Wind field measurement method of wind lidar based on single-stage amplification laser

Abstract

The present application relates to the technical field of wind field measurement, and more particularly to a wind field measurement method based on a single-stage amplification laser. The method obtains a pump driving current, a Doppler effective peak power, a wideband phase noise floor and a current spectrum accumulation time; subtracts the limit pump current threshold from the pump driving current to obtain a coherent current margin, and obtains a data effective rate based on the Doppler effective peak power and the wideband phase noise floor statistics; further calculates the effective rate time gain gradient and the bottom noise power evolution gradient, maps them to a two-dimensional state space to obtain the control demand effectiveness and the coherent conflict outbreak degree; finally, the working area is divided by threshold comparison, and the spectrum accumulation time adjustment and the pump driving current control instruction are generated. The method breaks the single variable control limitation, effectively improves the data effective rate under the premise of ensuring the safety of the device operation, and reduces the measurement delay and the computing power consumption caused by invalid accumulation.

Description

Technical Field

[0001] This invention relates to the field of wind field measurement technology, and in particular to a wind field measurement method based on a single-stage amplified laser lidar. Background Technology

[0002] In the application of coherent wind lidar, single-stage amplified lasers are often used as laser emitting units. However, existing equipment has significant limitations in its operation mode, with the transmitter power control and inversion end accumulation scheduling typically operating independently. Specifically, the transmitter adjusts the pump current only based on temperature or a fixed threshold, while the inversion end adjusts the accumulation time only based on the echo intensity, lacking state interaction and coordinated operation between the two. When the pump current approaches the critical value of stimulated Brillouin scattering, the nonlinear effect within the fiber intensifies, leading to a sharp decrease in signal coherence and a significant increase in noise floor. At this point, if the inversion end continues to blindly increase the accumulation time, it will not only fail to effectively improve data efficiency but will also introduce more noise components. Furthermore, excessively high pump currents pose a potential risk of damaging the gain fiber. Existing scheduling methods often employ fixed parameters or single-variable closed-loop control, either fixing the pump current and only adjusting the accumulation time, failing to fully utilize the pump current gain space, or fixing the accumulation time and only adjusting the pump current, which easily triggers coherence degradation. Therefore, how to quantify the coupling contradiction between the coherence risk on the transmitting side and the time gain on the scheduling side, and realize cross-end collaborative scheduling, has become a technical problem that urgently needs to be solved. Summary of the Invention

[0003] The main objective of this invention is to provide a wind field measurement method for wind lidar based on a single-stage amplified laser. By constructing a joint characterization system of the coherence state of the transmitter and the echo quality state, the efficient time gain gradient and the noise floor power evolution gradient are mapped to a two-dimensional conflict state space. This achieves the coordinated scheduling of pump drive current and spectrum accumulation time, breaking the single-variable control limitation caused by the independent operation of the transmitter and inversion end in existing equipment. This scheme can not only adaptively maximize the data efficiency of wind speed measurement while ensuring the safe operation of laser amplification devices and avoiding coherence degradation, but also effectively terminate the invalid integration waiting process of low-quality signals, significantly reduce measurement delay and computing power waste, and adapt to the parameter differences of different individuals without the need to establish a complex underlying mechanism model.

[0004] The technical solution of the present invention is as follows:

[0005] Firstly, a wind field measurement method based on a single-stage amplified laser lidar is proposed, which includes the following steps:

[0006] Obtain the pump drive current, Doppler effective peak power, broadband phase noise floor, and the current spectrum accumulation time;

[0007] The coherent current margin is obtained by subtracting the preset limit pump current threshold from the pump drive current.

[0008] The instantaneous signal-to-noise ratio is obtained by dividing the effective peak power of the Doppler by the broadband phase noise floor. The data efficiency is obtained by calculating the percentage of time during which the instantaneous signal-to-noise ratio is greater than the preset effective signal-to-noise ratio threshold.

[0009] The partial derivative of the data efficiency with respect to the spectrum accumulation time is calculated to obtain the efficiency time gain gradient, and the efficiency time gain gradient is used as the horizontal axis to control demand effectiveness; the partial derivative of the broadband phase noise floor with respect to the pump drive current is calculated to obtain the noise floor power evolution gradient, and the noise floor power evolution gradient is divided by the sum of the coherent current margin and the preset regularization constant, and used as the vertical axis to determine the coherent conflict burst degree.

[0010] When the effectiveness of the horizontal axis control requirement is greater than the preset efficiency gain determination threshold and the coherent conflict burst degree of the vertical axis is less than the preset coherent conflict tolerance threshold, the spectrum accumulation time is increased; when the effectiveness of the horizontal axis control requirement is less than or equal to zero and the coherent conflict burst degree of the vertical axis is greater than or equal to the preset coherent conflict tolerance threshold, the pump drive current is reduced and the spectrum accumulation time is locked.

[0011] A further improvement of the present invention is that the process of obtaining the pump drive current, the effective peak power of Doppler, and the broadband phase noise floor includes:

[0012] The voltage drop signal of the internal current sampling resistor of the laser amplifier is read by the driver board, and the voltage drop signal is converted into the pump drive current.

[0013] The thermistor resistance value attached to the heat sink surface of the amplifier gain fiber is read by the driver board, and the heat sink temperature is obtained by converting the thermistor resistance value.

[0014] The heterodyne beat frequency signal sequence output by the analog-to-digital converter is received, and a fast Fourier transform is performed on the heterodyne beat frequency signal sequence of fixed length to obtain the frequency domain spectrum;

[0015] The peak energy is extracted within the Doppler frequency shift interval of the frequency domain spectrum, and the peak energy is taken as the effective peak power of the Doppler.

[0016] The energy value is integrated within the high-frequency signal-free region of the frequency domain spectrum to obtain the broadband phase noise floor.

[0017] A further improvement of the present invention is that the step of subtracting the preset limiting pump current threshold from the pump drive current to obtain the coherent current margin includes:

[0018] Based on multiple sets of temperature and current data obtained from calibration, the preset limit pump current threshold is obtained by interpolation according to the heat sink temperature.

[0019] The coherent current margin is obtained by subtracting the preset limit pump current threshold from the pump drive current.

[0020] The thermal lens evolution rate is obtained by taking the first derivative of the heat sink temperature in the time dimension and using the difference between the temperature difference of two adjacent sampling times divided by the sampling interval.

[0021] The coherent current margin and the thermal lens evolution rate are combined into a two-dimensional column vector to generate the transmitter coherent execution state vector.

[0022] A further improvement of the present invention is that the step of dividing the effective peak power of the Doppler by the broadband phase noise floor to obtain the instantaneous signal-to-noise ratio, and calculating the percentage of time during which the instantaneous signal-to-noise ratio is greater than a preset effective signal-to-noise ratio threshold to obtain the data effectiveness includes:

[0023] The instantaneous signal-to-noise ratio is obtained by dividing the effective peak power of the Doppler by the broadband phase noise floor.

[0024] Within the time interval corresponding to the preset statistical time window, the percentage of time when the instantaneous signal-to-noise ratio is greater than the preset effective signal-to-noise ratio threshold is statistically analyzed using a step function to obtain the data effectiveness rate. When the input value of the step function is greater than zero, the step function outputs one; when the input value of the step function is less than or equal to zero, the step function outputs zero.

[0025] The data is efficiently combined with the current broadband phase noise floor to generate an echo quality state vector.

[0026] A further improvement of this invention is that the calculation of the partial derivative of the data efficiency with respect to the spectral accumulation time yields the efficiency time gain gradient, and the efficiency time gain gradient is used as the horizontal axis to control demand effectiveness; the calculation of the partial derivative of the broadband phase noise floor with respect to the pump drive current yields the noise floor power evolution gradient, and the noise floor power evolution gradient is divided by the sum of the coherent current margin and a preset regularization constant, and used as the vertical axis to determine the coherent conflict burst degree, including:

[0027] Obtain the difference in data efficiency and the difference in spectrum accumulation time within two adjacent scheduling periods, divide the difference in data efficiency by the difference in spectrum accumulation time, and calculate the efficiency time gain gradient.

[0028] The difference between the broadband phase noise floor and the difference between the pump drive current in two adjacent scheduling periods are obtained. The difference between the broadband phase noise floor and the difference between the pump drive current are divided to calculate the noise floor power evolution gradient.

[0029] The effective time gain gradient is directly used as the horizontal axis to control the effectiveness of the demand.

[0030] The noise floor power evolution gradient is used as the numerator, the sum of the coherent current margin and the preset regularization constant is used as the denominator, and the numerator is divided by the denominator to obtain the coherent conflict burst degree on the vertical axis.

[0031] The effectiveness of the control demand on the horizontal axis and the coherence conflict outbreak degree on the vertical axis are combined into two-dimensional coordinate points to generate coherence and scheduling conflict state space points.

[0032] A further improvement of the present invention is that, when the effectiveness of the horizontal axis control demand is greater than a preset efficiency gain determination threshold and the coherent collision burst degree of the vertical axis is less than a preset coherent collision tolerance threshold, the spectrum accumulation time is increased; when the effectiveness of the horizontal axis control demand is less than or equal to zero and the coherent collision burst degree of the vertical axis is greater than or equal to the preset coherent collision tolerance threshold, the pump drive current is reduced and the spectrum accumulation time is locked, including:

[0033] When the effectiveness of the horizontal axis control requirement is greater than the preset efficiency gain determination threshold and the coherent conflict burst degree of the vertical axis is less than the preset coherent conflict tolerance threshold, the radar is determined to be in a benign cooperation zone, and an instruction is generated that allows the spectrum accumulation time to be increased.

[0034] When the effectiveness of the horizontal axis control requirement is less than or equal to the preset efficiency gain determination threshold and the coherent conflict burst degree of the vertical axis is less than the preset coherent conflict tolerance threshold, the radar is determined to be in the target missing area, and an instruction to truncate the spectrum accumulation time is generated.

[0035] When the validity of the horizontal axis control requirement is less than or equal to zero and the coherent collision burst degree of the vertical axis is greater than or equal to the preset coherent collision tolerance threshold, the radar is determined to be in the coherent collision zone, and an instruction is generated to force the pump drive current to fall back to below the preset limit pump current threshold, while locking the spectrum accumulation time.

[0036] Secondly, a computer-readable storage medium is proposed, on which a computer program is stored. When the computer program is executed by a processor, it implements the above-mentioned wind field measurement method based on a single-stage amplified laser lidar.

[0037] Thirdly, an electronic device is proposed, including a memory for storing instructions and a processor for executing the instructions, causing the device to perform the above-described wind field measurement method based on a single-stage amplified laser lidar.

[0038] The technical effects of this invention are as follows:

[0039] A wind field measurement method based on a single-stage amplified laser lidar was constructed. By building a joint characterization system of the coherence state of the transmitter and the echo quality state, the efficiency time gain gradient and the noise floor power evolution gradient are mapped to a two-dimensional conflict state space. This enables the coordinated scheduling of pump drive current and spectrum accumulation time, breaking the single-variable control limitation caused by the independent operation of the transmitter and inversion end in existing equipment. This scheme can not only adaptively maximize the data efficiency of wind speed measurement while ensuring the safe operation of laser amplification devices and avoiding coherence degradation, but also effectively terminate the invalid integration waiting process of low-quality signals, significantly reduce measurement delay and computing power waste, and adapt to the parameter differences of different individuals without the need to establish a complex underlying mechanism model. Attached Figure Description

[0040] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0041] Figure 1 This is a flowchart illustrating the wind field measurement method of wind-measuring lidar based on a single-stage amplified laser according to Embodiment 1 of the present invention. Detailed Implementation

[0042] Example 1: This example is applied to a 1550nm coherent Doppler wind lidar with a single-stage, two-way erbium-doped fiber amplification structure. The hardware includes a narrow-linewidth seed laser, an acousto-optic frequency-shift modulation unit, a single-stage fiber amplifier, a pump driver board, a high-speed analog-to-digital converter acquisition unit, an FPGA real-time signal processing module, an embedded scheduling control unit, an optical transceiver antenna, and a heat sink temperature sampling circuit. The entire process is executed cyclically by the embedded scheduling control unit at a fixed scheduling cycle. The preferred scheduling cycle range is 10ms to 100ms; this example selects 20ms as the baseline scheduling cycle. (The technical field to which this example pertains is...) Personnel can adjust the scheduling cycle value based on the radar pulse repetition frequency and the number of detection range gates. All preset thresholds, regularization constants, statistical window lengths, interpolation calibration parameters, and normalization reference parameters are pre-stored in the embedded unit's non-volatile storage area and automatically loaded upon power-on, requiring no manual intervention throughout the process. All mathematical operations, feature vector generation, and state interval determination logic are solidified in the FPGA parallel computing core, ensuring real-time output of control commands during wind field measurement. This avoids conflicts between coherent nonlinear effects at the amplification end and the cumulative signal-to-noise ratio gain, balancing the effective data ratio of long-distance wind field detection with the coherent stability of a single-stage amplified light source. Specifically, such as... Figure 1 As shown, the wind field measurement method based on a single-stage amplified laser proposed in this embodiment includes the following specific steps:

[0043] The process of acquiring the pump drive current, Doppler effective peak power, broadband phase noise floor, and currently issued spectrum accumulation time includes: reading the voltage drop signal of the internal current sampling resistor of the laser amplifier through the driver board, and converting the voltage drop signal into the pump drive current; reading the resistance value of the thermistor attached to the heat sink surface of the amplifier gain fiber through the driver board, and converting the thermistor resistance value into the heat sink temperature; receiving the heterodyne beat frequency signal sequence output by the analog-to-digital converter, and performing a fast Fourier transform on the heterodyne beat frequency signal sequence of fixed length to obtain the frequency domain spectrum; extracting the peak energy in the Doppler frequency shift interval of the frequency domain spectrum, and using the peak energy as the Doppler effective peak power; and integrating the energy value in the high-frequency no-signal interval of the frequency domain spectrum to obtain the broadband phase noise floor.

[0044] In this embodiment, during the synchronous acquisition of radar underlying sensor parameters, the raw physical signals and configuration parameters output from the hardware sensing circuit, analog-to-digital conversion unit, and upper-level scheduling and distribution interface are uniformly received. This process completes the standardized numerical conversion of five core input quantities: pump drive current, Doppler effective peak power, broadband phase noise floor, spectrum accumulation time, and amplifier heat sink temperature. The output is a standardized sensor value sequence with dimensional calibration. This standardized value sequence is directly input to the coherent state quantization solution stage at the transmitter as the basic calculation input. This stage simultaneously completes four parallel acquisition processes: current sampling, temperature sampling, heterodyne beat frequency signal acquisition, and upper-level configuration reading. The four acquisition processes share the same hardware synchronization clock, with the clock reference frequency preferably set to 100MHz to ensure the alignment of timestamps of multi-source sensor data and eliminate calculation deviations caused by timing misalignment. For the acquisition and conversion process of the pump drive current, a high-precision metal film current sampling resistor is connected in series inside the pump drive board. The preferred resistance value is between 0.05Ω and 0.2Ω. In this embodiment, a 0.1Ω sampling resistor is used. The amplifier pump circuit current flows through the sampling resistor to generate a voltage drop signal. The board has a built-in 16-bit high-precision differential analog-to-digital sampling channel to continuously read the voltage drop analog signal. The embedded unit has a built-in current-voltage drop linear conversion calibration curve. The conversion calculation formula is written as: In the formula This represents the standardized pump drive current; The sampling voltage is the voltage drop across the sampling resistor. The nominal resistance value of the sampling resistor is used; after conversion, the real-time pump drive current value is obtained and cached in the memory register. For the amplifier heat sink temperature acquisition and conversion process, a single-stage amplification gain fiber is tightly attached to the surface of an aluminum heat sink, and an NTC negative temperature coefficient thermistor is attached to the surface of the heat sink. The thermistor and the precision voltage divider circuit built into the driver board form a temperature sampling loop. The driver board reads the thermistor's voltage divider voltage in a time-division multiplexing manner. The embedded unit pre-stores a thermistor resistance-temperature segmented calibration data table, and uses a lookup table and segmented linear interpolation to convert the resistance value to the heat sink temperature. The conversion of heat sink temperature units to degrees Celsius is now standardized. Each scheduling cycle synchronously records the current heat sink temperature sampling value and stores it in a time-series temperature buffer queue. The queue depth matches the preset statistical time window length for subsequent differential calculation of the thermal lens evolution rate. For the heterodyne beat frequency signal acquisition and spectrum analysis process, the optical antenna receives the backscattered echo from atmospheric aerosols and causes heterodyne interference with the local oscillator seed light. A balanced photodetector converts the interference light signal into a time-domain electrical signal. A high-speed analog-to-digital converter performs discrete sampling at a fixed sampling frequency, outputting a heterodyne beat frequency signal sequence. The length of a single signal sequence is fixed at 2048 sampling points, and the preferred sampling frequency is 250MHz. Each fixed-length beat frequency sequence is directly fed into the FPGA's built-in radix-4 parallel fast Fourier transform operation kernel to complete the time-domain to frequency-domain conversion, obtaining a two-dimensional frequency-domain power spectral density matrix. The horizontal axis of the spectrum matrix represents the discrete Doppler frequency points, and the vertical axis represents the power spectral density at the corresponding frequency points. The standard unit for power spectral density is After the frequency domain spectrum is generated, energy extraction is performed in two intervals. The first interval corresponds to the effective Doppler frequency shift. The interval range is calibrated according to the radar's preset maximum detectable wind speed. The preferred Doppler frequency shift interval for a 1550nm band wind measuring radar is set to... Within this interval, traverse all discrete frequency points to retrieve the maximum power spectral density. Combined with the discrete frequency point spacing of the spectrum The effective peak power of Doppler was calculated. The conversion formula is: In the formula The unit is watt (Wt) ), Units are , The unit is Hertz (Hz) The second interval is selected as the high-frequency, signal-free region of the spectrum. This signal-free region is defined as the frequency band ranging from 2MHz to 5MHz outside the upper limit of the effective Doppler frequency shift interval. This interval contains no atmospheric echo Doppler signals and only includes laser phase noise and circuit shot noise. The power spectral density corresponding to all discrete frequency points within this interval is accumulated and integrated point by point. The integration formula is written as: In the formula This is the broadband phase noise floor, measured in watts (W). ); Represents discrete frequency points The corresponding power spectral density, in units ; The discrete frequency point interval of the spectrum, in units ; , These are the start and end frequencies of the high-frequency no-signal interval, in units of... Multiplying the power spectral density at a single frequency point by the frequency interval yields the total noise power at that point. Summing this across the entire frequency range gives the total broadband phase noise floor, which is dimensionlessly consistent. After integration, the real-time broadband phase noise floor value is obtained and synchronously cached with the effective peak power of the Doppler signal. For the acquisition of the spectrum accumulation time parameter, the embedded scheduling unit receives the spectrum accumulation time configuration parameters issued by the upper-level wind field measurement and control platform in real time via an Ethernet communication interface. The parameter units are uniformly converted to microseconds. The parameters sent out will overwrite the current spectrum accumulation time variable stored in memory in real time. In each scheduling cycle, the value of the variable is refreshed synchronously. After the standardization and transformation of all five types of parameters are completed, a synchronous sensing parameter array is generated and passed to the next processing stage.

[0045] The coherent current margin is obtained by subtracting the pump drive current from the preset limiting pump current threshold; this includes: obtaining the preset limiting pump current threshold by interpolation based on the heat sink temperature using multiple sets of temperature and current correspondence data obtained from calibration; obtaining the coherent current margin by subtracting the preset limiting pump current threshold from the pump drive current; obtaining the thermal lens evolution rate by taking the first derivative of the heat sink temperature in the time dimension and using the difference between the temperature difference of two adjacent sampling times divided by the sampling interval; and combining the coherent current margin and the thermal lens evolution rate into a two-dimensional column vector to generate the transmitter coherent execution state vector.

[0046] In this embodiment, in the coherent state quantization solution stage at the transmitter, this stage receives the standardized pump drive current and heat sink temperature values ​​output from the preceding stage. Based on the pre-calibrated temperature-limiting current mapping relationship, it solves the limiting pump current threshold, sequentially calculating the coherent current margin and thermal lensing evolution rate, ultimately generating the transmitter coherent execution state vector. This two-dimensional vector serves as the denominator parameter for calculating the conflict burst intensity and is continuously passed downstream. All calculations in this stage are based on the nonlinear effect of thermal lensing in a single-stage fiber amplifier and the coherent failure mechanism of stimulated Brillouin scattering. The core logic lies in quantifying the remaining safe range between the current pumping condition and the critical coherent failure condition, while simultaneously quantifying the thermal lensing distortion evolution rate caused by changes in heat sink temperature. The complete calculation process consists of four steps: limiting pump current threshold interpolation solution, coherent current margin calculation, thermal lensing evolution rate differential solution, and coherent execution state vector concatenation. The limiting pump current threshold interpolation solution relies on multiple temperature-limiting current calibration datasets completed before the entire system leaves the factory. The calibration process is completed in a constant temperature test chamber, covering the entire operating temperature range of the radar. to Every A set of limiting pump current data under thermal equilibrium conditions was collected. The limiting pump current is defined as the pump drive current at which the single-stage amplifier just exhibits stimulated Brillouin scattering coherent distortion. The unit is uniformly ampere (A). All calibration data is stored in an embedded flash memory as a two-dimensional key-value pair data table. During real-time computation, the heat sink temperature collected in the current scheduling cycle is read. Search the data table and Two adjacent calibration temperature points , and its corresponding limiting current , A linear interpolation algorithm is used to solve for the limiting pump current threshold corresponding to the current temperature. The interpolation calculation formula is: In the formula , , All units are ; , , All units are After interpolation, the limiting pump current threshold, which dynamically varies with the heat sink temperature, is obtained. Unlike the fixed threshold scheme, this can adapt to the amplifier's nonlinear critical operating condition shift caused by temperature drift. The coherent current margin is directly calculated by subtracting the real-time pump drive current from the limiting pump current threshold. The calculation formula is written as: In the formula Represents the coherent current margin, measured in amperes ( ). The physical meaning of the value is the difference between the current pump current and the critical current for coherence failure. A value greater than zero indicates that the amplifier is within the coherence safety range, while a value less than zero indicates that the current pumping condition has exceeded the coherence threshold, posing a risk of a surge in phase noise and loss of echo coherence. The thermal lens evolution rate is approximated using a first-order difference in the time dimension. The embedded unit establishes a time-series temperature buffer queue to continuously store the heat sink temperatures collected in the two most recent adjacent scheduling cycles. , The time difference between adjacent sampling times is equal to the fixed scheduling period. The unit is uniformly set to milliseconds (milliseconds). The difference solution formula is: In the formula The thermal lens evolution rate is expressed in units of 1000 m / s. A larger absolute value indicates a faster change in heat sink temperature, a higher fiber gain lens distortion rate, and a correspondingly faster coherent stability attenuation rate. After calculating the coherent current margin and thermal lens evolution rate, the two scalar values ​​are concatenated column-wise to generate a two-dimensional column vector as the transmitter's coherent execution state vector. The vector expression is written as: This vector fully represents the current coherent safety margin and thermal distortion evolution rate of the single-stage amplification transmitter. The two components inside the vector are synchronously written into shared memory for real-time reading and calling in subsequent coherent conflict burst degree calculations.

[0047] The instantaneous signal-to-noise ratio (SNR) is obtained by dividing the effective peak power of the Doppler signal by the broadband phase noise floor. The data efficiency is obtained by statistically analyzing the percentage of time the instantaneous SNR is greater than a preset effective SNR threshold. This process includes: dividing the effective peak power of the Doppler signal by the broadband phase noise floor to obtain the instantaneous SNR; within a preset statistical time window, using a step function to statistically analyze the percentage of time the instantaneous SNR is greater than the preset effective SNR threshold to obtain the data efficiency, wherein the step function outputs one when the input value is greater than zero, and outputs zero when the input value is less than or equal to zero; and combining the data efficiency with the broadband phase noise floor at the current moment to generate an echo quality state vector.

[0048] In this embodiment, during the echo signal quality quantification and evaluation stage, this stage receives three types of parameters output from the radar's bottom-level sensor parameter synchronous acquisition stage: Doppler effective peak power, broadband phase noise floor, and current spectrum accumulation time. The instantaneous signal-to-noise ratio (SNR) is obtained through power division. Based on a step function, effective signal duration statistics are completed within a preset statistical time window to calculate the data effectiveness rate. Finally, the data is concatenated to generate an echo quality state vector for downstream transmission. This stage is used to quantify the effective detection ratio of the current atmospheric echo signal and characterize the potential gain of spectrum accumulation time on the effectiveness of measurement data. All statistical operations are synchronously updated via sliding windows. The window length is preferably between 50 and 200 scheduling cycles. This embodiment selects 100 scheduling cycles as the statistical time window. The instantaneous SNR value calculated in each scheduling cycle is cached within the window, and the earliest cycle data is cyclically overwritten to achieve sliding updates. The instantaneous SNR is directly calculated by dividing the Doppler effective peak power by the broadband phase noise floor. The calculation formula is: In the formula The instantaneous signal-to-noise ratio is dimensionless. , All units are A higher instantaneous signal-to-noise ratio (SNR) indicates a stronger echo signal relative to the noise floor in a single spectrum acquisition, resulting in higher reliability for wind field Doppler peak identification. The preset effective SNR threshold... Pre-calibrated and stored values ​​are preferably in the range of 3 to 8, dimensionless. In this embodiment, 5 is selected as the baseline threshold to distinguish between valid echo samples and invalid noise samples. A unit step function is introduced to quantify the effectiveness of single-cycle sampling. The step function expression is written as: The difference between the instantaneous signal-to-noise ratio and the effective signal-to-noise ratio threshold is used as the input variable of the step function, i.e. When the instantaneous signal-to-noise ratio (SNR) is higher than the threshold, the step function outputs a value of 1, marking the scheduling period as a valid sampling period; otherwise, it outputs a value of 0, marking it as an invalid sampling period. Within the time interval corresponding to the complete preset statistical time window, the step function output values ​​corresponding to all scheduling periods within the window are accumulated to obtain the number of periods corresponding to the total valid sampling duration within the window. This number is then divided by the total number of periods in the statistical window to obtain the data effectiveness rate. The formula is as follows: In the formula The representative data efficiency ranges from 0 to 1 and is dimensionless. The total number of scheduling cycles included in the statistical time window, a dimensionless integer; For the first in the window The instantaneous signal-to-noise ratio of each scheduling cycle is summed to obtain the effective cycle count. Dividing this count by the total number of cycles yields the dimensionless proportion. After solving for the data efficiency, the broadband phase noise floor and data efficiency of the current scheduling cycle are combined into a two-dimensional row vector to generate the echo quality state vector. The vector expression is written as: This vector fully represents the effective proportion of the echo signal and the noise floor level, providing time-series sample data for downstream gradient calculation. The vector value is refreshed synchronously in each scheduling cycle to ensure that the gradient calculation uses the latest time-series state data.

[0049] The efficiency time gain gradient is obtained by calculating the partial derivative of the data efficiency with respect to the spectrum accumulation time, and is used as the horizontal axis to control demand effectiveness; the noise floor is calculated by calculating the partial derivative of the broadband phase noise floor with respect to the pump drive current to obtain the noise floor power evolution gradient, and the noise floor power evolution gradient is divided by the sum of the coherent current margin and a preset regularization constant, and is used as the vertical axis to determine the coherent conflict outbreak degree; including: obtaining the difference between the data efficiency and the difference between the spectrum accumulation time in two adjacent scheduling periods, dividing the difference between the data efficiency and the difference between the spectrum accumulation time to calculate the efficiency time gain gradient; obtaining the phase noise floor... The difference between the broadband phase noise floor and the difference between the pump drive current in two adjacent scheduling cycles is used to calculate the noise floor power evolution gradient by dividing the difference between the broadband phase noise floor and the difference between the pump drive current. The efficiency time gain gradient is directly used as the horizontal axis control demand effectiveness. The noise floor power evolution gradient is used as the numerator, and the sum of the coherent current margin and the preset regularization constant is used as the denominator. The numerator is divided by the denominator to obtain the vertical axis coherent conflict burst degree. The horizontal axis control demand effectiveness and the vertical axis coherent conflict burst degree are combined into a two-dimensional coordinate point to generate coherent and scheduling conflict state space points.

[0050] In this embodiment, during the conflict state space feature construction stage, the time-series data of the transmitter's coherent execution state vector and echo quality state vector buffers are read simultaneously. Time-series samples from two adjacent scheduling cycles are selected to solve for the efficiency time gain gradient and the noise floor power evolution gradient, respectively. Normalization is performed on the two types of gradients to eliminate differences in physical dimensions, resulting in the dimensionless horizontal axis representing the effectiveness of control requirements and the dimensionless vertical axis representing the degree of coherent conflict. Finally, these are combined to generate a two-dimensional coherent and scheduling conflict state space point. This space point serves as the sole input for the multi-interval collaborative scheduling command output stage. The core of this stage is to construct a two-dimensional dimensionless feature coordinate system that considers both the echo quality gain potential and the degree of coherent nonlinear conflict at the transmitter, achieving decoupling and quantization of the two control objectives. All partial derivatives are solved using adjacent two-cycle difference approximations to avoid real-time losses caused by complex differential operations. Adjacent scheduling cycles are marked as the [number missing]. Period and the The system stores two sets of time-series sample data in two separate periods. The efficiency time gain gradient is used to characterize the improvement in data efficiency caused by extending the spectral accumulation time. First, the difference in data efficiency between two adjacent periods is read. Spectrum cumulative time difference The first-order partial derivative of the data efficiency with respect to the cumulative time of the spectrum is approximated using the difference ratio. The solution formula is written as follows: In the formula The original efficient time gain gradient has the dimension of . ; Dimensionless Units converted to seconds ( The calculation is then performed; to eliminate the numerical differences caused by the time scale, a system-preset baseline cumulative duration is used. (unit Preferred value Normalizing the original gradient yields the dimensionless control requirement effectiveness. Normalization formula: Normalized x-axis Completely dimensionless, a value greater than zero indicates that increasing the spectral accumulation time can improve data efficiency and has gain-adjusting value; a value less than or equal to zero indicates that extending the accumulation time cannot improve the effective echo ratio, and further increasing the accumulation time will only consume hardware processing resources and reduce the wind field measurement update rate. The noise floor power evolution gradient is used to characterize the impact of pump drive current changes on the broadband phase noise floor, and the difference in broadband phase noise floor between two adjacent cycles is read. Pump drive current difference The first-order partial derivative of the broadband phase noise floor with respect to the pump drive current, obtained by differential approximation, is given by the following formula: In the formula The original noise floor power evolution gradient, with dimensions of ;molecular unit denominator unit A larger gradient value indicates that even a small increase in pump current can lead to a rapid rise in phase noise floor, resulting in a rapid deterioration in the coherence stability of the single-stage amplifier. When solving for the dimensionless ordinate coherence conflict burst degree, the gradient of the noise floor power evolution is used as the numerator, and the denominator is the sum of the coherence current margin and a preset regularization constant. The unit is ampere (Ampere) The preferred value range is 0.01 to 0.1A. In this embodiment, 0.05A is selected as the reference value. The regularization constant is used to avoid the overflow of the burst value due to the extremely small denominator when the coherent current margin approaches zero. At the same time, the rated noise power reference of the system is introduced. (unit (Factory-calibrated fixed values) are normalized overall. The complete calculation formula is written as follows: In the formula Represents the degree of coherent conflict on the dimensionless vertical axis; The rated pump current of the amplifier, in units Used to offset Medium current dimension; Based on rated noise power, unit The power dimension cancels out the original composite dimension. through After normalization, physical units are completely eliminated, ultimately These are purely dimensionless values. Higher values ​​indicate that an increase in pump current under current operating conditions can easily trigger a surge in phase noise floor, leading to a higher risk of coherent failure. Lower values ​​indicate sufficient coherent safety margin at the transmitter, meaning small fluctuations in pump current will not worsen the echo signal-to-noise ratio. After solving the horizontal and vertical coordinates, the dimensionless control requirement effectiveness and the dimensionless coherent conflict outbreak degree are combined into two-dimensional coordinate points to generate coherent and scheduling conflict state space points. The coordinate point expression is written as: The two components of this two-dimensional coordinate point are both dimensionless values. It fully integrates two core features: the gain of echo accumulation time adjustment and the risk of coherent nonlinear conflict at the transmitting end. It eliminates the judgment benchmark deviation caused by different physical dimensions. The two components of the coordinate point are synchronously transmitted to the multi-interval collaborative scheduling command output link as the only input feature quantity for interval judgment. A new set of coordinate points is refreshed in each round of scheduling cycle to realize real-time update of the state space.

[0051] When the effectiveness of the horizontal axis control requirement is greater than the preset efficiency gain determination threshold and the coherent conflict burst degree of the vertical axis is less than the preset coherent conflict tolerance threshold, the spectrum accumulation time is increased; when the effectiveness of the horizontal axis control requirement is less than or equal to zero and the coherent conflict burst degree of the vertical axis is greater than or equal to the preset coherent conflict tolerance threshold, the pump drive current is reduced and the spectrum accumulation time is locked. This includes: when the effectiveness of the horizontal axis control demand is greater than the preset efficiency gain determination threshold and the coherent collision burst degree of the vertical axis is less than the preset coherent collision tolerance threshold, determining that the radar is in a benign cooperation zone and generating an instruction to allow the spectrum accumulation time to be increased; when the effectiveness of the horizontal axis control demand is less than or equal to the preset efficiency gain determination threshold and the coherent collision burst degree of the vertical axis is less than the preset coherent collision tolerance threshold, determining that the radar is in a target missing zone and generating an instruction to truncate the spectrum accumulation time; when the effectiveness of the horizontal axis control demand is less than or equal to zero and the coherent collision burst degree of the vertical axis is greater than or equal to the preset coherent collision tolerance threshold, determining that the radar is in a coherent collision zone, generating an instruction to force the pump drive current back to below the preset limit pump current threshold, and simultaneously locking the spectrum accumulation time.

[0052] In this embodiment, in the multi-interval collaborative scheduling instruction output stage, this stage divides three independent operating condition intervals based on the relationship between the coordinate points' horizontal and vertical coordinate values ​​and two types of thresholds. For each type of interval, corresponding spectrum accumulation time and pump drive current adjustment instructions are generated. These instructions are directly sent to the FPGA acquisition and control core and the pump drive board to perform hardware parameter updates. The efficiency gain determination threshold set in this stage... The preferred value range is to In this embodiment, we select Coherent conflict tolerance threshold The preferred value range is 0.1 to 1. In this embodiment, 0.5 is selected. The three types of operating conditions are divided into independent logical intervals with no overlap, covering all operating conditions of the radar. The complete judgment and instruction generation process executes the three types of interval logic judgments in sequence, prioritizing the judgment of the benign cooperation zone, followed by the judgment of the coherent conflict zone, and the remaining operating conditions are classified into the target missing zone. The first type of operating condition is the benign cooperation zone. The judgment condition is that the effectiveness of the control demand on the horizontal axis is greater than the efficiency gain judgment threshold, and the coherent conflict outbreak degree on the vertical axis is less than the coherent conflict tolerance threshold. In this range, extending the spectrum accumulation time can stably improve the data efficiency, and the coherent safety margin of the transmitter is sufficient. The pump current will not trigger a rapid rise in phase noise. The embedded unit generates a control command that allows the spectrum accumulation time to be increased. The command carries a fixed accumulation time step increment. The preferred value of the step increment is 50μs to 200μs. In this embodiment, 100μs is selected. The duration superposition update is performed once in each round of scheduling. The hardware upper limit protection threshold of the spectrum accumulation time is set synchronously. The upper limit is preferably 5000μs. When the accumulation time reaches the upper limit, the incremental update stops to avoid the single spectrum acquisition time being too long and reducing the wind field scanning update frequency. The second type of operating condition is the coherent conflict zone. The judgment condition is that the effectiveness of the control demand on the horizontal axis is less than or equal to zero, while the coherent conflict outbreak degree on the vertical axis is greater than or equal to the coherent conflict tolerance threshold. In this range, extending the spectrum accumulation time cannot improve the effective echo ratio, and the current pump current is very close to the coherent failure threshold. Even a small increase in current can cause a sharp rise in the phase noise floor and a significant decrease in the instantaneous signal-to-noise ratio of the echo. The embedded unit synchronously outputs two types of hardware control instructions. The first type of instruction is the pump drive current back-off instruction, which forcibly reduces the pump drive current to below the limit pump current threshold obtained by interpolating the current heat sink temperature. The preferred value range for the current back-off step is 0.02A to 0.1A. In this embodiment, 0.05A is selected. The current is continuously reduced every cycle until the pump current is lower than the limit threshold. The second type of instruction is the spectrum accumulation time lock instruction, which writes to the register to lock the current spectrum accumulation time value. No matter how the echo quality changes later, the duration will not be increased or decreased until the coordinate point leaves the coherent conflict zone and the lock is automatically released.The third type of operating condition is the target missing area, which covers all intermediate operating conditions that do not belong to the benign cooperation area or the coherent conflict area. The judgment condition is that the effectiveness of the control demand on the horizontal axis is less than or equal to the effective gain judgment threshold, and the coherent conflict outbreak degree on the vertical axis is less than the coherent conflict tolerance threshold. In this range, the coherent safety margin of the transmitter is sufficient, but extending the spectrum accumulation time cannot bring about an increase in the effective data ratio. Continuously increasing the accumulation time will only waste hardware computing power and measurement time. The embedded unit generates a control instruction to truncate the spectrum accumulation time. The instruction carries a fixed duration reduction step. The reduction step is preferably between 50μs and 200μs. In this embodiment, 100μs is selected. The hardware lower limit protection threshold for the spectrum accumulation time is set simultaneously. The lower limit is preferably 200μs. The reduction operation stops after the accumulation time drops to the lower limit to ensure the signal-to-noise ratio of the basic Doppler spectrum acquisition.

[0053] Example 2: This example provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-described wind field measurement method based on a single-stage amplified laser lidar by calling the computer program stored in the memory.

[0054] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the wind field measurement method based on a single-stage amplified laser provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.

[0055] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0056] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0057] This invention is described with reference to flowchart illustrations and block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and block diagrams, as well as combinations of blocks in the flowchart illustrations and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and boxes Figure 1 The steps of the function specified in one or more boxes.

[0059] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A wind field measurement method based on a single-stage amplified laser lidar, characterized in that: The specific steps include the following: Obtain the pump drive current, Doppler effective peak power, broadband phase noise floor, and the current spectrum accumulation time; The coherent current margin is obtained by subtracting the preset limit pump current threshold from the pump drive current. The instantaneous signal-to-noise ratio is obtained by dividing the effective peak power of the Doppler by the broadband phase noise floor. The data efficiency is obtained by calculating the percentage of time during which the instantaneous signal-to-noise ratio is greater than the preset effective signal-to-noise ratio threshold. The efficiency time gain gradient is obtained by calculating the partial derivative of the data efficiency with respect to the spectrum accumulation time, and is used as the horizontal axis to determine the control demand effectiveness; the noise floor power evolution gradient is obtained by calculating the partial derivative of the broadband phase noise floor with respect to the pump drive current, and is used as the vertical axis to determine the coherent conflict burst degree by dividing the noise floor power evolution gradient by the sum of the coherent current margin and the preset regularization constant; including: Obtain the difference in data efficiency and the difference in spectrum accumulation time within two adjacent scheduling periods, divide the difference in data efficiency by the difference in spectrum accumulation time, and calculate the efficiency time gain gradient. The difference between the broadband phase noise floor and the difference between the pump drive current in two adjacent scheduling periods are obtained. The difference between the broadband phase noise floor and the difference between the pump drive current are divided to calculate the noise floor power evolution gradient. The effective time gain gradient is directly used as the horizontal axis to control the effectiveness of the demand. The noise floor power evolution gradient is used as the numerator, the sum of the coherent current margin and the preset regularization constant is used as the denominator, and the numerator is divided by the denominator to obtain the coherent conflict burst degree on the vertical axis. The effectiveness of the control demand on the horizontal axis and the coherence conflict outbreak degree on the vertical axis are combined into two-dimensional coordinate points to generate coherence and scheduling conflict state space points. When the effectiveness of the horizontal axis control demand is greater than a preset efficiency gain determination threshold and the coherent collision burst degree of the vertical axis is less than a preset coherent collision tolerance threshold, the spectrum accumulation time is increased; when the effectiveness of the horizontal axis control demand is less than or equal to zero and the coherent collision burst degree of the vertical axis is greater than or equal to the preset coherent collision tolerance threshold, the pump drive current is decreased and the spectrum accumulation time is locked; including: When the effectiveness of the horizontal axis control requirement is greater than the preset efficiency gain determination threshold and the coherent conflict burst degree of the vertical axis is less than the preset coherent conflict tolerance threshold, the radar is determined to be in a benign cooperation zone, and an instruction is generated that allows the spectrum accumulation time to be increased. When the effectiveness of the horizontal axis control requirement is less than or equal to the preset efficiency gain determination threshold and the coherent conflict burst degree of the vertical axis is less than the preset coherent conflict tolerance threshold, the radar is determined to be in the target missing area, and an instruction to truncate the spectrum accumulation time is generated. When the validity of the horizontal axis control requirement is less than or equal to zero and the coherent collision burst degree of the vertical axis is greater than or equal to the preset coherent collision tolerance threshold, the radar is determined to be in the coherent collision zone, and an instruction is generated to force the pump drive current to fall back to below the preset limit pump current threshold, while locking the spectrum accumulation time.

2. The wind field measurement method based on a single-stage amplified laser lidar according to claim 1, characterized in that: The process of acquiring the pump drive current, the effective peak power of Doppler, and the broadband phase noise floor includes: The voltage drop signal of the internal current sampling resistor of the laser amplifier is read by the driver board, and the voltage drop signal is converted into the pump drive current. The thermistor resistance value attached to the heat sink surface of the amplifier gain fiber is read by the driver board, and the heat sink temperature is obtained by converting the thermistor resistance value. The heterodyne beat frequency signal sequence output by the analog-to-digital converter is received, and a fast Fourier transform is performed on the heterodyne beat frequency signal sequence of fixed length to obtain the frequency domain spectrum; The peak energy is extracted within the Doppler frequency shift interval of the frequency domain spectrum, and the peak energy is taken as the effective peak power of the Doppler. The energy value is integrated within the high-frequency signal-free region of the frequency domain spectrum to obtain the broadband phase noise floor.

3. The wind field measurement method based on a single-stage amplified laser lidar according to claim 2, characterized in that: The step of subtracting the preset limiting pump current threshold from the pump drive current to obtain the coherent current margin includes: Based on multiple sets of temperature and current data obtained from calibration, the preset limit pump current threshold is obtained by interpolation according to the heat sink temperature. The coherent current margin is obtained by subtracting the preset limit pump current threshold from the pump drive current. The thermal lens evolution rate is obtained by taking the first derivative of the heat sink temperature in the time dimension and using the difference between the temperature difference of two adjacent sampling times divided by the sampling interval. The coherent current margin and the thermal lens evolution rate are combined into a two-dimensional column vector to generate the transmitter coherent execution state vector.

4. The wind field measurement method based on a single-stage amplified laser lidar according to claim 3, characterized in that: The step of dividing the effective peak power of the Doppler by the broadband phase noise floor to obtain the instantaneous signal-to-noise ratio (SNR), and calculating the percentage of time the instantaneous SNR is greater than a preset effective SNR threshold to obtain the data effectiveness rate includes: The instantaneous signal-to-noise ratio is obtained by dividing the effective peak power of the Doppler by the broadband phase noise floor. Within the time interval corresponding to the preset statistical time window, the percentage of time when the instantaneous signal-to-noise ratio is greater than the preset effective signal-to-noise ratio threshold is statistically analyzed using a step function to obtain the data effectiveness rate. When the input value of the step function is greater than zero, the step function outputs one; when the input value of the step function is less than or equal to zero, the step function outputs zero. The data is efficiently combined with the current broadband phase noise floor to generate an echo quality state vector.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the wind field measurement method based on a single-stage amplified laser as described in any one of claims 1-4.

6. An electronic device, characterized in that, It includes a memory for storing instructions; and a processor for executing the instructions, causing the device to perform the wind field measurement method based on a single-stage amplified laser as described in any one of claims 1 to 4.

Images