microwave source signal modulation system includes a fully digital phase-locked loop, a direct digital synthesizer, and a frequency multiplier. The FPGA controls the generation, modulation, and amplification of microwave signals. The laser control system of the present invention has high microwave source frequency stability and strong noise suppression capabilities.

Description

Technical Field

[0001] The present invention relates to the technical field of magnetometers, and in particular to a laser control system for a coherent population trapping magnetometer. Background Art

[0002] Coherent population trapping (CPT) magnetometers use lasers to interact with alkali metal atoms (such as 87 Measuring magnetic field intensity using the quantum effect spectrum of Rb requires a highly stable laser control system. However, existing laser control systems, on the one hand, still rely on traditional bandgap reference circuits, which are characterized by high power consumption (requiring a startup circuit and high current bias), significant noise and temperature drift, and are unable to meet the requirements of ultra-low power sensors. On the other hand, existing temperature control uses conventional PID algorithms, which have high overshoot (>1.5%), slow recovery, and are unable to adapt to nonlinear temperature fluctuations. Furthermore, the microwave sources used in existing laser control systems are large in size, have high phase noise (-80 dBc / Hz @10 kHz), and have limited modulation capabilities, making it difficult to support the miniaturization and dynamic modulation requirements of high-precision magnetometers. Summary of the Invention

[0003] The object of the present invention is to provide a high-precision laser control system for a coherent population trapping magnetometer with high microwave source frequency stability and strong noise suppression capability.

[0004] The present invention provides a laser control system for a coherent population trapping magnetometer, comprising a core control unit, a driving circuit system, a microwave source signal modulation system, and an H-bridge bidirectionally connected to the core control unit;

[0005] The driving circuit system includes a bandgap reference voltage circuit, and the bandgap reference voltage circuit includes an input low-power bias startup circuit and a folding operational amplifier;

[0006] The core control unit includes an FPGA and a temperature control circuit system. The temperature control circuit system adopts an anti-integral windup algorithm combined with fuzzy PID control to dynamically adjust PID parameters to achieve temperature regulation;

[0007] The microwave source signal modulation system includes a full digital phase-locked loop, a direct digital synthesizer and a frequency multiplier, and realizes the generation, modulation and amplification of microwave signals through the FPGA control.

[0008] The bandgap reference voltage circuit (BGR) of the driving circuit system is optimized in terms of noise, temperature characteristics, and power, and accurately converts the bandgap reference voltage into a reference current.

[0009] Furthermore, the low-power bias startup circuit at the input end of the laser control system of the present invention is composed of MOS transistors M1, M2, and M3 to form a subthreshold bias circuit, and the MOS transistor M1 is grounded through a resistor R1;

[0010] Subthreshold current ,in is the electron mobility in the channel; is the capacitance per unit area of ​​the gate oxide layer; W is the channel width; L is the channel length; is the gate-source voltage; is the drain-source voltage; is the transistor threshold voltage; m is the inverse of the gate surface coupling coefficient; V T is the thermal voltage of the semiconductor;

[0011] ;in k is the Boltzmann constant, T It's the temperature, q is the elementary charge;

[0012] The folding operational amplifier includes MOS tubes M4 to M14, and a compensation capacitor C is provided between the source of the MOS tube M12 and the gate of the MOS tube M15. c ;

[0013] One end of the resistor R5 is connected to Vdd to provide an initial bias voltage for the folding operational amplifier.

[0014] Furthermore, the bandgap reference voltage circuit of the laser control system of the present invention further includes a bipolar transistor Q1 and a bipolar transistor Q2; one end of the bipolar transistor Q1 is connected to the drain of the MOS transistor M15, and the other end is connected to the resistor R2; the source of the MOS transistor M15 is connected to the resistor R6;

[0015] The emitter of the bipolar transistor Q1 is grounded through the resistor R2.

[0016] Feedback resistor R4 is connected in parallel with capacitor CFF and then output V out ;

[0017] and satisfied; ;

[0018] ;

[0019] in, V ref is the bandgap reference voltage; V out is the output voltage;V PTAT It is a voltage that is proportional to absolute temperature.

[0020] Furthermore, the anti-integral saturation algorithm of the laser control system of the present invention specifically includes: inputting a temperature measurement module NTC with a negative temperature coefficient to obtain the current temperature T in real time. current , and calculate the error e(t) and the error change rate ， t is time; if e(t)>e set , directly output pulse width modulation PWM duty cycle 100%, drive H bridge to run at full power, skip PID adjustment, and reduce error as quickly as possible; if e(t)≤e set , then enter the fuzzy PID control link, according to the error e (t) and the error change rate de(t) / dt Dynamically adjust the control parameters; where e set The temperature target group is set; the temperature measurement module NTC continuously monitors the current temperature T current , update, and form a closed-loop control;

[0021] The expression of the anti-integral windup algorithm is formula 1:

[0022] Formula I;

[0023] Formula II;

[0024] in is the proportional coefficient corrected by fuzzy PID, is the integral coefficient corrected by fuzzy PID, is the differential coefficient corrected by fuzzy PID, is the initial value of the proportional coefficient, is the initial value of the integral coefficient, is the initial value of the differential coefficient, is the proportional coefficient correction value, is the integral coefficient correction value, is the correction value of the differential coefficient, is the instantaneous value of the error fed back by the system, τ is the integration time constant, It is the output of the temperature control circuit system controlled by the PID algorithm.

[0025] Furthermore, the fuzzy PID control of the laser control system of the present invention includes a membership function optimized by a genetic algorithm, and the membership function optimized by the genetic algorithm includes:

[0026] S1: Encode the rule base of the fuzzy PID controller using a real number encoding method to generate an initial rule base;

[0027] S2: Use the evaluation function to evaluate the fitness of the initial rule base;

[0028] S3: Then the encoded rule base is subjected to selection, crossover and mutation processes to generate an optimized new rule base;

[0029] S4: Re-evaluate the fitness of the new rule base. If the fitness minimum condition is met, decode and output the optimal fuzzy rule table and update the rule base.

[0030] S5: Real-time acquisition of the error value of the controlled object, through fuzzy processing, fuzzy rule mapping and defuzzification operation, calculate the error value of the controlled object. ;

[0031] S6: 、 、 and K p0 ，K I0 ，K D0 Superposition, get K p ，K I ，K D ， Realize dynamic adjustment of PID control parameters.

[0032] Furthermore, the output end of the fully digital phase-locked loop of the laser control system of the present invention is connected to a second-order active loop filter and then connected to a frequency multiplier via a microstrip line;

[0033] The fully digital phase-locked loop has a built-in voltage-controlled oscillator (VCO) and a digital phase detector. The fully digital phase-locked loop uses a core control unit to configure a register (Register) via a serial peripheral interface (SPI). INT and FRAC , the calculation formula is:

[0034] ;

[0035] in, is the output frequency, is the input reference frequency, INT is an integer division ratio, FRAC is the fractional division ratio, MOD The modulus of the fractional frequency division is used to control the resolution of the decimal place. INT, FRAC, MOD Adjust according to the chip manual;

[0036] The output end of the direct digital synthesizer is connected in series with a bandpass filter and then connected to a fully digital phase-locked loop;

[0037] The frequency multiplier output microstrip line is connected to the input end of the microwave amplifier, and the output end of the microwave amplifier is directly connected to a three-port DC bias network Bias-Tee for biasing.

[0038] Furthermore, the microwave source signal modulation system of the laser control system of the present invention further includes a reference frequency source, a microwave amplifier and a modulation control unit;

[0039] The reference frequency source includes a high-precision temperature-compensated crystal oscillator, which is powered by a 3.3V regulated power supply. The output end of the high-precision temperature-compensated crystal oscillator is connected to the reference clock input pin of the direct digital synthesizer through a printed circuit board (PCB) trace.

[0040] Furthermore, the driving circuit system of the laser control system of the present invention further includes a digital-to-analog converter connected to the core control unit, and a low voltage difference linear regulator provided between the digital-to-analog converter and the bandgap reference voltage circuit.

[0041] Furthermore, the frequency multiplier of the laser control system of the present invention includes a first-level frequency multiplier and a second-level frequency multiplier; both the first-level frequency multiplier and the second-level frequency multiplier are frequency multipliers with doubled frequency; and both the first-level frequency multiplier and the second-level frequency multiplier are followed by a bandpass filter capable of filtering the fundamental frequency.

[0042] Furthermore, the output frequency of the high-precision temperature-compensated crystal oscillator of the laser control system of the present invention is 10 MHz, and the stability is ±0.1 ppm.

[0043] The laser control system for a coherent population trapping magnetometer described in the present invention has the following beneficial effects:

[0044] The laser control system for a coherent population-trapped magnetometer described in this invention utilizes a bandgap reference voltage circuit, which incorporates a low-power input bias startup circuit. This reduces system power consumption while also suppressing system temperature rise. The central op amp structure increases output impedance, thereby boosting gain. It also reduces the Miller effect, improves frequency response, and mitigates temperature-induced zero-point drift, ensuring the circuit operates at the optimal operating point. The use of two differential pairs further enhances the common-mode rejection ratio (CMRR) and improves noise immunity. Furthermore, external interference factors are eliminated, ensuring that the reference voltage is linearly dependent only on temperature.

[0045] The improved anti-saturation algorithm described in the present invention and its combination with fuzzy PID control solves the overshoot problem of traditional PID in nonlinear systems, while making up for the defect of insufficient steady-state accuracy of traditional control. It can greatly improve the temperature control effect of vertical cavity surface emitting laser (VCSEL) laser, for example, solving problems such as rapid recovery after overshoot and stable accuracy.

[0046] Output limiting, integral limiting, rate limiting, and nonlinear limiting logic are introduced into the integral link of the traditional PID. When the error exceeds the set threshold, the integral action is suspended to prevent overshoot or oscillation caused by excessive accumulation of the integral term; when the error returns to the normal range, the integral action is gradually restored to ensure steady-state accuracy. By improving the anti-integral saturation logic, when the temperature fluctuation is large, it is determined whether to perform full-power heating or cooling, effectively improving the speed of returning to the set temperature, which can exceed the speed of the traditional PID anti-integral saturation algorithm by about 10%; at the same time, PID is only enabled when there is a small range of fluctuations, and the values ​​of the three PID parameters can be reduced to reduce the single adjustment amount, thereby reducing the overshoot amount. The overshoot amount can reach ≤1%, and it can still converge stably in the fast temperature change scenario.

[0047] At the same time, in the process of PID adjustment, the improved fuzzy PID algorithm is used, and the fuzzy PID can dynamically adjust parameters and adapt to nonlinearity. The membership function in the fuzzy PID control is defined by the Lyapunov function to define the energy function. , under the constraints of the anti-saturation mechanism, the control rate satisfies ,in, e(t) is the error, is the error rate of change, and t is time. This ensures that the system converges quickly to an equilibrium point near steady-state. The steady-state error is controlled within ±0.01K, meeting the wavelength stability requirements of high-precision vertical-cavity surface-emitting lasers (VCSELs).

[0048] The theoretical steady-state error of the system approaches zero, but the actual accuracy is limited by the resolution of the NTC temperature sensor and the circuit noise (equivalent to ±0.005°C).

[0049] The microwave source signal modulation system of the laser control system of the present invention exhibits significant advantages in terms of frequency stability, phase noise, modulation capability, volume power consumption, and integration.

[0050] Improved frequency stability and accuracy: The microwave source's frequency stability is better than ±0.1 ppm, far exceeding the ±1 ppm of traditional microwave sources. The temperature-compensated crystal oscillator (TCXO) provides a highly stable reference of ±0.1 ppm. The direct digital frequency synthesizer (DDS) achieves microhertz accuracy through a 32-bit control word, and the digital locking mechanism of the fully digital phase-locked loop (DPLL) further eliminates drift. In comparison, traditional microwave sources are limited by analog filters and environmental interference, resulting in poor stability.

[0051] Reduced phase noise: Phase noise is better than -100 dBc / Hz at 10 kHz offset, compared to -80 dBc / Hz for traditional microwave sources. The fully digital phase-locked loop (DPLL) utilizes a digital phase detector and optimized loop filter design for enhanced noise suppression. The direct digital synthesizer (DDS) generates a pure low-frequency signal, providing a low-noise foundation for subsequent frequency multiplication. The analog circuitry of traditional microwave sources is prone to introducing additional noise.

[0052] Low phase noise reduces laser spectrum broadening, improves atomic energy level excitation efficiency, and enhances the sensitivity of the atomic magnetometer.

[0053] Enhanced modulation capabilities: Incorporating an FPGA module, it supports modulation modes such as frequency modulation (FM) and phase modulation (PM), offering fast response times. Analysis: The FPGA's high-speed control capabilities, combined with the frequency flexibility of a direct digital synthesizer (DDS) and the phase adjustment capabilities of a fully digital phase-locked loop (DPLL), enable dynamic modulation. Traditional microwave sources only support fixed outputs and rely on external equipment for modulation, resulting in low efficiency and complexity.

[0054] Flexible modulation can be used for phase-locked amplification or noise suppression technology to further improve measurement accuracy and adapt to diverse experimental needs.

[0055] Miniaturization and low power consumption are achieved: The system size is reduced to 10cm x 10cm, with power consumption of 5-10W, far lower than the tens of watts of traditional microwave sources. Surface-mount components and on-chip amplifiers reduce size, while the integrated design optimizes power consumption. Traditional technologies rely on discrete devices, which are both bulky and energy-intensive.

[0056] Miniaturization makes microwave sources suitable for portable atomic magnetometers, such as those used in field geological exploration or integration into medical devices.

[0057] Improved system integration and usability: The integrated microwave source and modulation unit simplify debugging and enhance anti-interference capabilities. Printed circuit board (PCB) integration eliminates the complex connections between discrete modules. FPGA centralized control allows for software-based parameter adjustment; traditional systems require manual calibration and are susceptible to interference. This simplifies the development and maintenance of atomic magnetometers, improving engineering practicality.

[0058] This approach, by building a vertical-cavity surface-emitting laser (VCSEL) circuit control system and combining it with FPGA dynamic control and temperature feedback, effectively addresses the limitations of traditional magnetometers in frequency stability, phase noise, modulation flexibility, size, power consumption, and system integration. This technical result significantly improves the measurement accuracy and sensitivity of coherently arranged trapped-effect magnetometers, while also promoting their development in portable and integrated applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1This is a structural block diagram of the laser control system of the present invention;

[0060] Figure 2 is a structural block diagram of the driving circuit system of the present invention;

[0061] Figure 3 This is the bandgap reference voltage circuit diagram of the present invention;

[0062] Figure 4 This is the logic diagram of the anti-integral windup algorithm of the present invention;

[0063] Figure 5 This is a structural diagram of the fuzzy PID algorithm control of the present invention;

[0064] Figure 6 is a schematic diagram of the microwave source signal modulation system of the present invention;

[0065] Figure 7 This is the fuzzy rule table described in Example 1 of the present invention. DETAILED DESCRIPTION

[0066] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific embodiments of the present invention are now described in detail with reference to the accompanying drawings. The described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. If specific conditions are not specified in the specific implementation method, they shall be carried out according to conventional conditions or conditions provided by the manufacturer.

[0067] When the following description refers to the accompanying 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 the present disclosure. The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the present disclosure.

[0068] A laser control system for a coherent population trapping magnetometer, comprising a core control unit, a drive circuit system, a microwave source signal modulation system, and an H-bridge bidirectionally connected to the core control unit;

[0069] The driving circuit system includes a bandgap reference voltage circuit, and the bandgap reference voltage circuit includes an input low-power bias startup circuit and a folding operational amplifier;

[0070] The core control unit includes an FPGA and a temperature control circuit system. The temperature control circuit system adopts an anti-integral windup algorithm combined with fuzzy PID control to dynamically adjust PID parameters to achieve temperature regulation;

[0071] The microwave source signal modulation system includes a full digital phase-locked loop, a direct digital synthesizer and a frequency multiplier, and realizes the generation, modulation and amplification of microwave signals through the FPGA control.

[0072] In other embodiments, the input-end low-power bias startup circuit is composed of MOS transistors M1, M2, and M3 to form a sub-threshold bias circuit, and the MOS transistor M1 is grounded via a resistor R1;

[0073] Subthreshold current ,in is the electron mobility in the channel; is the capacitance per unit area of ​​the gate oxide layer; W is the channel width; L is the channel length; is the gate-source voltage; is the drain-source voltage; is the transistor threshold voltage; m is the inverse of the gate surface coupling coefficient; V T is the thermal voltage of the semiconductor;

[0074] ;in k is the Boltzmann constant, T It's the temperature, q is the elementary charge;

[0075] The folding operational amplifier includes MOS tubes M4 to M14, and a compensation capacitor C is provided between the source of the MOS tube M12 and the gate of the MOS tube M15. c ;

[0076] One end of the resistor R5 is connected to Vdd to provide an initial bias voltage for the folding operational amplifier.

[0077] In other embodiments, the bandgap reference voltage circuit further includes a bipolar transistor Q1 and a bipolar transistor Q2; one end of the bipolar transistor Q1 is connected to the drain of the MOS transistor M15, and the other end is connected to the resistor R2; the source of the MOS transistor M15 is connected to the resistor R6;

[0078] The emitter of the bipolar transistor Q1 is grounded through the resistor R2.

[0079] Feedback resistor R4 is connected in parallel with capacitor CFF and then output V out ;

[0080] and satisfy ;

[0081] ;

[0082] in, V ref is the bandgap reference voltage; V out is the output voltage; V PTAT It is a voltage that is proportional to absolute temperature.

[0083] In other embodiments, the anti-integral saturation algorithm specifically includes: inputting a negative temperature coefficient temperature measurement module NTC to obtain the current temperature T in real time current , and calculate the error e(t) and the error change rate ， t is time; if e(t)>e set , directly output pulse width modulation PWM duty cycle 100%, drive H bridge to run at full power, skip PID adjustment, and reduce error as quickly as possible; if e(t)≤e set , then enter the fuzzy PID control link, according to the error e (t) and the error change rate de (t) / dt Dynamically adjust the control parameters; where e set The temperature target group is set; the temperature measurement module NTC continuously monitors the current temperature T current , update, and form a closed-loop control;

[0084] The expression of the anti-integral windup algorithm is formula 1:

[0085] Formula I;

[0086] Formula II;

[0087] in is the proportional coefficient corrected by fuzzy PID, is the integral coefficient corrected by fuzzy PID, is the differential coefficient corrected by fuzzy PID, is the initial value of the proportional coefficient, is the initial value of the integral coefficient, is the initial value of the differential coefficient, is the proportional coefficient correction value, is the integral coefficient correction value, is the correction value of the differential coefficient, is the instantaneous value of the error fed back by the system, τ is the integration time constant, It is the output of the temperature control circuit system controlled by the PID algorithm.

[0088] In other embodiments, the fuzzy PID control includes a membership function optimized using a genetic algorithm, and the membership function optimized using the genetic algorithm includes:

[0089] S1: Encode the rule base of the fuzzy PID controller using a real number encoding method to generate an initial rule base;

[0090] S2: Use the evaluation function to evaluate the fitness of the initial rule base;

[0091] S3: Then the encoded rule base is subjected to selection, crossover and mutation processes to generate an optimized new rule base;

[0092] S4: Re-evaluate the fitness of the new rule base. If the fitness minimum condition is met, decode and output the optimal fuzzy rule table and update the rule base.

[0093] S5: Real-time acquisition of the error value of the controlled object, through fuzzy processing, fuzzy rule mapping and defuzzification operation, calculate the error value of the controlled object. ;

[0094] S6: 、 、 and K p0 ，K I0 ，K D0 Superposition, get K p ，K I ，K D ， Realize dynamic adjustment of PID control parameters.

[0095] In other embodiments, the output end of the fully digital phase-locked loop is connected to a second-order active loop filter and then connected to a frequency multiplier via a microstrip line;

[0096] The fully digital phase-locked loop has a built-in voltage-controlled oscillator (VCO) and a digital phase detector. The fully digital phase-locked loop uses a core control unit to configure a register (Register) via a serial peripheral interface (SPI) to set an integer frequency division ratio. INT and fractional frequency ratios FRAC , the calculation formula is:

[0097] ;

[0098] in, is the output frequency, is the input reference frequency, INT is an integer division ratio, FRAC is the fractional division ratio, MOD The modulus of the fractional frequency division is used to control the resolution of the decimal place. INT, FRAC, MOD Adjust according to the chip manual;

[0099] The output end of the direct digital synthesizer is connected in series with a bandpass filter and then connected to a fully digital phase-locked loop;

[0100] The frequency multiplier output microstrip line is connected to the input end of the microwave amplifier, and the output end of the microwave amplifier is directly connected to a three-port DC bias network Bias-Tee for biasing.

[0101] In other embodiments, the microwave source signal modulation system further includes a reference frequency source, a microwave amplifier, and a modulation control unit;

[0102] The reference frequency source includes a high-precision temperature-compensated crystal oscillator, which is powered by a 3.3V regulated power supply. The output end of the high-precision temperature-compensated crystal oscillator is connected to the reference clock input pin of the direct digital synthesizer through a printed circuit board (PCB) trace.

[0103] In other embodiments, the driving circuit system further includes a digital-to-analog converter connected to the core control unit, and a low voltage difference linear regulator provided between the digital-to-analog converter and the bandgap reference voltage circuit.

[0104] In other embodiments, the frequency multiplier includes a primary frequency multiplier and a secondary frequency multiplier; the primary frequency multiplier and the secondary frequency multiplier are both frequency multipliers with doubled frequency; and a bandpass filter capable of filtering the fundamental frequency is provided after the primary frequency multiplier and the secondary frequency multiplier.

[0105] In other embodiments, the high-precision temperature-compensated crystal oscillator has an output frequency of 10 MHz and a stability of ±0.1 ppm.

[0106] In other embodiments, the folded operational amplifier includes two differential pairs, uses cascode transistors to improve output impedance and common mode rejection ratio (CMRR), and uses cascade compensation capacitors to improve frequency response. The output impedance is increased to the megaohm level by the cascode transistors, and the gain bandwidth product (GBW) meets ,in g m5 is the input stage transconductance, and Cc is the compensation capacitor.

[0107] In other implementations, in the printed circuit board (PCB) design, decoupling capacitors are added around the temperature compensation crystal oscillator (TCXO) to filter out power supply noise and ensure signal purity.

[0108] In other embodiments, the membership function in the fuzzy PID control is defined by the Lyapunov function to define the energy function , under the constraints of the anti-saturation mechanism, the control rate satisfies ,in, e(t) is the error, is the error change rate, and t is the time. Ensure that the system converges quickly to the equilibrium point near the steady state.

[0109] Example 1:

[0110] A laser control system for a coherent population trapping magnetometer, such as Figure 1 As shown, it includes a core control unit, a driving circuit system, a microwave source signal modulation system, and an H bridge bidirectionally connected to the core control unit.

[0111] The driving circuit system includes a bandgap reference voltage circuit, which includes an input low-power bias startup circuit and a folding operational amplifier; the driving circuit system also includes a digital-to-analog converter connected to the core control unit, and a low-voltage difference linear regulator arranged between the digital-to-analog converter and the bandgap reference voltage circuit. Figure 2 As shown, the driving circuit system is mainly controlled by FPGA to input digital signals, which are converted into analog signals through the digital-to-analog converter DAC, and processed by a low-voltage difference linear regulator to obtain a stable electrical signal with low noise. Finally, the bandgap reference voltage is output to provide drive for the vertical cavity surface emitting laser (VCSEL).

[0112] In this embodiment 1, Figure 3 As shown, the input-end low-power bias startup circuit is composed of MOS transistors M1, M2, and M3 to form a sub-threshold bias circuit, and the MOS transistor M1 is grounded through a resistor R1;

[0113] Subthreshold current ,in is the electron mobility in the channel; is the capacitance per unit area of ​​the gate oxide layer; W is the channel width; L is the channel length; is the gate-source voltage; is the drain-source voltage; is the transistor threshold voltage; m is the inverse of the gate surface coupling coefficient; V T is the thermal voltage of the semiconductor;

[0114] ;in k is the Boltzmann constant, T It's the temperature, q is the elementary charge;

[0115] The folding operational amplifier includes MOS tubes M4 to M14, and a compensation capacitor C is provided between the source of the MOS tube M12 and the gate of the MOS tube M15. c ;

[0116] One end of resistor R5 is connected to Vdd, which provides an initial bias voltage for the folding operational amplifier. Vdd is the chip operating voltage.

[0117] The bandgap reference voltage circuit further includes a bipolar transistor Q1 and a bipolar transistor Q2; one end of the bipolar transistor Q1 is connected to the drain of the MOS transistor M15, and the other end is connected to the resistor R2; the source of the MOS transistor M15 is connected to the resistor R6;

[0118] The emitter of the bipolar transistor Q1 is grounded through the resistor R2.

[0119] Feedback resistor R4 is connected in parallel with capacitor CFF and then output V out ;

[0120] and satisfy ;

[0121] According to Kirchhoff's current law, we can get:

[0122] ;

[0123] in, V ref is the bandgap reference voltage; V out is the output voltage; V PTAT It is a voltage that is proportional to absolute temperature.

[0124] The folded operational amplifier comprises two differential pairs, which improve the output impedance and common mode rejection ratio (CMRR) through cascode transistors, and use cascade compensation capacitors to improve the frequency response. The output impedance is increased to the megaohm level through cascode transistors, and the gain bandwidth product (GBW) meets ,in g m5 is the input stage transconductance, and Cc is the compensation capacitor.

[0125] The core control unit includes an FPGA and a temperature control circuit system. The temperature control circuit system adopts an anti-integral windup algorithm combined with fuzzy PID control to dynamically adjust PID parameters to achieve temperature regulation.

[0126] like Figure 4 As shown, the anti-integral saturation algorithm specifically includes: inputting a temperature measurement module NTC with a negative temperature coefficient to obtain the current temperature T in real time current , and calculate the error e(t) and the error change rate ， t is time; if e(t)>e set , directly output pulse width modulation PWM duty cycle 100%, drive H bridge to run at full power, skip PID adjustment, and reduce error as quickly as possible; if e(t)≤e set , then enter the fuzzy PID control link, according to the error e (t) and the error change rate de(t) / dt Dynamically adjust the control parameters; where e set The temperature target group is set; the temperature measurement module NTC continuously monitors the current temperature T current , update, and form a closed-loop control;

[0127] The expression of the anti-integral windup algorithm is formula 1:

[0128] Formula I;

[0129] Formula II;

[0130] in is the proportional coefficient corrected by fuzzy PID, is the integral coefficient corrected by fuzzy PID, is the differential coefficient corrected by fuzzy PID, is the initial value of the proportional coefficient, is the initial value of the integral coefficient, is the initial value of the differential coefficient, is the proportional coefficient correction value, is the integral coefficient correction value, is the correction value of the differential coefficient, is the instantaneous value of the error fed back by the system, τ is the integration time constant, It is the output of the temperature control circuit system controlled by the PID algorithm.

[0131] like Figure 5 As shown, the fuzzy PID control link specifically includes:

[0132] Fuzzification: Map the error e and the error change rate de / dt into fuzzy linguistic variables. According to the law of laser temperature change, the domain range of the error e and the error change rate de / dt is determined, and the domain is evenly divided into eight fuzzy sets, namely negative large (NB), negative medium (NM), negative small (NS), negative zero (NO), positive zero (PO), positive small (PS), positive medium (PM), and positive large (PB). Among them, negative large and positive large use the activation function Sigmoid membership function, and other fuzzy sets use triangular membership functions, that is, the error e and the error change rate de / dt can be fuzzified.

[0133] Rule base: Generates control decisions based on preset rules. When the error is large and the error change rate is large, the proportional coefficient is increased. K P , increase the integral coefficient K I , increase the differential coefficient K D ; The error is small and the error change rate is large, reduce the proportional coefficient K P , increase the integral coefficient K I , reduce the differential coefficient K D ; If the error is large and the error change rate is very small, increase the proportional coefficient K P , increase the integral coefficient K I , increase the differential coefficient K D ; When the error is small and the error change rate is very small, reduce the proportional coefficient K P , increase the integral coefficient K I , reduce the differential coefficient K D ;

[0134] Specific as Figure 7 The fuzzy rule table shown.

[0135] Defuzzification: The fuzzy output values ​​obtained according to the fuzzy rules correspond to seven output fuzzy sets. The membership functions all use trigonometric functions. The centroid method is used to convert the fuzzy output values ​​into three clear values ​​under the joint action of multiple output sets. They are the corrected values ​​of the proportional integral differential coefficients obtained in three different domains.

[0136] Output: Converted into a precise pulse-width modulation (PWM) duty cycle adjustment according to Formula I. The output value range is limited to [-1000, 1000], where 1000 corresponds to a forward cooling duty cycle of 100%, and -1000 corresponds to a negative heating duty cycle of 1000. The forward and reverse current conversion is achieved through an H-bridge. The dynamically adjusted pulse-width modulation (PWM) signal drives the vertical cavity surface emitting laser (VCSEL) through the H-bridge to precisely control the temperature.

[0137] The fuzzy PID control adopts the Mamdani type fuzzy controller, the core of which adopts the inference method of approximate inference, superimposing the results of multiple fuzzy sets to obtain the fuzzy set corresponding to the output, and using the centroid method to convert it into a clear value to replace this fuzzy set; the input is defined as the error e(t) and the error change rate de(t) / dt ;;The output of fuzzy PID control is ΔK p , ΔKI , ΔK D .

[0138] The fuzzy PID control realizes the output function through the multiplier and accumulator embedded in the core control unit, and the multiplication and addition operations are respectively realized by the multiplier and the adder.

[0139] The fuzzy PID control includes a membership function optimized by a genetic algorithm, and the membership function optimized by the genetic algorithm includes:

[0140] S1: Encode the rule base of the fuzzy PID controller using a real number encoding method to generate an initial rule base;

[0141] S2: Use the evaluation function to evaluate the fitness of the initial rule base;

[0142] S3: Then the encoded rule base is subjected to selection, crossover and mutation processes to generate an optimized new rule base;

[0143] S4: Re-evaluate the fitness of the new rule base. If the fitness minimum condition is met, decode and output the optimal fuzzy rule table and update the rule base.

[0144] S5: Real-time acquisition of the error value of the controlled object, through fuzzy processing, fuzzy rule mapping and defuzzification operation, calculate the error value of the controlled object. ;

[0145] S6: 、 、 and K p0 ，K I0 ，K D0 Superposition, get K p ，K I ，K D ， Realize dynamic adjustment of PID control parameters.

[0146] The fuzzy PID control optimizes the rule base through experimental data or genetic algorithm, merges similar rules, and constructs a partition control rule set for rapid heating in low temperature areas and fine adjustment in high temperature areas.

[0147] like Figure 6 As shown, the microwave source signal modulation system includes a full digital phase-locked loop, a direct digital synthesizer and a frequency multiplier, and realizes the generation, modulation and amplification of microwave signals through the FPGA control.

[0148] The output end of the direct digital synthesizer is connected in series with a bandpass filter and then connected to a fully digital phase-locked loop;

[0149] The frequency multiplier output microstrip line is connected to the input end of the microwave amplifier, and the output end of the microwave amplifier is directly connected to a three-port DC bias network Bias-Tee for biasing.

[0150] In this embodiment 1, the microwave source signal modulation system further includes a reference frequency source, a microwave amplifier and a modulation control unit;

[0151] The reference frequency source is responsible for providing a stable reference frequency for the direct digital frequency synthesizer (DDS) module. It includes a high-precision temperature-compensated crystal oscillator (TCXO), which is soldered to a reserved location on the printed circuit board (PCB). A 3.3V regulated power supply powers the TCXO, and the output of the TCXO is connected to the reference clock input pin of the DDS via a PCB trace. The TCXO has an output frequency of 10 MHz and a stability of ±0.1 ppm.

[0152] In printed circuit board (PCB) design, decoupling capacitors are added around the temperature compensation crystal oscillator (TCXO) to filter out power supply noise and ensure signal purity.

[0153] The output end of the fully digital phase-locked loop is connected to a second-order active loop filter and then connected to a frequency multiplier via a microstrip line;

[0154] The fully digital phase-locked loop has a built-in voltage-controlled oscillator (VCO) and a digital phase detector. The fully digital phase-locked loop uses a core control unit to configure a register (Register) via a serial peripheral interface (SPI) to set an integer frequency division ratio. INT and fractional frequency ratios FRAC , the calculation formula is:

[0155] ;

[0156] In this embodiment 1, is the output frequency, is the input reference frequency, INT is an integer division ratio, FRAC is the fractional division ratio, MOD The modulus of the fractional frequency division is used to control the resolution of the decimal place. INT, FRAC, MOD Adjust according to the chip manual and control the chip to output a 1.70875GHz signal.

[0157] In this embodiment 1, the chip manual uses the ADF4351 chip of Analog Devices, and the registers of the ADF4351 are written in sequence through the FPGA (the following is in hexadecimal):

[0158] Register 5: 0x00580005;

[0159] Register 4: 0x00800042 (enable divider ÷2 and output);

[0160] Register 3: 0x00000003 (default configuration);

[0161] Register 2: 0x0000401F (R = 1, charge pump current 5 mA);

[0162] Register 1: 0x00008004 (MOD = 4, 8 / 9 divider);

[0163] Register 0: 0x00155003 (INT=341, FRAC=3);

[0164] By configuring the integer divide ratio of the ADF4351 ( INT Set to INT =341), fractional division ratio (set FRAC to FRAC =3, MOD Set to MOD =4) and an RF divider (÷2) can multiply the 10 MHz reference signal to 1.70875 GHz. In actual applications, the loop filter and output matching must be adjusted according to the hardware environment.

[0165] A fully digital phase-locked loop (DPLL) multiplies a 100 MHz signal to 1.70875 GHz. Using a DPLL chip with a built-in VCO and digital phase detector, the 100 MHz signal from the direct digital synthesizer (DDS) module is connected to the DPLL's reference input pin. By setting the frequency division ratio, a 1.70875 GHz signal is output. A second-order active loop filter is connected to the output to suppress noise, and then the signal is connected to the subsequent frequency multiplier via a microstrip line.

[0166] The frequency multiplier includes a first-stage frequency multiplier and a second-stage frequency multiplier; both the first-stage frequency multiplier and the second-stage frequency multiplier are frequency multipliers with double frequency; and both the first-stage frequency multiplier and the second-stage frequency multiplier are provided with a bandpass filter capable of filtering the fundamental frequency. In this embodiment 1, two cascaded frequency multipliers are used to quadruple the frequency of the 1.70875 GHz signal, and after frequency multiplication, the desired 6.835 GHz signal is obtained. Two double-frequency multipliers are cascaded, and after the first stage, the 1.7 GHz fundamental frequency is filtered through a bandpass filter, retaining a frequency centered at 3.4 GHz. After the second stage, the 3.4 GHz fundamental frequency is filtered through a bandpass filter, retaining a frequency centered at 6.8 GHz.

[0167] The microwave amplifier amplifies the microwave signal and feeds it into a three-port DC bias network, Bias-Tee, to modulate the laser. A microwave amplifier chip with a gain of approximately 15 dB is used to boost the 6.835 GHz signal power to the required range. The output of the fully digital phase-locked loop (DPLL) is connected to the amplifier chip input via a 50-ohm microstrip line, while the output is directly connected to the three-port DC bias network, Bias-Tee, with a 5V supply voltage.

[0168] Verilog code is written using an FPGA development board to generate modulation signals, and the direct digital synthesizer (DDS) chip and the fully digital phase-locked loop (DPLL) chip are controlled through the serial peripheral interface (SPI) to achieve microwave frequency modulation and phase modulation.

[0169] In addition, in this embodiment 1, the components are packaged in SMD, the temperature compensated crystal oscillator TCXO is placed near the power supply, the direct digital frequency synthesizer DDS and the fully digital phase-locked loop DPLL are close to the center, the amplifier is close to the output end, and the FPGA is located in the control area; the high-frequency signal is routed through a 50-ohm microstrip line, the power layer and the ground layer are separated to reduce interference, and after the components are soldered, a metal shield is used to cover the high-frequency module and connected to the ground layer to reduce electromagnetic interference, and the output end is connected to the three-port DC bias network Bias-Tee through the RF interface.

[0170] In summary, a miniaturized microwave source controlled by FPGA can be realized, which can output a microwave signal with a frequency of 6.835 GHz and adjustable frequency and phase.

[0171] The embodiments described above are some of the embodiments of the present invention, rather than all of them. The detailed description of the embodiments of the present invention is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort are also within the scope of protection of the present invention.

Claims

1. A laser control system for a coherent population trapping magnetometer, characterized in that: It includes a core control unit, a driving circuit system, a microwave source signal modulation system, and an H-bridge bidirectionally connected to the core control unit; The driving circuit system includes a bandgap reference voltage circuit, and the bandgap reference voltage circuit includes an input low-power bias startup circuit and a folding operational amplifier; The core control unit includes an FPGA and a temperature control circuit system. The temperature control circuit system adopts an anti-integral windup algorithm combined with fuzzy PID control to dynamically adjust PID parameters to achieve temperature regulation; The microwave source signal modulation system includes a fully digital phase-locked loop, a direct digital synthesizer and a frequency multiplier, and realizes the generation, modulation and amplification of microwave signals through the control of the FPGA; The fuzzy PID control includes a membership function optimized by a genetic algorithm, and the membership function optimized by the genetic algorithm includes: S1: Encode the rule base of the fuzzy PID controller using a real number encoding method to generate an initial rule base; S2: Use the evaluation function to evaluate the fitness of the initial rule base; S3: Then the encoded rule base is subjected to selection, crossover and mutation processes to generate an optimized new rule base; S4: Re-evaluate the fitness of the new rule base. If the fitness minimum condition is met, decode and output the optimal fuzzy rule table and update the rule base. S5: Real-time acquisition of the error value of the controlled object, and calculation of the control parameters through fuzzy processing, fuzzy rule mapping and defuzzification operations. ; S6: 、 、 and K p0 ，K I0 ，K D0 Superposition, get K p ，K I ，K D ， Realize dynamic adjustment of PID control parameters; The membership function in the fuzzy PID control is defined by the Lyapunov function to define the energy function , under the constraints of the anti-saturation mechanism, the control rate satisfies ,in, e(t) is the error, is the error change rate, t is the time; The anti-integral saturation algorithm specifically includes: obtaining the current temperature T in real time current , and calculate the error e(t) and the error change rate , t is time; if e(t)>e set , directly output pulse width modulation PWM duty cycle 100%, drive H bridge to run at full power, skip PID adjustment, and reduce error as quickly as possible; if e(t)≤e set , then enter the fuzzy PID control link, according to the error e (t) and the error change rate Dynamically adjust the control parameters; where e set The target temperature group is set.

2. The laser control system according to claim 1, characterized in that: The bandgap reference voltage circuit further includes a bipolar transistor Q1 and a bipolar transistor Q2; one end of the bipolar transistor Q1 is connected to the drain of the MOS transistor M15, and the other end is connected to the resistor R2; the source of the MOS transistor M15 is connected to the resistor R6; The emitter of the bipolar transistor Q1 is grounded through the resistor R2. Feedback resistor R4 is connected in parallel with capacitor CFF and then output V out ; and satisfy ; ； in, V ref is the bandgap reference voltage; V out is the output voltage; V PTAT It is a voltage that is proportional to absolute temperature.

3. The laser control system according to claim 1, characterized in that: The anti-integral saturation algorithm specifically includes: inputting a negative temperature coefficient temperature measurement module NTC to obtain the current temperature T in real time current , the temperature measurement module NTC continuously monitors the current temperature T current , update, and form a closed-loop control; The expression of the anti-integral windup algorithm is formula 1: Formula I; Formula II; in is the proportional coefficient corrected by fuzzy PID, is the integral coefficient corrected by fuzzy PID, is the differential coefficient corrected by fuzzy PID, is the initial value of the proportional coefficient, is the initial value of the integral coefficient, is the initial value of the differential coefficient, is the proportional coefficient correction value, is the integral coefficient correction value, is the correction value of the differential coefficient, is the instantaneous value of the error fed back by the system, τ is the integration time constant, go through PID Algorithm-controlled temperature control circuit system output.

4. The laser control system according to claim 1, characterized in that: The output end of the fully digital phase-locked loop is connected to a second-order active loop filter and then connected to a frequency multiplier via a microstrip line; The fully digital phase-locked loop has a built-in voltage-controlled oscillator (VCO) and a digital phase detector. The fully digital phase-locked loop uses a core control unit to configure a register (Register) via a serial peripheral interface (SPI). INT and FRAC , the calculation formula is: ； in, is the output frequency, is the input reference frequency, INT is an integer division ratio, FRAC is the fractional division ratio, MOD The modulus of the fractional frequency division is used to control the resolution of the decimal place. INT, FRAC, MOD Adjust according to the chip manual; The frequency multiplier output microstrip line is connected to the input end of the microwave amplifier, and the output end of the microwave amplifier is directly connected to a three-port DC bias network Bias-Tee for biasing.

5. The laser control system according to claim 1, characterized in that: The microwave source signal modulation system further includes a reference frequency source, a microwave amplifier and a modulation control unit; The reference frequency source includes a high-precision temperature-compensated crystal oscillator, which is powered by a 3.3V regulated power supply. The output end of the high-precision temperature-compensated crystal oscillator is connected to the reference clock input pin of the direct digital synthesizer through a printed circuit board (PCB) trace.

6. The laser control system according to claim 1, characterized in that: The driving circuit system further comprises a digital-to-analog converter connected to the core control unit, and a low voltage difference linear regulator arranged between the digital-to-analog converter and the bandgap reference voltage circuit.

7. The laser control system according to claim 1, characterized in that: The frequency multiplier includes a primary frequency multiplier and a secondary frequency multiplier; both the primary frequency multiplier and the secondary frequency multiplier are frequency multipliers with doubled frequency; and both the primary frequency multiplier and the secondary frequency multiplier are followed by a bandpass filter capable of filtering the fundamental frequency.

8. The laser control system according to claim 5, wherein: The output frequency of the high-precision temperature-compensated crystal oscillator is 10 MHz and the stability is ±0.1 ppm.

Images