Polarization difference detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring

Abstract

This invention relates to a zero-point drift compensation method for polarization differential detection based on in-situ temperature-strain synchronous monitoring. Without altering the existing optical architecture, it achieves online compensation for slowly varying drifts of polarization optical devices during long-term continuous operation, thereby improving the long-term stability of the system. Specifically, this invention addresses the output drift of key polarization optical devices in the polarization detection optical path caused by the coupling effect of environmental temperature fluctuations and structural stress. Firstly, platinum resistance temperature sensors and resistance strain gauges are deployed around the key polarization optical devices, and a synchronous data acquisition module is used to achieve high-precision synchronous acquisition of temperature, strain, and system output signals. The long-term acquired temperature-strain-system output data sequence is used to drive a drift compensation model, which estimates the output drift in real time and generates a compensation amount to correct the system output online.

Description

Technical Field

[0001] This invention relates to the field of precision optical detection and signal compensation technology, specifically to a method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection. This method is applicable to the suppression of long-term slow drift in precision measurement systems that employ polarization differential detection structures, such as atomic magnetic measurement and atomic inertial measurement. Background Technology

[0002] With the development of new theories and technologies in the quantum field, precision measurement devices based on quantum effects are constantly breaking through the measurement limits of traditional instruments. In precision measurement systems such as atomic magnetic measurement and atomic inertial measurement, polarization beam-splitting optical devices are usually used for dual-path differential detection to achieve signal output. The long-term stability of the polarization detection optical path has a significant impact on the long-term continuous operation and output stability of the system.

[0003] The stability of the polarization detection optical path is a prerequisite for the precision measurement system described above. The long-term stability of the system output is greatly affected by the ambient temperature and the stress deformation of the polarization optical components, which are important factors causing slow drift in the system. Conventional solutions mainly rely on high-precision temperature control systems and have not yet established a correlation compensation model between multi-dimensional environmental parameters and system output drift. Due to the lack of real-time sensing of the strain dimension and dynamic correction at the software level, the system cannot effectively compensate for the polarization error induced by stress fluctuations in the optical components, making it difficult to fundamentally suppress the long-term drift of the polarization detection signal. Therefore, it is necessary to propose a zero-point drift compensation method for polarization differential detection based on in-situ temperature-strain synchronous monitoring to improve the long-term stability of the system without changing the existing polarization detection structure. Summary of the Invention

[0004] This invention provides a method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring of polarization differential detection. It is applicable to suppressing the long-term slow drift of polarization detection signals in precision measurement systems that employ polarization differential detection structures, such as atomic magnetic measurement and atomic inertial measurement. The technical problem it solves is that during long-term continuous operation of related precision measurement systems, the output signal exhibits slow drift due to gradual changes in the environment, which affects the long-term stability of the system.

[0005] The technical solution of the present invention is as follows:

[0006] A method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection is characterized by the following steps:

[0007] Step 1: In a precision measurement system employing a polarization differential detection structure, a platinum resistance temperature sensor and a resistance strain gauge are respectively installed on the bottom surface of the polarization beam splitter on the outgoing light path of the atomic gas cell. The platinum resistance temperature sensor and the resistance strain gauge are respectively connected to the host computer through a synchronous data acquisition module. The transmission side of the polarization beam splitter is connected to the first input terminal of the differential amplifier circuit through a second photodetector. The reflection side of the polarization beam splitter is connected to the second input terminal of the differential amplifier circuit through a first photodetector. The output terminal of the differential amplifier circuit is connected to the host computer through the synchronous data acquisition module. The reflection side of the non-polarization beam splitter on the incident light path of the atomic gas cell is connected to the host computer sequentially through a third photodetector and the synchronous data acquisition module. The liquid crystal variable phase delay device on the incident light path of the atomic gas cell is connected to the host computer through the synchronous data acquisition module.

[0008] Step 2: Establish the drift equivalent equation of the differential output signal with respect to temperature change, strain and cross-coupling terms, and introduce random noise term to characterize measurement uncertainty and unmodeled disturbance;

[0009] Step 3: The drift model parameters are updated online using recursive estimation, and the drift estimate is obtained in real time. Specifically, based on the drift equivalent equation, the drift model parameters are updated online using adaptive Kalman filtering based on the synchronous observation data of temperature, strain and differential output, so as to obtain the differential output drift estimate caused by temperature-strain coupling in real time.

[0010] Step 4: Generate a compensation amount based on the drift estimate, perform online correction on the differential output signal to obtain the online compensated differential output signal, and use the online compensated differential output signal for system output calculation to suppress the slow drift of the polarization detection optical path during long-term continuous operation.

[0011] In step 1, the input side of the polarizing beam splitter is connected to the laser emission side of the atomic gas chamber via a third half-wave plate. The atomic gas chamber is located inside a non-magnetic oven, which is located inside a triaxial magnetic field coil. The laser incident side of the atomic gas chamber is connected to the DFB laser in sequence via the non-polarizing beam splitter, the second half-wave plate, the second linear polarizer, the liquid crystal variable phase delayer, the first linear polarizer, and the first half-wave plate.

[0012] Step 2 includes the following expression:

[0013] ;

[0014] ;

[0015] ;

[0016] in Let k be the differential output observation at time k, where k is a positive integer. It is the current input vector. It is a parameter vector. For random noise, It is the change in temperature. It is the dependent variable. It is a cross-coupling term of temperature and strain. It is a bias term. It is the temperature sensitivity coefficient. It is the sensitivity coefficient of the strain term. It is the sensitivity coefficient of the cross-coupling term. , , All of these are online estimates, which enable dynamic characterization of output drift characteristics and adaptive parameter mapping under multiphysics coupling environments.

[0017] Step 3 includes the following expression:

[0018] ;

[0019] in It is the output drift estimate.

[0020] Step 4 includes the following expression:

[0021] ;

[0022] in It is the differential output signal quantity after online compensation. The differential output observations are at time k.

[0023] The technical effects of this invention are as follows: This invention, based on an in-situ temperature-strain synchronous monitoring method for zero-point drift compensation in polarization differential detection, provides a feasible solution for suppressing slow drift of polarization optical devices during long-term continuous operation and improving the long-term stability of the system without changing the existing optical structure. This invention takes the output drift of key polarization optical devices in the polarization detection optical path caused by the coupling effect of environmental temperature fluctuations and structural stress as the compensation object. First, platinum resistance temperature sensors and resistance strain gauges are deployed around the key polarization optical devices, and a synchronous data acquisition module is used to achieve high-precision synchronous acquisition of temperature, strain, and system output signals. Second, this invention constructs an equivalent observation model for the differential output drift, introducing temperature change, strain, and their cross-coupling terms into the drift parameterization expression, and establishing a recursive estimation model between differential output observations and multi-physical quantity joint observation data. Then, the long-term acquired temperature-strain-system output data sequence is used to drive the drift compensation model, estimate the output drift in real time, generate compensation amounts, and perform online correction of the system output. The drift compensation scheme of the present invention is an online, low-intrusion compensation method that can improve the resistance to environmental disturbances and long-term stability of the polarization detection optical path in complex and slowly changing environments.

[0024] The advantages of this invention compared to existing technologies are as follows: This invention achieves the construction of drift driving quantities without changing the existing polarization detection optical structure by synchronously acquiring temperature-strain in-situ monitoring and differential output; and by combining a drift equivalent model including temperature, strain, and cross-coupling terms with recursive estimation, it can track slow time-varying drifts online; and the compensation process of this invention is an online compensation method, which facilitates engineering implementation, system integration, and long-term operation deployment. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the atomic precision measurement system involved in the polarization difference detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of the present invention.

[0026] Figure 2 This is a schematic diagram of the output drift estimation process involved in the polarization difference detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of the present invention. Figure 2 This includes step 1, synchronous data acquisition. , and , Let k be the differential output observation at time k, where k is a positive integer. It is the temperature resistance measurement value obtained synchronously at time k. It is the measured value of the strain gauge at time k obtained through synchronous acquisition; Step 2, construct the current input vector. , , It is the change in temperature. It is the dependent variable. It is a cross-coupling term of temperature and strain; Step 3, establish an equivalent observation model, , For the differential output observation at time k, It is the current input vector. It is a parameter vector. For the random noise term; in step 4, the Kalman recursion online estimation is used to update the result. ,and , It predicts the covariance matrix, which is used to characterize the recursive propagation of parameter estimation uncertainty; step 5, outputs the drift estimate. , , It is the temperature sensitivity coefficient. It is the sensitivity coefficient of the strain term. It is the sensitivity coefficient of the cross-coupling term. , , All of these are online estimates, which enable dynamic characterization of output drift characteristics and adaptive parameter mapping under multiphysics coupling environments.

[0027] Figure 3 This is a schematic diagram of the online compensation process involved in the polarization differential detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of the present invention. It includes step 1, input... , and Step 2: Read the recursive parameter estimation results. , , Step 3: Calculate the output drift estimate. , Step 4: Output the online compensation amount for drift. , Step 5: Return to step 1 and begin the next cycle.

[0028] The reference numerals in the attached figures are explained as follows: 1-DFB laser (Distributed Feedback Laser); 2-First half-wave plate (HWP1); 3-First linear polarizer (LP1); 4-Liquid Crystal Variable Retarder (LCVR); 5-Second linear polarizer (LP2); 6-Second half-wave plate (HWP2); 7-Non-polarizing beamsplitter (NPBS); 8-Triaxial magnetic field coil; 9-Non-magnetic oven; 10-Atomic gas chamber; 11-Third half-wave plate (HWP3); 12-First photoelectric detector (PD1); 13-Differential amplifier circuit; 14-Second photoelectric detector (PD2); 15-Platinum resistance temperature sensor; 16-Resistance strain gauge; 17-Polarizing beamsplitter (NPBS) 18-Splitter (PBS); 19-Third photodetector (PD3); 20-Synchronous data acquisition module; 21-Host computer. Detailed Implementation

[0029] The following is in conjunction with the attached diagram ( Figures 1-3 The present invention will be described in conjunction with the examples.

[0030] Figure 1 This is a schematic diagram of the atomic precision measurement system involved in the polarization difference detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of the present invention. Figure 2 This is a schematic diagram of the output drift estimation process involved in the polarization difference detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of the present invention. Figure 3 This is a schematic diagram of the online compensation process involved in the polarization difference detection zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of the present invention. (Reference) Figures 1 to 3As shown, the zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring of polarization differential detection includes the following steps: Step 1, in a precision measurement system employing a polarization differential detection structure, a platinum resistance temperature sensor 15 and a resistance strain gauge 16 are respectively installed on the bottom surface of the polarization beam splitter 17 on the outgoing light path of the atomic gas cell 10. The platinum resistance temperature sensor 15 and the resistance strain gauge 16 are respectively connected to the host computer 20 through a synchronous data acquisition module 19. The transmission side of the polarization beam splitter 17 is connected to the first input terminal of the differential amplifier circuit 13 through a second photodetector 14. The reflection side of the polarization beam splitter 17 is connected to the second input terminal of the differential amplifier circuit 13 through a first photodetector 12. The output terminal of the differential amplifier circuit 13 is connected to the host computer 20 through the synchronous data acquisition module 19. The reflection side of the non-polarization beam splitter 7 on the incident light path of the atomic gas cell 10 is connected to the third photodetector 18 and the synchronous data acquisition module 19 in sequence. The host computer 20 is connected to the liquid crystal variable phase delay device 4 on the incident light path of the atomic gas chamber 10 through the synchronous data acquisition module 19. Step 2: Establish the drift equivalent equation of the differential output signal with respect to temperature change, strain and cross-coupling terms, and introduce random noise terms to characterize measurement uncertainty and unmodeled disturbances. Step 3: Update the drift model parameters online using recursive estimation and obtain the drift estimate in real time. Specifically, based on the drift equivalent equation, according to the synchronous observation data of temperature, strain and differential output, the drift model parameters are updated online using adaptive Kalman filtering to obtain the differential output drift estimate caused by temperature-strain coupling in real time. Step 4: Generate a compensation amount based on the drift estimate, and correct the differential output signal online to obtain the online compensated differential output signal. Use the online compensated differential output signal for system output calculation to suppress the slow drift of the polarization detection optical path during long-term continuous operation.

[0031] In step 1, the input side of the polarizing beam splitter 17 is connected to the laser emission side of the atomic gas chamber 10 through the third half-wave plate 11. The atomic gas chamber 10 is located inside the non-magnetic oven 9, which is located inside the triaxial magnetic field coil 8. The laser incident side of the atomic gas chamber 10 is connected to the DFB laser 1 in sequence through the non-polarizing beam splitter 7, the second half-wave plate 6, the second linear polarizer 5, the liquid crystal variable phase delay 4, the first linear polarizer 3, and the first half-wave plate 2.

[0032] Step 2 includes the following expression:

[0033] ;

[0034] ;

[0035] ;

[0036] in Let k be the differential output observation at time k, where k is a positive integer. It is the current input vector. It is a parameter vector. For random noise, It is the change in temperature. It is the dependent variable. It is a cross-coupling term of temperature and strain. It is a bias term. It is the temperature sensitivity coefficient. It is the sensitivity coefficient of the strain term. It is the sensitivity coefficient of the cross-coupling term. , , All of these are online estimates, which enable dynamic characterization of output drift characteristics and adaptive parameter mapping under multiphysics coupling environments.

[0037] Step 3 includes the following expression:

[0038] ;

[0039] in It is the output drift estimate.

[0040] Step 4 includes the following expression:

[0041] ;

[0042] in It is the differential output signal quantity after online compensation. The differential output observations are at time k.

[0043] This invention discloses a zero-point drift compensation method for polarization differential detection based on in-situ temperature-strain synchronous monitoring, providing a feasible solution to suppress the slow drift of polarization optical devices during long-term continuous operation and improve the long-term stability of the system without changing the existing optical architecture. This invention focuses on the output drift of key polarization optical devices in the polarization detection optical path caused by the coupling effect of environmental temperature fluctuations and structural stress. First, platinum resistance temperature sensors and resistance strain gauges are deployed around the key polarization optical devices, and a synchronous data acquisition module is used to achieve high-precision synchronous acquisition of temperature, strain, and system output signals. Second, this invention constructs an equivalent observation model for the differential output drift, incorporating temperature change, strain, and their cross-coupling terms into the drift parameterization expression, and establishing a recursive estimation model between the differential output observation and the joint observation data of multiple physical quantities. Then, the long-term acquired temperature-strain-system output data sequence is used to drive the drift compensation model, estimating the output drift in real time and generating compensation amounts for online correction of the system output. The drift compensation scheme of this invention is an online, low-intrusion compensation method that can improve the resistance to environmental disturbances and long-term stability of the polarization detection optical path under complex, slowly changing environments.

[0044] A zero-point drift compensation method for polarization differential detection based on in-situ temperature-strain synchronous monitoring includes: a DFB laser (Distributed Feedback Laser, DFB) (1), a first half-wave plate (HWP1) (2), a first linear polarizer (LP1) (3), a liquid crystal variable phase retarder (LCVR) (4), a second linear polarizer (LP2) (5), a second half-wave plate (HWP2) (6), a non-polarizing beamsplitter (NPBS) (7), a triaxial magnetic field coil (8), a non-magnetic oven (9), an atomic gas chamber (10), a third half-wave plate (HWP3) (11), a first photoelectric detector (PD1) (12), a differential amplifier circuit (13), a second photoelectric detector (PD2) (14), a platinum resistance temperature sensor (15), a resistance strain gauge (16), and a polarizing beamsplitter (NPBS). splitter (PBS) (17), third photodetector (PD3) (18), synchronous data acquisition module (19), host computer (20).

[0045] Step 1: Construct the equivalent equation for the polarization differential detection optical path drift, including a random noise term, and complete the input construction and parameter initialization. A platinum resistance temperature sensor and a strain gauge are integrated in situ on the surface of the polarization beam splitter to acquire temperature and strain signals. The differential output signal is obtained by a differential amplifier circuit. Temperature, strain, and differential output signals are simultaneously acquired via a synchronous data acquisition module. The platinum resistance measurement value and the strain resistance measurement value are converted into temperature change and strain, respectively, and the temperature change, strain, and their cross-coupling term are constructed as drift estimation inputs. Under conditions without external excitation, the system's differential output signal is acquired and averaged to determine the initial zero-bias estimate of the system, which is used as the initial value of the bias term in the temperature-strain drift model. Based on this, the drift equivalent equation for the differential output signal with respect to temperature change, strain, and cross-coupling term is established, and a random noise term is introduced to characterize measurement uncertainty and unmodeled disturbances.

[0046] Step 2: Update the drift model parameters online using recursive estimation and obtain the drift estimate in real time. Based on the equivalent equation in Step 1, and using synchronous observation data of temperature, strain, and differential output, an adaptive Kalman filter is used to update the drift model parameters online, thereby obtaining the differential output drift estimate caused by temperature-strain coupling in real time.

[0047] Step 3: Generate compensation amount based on drift estimate and implement online compensation. The compensation amount is generated based on the drift estimate obtained in Step 2, and the differential output signal is corrected online to obtain the compensated differential output signal. The compensated differential output signal is used for system output calculation to suppress the slow drift of the polarization detection optical path during long-term continuous operation.

[0048] The platinum resistance thermometer and strain gauge serve as multi-physical quantity monitoring sensors for the polarization differential detection optical path, and are mounted on the surface of the polarization beam splitter. The strain gauge is used to characterize the structural deformation state of the polarization optical device caused by assembly clamping force and environmental changes, while the platinum resistance thermometer is used to characterize the local thermal state of the device. The synchronous data acquisition module synchronously acquires temperature signals, strain signals, and differential output signals under a unified sampling clock, and forms a joint observation data sequence of temperature-strain-differential output for recursive estimation in step two.

[0049] The drift equivalent equation established in step one is: Where k represents the k-th time point. The differential output observations at time k; For the input vector, The change in temperature As the dependent variable, This is a cross-coupling term of temperature and strain; Let be the vector of parameters to be estimated, where For bias terms, , , These are the sensitivity coefficients for the temperature term, strain term, and cross-coupling term, respectively. This represents random noise.

[0050] In step two, based on the observation residuals of the drift equivalent equation, a Kalman filter is used to apply the parameter vector. Perform online recursive updates; obtain the drift estimate at time k as follows: And in step three, through The compensated differential output signal is obtained.

[0051] A zero-point drift compensation method based on in-situ temperature-strain synchronous monitoring and polarization differential detection includes the following steps:

[0052] Step 1: The differential output observations are synchronously acquired by the synchronous data acquisition module. Temperature resistance measurement value With strain resistance measurement value The average value of the initial stable data segment was used as the reference resistance. , Convert the resistance measurement value into temperature change. With dependent variable And construct a cross-coupling term for temperature and strain. Construct the input vector Establish the drift equivalent equation ;where the parameter vector For the quantity to be estimated, where For bias terms, , , These are the sensitivity coefficients for the temperature term, strain term, and cross-coupling term, respectively. This represents the random noise term (the subscript k above represents the k-th time). The drift model parameters are initialized based on the initial data segment to obtain initial parameter values. Then, in step two, the parameter vector is updated online using Kalman recursion.

[0053] Step 2: Convert the parameter vector It is treated as a slowly varying stochastic process over time, and an online recursive estimation is performed using Kalman filtering. Let the process noise covariance be Q and the measurement noise covariance be R. Then, at time k, covariance prediction is first performed. Then based on the current input vector Perform observation updates and calculate observation residuals. ; Calculate Kalman gain And recursively update the parameter vector With covariance The above recursive update yields... , , The online estimation value enables dynamic characterization and adaptive parameter mapping of output drift characteristics under multiphysics coupling environment.

[0054] Step 3: Based on the parameter vector obtained in Step 2 Calculate the drift estimate caused by the temperature term, strain term, and cross-coupling term. The differential output observations are then corrected online to obtain the compensated differential output signal. Among them, the drift estimator Excluding bias terms That is, during compensation, the DC baseline component of the differential output is retained, while only the slowly varying drift components corresponding to the temperature, strain, and cross-coupling terms are suppressed.

[0055] refer to Figure 1 The precision measurement system employs a polarization differential detection structure for polarization detection. The detection light, emitted from the laser, passes sequentially through a pre-stage polarization modulation and shaping optical component to precisely set and modulate the polarization state of the incident light. Subsequently, a non-polarizing beam splitter separates one beam for photodetector PD3, which is used for laser power monitoring and forms a power stabilization loop with the laser power control system. The main optical path then enters the atomic gas chamber. Following this, the detection light carrying atomic information exits the gas chamber and is separated into two polarized beams by a subsequent polarization analysis component, which are received by photodetectors PD1 and PD2, respectively. The two detection signals are differentially amplified to obtain a differential output signal. A platinum resistance temperature sensor and a resistance strain gauge are arranged on the surface of a polarizing beam splitter to acquire temperature and resistance signals. With strain resistance signal The synchronous data acquisition module performs data acquisition under a unified sampling clock. , , Multi-channel synchronous acquisition and recording are performed, and the joint observation data is transmitted to the host computer for subsequent drift analysis and compensation processing.

[0056] refer to Figure 2 The output drift estimation process of a polarization detection system based on temperature-strain joint monitoring. The differential output observations at time k; and These are the measured values ​​of the temperature resistance and strain resistance at time k, obtained synchronously. To eliminate differences in the initial state of the devices and establish a reference for physical quantity conversion, this invention selects an initial data interval (denoted as k=1,2,…N0) where the output is in a quasi-steady state during the initial stage after system startup, and takes the average value of the temperature sensing resistance and strain sensing resistance within this interval, defining the reference resistance as... , .in , The resistance values ​​of the temperature sensing resistor and the strain sensing resistor under reference conditions are respectively characterized within the initialization interval. The resistance signals are further converted into temperature changes. With dependent variable .in The temperature coefficient of the temperature sensing resistor. The sensitivity coefficient of the strain sensor resistor can be obtained from the device's nominal parameters, and a cross-coupling term for temperature and strain can be constructed. .

[0057] Based on this, construct the input vector. And define the parameter vector to be estimated as ,in For bias terms, , , These are the sensitivity coefficients for the temperature term, strain term, and cross-coupling term, respectively. A drift equivalent equation is established. ,in For the differential output observation at time k, This is a random noise term used to characterize measurement noise and unmodeled disturbances. Further, based on this equivalent equation, the parameter vector... The model is a stochastic process that changes slowly over time, and Kalman filtering is used to achieve online recursive estimation.

[0058] At time k, covariance prediction is performed first. Let the process noise covariance be Q, and the parameter estimation error covariance matrix be P:

[0059]

[0060] in The parameter estimation error covariance matrix is ​​the value from the previous time step. To predict the covariance matrix, which is used to characterize the recursive propagation of parameter estimation uncertainty; then based on the current input vector... Compared with the parameter estimate at the previous time step Construction observation residuals :

[0061]

[0062] Among the observation residuals The scalar residual characterizes the deviation between the current observation and the observation predicted based on the parameters from the previous time step. Further, the Kalman gain is calculated:

[0063]

[0064] in, This is the Kalman gain vector, used to implement the residual. Optimal allocation of parameter corrections; R is the measurement noise covariance, which is taken as a scalar in this invention to characterize the differential output observations. The impact of measurement noise and unmodeled high-frequency disturbances on the observations is considered. Finally, the parameter vector and covariance matrix are updated based on the Kalman recursive structure, and the parameter vector and covariance are updated recursively:

[0065] ,

[0066] The above recursive update yields the following results. , , The online estimation value enables dynamic characterization and adaptive parameter mapping of output drift characteristics under multiphysics coupling environment.

[0067] refer to Figure 3 The differential output online compensation process of the polarization detection system. Based on... Figure 2 From the recursive estimation results, the sensitivity coefficients corresponding to the temperature, strain, and cross-coupling terms are extracted, and the drift estimate caused by temperature-strain coupling is calculated. And generate the compensated differential output signal. Among them, the drift estimator It only addresses the slowly varying drift components caused by temperature, strain, and cross-coupling terms, and retains the baseline component of the differential output during the compensation process to meet the engineering requirements of avoiding miscompensation of the real signal and suppressing slowly varying drift under long-term system operation conditions.

[0068] Contents not described in detail in this specification are existing technologies known to those skilled in the art. It is hereby indicated that the above description is intended to help those skilled in the art understand the present invention, but does not limit the scope of protection of the present invention. Any equivalent substitutions, modifications, improvements, and / or simplifications of the above descriptions that do not depart from the essence of the present invention fall within the scope of protection of the present invention.

Claims

1. A method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection, characterized in that, Includes the following steps: Step 1: In a precision measurement system employing a polarization differential detection structure, a platinum resistance temperature sensor and a resistance strain gauge are respectively installed on the bottom surface of the polarization beam splitter on the outgoing light path of the atomic gas cell. The platinum resistance temperature sensor and the resistance strain gauge are respectively connected to the host computer through a synchronous data acquisition module. The transmission side of the polarization beam splitter is connected to the first input terminal of the differential amplifier circuit through a second photodetector. The reflection side of the polarization beam splitter is connected to the second input terminal of the differential amplifier circuit through a first photodetector. The output terminal of the differential amplifier circuit is connected to the host computer through the synchronous data acquisition module. The reflection side of the non-polarization beam splitter on the incident light path of the atomic gas cell is connected to the host computer sequentially through a third photodetector and the synchronous data acquisition module. The liquid crystal variable phase delay device on the incident light path of the atomic gas cell is connected to the host computer through the synchronous data acquisition module. Step 2: Establish the drift equivalent equation of the differential output signal with respect to temperature change, strain and cross-coupling terms, and introduce random noise term to characterize measurement uncertainty and unmodeled disturbance; Step 3: The drift model parameters are updated online using recursive estimation, and the drift estimate is obtained in real time. Specifically, based on the drift equivalent equation, the drift model parameters are updated online using adaptive Kalman filtering based on the synchronous observation data of temperature, strain and differential output, so as to obtain the differential output drift estimate caused by temperature-strain coupling in real time. Step 4: Generate a compensation amount based on the drift estimate, perform online correction on the differential output signal to obtain the online compensated differential output signal, and use the online compensated differential output signal for system output calculation to suppress the slow drift of the polarization detection optical path during long-term continuous operation.

2. The method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection according to claim 1, characterized in that, In step 1, the input side of the polarizing beam splitter is connected to the laser emission side of the atomic gas chamber via a third half-wave plate. The atomic gas chamber is located inside a non-magnetic oven, which is located inside a triaxial magnetic field coil. The laser incident side of the atomic gas chamber is connected to the DFB laser in sequence via the non-polarizing beam splitter, the second half-wave plate, the second linear polarizer, the liquid crystal variable phase delayer, the first linear polarizer, and the first half-wave plate.

3. The method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection according to claim 1, characterized in that, Step 2 includes the following expression: ； ； ； in Let k be the differential output observation at time k, where k is a positive integer. It is the current input vector. It is a parameter vector. For random noise, It is the change in temperature. It is the dependent variable. It is a cross-coupling term of temperature and strain. It is a bias term. It is the temperature sensitivity coefficient. It is the sensitivity coefficient of the strain term. It is the sensitivity coefficient of the cross-coupling term. , , All of these are online estimates, which enable dynamic characterization of output drift characteristics and adaptive parameter mapping under multiphysics coupling environments.

4. The method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection according to claim 1, characterized in that, Step 3 includes the following expression: ； in It is the output drift estimate.

5. The method for zero-point drift compensation based on in-situ temperature-strain synchronous monitoring and polarization differential detection according to claim 1, characterized in that, Step 4 includes the following expression: ； in It is the differential output signal quantity after online compensation. The differential output observations are at time k.

Images