CN-122026857-A - Calibration value acquisition system and bandwidth optimization method for input filter of sampling voltmeter

CN122026857ACN 122026857 ACN122026857 ACN 122026857ACN-122026857-A

Abstract

The invention provides a calibration value acquisition system and a sampling voltmeter input filter bandwidth optimization method, which utilize an Alternating Current Josephson (ACJVS) quantum voltage system as an alternating current source, a Programmable Josephson (PJVS) quantum reference voltage system (serving as a provider of reference voltage, a differential sampling method is adopted to acquire a modified sampling voltmeter input filter bandwidth calibration value, a filter bandwidth error compensation model is designed, a data driving method is adopted to select the filter bandwidth error compensation model and a super parameter thereof, and model parameters are estimated by a nonlinear least square fitting method and a data cross verification method.

Inventors

WANG YANPING
YANG SHUIWANG
LI ZONGKUN
QIU SHAOJUN
WANG CHENYAN
ZOU CHUNHUI

Assignees

北京振兴计量测试研究所

Dates

Publication Date: 20260512
Application Date: 20251225

Claims (8)

1. A calibration value acquisition system for bandwidth optimization of an input filter of a sampling voltmeter is characterized by comprising a Programmable Josephson (PJVS) quantum reference voltage system (1), a Josephson (ACJVS) alternating current quantum voltage system (2), a sampling voltmeter (3), a signal generator (4), a rubidium clock (5), four photoelectric isolators (6), (7), (8) and (9), wherein the Programmable Josephson (PJVS) quantum reference voltage system (1) is connected with the high end of the Alternating Current Josephson (ACJVS) quantum voltage system (2), the Programmable Josephson (PJVS) quantum reference voltage system (1) is connected with the low end and the high end of the Alternating Current Josephson (ACJVS) quantum voltage system (2) respectively, a rubidium clock (5) is connected with the PJVS quantum reference voltage system (1), ACJVS quantum alternating current voltage system (2) and the signal generator (4) respectively through three photoelectric isolators (8), (6) and (7), and the Programmable Josephson (PJVS) quantum reference voltage system (1) and the low end of the Alternating Current Josephson (ACJVS) quantum voltage system (2) are connected with the low end and the high end of the sampling voltmeter (3) quantum voltage system (35) respectively, the Programmable Josephson (PJVS) quantum reference voltage system (1) is used as a provider of reference voltage, and a differential sampling method is adopted to collect the bandwidth calibration value of the input filter of the sampling voltmeter (3).
2. The calibration value acquisition system for bandwidth optimization of a sampling voltmeter input filter according to claim 1, wherein the sampling voltmeter (3) is designed such that the sampling voltmeter (3) is a double-integration analog-to-digital converter.
3. A calibration value acquisition system for bandwidth optimization of a sampling voltmeter input filter according to claim 1 or 2, characterized in that the trigger signal of the sampling voltmeter (3) is provided by a josephson (ACJVS) ac quantum voltage system (2) through a photo-isolator (9).
4. A method for optimizing bandwidth of a sampling voltmeter input filter, the method comprising: Designing a calibration value acquisition system for bandwidth optimization of the input filter of the sampling voltmeter according to claims 1-3; the system is utilized, and a differential sampling method is adopted to collect the bandwidth calibration value of the input filter of the sampling voltmeter (3); the bandwidth error compensation coefficient of the filter is input at the front end of the sampling voltmeter is designed, and the bandwidth error compensation coefficient is shown as the following formula: Wherein f is the frequency of the input signal, p is the filter micro-parameter, h is the filter non-micro-parameter, and V out 、V in is the output and input voltage respectively; acquiring a filter micro-parameter and a filter non-micro-parameter based on a calibration value acquisition result, and obtaining a final bandwidth error compensation coefficient on the basis; and performing bandwidth optimization compensation based on the final bandwidth error compensation coefficient.
5. The method for optimizing bandwidth of a sampling voltmeter input filter according to claim 4, wherein the filter non-microcrop h is obtained by: (1) Sorting the calibration data according to the frequency, sequentially selecting the data in a crossing way, dividing the data into 5 groups, and respectively marking the data as d1, d2, d3, d4 and d5; (2) Selecting one group from the 5 groups of data, namely the group of data as a verification set, and the data set formed by the other four groups of data is called a training set; (3) Optionally selecting a possible h0 value, solving the parameter p by using the data of the training set to obtain a filter model, then taking the data of the verification set into the filter model, and calculating a loss value of the verification set, wherein the loss value is marked as l1; (4) Repeating steps (2) - (3) for d 2-d 5 to obtain corresponding loss values l 2-l 5, respectively, and calculating to obtain h0 score (5) And (3) taking all possible values of h, repeating the steps (2) - (4), obtaining the score of each value, and selecting the h value with the smallest score as a final model solving result.
6. The method of optimizing bandwidth of a sampling voltmeter input filter according to claim 5, wherein in step (3), the parameter p is solved by using data of a training set in the following manner: For the micro parameter p, a nonlinear least square fitting processing mode is adopted for estimation, and for a single calibration value sampling point, a loss function is defined as Loss i (p)＝[Filter(f i ,p,h)-c i ] 2 (2) Where (f i ,c i ) is the calibration value, the final optimization problem can be expressed as P is solved according to equation (3) based on the value h0 and the training set.
7. The method of optimizing bandwidth of a sample voltmeter input filter of claim 6, wherein in step (3), validation set data is substituted into said filter model, and a loss value thereof is calculated according to formula (3).
8. The method for optimizing bandwidth of a filter according to any one of claims 6 to 7, wherein an optimal parameter p is obtained, and further, a final bandwidth error compensation coefficient of the filter of the sampled voltmeter is obtained based on the optimal parameters h and p, and the optimal parameter p is obtained by: And after the optimal super-parameter h is obtained, solving the p by using the optimal super-parameter h and the full calibration data according to a formula (3).

Description

Calibration value acquisition system and bandwidth optimization method for input filter of sampling voltmeter Technical Field The invention belongs to the technical field of metering test instruments, relates to a calibration value acquisition system and a bandwidth optimization method of an input filter of a sampling voltmeter, and particularly relates to a calibration value acquisition system and a bandwidth optimization method of the input filter of the sampling voltmeter. Background In recent years, ac signal precision measurement by differential sampling method has become a hot topic of research in various countries and regions. The sampling voltmeter is an important component part in an alternating current signal precision measurement system based on a differential sampling method, and the sampled difference signal of the sampling voltmeter directly influences the accuracy of an alternating current signal precision measurement final result. The input bandwidth of the front-end filter of the sampling voltmeter is limited, and as the frequency of the alternating current signal increases, the accuracy of the measured difference signal also decreases. However, while differential sampling methods can achieve small amounts of measurement, they cannot reduce the impact of the sampling voltmeter filter bandwidth because the sampling voltmeter front-end filter bandwidth itself is slope dependent. In the existing method for compensating the limited bandwidth error of the front-end filter of the sampling voltmeter, the front-end filter model of the sampling voltmeter needs to be designed, and uncertainty evaluation cannot be well carried out on the designed filter compensation model. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides a calibration value acquisition system and a bandwidth optimization method for an input filter of a sampling voltmeter. The invention can solve the technical problem that the input bandwidth of the integral sampler at high frequency affects the magnitude value transfer accuracy. The technical scheme of the invention is as follows: According to one aspect, a calibration value acquisition system for bandwidth optimization of a sampling voltmeter input filter is provided, the acquisition system comprises a Programmable Josephson (PJVS) quantum reference voltage system, a Josephson (ACJVS) alternating current quantum voltage system, a sampling voltmeter, a signal generator, a rubidium clock and four photoelectric isolators, wherein the Programmable Josephson (PJVS) quantum reference voltage system is connected with the high end of the Alternating Current Josephson (ACJVS) quantum voltage system (2), the Programmable Josephson (PJVS) quantum reference voltage system is connected with the low end and the high end of the Alternating Current Josephson (ACJVS) quantum voltage system respectively, the rubidium clock is connected with the PJVS quantum reference voltage system, the ACJVS alternating current quantum voltage system and an external time base of the signal generator respectively, the Josephson (ACJVS) alternating current quantum voltage system is also connected to a trigger signal end of the sampling voltmeter through the photoelectric isolators (9), the Programmable Josephson (PJVS) quantum reference voltage system is utilized as a current source system, and the sampling voltage of the sampling voltmeter input filter can be provided by adopting a differential method of sampling voltmeter input filter. Further, the sampling voltmeter is designed to be a double-integration analog-to-digital converter. Further, the trigger signal of the sampling voltmeter is provided by a Josephson (ACJVS) alternating current quantum voltage system through a photoelectric isolator. According to another aspect, there is provided a method of bandwidth optimization of a sampling voltmeter input filter, the method comprising: designing the calibration value acquisition system for bandwidth optimization of the input filter of the sampling voltmeter; The system is utilized, and a differential sampling method is adopted to collect the bandwidth calibration value of the input filter of the sampling voltmeter; the bandwidth error compensation coefficient of the filter is input at the front end of the sampling voltmeter is designed, and the bandwidth error compensation coefficient is shown as the following formula: (1) wherein f is the frequency of the input signal, p is the filter micro-parameter, h is the filter non-micro-parameter; 、 Output and input voltages, respectively; acquiring a filter micro-parameter and a filter non-micro-parameter based on a calibration value acquisition result, and obtaining a final bandwidth error compensation coefficient on the basis; and performing bandwidth optimization compensation based on the final bandwidth error compensation coefficient. Further, the filter non-mi