A temperature compensation method for constant current power supply current measurement

Abstract

The present application relates to the technical field of constant current power supply current measurement, and discloses a temperature compensation method for constant current power supply current measurement, comprising: synchronously acquiring temperature measurement parameters and electric energy sampling quantity in a current sampling window, extracting average temperature and longitudinal temperature difference; analyzing the electric energy sampling quantity to obtain self-heating power and electric input power, and combining the average light power to calculate waste heat power; then, identifying a thermal response core based on the delay performance of the longitudinal temperature difference and the waste heat power, constructing return heat flow; further fusing the return heat flow, the self-heating power and the average temperature to reconstruct compensation resistance and thermoelectric potential, and finally correcting the electric energy sampling quantity and outputting a compensation current. The present application can overcome the interference caused by the superposition of external return heat and internal self-heating without changing the original control loop, and improve the accuracy of system current measurement.

Description

Technical Field

[0001] This invention relates to the field of constant current power supply current measurement technology, and more specifically, to a temperature compensation method for constant current power supply current measurement. Background Technology

[0002] In laser energy closed-loop control systems, to maintain stable light output, photodetectors or energy meters feed back the captured laser pulse energy deviation to the drive system, causing the constant current power supply's output current to be dynamically adjusted in small amplitudes and continuously. In practical engineering applications, the current of the constant current power supply is usually detected by a sampling resistor connected in series in the loop. However, the thermal environment of high-power laser equipment during operation is extremely complex. On the one hand, the large current flowing through the sampling resistor generates significant Joule self-heating; on the other hand, the laser chip generates a large amount of waste heat during electro-optic conversion. This waste heat is conducted along the heat sink, cooler, and large-section conductive busbar, ultimately returning to the sampling resistor in the form of a unidirectional heat flow with a significant delay effect. Due to the different thermal inertia of each component along the way, this external delayed return heat and the symmetrical heating of the sampling resistor itself are superimposed, resulting in a significant temperature gradient on the surface of the sampling resistor, which is no longer an ideal isothermal body. Currently, the common practice in the industry is to place a single temperature sensor near the sampling resistor and simply correct the temperature coefficient (TCR) of the resistor based on the acquired average temperature. This approach only considers the overall static temperature drift and completely fails to distinguish between current self-heating and external unidirectional backheating. In actual commissioning and long-term equipment testing, it was found that when the energy closed-loop operates, the hysteretic heating of the sampling resistor due to thermal inertia and the parasitic thermoelectric potential generated at the contact surface introduce significant bias errors into the current measurement channel. More problematic is that the control system misinterprets this thermal drift of the measurement channel itself as fluctuations in the actual output current of the laser, leading to incorrect compensation adjustments. Forcibly introducing complex temperature control actions to intervene in the main control layer to solve this problem easily disrupts the original control cycle, causing oscillations or even lockouts in the system closed loop. How to isolate these superimposed non-isothermal measurement errors without changing the existing logic of the main control loop is a key technical challenge in improving the closed-loop control accuracy of high-power lasers. Summary of the Invention

[0003] This invention provides a temperature compensation method for measuring the current of a constant current power supply, which solves the technical problems mentioned in the background art.

[0004] This invention provides a temperature compensation method for constant current power supply current measurement, applied to a metering assembly equipped with a constant temperature hot plate, a sampling resistor, and a laser, comprising: Establish a synchronous sampling sequence, acquire temperature measurement parameters and electrical energy sampling amount within the current sampling window, and extract the average temperature characterizing the spatial heat distribution of the sampling resistor and the longitudinal temperature difference characterizing the directional thermal gradient. The electrical energy sampling data is analyzed to extract the self-heating power characterizing the electrothermal conversion and the electrical input power reflecting the energy consumption state of the laser; The average optical power converted into light energy is stripped from the electrical input power to extract the waste heat power characterizing the feedback-induced heat source; Based on the delay characteristics of the longitudinal temperature difference and the waste heat power, a thermal response kernel is identified, and a reheated heat flow characterizing the delayed feedback of the heat source is constructed. The regenerated heat flow and the self-heating power are combined to extract the normalized heat flow that reflects the dominant state of the heat source, and combined with the average temperature to form a temperature feature set that reflects the dynamic heat distribution. Based on the temperature feature set and the normalized heat flux, the compensation resistance for eliminating non-isothermal errors and the thermoelectric potential for stripping parasitic voltages are reconstructed. The electrical energy sampling quantity is corrected by using the compensation resistance and the thermoelectric potential to eliminate the superposition interference of structural backheating and self-heating, and a compensation current is output.

[0005] The beneficial effects of this invention include: by synchronously acquiring temperature measurement parameters and electrical energy sampling on the time axis, and by uniformly incorporating the waste heat induced by the laser closed-loop feedback and the Joule heat of the sampling resistor itself into the calculation model, the non-isothermal measurement error caused by the superposition of asymmetric backheating and endogenous self-heating is eliminated. This invention can achieve accurate correction of current parameters without interfering with the existing constant current control main loop, effectively avoiding the closed-loop system misinterpreting temperature drift as fluctuations in actual laser energy, and improving the accuracy and stability of current measurement under complex heat source interference. Attached Figure Description

[0006] Figure 1 This is a flowchart of a temperature compensation method for constant current power supply current measurement according to the present invention. Detailed Implementation

[0007] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0008] like Figure 1As shown, a temperature compensation method for constant current power supply current measurement is applied to a metering assembly equipped with a constant temperature hot plate, a sampling resistor, and a laser, comprising: Establish a synchronous sampling sequence, acquire temperature measurement parameters and electrical energy sampling amount within the current sampling window, and extract the average temperature characterizing the spatial heat distribution of the sampling resistor and the longitudinal temperature difference characterizing the directional thermal gradient. The electrical energy sampling data is analyzed to extract the self-heating power characterizing the electrothermal conversion and the electrical input power reflecting the energy consumption state of the laser; The average optical power converted into light energy is stripped from the electrical input power to extract the waste heat power characterizing the feedback-induced heat source; Based on the delay characteristics of the longitudinal temperature difference and the waste heat power, a thermal response kernel is identified, and a reheated heat flow characterizing the delayed feedback of the heat source is constructed. The regenerated heat flow and the self-heating power are combined to extract the normalized heat flow that reflects the dominant state of the heat source, and combined with the average temperature to form a temperature feature set that reflects the dynamic heat distribution. Based on the temperature feature set and the normalized heat flux, the compensation resistance for eliminating non-isothermal errors and the thermoelectric potential for stripping parasitic voltages are reconstructed. The electrical energy sampling quantity is corrected by using the compensation resistance and the thermoelectric potential to eliminate the superposition interference of structural backheating and self-heating, and a compensation current is output.

[0009] This embodiment is applied to a metering assembly equipped with a constant-temperature hot plate, a sampling resistor, and a laser. The operating platform uses a 32-bit industrial-grade microcontroller or digital signal processor with a main frequency of no less than 100MHz, a random access memory capacity of no less than 64KB, and a flash memory capacity of no less than 256KB. The constant-temperature hot plate uses a temperature control method combining semiconductor cooling and heating, with a temperature control accuracy of no less than ±0.05 degrees Celsius, a temperature stability of no less than ±0.02 degrees Celsius per hour, and a temperature response time of no more than 10 seconds. The sampling resistor is a high-power metal film resistor with a four-lead structure, a nominal resistance range of 1 milliohm to 100 milliohms, and a rated power of no less than 10 watts. The laser is a pulsed semiconductor laser with a pulse width range of 1 nanosecond to 1 millisecond and a repetition frequency range of 1 Hz to 100 kHz. The specific implementation steps are as follows: S201 simultaneously acquires electrical energy samples including sampling voltage, terminal voltage, and pulse energy at the current sampling moment, as well as temperature parameters including the first terminal temperature, center temperature, second terminal temperature, and the temperature of the constant-temperature hot plate, locking the electrothermal state of the same transient state. The sampling voltage is the voltage drop across the sampling resistor, acquired using a 16-bit resolution analog-to-digital converter with a sampling accuracy of no less than ±0.1% of full scale and a sampling bandwidth of no less than 100 kHz. The terminal voltage is the supply voltage across the laser, acquired using a 16-bit resolution analog-to-digital converter with a sampling accuracy of no less than ±0.1% of full scale and a sampling bandwidth of no less than 100 kHz. The pulse energy is the light energy output of a single laser pulse, acquired using a thermopile laser energy meter with a spectral response range covering the laser output wavelength, an energy measurement accuracy of no less than ±0.5%, and a response time of no more than 1 microsecond. The first end temperature is the surface temperature of the sampling resistor closest to the laser, the center temperature is the surface temperature at the geometric center of the sampling resistor, the second end temperature is the surface temperature of the sampling resistor furthest from the laser, and the hot plate temperature is the surface temperature of the constant-temperature hot plate supporting the sampling resistor and the laser. All temperatures are collected using platinum resistance temperature sensors with a measurement accuracy of ±0.01 degrees Celsius and a response time of no more than 100 milliseconds. The trigger output signal from the laser is simultaneously acquired as the main trigger source, triggering the acquisition action of all sampling channels simultaneously. The timestamp error of each sampled data does not exceed 10% of the sampling period. The sampling period is set from 10 milliseconds to 100 milliseconds, adjusted according to the laser's pulse repetition frequency. When the laser repetition frequency is higher than 1 kHz, the sampling period is set to 10 milliseconds; when the laser repetition frequency is lower than 100 Hz, the sampling period is set to 100 milliseconds.

[0010] The sampling resistor uses a four-lead connection. The current lead is connected to the two current terminals of the sampling resistor, and the voltage lead is connected to the two voltage terminals. The voltage and current leads are independent to avoid the lead resistance affecting the sampling voltage measurement. The input range of the analog-to-digital converter is determined based on the nominal resistance and maximum operating current of the sampling resistor. The preamplifier gain is set so that the full-scale output of the sampling voltage corresponds to 80% to 90% of the full-scale input of the analog-to-digital converter. The pulse energy meter is installed directly in front of the laser output optical axis, 0.5 to 2 meters away from the laser output window. The center of the energy meter's sensitive surface coincides with the optical axis, with a tilt angle not exceeding 0.5 degrees, and the area of ​​the sensitive surface is greater than 1.2 times the area of ​​the laser spot.

[0011] S202 fuses the temperatures of the first end, center, and second end based on their spatial distribution weights, and reconstructs the average temperature reflecting the heating state of the sampling resistor with the center temperature as the dominant weight. The calculation formula is as follows: ; Where n is the index number of the current sampling time, which increments sequentially from the start of the system. The temperature at the first end at the current sampling time. The center temperature at the current sampling time. This represents the temperature at the second end at the current sampling moment. The first end temperature measuring point is located on the side of the sampling resistor closer to the laser along its length, 0.5 mm from the edge of the resistor. The center temperature measuring point is located at the geometric center of the sampling resistor's length and width. The second end temperature measuring point is located on the side of the sampling resistor furthest from the laser along its length, 0.5 mm from the edge of the resistor. The temperature sensor is mounted on the surface of the sampling resistor using thermally conductive silicone grease, with a thickness not exceeding 0.1 mm. After mounting, it is secured with polyimide tape, with the tape covering no more than 20% of the area of ​​the temperature sensor's sensitive element.

[0012] The sampling resistor is mounted on the surface of the constant-temperature hot plate via an insulating thermally conductive pad. The pad's thickness does not exceed 0.2 mm, and its thermal conductivity is not less than 2 watts per meter (Kelvin). It is secured with four evenly distributed screws at a torque of 0.5 Nm to 1 Nm. The temperature sensor leads are twisted-pair cables with an outer aluminum foil shielding layer. The shielding layer is grounded at one end, and the leads are routed away from high-current wires and high-frequency interference sources to avoid electromagnetic interference introducing measurement noise.

[0013] S203 extracts the temperature difference between the first and second ends to construct a longitudinal temperature difference characterizing the direction of heat flow transfer. The calculation formula is as follows: ; S204 extracts the time span between the current sampling time and the pre-cached previous sampling time, which is used as the sampling duration for subsequent energy integration. The calculation formula is as follows: ; in The absolute time corresponding to the current sampling moment is obtained using the system's internal high-precision real-time clock, with a clock accuracy of no less than ±10ppm. The absolute time corresponding to the previous sampling moment is pre-cached in the system random access memory.

[0014] S301 extracts the nominal current, uncorrected for temperature drift, based on the impedance mapping relationship between the sampling voltage and the nominal resistance value of the sampling resistor at a preset reference temperature. The calculation formula is as follows: ; in This is the sampled voltage at the current sampling time. The nominal resistance value of the sampling resistor is set at a preset reference temperature of 25 degrees Celsius. The nominal resistance value is provided by the factory calibration certificate of the sampling resistor.

[0015] S302 integrates the power mapping relationship between the sampling voltage and the nominal current, extracts the self-heating power characterizing the intrinsic heat source properties of the sampling resistor, and the calculation formula is as follows: ; By combining the S303 terminal voltage and nominal current, the electrical input power, which characterizes the overall energy consumption state of the laser, is extracted using the following formula: ; in This represents the terminal voltage at the current sampling time.

[0016] S401 performs discrete integration and convergence on the pulse energy along the sampling duration to extract the average optical power characterizing the true optical energy output state of the laser. The calculation formula is as follows: ; in Let p be the set of discrete-time series of pulse energies within the current sampling window, where p is the valid index number within this set. Let be the output energy of the p-th laser pulse. Discrete integral convergence is used to converge all trigger moments within the current sampling window. The laser pulse energy within the interval is accumulated and then divided by the sampling duration.

[0017] S402 extracts the waste heat power, representing the heating state residing inside the laser, by subtracting the average optical power from the electrical input power. The calculation formula is as follows: ; S403 performs dimensionless processing based on the relative intensity mapping relationship between waste heat power and electrical input power, reconstructing a normalized heat source characterizing the proportion of waste heat to electrical input, in order to eliminate dimensional interference from the absolute value fluctuations of the external heat source. The calculation formula is as follows: ; S501 extracts the actual heat flow characteristic of physical conduction along the sampling resistor based on the thermal impedance characteristics formed by the equivalent thermal conductivity and equivalent cross-sectional area of ​​the sampling resistor, combined with the longitudinal temperature difference and the preset temperature measurement distance. The calculation formula is as follows: ; in The equivalent thermal conductivity of the sampling resistor is obtained by measuring the steady-state heat flow method. The measurement process involves placing the sampling resistor in a constant temperature environment, applying a known stable temperature difference across its two ends, measuring the heat flow through the sampling resistor, and calculating the equivalent thermal conductivity according to Fourier's law of heat conduction. The typical value range is from 10 W / m Kelvin to 100 W / m Kelvin. The equivalent thermal conductivity cross-sectional area of ​​the sampling resistor is denoted as , and the cross-sectional area of ​​the effective thermal conductivity portion of the sampling resistor is denoted as . It is calculated through three-dimensional laser scanning measurement or by precise dimensional parameters provided by the manufacturer. The preset temperature measurement distance is defined as the center distance between the first end temperature measurement point and the second end temperature measurement point along the length of the sampling resistor.

[0018] The specific steps for measuring equivalent thermal conductivity using the steady-state heat flow method are as follows: Place the sampling resistor horizontally on an insulated support. Attach a heating element to one end of the resistor and a heat sink to the other end, connecting the heat sink to a constant-temperature cold source. Apply a constant power to the heating element. After the temperatures at both ends of the sampling resistor stabilize, record the temperatures at the first and second ends, as well as the input power to the heating element. According to Fourier's law of heat conduction, the formula for calculating equivalent thermal conductivity is: ; in The input power of the heating element. The distance between the center of the heating element and the center of the heat sink. Let be the equivalent thermal conductivity cross-sectional area of ​​the sampling resistor. For a rectangular cross-section sampling resistor, the formula for calculating the equivalent thermal conductivity cross-sectional area is: ; in The width of the sampling resistor, Let be the thickness of the sampling resistor. For a cylindrical cross-section sampling resistor, the formula for calculating the equivalent thermal conductivity cross-sectional area is: ; in The diameter of the sampling resistor.

[0019] S502 constructs a residual model that includes the mapping relationship between actual heat flow and waste heat power, extracts a thermal response kernel that reflects the physical thermal inertia transfer delay characteristics of the metering components, and the calculation formula is as follows: ; Where m is the discrete vernier variable from the initial time to the current time, r is the integral variable of the delay memory step, and h[r] is the sequence of physical thermal delay coefficients to be solved. The maximum delay step of the thermal response kernel. The sampling step size is set to 100 to 500, adjusted according to the thermal inertia of the metering component; a larger value is used when the thermal inertia is high, and a smaller value is used when the thermal inertia is low. The solution process uses the recursive least squares method to avoid the numerical stability problem caused by direct matrix inversion. The recursive formula is as follows: ; ; ; in For the input vector, This is an estimate of the thermal response kernel. Here is the gain matrix. Let covariance matrix be the variance matrix. It is the identity matrix. The initial conditions are set as follows: , ,in The initial covariance coefficients range from 1000 to 10000. When missing historical waste heat power data in the input vector, zeros are used to fill the gaps.

[0020] The update strategy for the thermal response kernel is as follows: during the first 1000 sampling periods after system startup, recursive least squares updates are performed continuously. The update continues until the changes in all coefficients of the thermal response kernel are less than a certain value within 100 consecutive sampling periods. When the thermal response kernel converges, online updates are stopped, and the converged thermal response kernel is fixed for use. If laser replacement or metering component disassembly / reassembly occurs during system operation, the online identification process of the thermal response kernel needs to be retried.

[0021] S503 uses a thermal response kernel to perform discrete convolution integrals on the memory attenuation characteristics of waste heat power within each prior sampling window to reconstruct the return heat flow characterizing the laser waste heat delayed by physical components. The calculation formula is as follows: ; in Let r be the thermal response coefficient corresponding to the r-th delay step. Let be the waste heat power at the nr-th sampling time. hour, Set to 0. Historical waste heat power data is buffered using a circular buffer with a depth of [value missing]. After each sampling, the latest waste heat power data is written to the buffer, overwriting the oldest data.

[0022] The circular buffer employs a separate read and write pointer management method. The write pointer points to the position in the buffer where new data is about to be written, while the read pointer points to the position of the earliest valid data in the buffer. Each time new waste heat power data is written, the data is written to the position pointed to by the write pointer, and then the write pointer is incremented by 1. If the write pointer exceeds the maximum index of the buffer, it wraps back to 0. When reading historical waste heat power data, the index of the corresponding data is calculated based on the delay step size r. .

[0023] S601, based on the dimensionless orthogonal vector mapping relationship between reheated heat flux and self-heating power, extracts the normalized heat flux characterizing the degree of influence of external reheating. The calculation formula is as follows: ; S602 integrates the directional properties of normalized heat flux and longitudinal temperature difference to construct a weighted temperature difference for correcting structural asymmetric thermal bias. The calculation formula is as follows: ; S603 extracts the trend of the average temperature change between the current sampling time and the previous sampling time over the sampling duration, and constructs the temperature change rate characterizing the thermal hysteresis state of the sampling resistor. The calculation formula is as follows: ; in The average temperature corresponding to the previous sampling time is pre-cached in the system's random access memory.

[0024] S604 extracts the deviation of the hot plate temperature from the reference temperature to construct the background temperature difference characterizing the system's environmental noise level. The calculation formula is as follows: ; in The temperature of the hot plate at the current sampling moment. The preset reference temperature is 25 degrees Celsius.

[0025] The S605 collects average temperature, weighted temperature difference, temperature change rate, and background temperature difference to form a temperature feature set of multi-dimensional environmental state parameters, which is stored in the system random access memory for subsequent calculations of compensation resistance and thermoelectric potential.

[0026] The S701 acquires the pre-calibrated first-order temperature coefficient, second-order temperature coefficient, power coefficient, regenerative coefficient, thermal hysteresis coefficient, background coefficient, thermoelectric potential coefficient, and correction coefficient for the metering components. These coefficients are obtained through offline calibration experiments. The equipment required for the calibration experiments includes a high and low temperature test chamber, a precision DC power supply, a standard ammeter, a standard voltmeter, a laser energy meter, a data acquisition system, and a computer. The calibration experiment steps are as follows: The first step is to calibrate the first-order and second-order temperature coefficients. The metrology unit is placed in a high-low temperature test chamber, and the laser power supply circuit is disconnected to ensure no current flows through the sampling resistor. The temperature of the high-low temperature test chamber is set from -10°C to 60°C, varying in 5-degree increments. After each temperature point stabilizes for 30 minutes, the hot plate temperature, the first end temperature, the center temperature, the second end temperature, and the voltage across the sampling resistor are recorded. The standard current flowing through the sampling resistor is measured using a standard ammeter, and the actual resistance value of the sampling resistor at each temperature point is calculated. The first-order temperature coefficient is obtained using quadratic polynomial fitting. With second-order temperature coefficient Typical value ranges are as follows: per degree Celsius to per degree Celsius, per square degree Celsius per square degree Celsius.

[0027] The second step is to calibrate the power coefficient. The high and low temperature test chamber temperature is set to 25 degrees Celsius and kept stable, while the laser power supply circuit is disconnected. Different currents are applied to the sampling resistor using a precision DC power supply, ranging from 10% to 100% of the rated current, varying in 10% increments. After each current point stabilizes for 10 minutes, the sampling voltage, first-terminal temperature, center temperature, second-terminal temperature, and standard current are recorded. The actual resistance value of the sampling resistor at each power point is calculated. Using a multiple linear regression method, the power coefficient is obtained by fitting the calibrated temperature coefficient. The typical value range is per watt to per watt.

[0028] The third step is to calibrate the heat return coefficient. The high and low temperature test chamber temperature is set to 25 degrees Celsius and maintained stable. The power supply circuit of the sampling resistor is disconnected, and different driving currents are applied to the laser to generate waste heat of varying powers. After each driving current point stabilizes for 30 minutes, the temperatures of the hot plate, the first end, the center, and the second end, as well as the electrical input power and optical output power of the laser, are recorded. The weighted temperature difference corresponding to each waste heat power point and the actual resistance value of the sampling resistor are calculated. Using a multiple linear regression method, based on the calibrated temperature and power coefficients, the heat return coefficient is fitted to obtain the coefficient. The typical value range is per degree Celsius to Every degree Celsius.

[0029] The fourth step is to calibrate the thermal hysteresis coefficient. The high and low temperature test chamber temperature is set to 25 degrees Celsius and maintained stable, while the laser power supply circuit is disconnected. A step current is applied to the sampling resistor, increasing from 0 to 50% of the rated current. The average temperature change curve of the sampling resistor over time and the actual resistance value of the sampling resistor are recorded during the step process. Using a multiple linear regression method, based on the calibrated temperature coefficient, power coefficient, and recrystallization coefficient, the thermal hysteresis coefficient is fitted to obtain the coefficient. The typical value range is from 0.01 degrees Celsius per second to 0.1 degrees Celsius per second.

[0030] Step 5: Calibrate the background coefficient. Disconnect the power supply to both the laser and the sampling resistor. Set the temperature of the high and low temperature test chamber from -10°C to 60°C in 5°C increments. After each temperature point stabilizes for 30 minutes, record the hot plate temperature and the actual resistance value of the sampling resistor. Using a multiple linear regression method, based on the calibrated temperature coefficient, power coefficient, reheat coefficient, and thermal hysteresis coefficient, fit the background coefficient to obtain it. The typical value range is per degree Celsius to Every degree Celsius.

[0031] Step 6: Calibrate the thermoelectric potential coefficient and correction coefficient. Disconnect the power supply circuit of the sampling resistor and apply a known stable temperature difference across the resistor, ranging from -5°C to 5°C, in 1°C increments. After each temperature difference stabilizes for 10 minutes, record the open-circuit voltage across the sampling resistor. Then, apply different driving currents to the laser to generate different intensities of backheating, and record the open-circuit voltage and longitudinal temperature difference across the sampling resistor at each backheating intensity. The thermoelectric potential coefficient is obtained using a linear fitting method. With correction factor Typical value ranges are as follows: volts per degree Celsius 0 to 0.5 volts per degree Celsius.

[0032] After calibration, all coefficients are stored in the system flash memory and automatically loaded into the random access memory upon system power-on. The verification standard for the calibration results is that the current measurement compensation error does not exceed ±0.05% across the entire operating temperature range and power range.

[0033] The calibration experiment data preprocessing employed a combination of moving average filtering and outlier removal. 100 data sets were collected for each temperature or power point; 10 sets each of maximum and minimum values ​​were removed, and the arithmetic mean of the remaining 80 sets was taken as the valid data for that point. The independent variable matrix X for the multiple linear regression was constructed such that each row corresponds to a data point and each column corresponds to a feature term. The feature terms include... , , , , , The dependent variable vector Y is the ratio of the actual resistance value to the nominal resistance value of the sampling resistor. The formula for calculating the regression coefficient is: ; in This is a vector containing constant terms and coefficients. The calibration results need to be repeated three times, with the deviation of each coefficient not exceeding ±5% in each experiment. The average of the three experiments is taken as the final calibration coefficient.

[0034] Based on the nominal resistance, S702 integrates various parameters in the temperature feature set with the corresponding quantities of the aforementioned systems to perform multi-dimensional physical effect superposition, reconstructing the compensation resistance value to eliminate interference from non-isothermal structural features. The calculation formula is as follows: ; in It is a first-order temperature coefficient. It is a second-order temperature coefficient. For power coefficient, The reheat coefficient, This is the thermal hysteresis coefficient. This is the background coefficient.

[0035] S703 integrates the distribution characteristics of normalized heat flux and longitudinal temperature difference, and combines correction coefficients and thermoelectric potential coefficients to reconstruct the thermoelectric potential that generates parasitic potential differences along the temperature gradient. The calculation formula is as follows: ; in Thermoelectric potential coefficient, This is a correction factor.

[0036] S801 extracts the thermoelectric potential representing parasitic voltage interference from the sampled voltage, and obtains the effective voltage that eliminates the influence of thermoelectric effect voltage drop. The calculation formula is as follows: ; Based on the impedance mapping relationship between effective voltage and compensation resistance, S802 extracts the compensation current to remove environmental temperature drift and structural backheating interference. The calculation formula is as follows: ; S803 compares the difference between the nominal current and the compensation current and performs a normalized mapping based on the compensation current to extract the relative correction amount used to assess the deviation of the environmental thermal drift state. The calculation formula is as follows: ; S804 constructs an impedance-heat mapping model by combining compensation current and compensation resistance, and extracts the actual self-heat for closed-loop metering record status verification. The calculation formula is as follows: ; During system operation, the abnormal data handling mechanism is as follows: When the sampling voltage, terminal voltage, or temperature parameter exceeds the preset range, the valid value of the previous moment is used to replace the current abnormal value; when the pulse energy data exceeds the range or is 0 for three consecutive times, it is determined that the laser energy meter is faulty, triggering a system alarm and recording the fault information; when the hot plate temperature deviates from the set value by more than ±0.5 degrees Celsius and lasts for more than 1 minute, it is determined that the constant temperature hot plate is faulty, triggering a system alarm and stopping the laser output. The initial power-on state of the system is set as follows: all temperature parameters take the first sampling value, the backheat flux is initialized to 0, the temperature change rate is initialized to 0, the relative correction is initialized to 0, and the actual self-heating is initialized to 0.

[0037] The method for verifying actual self-heating capacity is to compare the actual self-heating capacity with the theoretical self-heating power. The formula for calculating the theoretical self-heating power is: ; When the relative deviation between the actual self-heating heat and the theoretical self-heating power exceeds ±1% and lasts for more than 10 sampling cycles, the system is determined to be abnormal, triggering a status warning and recording relevant parameters.

[0038] Supplementing the overall operating mechanism: All sampled data undergoes first-order low-pass filtering before entering the calculation process. The filter cutoff frequency is set to 1 / 10 of the sampling frequency, and the filtering formula is: ; in This is the original sampled data. The filtered data, The filter coefficient ranges from 0.1 to 0.3. The system calibration cycle is 12 months. If the metering components have experienced severe vibration, high temperature environments, or maintenance and disassembly, recalibration is required in advance. For continuous wave lasers, pulse energy is replaced by continuous optical power, and the average optical power is directly taken from the real-time measurement value of the laser power meter, without the need for discrete integration and focusing.

[0039] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. A temperature compensation method for constant current power supply current measurement, applied to a metering assembly equipped with a constant temperature hot plate, a sampling resistor, and a laser, characterized in that, include: Establish a synchronous sampling sequence, acquire temperature measurement parameters and electrical energy sampling amount within the current sampling window, and extract the average temperature characterizing the spatial heat distribution of the sampling resistor and the longitudinal temperature difference characterizing the directional thermal gradient. The electrical energy sampling data is analyzed to extract the self-heating power characterizing the electrothermal conversion and the electrical input power reflecting the energy consumption state of the laser; The average optical power converted into light energy is stripped from the electrical input power to extract the waste heat power characterizing the feedback-induced heat source; Based on the delay characteristics of the longitudinal temperature difference and the waste heat power, a thermal response kernel is identified, and a reheated heat flow characterizing the delayed feedback of the heat source is constructed. The regenerated heat flow and the self-heating power are combined to extract the normalized heat flow that reflects the dominant state of the heat source, and combined with the average temperature to form a temperature feature set that reflects the dynamic heat distribution. Based on the temperature feature set and the normalized heat flux, the compensation resistance for eliminating non-isothermal errors and the thermoelectric potential for stripping parasitic voltages are reconstructed. The electrical energy sampling quantity is corrected by using the compensation resistance and the thermoelectric potential to eliminate the superposition interference of structural backheating and self-heating, and a compensation current is output.

2. The temperature compensation method for constant current power supply current measurement according to claim 1, characterized in that, Establish a synchronous sampling sequence, acquire temperature measurement parameters and electrical energy sampling within the current sampling window, and extract the average temperature characterizing the spatial heat distribution of the sampling resistor and the longitudinal temperature difference characterizing the directional thermal gradient, including: At the current sampling moment, the electrical energy sample quantity, including the sampling voltage, terminal voltage and pulse energy, and the temperature measurement parameters, including the first terminal temperature, center temperature, second terminal temperature and the hot plate temperature of the constant temperature hot plate, are simultaneously acquired to lock the same transient electrothermal state. The average temperature reflecting the heating state of the sampling resistor is reconstructed by fusing the spatial distribution weights of the first end temperature, the center temperature and the second end temperature, with the center temperature as the dominant weight. Extract the difference feature between the first end temperature and the second end temperature to construct the longitudinal temperature difference characterizing the direction of heat flow transfer; Extract the time span between the current sampling time and the pre-cached previous sampling time, and use it as the sampling duration for subsequent energy integration.

3. The temperature compensation method for constant current power supply current measurement according to claim 2, characterized in that, The analysis of the electrical energy sampling data extracts the self-heating power characterizing the electrothermal conversion and the electrical input power reflecting the energy consumption state of the laser, including: Based on the impedance mapping relationship between the sampling voltage and the nominal resistance value of the sampling resistor at a preset reference temperature, the nominal current uncorrected for temperature drift is extracted. By integrating the power mapping relationship between the sampling voltage and the nominal current, the self-heating power characterizing the endogenous heat source attribute of the sampling resistor is extracted; By combining the terminal voltage and the nominal current, the electrical input power, which characterizes the overall energy consumption state of the laser, is extracted.

4. The temperature compensation method for constant current power supply current measurement according to claim 3, characterized in that, The average optical power converted into light energy is stripped from the electrical input power, and the waste heat power characterizing the feedback-induced heat source is extracted, including: The pulse energy is discretely integrated and converged along the sampling duration to extract the average optical power that characterizes the true optical energy output state of the laser; The waste heat power, which characterizes the heating state residing inside the laser, is extracted by subtracting the average optical power from the electrical input power. Dimensionless processing is performed based on the relative intensity mapping relationship between the waste heat power and the electrical input power to reconstruct the normalized heat source that characterizes the proportion of waste heat to electrical input, thereby eliminating dimensional interference from the absolute value fluctuation of the external heat source.

5. A temperature compensation method for constant current power supply current measurement according to claim 4, characterized in that, Based on the delayed characteristics of the longitudinal temperature difference and the waste heat power, a thermal response kernel is identified, and a reheated heat flux characterizing the delayed feedback of the heat source is constructed, including: Based on the thermal impedance characteristics formed by the equivalent thermal conductivity and equivalent cross-sectional area of ​​the sampling resistor, combined with the longitudinal temperature difference and the preset temperature measurement distance, the actual heat flow characterizing the physical conduction along the sampling resistor is extracted. Construct a residual model that includes the mapping relationship between the actual heat flow and the waste heat power, and extract the thermal response kernel that reflects the physical thermal inertia transfer delay characteristics of the metering component; Based on the thermal response kernel, discrete convolution integrals are performed on the memory decay characteristics of the waste heat power within each prior sampling window to reconstruct the reheated heat flow that represents the waste heat from the laser returning after a delay through physical components.

6. The temperature compensation method for constant current power supply current measurement according to claim 5, characterized in that, The normalized heat flux reflecting the dominant state of the heat source is extracted by fusing the regenerated heat flux with the self-heating power, and combined with the average temperature to form a temperature feature set reflecting the dynamic heat distribution, including: Based on the dimensionless orthogonal vector mapping relationship between the reheated heat flux and the self-heating power, the normalized heat flux characterizing the degree of external reheating influence is extracted; By integrating the normalized heat flux and the directional properties of the longitudinal temperature difference, a weighted temperature difference is constructed to correct the structural asymmetric thermal bias. Extract the trend of the average temperature change between the current sampling time and the previous sampling time within the sampling duration, and construct the temperature change rate characterizing the thermal hysteresis state of the sampling resistor; Extract the deviation of the hot plate temperature from the reference temperature to construct a background temperature difference characterizing the system's environmental noise. The average temperature, the weighted temperature difference, the rate of temperature change, and the background temperature difference are aggregated to form a temperature feature set of multi-dimensional environmental state parameters.

7. A temperature compensation method for constant current power supply current measurement according to claim 6, characterized in that, Based on the temperature feature set and the normalized heat flux, the compensation resistance for eliminating non-isothermal errors and the thermoelectric potential for stripping parasitic voltages are reconstructed, including: Obtain the first-order temperature coefficient, second-order temperature coefficient, power coefficient, reheat coefficient, thermal hysteresis coefficient, background coefficient, thermoelectric potential coefficient, and correction coefficient that have been pre-calibrated for the metering component; Based on the nominal resistance value, the parameters in the temperature feature set are combined with the corresponding quantities of the aforementioned systems to perform multidimensional physical effect superposition, and the compensation resistance value that eliminates interference from non-isothermal structural features is reconstructed. By integrating the distribution characteristics of the normalized heat flux and the longitudinal temperature difference, and combining the correction coefficient and the thermoelectric potential coefficient, the thermoelectric potential that generates a parasitic potential difference along the temperature gradient is reconstructed.

8. A temperature compensation method for constant current power supply current measurement according to claim 7, characterized in that, The electrical energy sampling quantity is corrected using the compensation resistance and the thermoelectric potential to eliminate the superposition interference of structural backheating and self-heating, and a compensation current is output, including: The thermoelectric potential characterizing parasitic voltage interference is stripped from the sampled voltage to extract the effective voltage that eliminates the influence of thermoelectric effect voltage drop; Based on the impedance mapping relationship between the effective voltage and the compensation resistance, the compensation current that removes environmental temperature drift and structural backheating interference is extracted. The difference between the nominal current and the compensation current is compared and a normalized mapping is performed based on the compensation current to extract a relative correction amount for evaluating the environmental thermal drift deviation state. An impedance-heating mapping model is constructed by combining the compensation current and the compensation resistance value, and the actual self-heating is extracted for closed-loop metering record status verification.

Images