Automatic verification method and device for intelligent measurement product

Abstract

The invention discloses an automatic verification method and device for an intelligent measurement product, and particularly relates to the technical field of intelligent measurement and impedance parameter automatic verification, and the method comprises the steps: reading an impedance item, a frequency point, a measuring range and a branch scale value of a to-be-verified product, combining the frequency point with an impedance branch, carrying out the two-time wiring switching of the same impedance branch, and generating a verification sequence. And outputting excitation parameters, branch parameters and wiring parameters. A verification sequence is generated according to the combination of frequency points and impedance branches, two times of wiring switching are configured under the same impedance branch, and the internal measurement error of a detected product and the additional error of a verification link can be separated by combining the division calculation of wiring difference, product deviation and branch drift. Therefore, result inversion and misjudgment caused by total error aliasing can be suppressed relatively.

Description

Technical Field

[0001] This invention relates to the field of intelligent measurement and automatic impedance parameter verification technology, and more specifically, to an automated verification method and apparatus for intelligent measurement products. Background Technology

[0002] In the field of automated verification of intelligent measurement products, the mainstream practice in the industry is to solve the problem of whether the impedance measurement function of the tested product meets the accuracy requirements. Standard resistors, standard capacitors, standard inductors or complex impedance standard parts are usually used as references. The automatic switching unit is connected to the tested product in sequence according to the preset verification points. The measurement results are collected under fixed frequency, fixed wiring method and fixed range conditions. The error judgment, qualified screening or periodic calibration is completed by comparing with the standard value. For example, when performing automated batch inspection on a production line using an intelligent measurement terminal that has the functions of measuring resistance, capacitance, inductance and equivalent impedance, the system needs to continuously complete multi-station switching, fixture contact, reference impedance calling and multi-range testing within a limited cycle time, while also meeting the constraints of high repeatability and low re-inspection rate. Under this constraint, the existing practice will consistently expose a key flaw: when changes in the contact state of the test fixture, fluctuations in parasitic parameters of the leads, aging of the switching channel, drift of the reference impedance network, and degradation of the internal impedance measurement link of the product under test coexist, the final verification result will show the total error after superposition. This leads to observable phenomena such as result reversal, increased misjudgment rate of boundary samples, and overall reversion of batch data after station maintenance for the same product at different workstations, different shifts, or before and after re-inspection. The root cause is that the existing verification method can only determine whether the error is out of tolerance, but cannot effectively separate the error of the product under test itself from the error introduced by the verification link during the verification process. The technical problem this application aims to solve is: how to effectively decouple the internal measurement error of the tested product from the additional error of the verification link for impedance measurement functions during the automated verification process of intelligent measurement products, thereby improving the verification accuracy and supporting fault prediction and health management while ensuring the verification cycle time. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an automated verification method and apparatus for intelligent measurement products. By constructing a verification sequence that includes multiple frequency points, multiple impedance branches, and dual-connection switching, and performing fractional calculations and cross-verifications of the measurement results for wiring differentials, branch deviations, and timing drift, the internal measurement errors of the tested product and the additional errors of the verification link are effectively decoupled, thereby solving the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an automated verification method for intelligent measurement products, comprising: S1. Read the impedance item, frequency point, range and branch nominal value of the product under test, generate the verification sequence according to the combination of frequency point and impedance branch, and the two wiring switches corresponding to the same impedance branch, and output the excitation parameters, branch parameters and wiring parameters. S2. According to the excitation parameters, branch parameters and wiring parameters, control the verification device to connect the impedance branch and switch the wiring mode, drive the tested product to perform impedance measurement, collect the measured value, measurement time and branch identification, and output the raw data. S3. Pair the raw data according to the same frequency point, the same impedance branch and different wiring methods, calculate the difference between the two measurements to obtain the wiring difference value, take the arithmetic mean of the wiring difference value according to the same frequency point to obtain the link error component, calculate the standard deviation of the difference between the measured value and the nominal value according to the same impedance branch to obtain the product error component, and obtain the branch drift component according to the difference between the first and last measurements of the same impedance branch. Output the error component. S4. Sort the error components by frequency and measurement time, calculate the arithmetic mean, range and adjacent difference of the product error components by frequency, calculate the arithmetic mean of the link error components by branch and wiring method, calculate the arithmetic mean of the first and last difference of the branch drift components, and output the product health parameters and workstation health parameters. S5. Map the product health parameters to impedance deviation results, map the workstation health parameters to workstation deviation results, write the branch compensation amount by branch, write the wiring compensation amount by wiring method, write the maintenance instructions by workstation, and output the verification conclusion and final verification result.

[0005] In a preferred embodiment, S1 includes: S1-1: Read the nominal values ​​of each frequency point, each range and each impedance branch corresponding to the impedance item, generate the first test unit according to the pairwise combination of frequency point and impedance branch, and output the first test unit set. S1-2. Configure two wiring switches for each first test unit in the first test unit set, and generate a second test unit set by alternating different impedance branches under the same frequency point and arranging the wiring switches of the same impedance branch at intervals, and output the verification sequence. S1-3. Based on each second test unit in the verification sequence, extract the corresponding frequency point, range, impedance branch and wiring switching sequence, and generate excitation parameters, branch switching parameters and wiring switching parameters that correspond one-to-one with the verification sequence.

[0006] In a preferred embodiment, S2 includes: S2-1. According to the excitation parameters, branch switching parameters and wiring switching parameters, control the verification device to connect the corresponding impedance branches and switch the corresponding wiring methods in sequence, drive the tested product to perform impedance measurement one by one, and output the measurement value corresponding to each measurement. S2-2. While outputting each measurement value, read the branch identifier of the corresponding impedance branch and the measurement time of the corresponding measurement value, and bind the measurement value, branch identifier and measurement time according to the same impedance measurement, and output a single measurement record set. S2-3. Arrange each single measurement record set sequentially according to the execution order in the verification sequence to generate raw data.

[0007] In a preferred embodiment, S3 includes: S3-1. Pair the original data according to the same frequency point, the same impedance branch and different wiring methods, calculate the difference between the two measurements in the group to obtain the wiring difference value, and write the wiring difference value, frequency point, impedance branch, wiring method and measurement time into the difference record set. S3-2. Construct frequency point windows for the differential record set according to the same frequency point, calculate the mean, range and sign inversion number of the wiring differential values ​​in each frequency point window, use the wiring consistency checker to perform cross-checking on the mean, range and sign inversion number, output the link token and write the corresponding frequency point into the link candidate set.

[0008] In a preferred embodiment, S3 further includes: S3-3. Construct branch windows for the original data according to the same impedance branches, calculate the dispersion, timing slope and front-to-back difference of the difference between the measured value and the nominal value in each branch window, and use the branch stability checker to cross-check the dispersion, timing slope and front-to-back difference, output the branch token and write the corresponding impedance branch into the branch candidate set. S3-4. Establish a timing propagation chain for the link candidate set and tributary candidate set at the same measurement time. Calculate the differential aggregation value across impedance tributaries at the same frequency point, the drift aggregation value across frequency points of the same impedance tributary, and the collision count at the same measurement time. Then, use the collision resolver to update the link token and tributary token in the manner of prioritizing the locking of frequency point paths with the link token, prioritizing the locking of tributary paths with the tributary token, and triggering backoff reassembly when the collision count reaches the backoff condition. Write the results to the token status table.

[0009] In a preferred embodiment, S3 further includes: S3-5. The link error component is obtained by summing and averaging the wiring differential values ​​of the locked frequency path in the token status table according to the frequency point. The product error component is obtained by calculating the dispersion of the difference between the measured value and the nominal value of the locked branch path. The branch drift component is obtained by averaging the difference between the two measurements of the locked branch path. The link error component, product error component and branch drift component are written into the error component set. S3-6. Perform cross-view consistency verification on the error component set. When the link error component and the branch drift component cross the corresponding constraint interval at the same measurement time, a re-detection token is generated and the re-detection flag is written back. When the link error component, product error component and branch drift component satisfy the corresponding constraint interval, an acknowledgment token is generated and the error component is output.

[0010] In a preferred embodiment, S4 includes: S4-1. Rearrange the error components according to frequency point order and measurement time order to construct frequency point sequence, branch sequence and time sequence, and output sequence dataset; S4-2. Calculate the mean and range of frequency points for the product error components in the sequence dataset according to the frequency point sequence, and calculate the time-series increment by calculating the difference between two adjacent product error components according to the time-series sequence, and output the product feature set.

[0011] In a preferred embodiment, S4 further includes: S4-3. Calculate the branch mean value of the link error component in the sequence dataset according to the branch sequence and the wiring mean value according to the wiring method. Calculate the mean value of the difference between the branch drift component and the time sequence according to the time sequence. Output the workstation feature set. S4-4. Combine the frequency point average, frequency point range, and timing increment of the product feature set according to the frequency point location to generate product health parameters, and combine the branch average, wiring average, and front-to-back difference average of the workstation feature set according to the branch location to generate workstation health parameters.

[0012] In a preferred embodiment, S5 includes: S5-1. Map the product health parameters to each impedance item according to the frequency point, calculate the aggregate value of the frequency point deviation of each impedance item to obtain the impedance deviation result, and map the station health parameters to each verification station according to the impedance branch and wiring method, calculate the aggregate value of the station deviation of each verification station to obtain the station deviation result. S5-2. Cross-combine the impedance deviation results and the workstation deviation results, and decompose each source of deviation according to the correspondence between impedance items, impedance branches and wiring methods. Generate the branch compensation amount corresponding to each impedance branch, the wiring compensation amount corresponding to each wiring method and the workstation maintenance instruction corresponding to each calibration workstation, and output the compensation instruction set. S5-3. Generate verification conclusions by matching the impedance deviation results and compensation instruction set with the impedance items, and combine the verification conclusions, branch compensation amounts, wiring compensation amounts, and workstation maintenance instructions into the final verification result.

[0013] An automated verification device for intelligent measurement products includes: The sequence arrangement module reads the impedance items, frequency points, ranges, and branch nominal values ​​of the product under test, generates a verification sequence according to the combination of frequency points and impedance branches, and the two wiring switches corresponding to the same impedance branch, and outputs excitation parameters, branch parameters, and wiring parameters. The parameter execution module controls the verification device to connect the impedance branch and switch the wiring mode according to the excitation parameters, branch parameters and wiring parameters, drives the tested product to perform impedance measurement, collects the measured value, measurement time and branch identification, and outputs the raw data. The error decoupling module pairs the raw data according to the same frequency point, the same impedance branch and different wiring methods, calculates the difference between two measurements to obtain the wiring difference value, takes the arithmetic mean of the wiring difference value according to the same frequency point to obtain the link error component, calculates the standard deviation of the difference between the measured value and the nominal value according to the same impedance branch to obtain the product error component, and obtains the branch drift component according to the difference between the first and last measurements of the same impedance branch, and outputs the error component. The health assessment module sorts error components by frequency and measurement time, calculates the arithmetic mean, range, and adjacent difference of product error components by frequency, calculates the arithmetic mean of link error components by branch and wiring method, calculates the arithmetic mean of the first and last difference of branch drift components, and outputs product health parameters and workstation health parameters. The results decision module maps product health parameters to impedance deviation results, and workstation health parameters to workstation deviation results. It writes branch compensation amounts by branch, wiring compensation amounts by wiring method, and maintenance instructions by workstation, and outputs the verification conclusion and final verification result.

[0014] The technical effects and advantages of this invention are as follows: By generating a verification sequence by combining frequency points with impedance branches, and configuring two wiring switches under the same impedance branch, and combining the fractional calculation of wiring differential, product deviation and branch drift, the internal measurement error of the product under test and the additional error of the verification link can be separated, thus helping to relatively suppress the result reversal and misjudgment caused by the total error aliasing. By pairing the original data with branches of the same frequency and impedance and different wiring methods, and calculating the wiring difference value, frequency mean, range and sign reversal number, a verification basis for wiring link consistency can be formed, which helps to identify the impact of contact state changes, wiring abnormalities or link disturbances on the verification results. By calculating the standard deviation, timing slope, and difference between the measured value and the nominal value of the branch, and combining the branch window, timing propagation chain, and conflict resolution processing, the drift characteristics and stability characteristics of the impedance branch can be extracted. This helps to relatively distinguish the branch aging, reference network drift, and deviation of the tested product itself. By rearranging the error components according to frequency and measurement time, and generating product health parameters and workstation health parameters respectively, a single measurement result can be converted into a continuously comparable characteristic quantity, which helps to support fault trend judgment and health status assessment within a certain range. By merging product health parameters into impedance deviation results and workstation health parameters into workstation deviation results, and further generating branch compensation, wiring compensation, and workstation maintenance instructions, a correspondence can be established between deviation sources and handling actions, thus helping to relatively shorten the re-inspection and positioning process and improve maintenance targeting. By maintaining a one-to-one correspondence between frequency points, branches, wiring methods, and measurement times in the verification sequence, and completing pairing, sorting, aggregation, and output along the same field chain in subsequent steps, data traceability and process coherence can be improved, thus helping to balance verification execution and subsequent analysis under limited cycle time conditions. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method steps of the present invention.

[0016] Figure 2 This is a block diagram of the device system of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Refer to the instruction manual appendix Figure 1-2 The present invention provides an automated verification method for intelligent measurement products, comprising: S1. Read the impedance item, frequency point, range and branch nominal value of the product under test, generate the verification sequence according to the combination of frequency point and impedance branch, and the two wiring switches corresponding to the same impedance branch, and output the excitation parameters, branch parameters and wiring parameters. In this specific embodiment, S1 is used to organize the impedance items, frequency points, ranges, and impedance branches corresponding to the product under test into an executable verification sequence before performing impedance measurement, and further generate excitation parameters, branch switching parameters, and wiring switching parameters corresponding to each item in the verification sequence; to ensure that subsequent steps can directly call and form a stable basis for calculating differential, mean, standard deviation, and first-to-last difference, the value scope of the relevant fields in this specific embodiment is limited as follows: the impedance item is a type of impedance measurement function selected under the current verification task; the frequency point is the discrete frequency value that needs to be measured under the impedance item; the range... The program is the measurement range selected for the impedance item at the corresponding frequency point; the impedance branch is a switchable branch in the reference impedance network; the nominal value of the impedance branch is the standard value corresponding to the impedance branch, and it uses the same metrological caliber as the measured value output under the current impedance item; the measured value in this specific embodiment is represented by a single scalar value, which is the resistance value under the resistance item, the capacitance value under the capacitance item, the inductance value under the inductance item, and the pre-converted equivalent impedance value under the equivalent impedance item; therefore, the measured value in subsequent steps can be directly subtracted from the nominal value of the impedance branch; this specific implementation process includes the following steps: In S1-1, first read the impedance item corresponding to the current product under test, then read the nominal values ​​of each frequency point, each range, and each impedance branch corresponding to the impedance item; the frequency point, range, and nominal value of the impedance branch can all be obtained from the verification configuration table. Each impedance branch corresponds to a unique branch identifier, and each branch identifier corresponds to only one nominal value of the impedance branch; after reading, the first test unit is generated by combining the frequency point and the impedance branch in pairs; when the same frequency point corresponds to multiple ranges, the range to which the current combination belongs is first determined within the range, and then the corresponding first test unit is formed; each first test unit contains at least the impedance item identifier, frequency point value, range value, branch identifier, and nominal value of the impedance branch, and is numbered sequentially according to the generation order to obtain the first test unit set; if the current impedance item contains three frequency points, two ranges, and four impedance branches, then the three frequency points and four impedance branches are combined under each range to generate twelve first test units; a total of twenty-four first test units are generated for the two ranges, and are written into the first test unit set in numerical order; In S1-2, each first test unit in the first test unit set is configured with two wiring switches to form two measurement records under the same frequency, the same range, and the same impedance branch. The two wiring switches correspond to the first wiring method and the second wiring method, respectively. Both are valid wiring states predefined by the verification device and are distinguished by the wiring method number. After configuration, each first test unit is expanded into two second test unit records. The two records are identical in all fields except for the wiring method number and the wiring switch order. Then, a verification sequence is generated according to a fixed rule: under the same frequency, the second test units corresponding to different impedance branches are arranged alternately; for the two second test units under the same impedance branch, Instead of arranging them consecutively, at least one second test unit corresponding to another impedance branch is inserted between them. This ensures a time interval between two subsequent measurements of the same impedance branch and reduces the instantaneous impact caused by continuous switching of the same impedance branch. After the arrangement is completed, all second test units are arranged in sequence to form a verification sequence. Taking branches A, B, and C at the same frequency point as an example, the verification sequence can be written in the order of branch A first connection, branch B first connection, branch C first connection, branch A second connection, branch B second connection, and branch C second connection. If there are multiple frequency points, the above arrangement is completed within one frequency point first, and then the next frequency point is moved on. The frequency point order is determined according to the frequency point order in the verification configuration table. In S1-3, based on each second test unit in the verification sequence, the corresponding frequency point, range, impedance branch, and wiring switching sequence are extracted to generate excitation parameters, branch switching parameters, and wiring switching parameters that correspond one-to-one with the verification sequence. Specifically, for each second test unit, its frequency point value and range value are first read to generate excitation parameters. The excitation parameters include at least the excitation frequency and range number, and if necessary, the excitation amplitude corresponding to that range, which is taken from a fixed value in the range configuration table. Then, its branch identifier is read to generate branch switching parameters. The branch switching parameters include at least the branch identifier and branch switching sequence number. Finally, its wiring method number and wiring switching sequence are read to generate... Wiring switching parameters; the wiring switching parameters include at least the wiring method number and the wiring execution sequence number; all three types of parameters are written into the parameter record table with the same sequence number as the corresponding second test unit, so as to ensure that when the subsequent verification device is called in sequence, the excitation parameters, branch switching parameters and wiring switching parameters can correspond to each item of the second test unit in the verification sequence; if the frequency point corresponding to a certain second test unit is 1kHz, the range is medium range, the branch identifier is B, and the wiring method number is 2, then the excitation parameters generated for this record are 1kHz and medium range, the branch switching parameters are branch B, the wiring switching parameters are wiring method 2, and they share the same sequence number with the second test unit; Through the above processing, impedance items, frequency points, ranges, and nominal values ​​of impedance branches can be converted into a complete, sequential, and directly executable verification sequence, generating excitation parameters, branch switching parameters, and wiring switching parameters corresponding to each item in the verification sequence. Since each second test unit simultaneously retains the impedance item, frequency point, range, branch identifier, nominal value of the impedance branch, wiring method number, and execution sequence, subsequent steps can directly complete the measurement value binding, frequency point pairing, standard deviation calculation for the same branch, and first-to-last difference calculation based on these fields, avoiding problems such as inability to retrieve values, mismatched fields, or uncertain timing. In practical applications: when a machine has resistance, capacitance, and inductance measurement functions... When the product under test enters the verification station, one impedance item can be selected as the current verification object. Then, the nominal values ​​of the three frequency points, two ranges, and four impedance branches corresponding to the impedance item are read from the verification configuration table. The first test unit set is generated by combining the frequency points and impedance branches. Subsequently, a first wiring method and a second wiring method are configured for each first test unit. The verification sequence is generated by alternating different impedance branches at the same frequency point and by the interval between two wirings of the same impedance branch. Finally, the frequency point, range, impedance branch, and wiring switching sequence are extracted one by one according to the verification sequence. The corresponding excitation parameters, branch switching parameters, and wiring switching parameters are generated and written into the parameter record table for subsequent verification devices to call in sequence.

[0019] S2. According to the excitation parameters, branch parameters and wiring parameters, control the verification device to connect the impedance branch and switch the wiring mode, drive the tested product to perform impedance measurement, collect the measured value, measurement time and branch identification, and output the raw data. In this specific embodiment, S2 is used to drive the verification device to perform impedance measurements on the tested product item by item according to the excitation parameters, branch switching parameters, and wiring switching parameters generated by S1, and to organize the measured value, branch identifier, and measurement time corresponding to each measurement into raw data that can be directly calculated later. The purpose of this process is to ensure that each impedance measurement strictly corresponds to a unique frequency point, range, impedance branch, and wiring method, and to ensure that the output data record can be aligned with the verification sequence item by item, thereby providing a complete data foundation for the subsequent calculation of wiring differential value, link error component, product error component, and branch drift component. To avoid problems such as unclear measurement objects, untraceable timing, or unmatched records in subsequent steps, this specific embodiment stipulates that: each impedance measurement corresponds to a unique single measurement record; the measured value adopts a single scalar value consistent with the current impedance item; the measurement time adopts the timestamp or execution sequence number recorded by the verification device when the measured value is output; the branch identifier adopts a unique number corresponding one-to-one with the impedance branch; a single measurement record includes at least the measured value, branch identifier, measurement time, frequency point, range, and wiring method. This specific implementation process includes the following steps: In S2-1, the verification device performs measurements sequentially according to the order of excitation parameters, branch switching parameters, and wiring switching parameters. For each set of parameters, the excitation conditions for the current impedance measurement are first set according to the excitation parameters, including at least the excitation frequency and range. Then, the reference impedance network is controlled to connect the corresponding impedance branch according to the branch switching parameters. Subsequently, the wiring switching unit is controlled to switch to the corresponding wiring mode according to the wiring switching parameters. After the above three actions are completed, the verification device sends a current measurement trigger signal to the product under test, driving the product under test to perform an impedance measurement at the current frequency, current range, current impedance branch, and current wiring mode, and reads the measured value output by the product under test. In this specific embodiment, the measured value is a single scalar. The value is consistent with the nominal value of the current impedance item and impedance branch, so it can be directly subtracted from the nominal value of the impedance branch in the future. After each set of parameters is executed, the corresponding measured value is output and the next set of parameters is executed. If there are six parameter records in the current verification sequence, the verification device will complete six independent measurements in sequence and output six measured values ​​in order. In practical applications, for example, if the frequency corresponding to the current parameter record is 1kHz, the range is medium range, the impedance branch is branch B, and the wiring method is the second wiring method, the verification device will first set the excitation frequency to 1kHz, switch the range to medium range, then connect branch B and switch to the second wiring method, and then trigger the tested product to complete an impedance measurement and output the corresponding measured value. In S2-2, simultaneously with each measured value output, the branch identifier and measurement time corresponding to that measurement are read, and the measured value, branch identifier, and measurement time are bound together according to the same impedance measurement to form a single measurement record. Here, the branch identifier is directly taken from the branch number in the current branch switching parameters, and the measurement time is recorded by the verification device immediately upon receiving the measured value. If the system uses a timestamp method, the measurement time is the current clock value; if the system uses an execution sequence number method, the measurement time is the execution number of the current measurement in the verification sequence. In addition to the above three items, to ensure that subsequent steps can be directly paired and grouped according to frequency point, range, and wiring method, this specific embodiment also includes the frequency point, range, and wiring method corresponding to the current measurement. All data are written to the same record, ensuring that each single measurement record contains at least six items: frequency, range, branch identifier, wiring method, measured value, and measurement time. When binding data, the same sequence number is used, meaning the current measured value, current branch identifier, current measurement time, current frequency, current range, and current wiring method share the same record number, thus guaranteeing that these fields all point to the same impedance measurement. All single measurement records are grouped into a single measurement record set according to the order of output. In practical applications, if the third measurement corresponds to a measured value of 100.25, a branch identifier of B, a measurement time of sequence number 3, a frequency of 1kHz, a range of medium, and a wiring method of the second wiring method, then these six fields are written together into the third single measurement record. In S2-3, the single measurement records in the single measurement record set are arranged sequentially according to the execution order in the verification sequence to generate raw data. The execution order here is consistent with the arrangement order of the verification sequence in S1, and no reordering or change of the field content in any single measurement record is performed. Only the single measurement records are continuously written into the raw data table according to their execution sequence number. In the raw data generated in this way, each record can be traced back to the unique second test unit in the verification sequence and can be directly used for subsequent pairing according to the same frequency point, the same impedance branch, and different wiring methods. To ensure that the first and last difference calculation can be executed, this specific implementation method stipulates that: the first and last records of the same impedance branch are based on the impedance branch in the raw data. The first and last records are determined. To ensure the execution of the time sequence, this specific implementation specifies that the measurement time order in subsequent steps shall be based on the record arrangement order in the original data. When the measurement time uses timestamps, the timestamp ascending order shall be consistent with the record order. In practical applications, if the execution order of the verification sequence is branch A first connection, branch B first connection, branch C first connection, branch A second connection, branch B second connection, branch C second connection, then the original data is written into six records in the same order. Subsequently, the first and fourth records of branch A can be directly paired, the second and fifth records of branch B can be paired, and the third and sixth records of branch C can be paired. Through the above processing, S2 transforms the excitation parameters, branch switching parameters, and wiring switching parameters output by S1 into raw data that corresponds item by item to the verification sequence. This ensures that each impedance measurement has a clear measurement object, a clear execution order, and a clear data attribution. Simultaneously, the frequency point, range, branch identification, wiring method, measured value, and measurement time in the raw data maintain a one-to-one correspondence. This guarantees that subsequent steps can directly complete frequency point pairing, standard deviation calculation for the same branch, first-to-last difference calculation, and sorting by frequency point and measurement time, without issues such as missing fields, mismatched fields, or discontinuous timing. In practical applications… Once the verification sequence has been generated according to the following steps: first wiring of branch A, first wiring of branch B, first wiring of branch C, second wiring of branch A, second wiring of branch B, and second wiring of branch C, the verification device first sets the frequency and range according to the first parameter record, connects branch A, switches to the first wiring mode, and triggers the tested product to output the first measurement value. Then, it synchronously writes the branch identifier, measurement time, frequency, range, and wiring mode for this measurement. Subsequently, it continues to execute the second to sixth parameter records in the same way, and finally forms a raw data table arranged in the order of the verification sequence, which can be directly called for subsequent error component calculations.

[0020] S3. Pair the raw data according to the same frequency point, the same impedance branch and different wiring methods, calculate the difference between the two measurements to obtain the wiring difference value, take the arithmetic mean of the wiring difference value according to the same frequency point to obtain the link error component, calculate the standard deviation of the difference between the measured value and the nominal value according to the same impedance branch to obtain the product error component, and obtain the branch drift component according to the difference between the first and last measurements of the same impedance branch. Output the error component. In this specific embodiment, S3 is used to further decompose the raw data generated by S2 into error components that can be used for subsequent health assessment and outcome decision-making. This process does not simply average all measurements; instead, it first extracts differential information caused by changes in wiring method from the raw data, then extracts deviation information caused by changes in impedance branches. Subsequently, it combines frequency, impedance branch, and measurement time dimensions to perform cross-validation, path locking, and conflict resolution on the two types of information, finally outputting link error components, product error components, and branch drift components. To ensure the process is executable, this specific embodiment defines the relevant terminology and value definitions as follows: each record in the raw data must include at least frequency, range, impedance branch, wiring method, measured value, and measurement time; wiring method... It includes at least a first wiring method and a second wiring method; the measured value and the nominal value of the impedance branch use the same measurement caliber; the dispersion is calculated using the standard deviation; the timing slope is calculated by dividing the difference between the measured values ​​of the first and last records within the same impedance branch by the difference in measurement time; the difference between consecutive records is calculated by the difference between the measured values ​​of the first and last records within the same impedance branch; the mean is the arithmetic mean; the range is the difference between the maximum and minimum values; the number of sign inversions is the number of sign changes of the difference values ​​of two adjacent items arranged in sequence; the collision count is the number of records simultaneously locked by different paths at the same measurement time; to ensure traceability of subsequent operations, both link tokens and branch tokens are generated in the form of record items, which include at least the object identifier, token type, generation time, and current status; the specific implementation process includes the following steps: In S3-1, the raw data is first paired according to the same frequency point, the same impedance branch, and different wiring methods. Specifically, the raw data is first sorted in ascending order of frequency point and measurement time. Then, under the same frequency point, the records are grouped according to the impedance branch. For each group, the two records with the first wiring method and the second wiring method are paired one-to-one. If there are more than two records under the same frequency point and the same impedance branch, the earliest first wiring method record is paired with the earliest second wiring method record according to the measurement time, and the remaining records are paired according to the same rule. After pairing, the wiring difference value is calculated for each pair of records in a fixed direction, which is the measured value under the first wiring method minus... The measured value under the second wiring method is then written into the differential record set along with the corresponding frequency point, impedance branch, wiring method identifier, and measurement time of the pair of records. The measurement time is taken as the measurement time of the last record in the pair of records, and the wiring method identifier is written as the differential pair identifier. After this processing, each record in the differential record set uniquely corresponds to a wiring differential calculation under the same frequency point and the same impedance branch. In practical applications, if the measured value of the first wiring method of branch A under a certain frequency point is 100.20 and the measured value of the second wiring method is 100.05, then the corresponding wiring differential value is 0.15, and it is written into the differential record set along with the frequency point, branch A, and the corresponding measurement time. In S3-2, frequency windows are constructed for the differential record set according to the same frequency point, and the mean, range, and sign inversion count of the wiring differential values ​​are calculated within each frequency window. Specifically, all wiring differential values ​​under the same frequency point are arranged in the order of writing. First, all wiring differential values ​​are added together and divided by the number of differential values ​​to obtain the mean. Then, the maximum value is taken and the minimum value is subtracted to obtain the range. Next, the signs of two adjacent wiring differential values ​​are compared in turn. If the previous value is positive and the next value is negative, or the previous value is negative and the next value is positive, it is recorded as one sign inversion. All signs are accumulated to obtain the sign inversion count. Then, a wiring consistency checker is used to cross-check the mean, range, and sign inversion count. In this specific implementation, the wiring consistency checker is not an abstract module, but rather executes the following fixed rules: first, it is determined whether the positive and negative directions of the mean of the wiring differential values ​​within the frequency window are consistent; then, the range is determined. The system checks whether the mean is greater than the absolute value of the mean; then it checks whether the number of sign inversions is greater than zero; if the mean direction is stable, the range is not greater than the absolute value of the mean, and the number of sign inversions is equal to zero, a stable link token is generated; if the mean direction is stable but the range is greater than the absolute value of the mean, or the number of sign inversions is greater than zero, a link token to be verified is generated; if the mean direction is unstable and the number of sign inversions is greater than zero, no link token is generated; for frequency points where link tokens have been generated, the frequency point is written into the link candidate set; thus, each item in the link candidate set corresponds to a frequency point object that has passed the wiring differential consistency screening; in practical applications, if there are four wiring differential values ​​under a certain frequency point, namely 0.12, 0.10, 0.11, and 0.09, then its mean is 0.105, its range is 0.03, and its number of sign inversions is zero, a stable link token can be generated and the frequency point can be written into the link candidate set; In S3-3, branch windows are constructed for the original data according to branches with the same impedance. Within each branch window, the standard deviation, timing slope, and front-to-back difference are calculated for the difference between the measured value and the nominal value of the impedance branch. Specifically, the original data is first grouped by impedance branch, and then the records within each group are arranged in ascending order of measurement time. Then, the difference between each measured value and the corresponding nominal value of the impedance branch within the branch window is calculated for each record, resulting in a deviation sequence. For this deviation sequence, its arithmetic mean is first calculated, then the variance is obtained by dividing the sum of squares of each deviation value minus the arithmetic mean by the number of deviations. Finally, the square root of the variance is taken to obtain the standard deviation. The timing slope is obtained by dividing the difference between the measured value of the first record and the measured value of the last record by the difference between the first and last measurement times. The front-to-back difference is obtained by taking the difference between the measured value of the first record and the measured value of the last record. Then, a branch stability checker is used to cross-check the standard deviation, timing slope, and front-to-back difference. The branch stability checker has its own... In the implementation, the following rules are consistently applied: First, determine if the timing slope is zero; then, determine if the positive or negative direction of the difference between the preceding and following points is consistent with the timing slope; then, determine if the standard deviation is greater than the absolute value of the difference between the preceding and following points. If the timing slope is zero and the difference between the preceding and following points is zero, a stable branch token is generated; if the timing slope is consistent with the direction of the difference between the preceding and following points and the standard deviation is not greater than the absolute value of the difference between the preceding and following points, a drift branch token is generated; if the timing slope is inconsistent with the direction of the difference between the preceding and following points, or the standard deviation is greater than the absolute value of the difference between the preceding and following points, a branch token to be verified is generated. For impedance branches that have generated branch tokens, the impedance branch is written into the branch candidate set. In practical applications, if the first and last measurement values ​​of an impedance branch in the entire verification sequence are 100.00 and 100.30 respectively, and the time difference between the first and last measurement is 6, then the timing slope is 0.05 and the difference between the preceding and following points is 0.30; if the standard deviation of the corresponding deviation sequence is 0.08, then the branch can generate a drift branch token and be written into the branch candidate set. In S3-4, a time-series propagation chain is established for the link candidate set and the branch candidate set at the same measurement time, and the differential aggregation value, drift aggregation value, and collision count are calculated on this time-series propagation chain. Specifically, firstly, using the measurement time as the primary key, the frequency point objects corresponding to the link candidate set and the impedance branch objects corresponding to the branch candidate set are mapped to the recording nodes at the same measurement time; then, nodes with adjacent measurement times that share the same frequency point or the same impedance branch are connected to form a propagation edge, forming a time-series propagation chain; subsequently, at each measurement time, the arithmetic mean of all wiring differential values ​​across the impedance branch at the same frequency point is taken to obtain the differential aggregation value at that measurement time; the arithmetic mean of all front and back differences across the frequency point of the same impedance branch is taken to obtain the drift aggregation value at that measurement time; then, the number of nodes simultaneously pointed to by the link candidate set and the branch candidate set at that measurement time and with inconsistent states is counted to obtain the collision count; finally, the link token and branch token are updated using a collision resolver; the collision resolver... In this specific implementation, the following rules are consistently enforced: When a node possesses both a link token and a tributary token, the frequency path corresponding to the link token is locked first, followed by the tributary path corresponding to the tributary token. If the collision count at the same measurement time is zero, the states of both tokens remain unchanged. If the collision count at the same measurement time is greater than zero, a rollback and reassembly are triggered. Rollback and reassembly involves reconstructing the pairing relationship and propagation edge based on the records from one measurement time before and after the current measurement time, without re-collecting measurement data. The updated link token and tributary token are written into the token state table, which includes at least the measurement time, frequency path state, tributary path state, and rollback flag. In practical applications, if the same record falls into both the link candidate set and the tributary candidate set at a given measurement time, and the states given by the two types of paths are different, then the record is counted as one collision. If there are two such records at the current measurement time, the collision count is 2, and a rollback and reassembly corresponding to that measurement time is triggered. In S3-5, the connection differential values ​​of the locked frequency paths in the token status table are summed and averaged by frequency point to obtain the link error component; the standard deviation of the difference between the measured value and the nominal value of the impedance branch of the locked branch path is calculated to obtain the product error component; the arithmetic mean of the difference between two consecutive measurements of the locked branch path is calculated to obtain the branch drift component; specifically, firstly, all records with the status of locked frequency paths are extracted from the token status table, then grouped by frequency point, and the corresponding connection differential values ​​under the same frequency point are added and divided by the number of records to obtain the link error component of that frequency point; then, all records with the status of locked branch paths are extracted, then grouped by impedance branch, and the difference between each measured value and the corresponding nominal value of the impedance branch under the same impedance branch is calculated using the standard deviation formula to obtain the product error component of that impedance branch; finally, the first record of the same impedance branch sorted by measurement time is... The difference between the first and last records is calculated, and the arithmetic mean of the first and last differences for all locked branch paths is taken to obtain the branch drift component. The above link error component, product error component, and branch drift component are all written into the error component set, where each item retains the corresponding frequency point identifier or impedance branch identifier and measurement time range for direct use in subsequent health assessments. In practical applications, if the three locked connection difference values ​​at a certain frequency point are 0.12, 0.10, and 0.11, then the link error component at that frequency point is 0.11; if the four deviation values ​​at a certain impedance branch are 0.20, 0.18, 0.22, and 0.19, then the product error component of that branch is the standard deviation of the four deviation values; if the first and last differences of the three locked impedance branches are 0.30, 0.24, and 0.36, then the branch drift component is the arithmetic mean of the three, 0.30. In S3-6, a cross-view consistency check is performed on the error component set, and a retest token or confirmation token is generated based on the check result. This cross-view consistency check compares the correspondence between link error components, product error components, and branch drift components at the same measurement time or on the same object. To ensure the check is executable, this specific implementation defines the constraint interval as the upper and lower bounds determined by the nominal impedance branch value and the repeated measurement deviation: For the link error component, its constraint interval is the average of all wiring difference values ​​at the same frequency point plus or minus half of the corresponding range; for the product error component, its constraint interval is the average of all deviation values ​​under the same impedance branch plus or minus the corresponding standard deviation; for the branch drift component, its constraint interval is the average of the first and last differences of all locked branches plus or minus the corresponding standard deviation. During the check, it is first determined whether the link error component and the branch drift component simultaneously fall outside their respective constraint intervals at the same measurement time; if they do... If the error is outside the specified range, a retest token is generated, and the retest flag is written back to the error component set. Then, it is determined whether the link error component, product error component, and branch drift component simultaneously fall within their respective constraint ranges. If they do, an acknowledgment token is generated, and the corresponding link error component, product error component, and branch drift component are used as the output error components of S3. If neither the retest condition nor the acknowledgment condition is met, the status is retained as pending verification, awaiting further processing. In practical applications, if at a certain measurement moment the link error component is 0.11, falling within the corresponding constraint range, the product error component is 0.02, falling within the corresponding constraint range, and the branch drift component is 0.30, also falling within the corresponding constraint range, then an acknowledgment token can be generated and the three error components can be output. If the link error component and branch drift component simultaneously fall outside their respective constraint ranges, a retest token is generated and the retest flag is written back. Through the above processing, S3 transforms the frequency point information, impedance branch information, wiring method information, and measurement timing information in the original data layer by layer into link error components, product error components, and branch drift components. This ensures that subsequent health assessments and result decisions no longer directly rely on single measurement values, but rather on error components that have undergone pairing, filtering, locking, and consistency verification. Simultaneously, through explicit calculations of wiring differential values, standard deviation, first-to-last difference, differential aggregation value, drift aggregation value, and collision count, as well as fixed generation rules for link tokens, branch tokens, retest tokens, and acknowledgment tokens, it guarantees that each result has a clear input object, a clear calculation process, and a clear output status, preventing issues such as unclear objects, incomplete fields, or unclear paths. In practical applications: when the original data table has already written each frequency point, each impedance branch, and each wiring according to the verification sequence... After recording measurements under the specified conditions, the first wiring method record is paired with the second wiring method record according to the same frequency and impedance branch to obtain the wiring differential value and form a differential record set. Then, the mean, range, and sign reversal number of the wiring differential value are calculated at the same frequency to generate link tokens and link candidate sets. At the same time, the standard deviation, timing slope, and front-to-back difference between the measured value and the nominal value of the impedance branch are calculated under the same impedance branch to generate branch tokens and branch candidate sets. Subsequently, the two types of candidate sets are connected into a timing propagation chain according to the same measurement time, and the differential aggregation value, drift aggregation value, and collision count are calculated to update the token status table. Finally, the link error component, product error component, and branch drift component are obtained for the locked path, and cross-view consistency verification is performed through the constraint interval. The confirmed error component or re-check mark is output, thus completing the entire implementation of S3.

[0021] S4. Sort the error components by frequency and measurement time, calculate the arithmetic mean, range and adjacent difference of the product error components by frequency, calculate the arithmetic mean of the link error components by branch and wiring method, calculate the arithmetic mean of the first and last difference of the branch drift components, and output the product health parameters and workstation health parameters. In this specific embodiment, S4 is used to further organize the link error components, product error components, and branch drift components output by S3 into product health parameters and workstation health parameters that can be directly used for result judgment and compensation generation. The basic idea of ​​this process is: first, rearrange the error components in a unified order to eliminate the arrangement differences of error components from different sources in the frequency point dimension, impedance branch dimension, and measurement time dimension; then, extract the frequency point mean, frequency point range, and time sequence increment reflecting the measurement stability of the tested product from the product error components, and extract the frequency point mean, frequency point range, and time sequence increment reflecting the stability of the calibration workstation from the link error components and branch drift components. Qualitative branch mean, wiring mean, and mean difference before and after are obtained. Finally, these characteristics are combined into product health parameters and workstation health parameters at fixed locations. To ensure the process is executable, the following limitations are made in this specific implementation: the frequency order follows the frequency arrangement order corresponding to the calibration configuration table in S1; the measurement time order follows the recording order of the original data in S2; the branch sequence follows the ascending order of impedance branch identifiers; the timing increment is the difference between two adjacent product error components at the same frequency point; the mean difference before and after is the arithmetic mean of the drift components of each impedance branch. This specific implementation process includes the following steps: In S4-1, the error components are first rearranged to construct frequency point sequences, branch sequences, and time series, and the sequence dataset is output. Specifically, each record in the error component set is read first, and the corresponding frequency point, impedance branch, wiring method, measurement time, link error component, product error component, and branch drift component are extracted. Then, all records are sorted in primary order by frequency point, and then sorted in secondary order by measurement time within the same frequency point, thus forming the frequency point sequence and time series. At the same time, the impedance branches in all records are sorted in ascending order by branch identifier to form the branch sequence. After completing the above three sorting, each record is sorted according to a unified field order. The data is written into a sequence dataset, where the unified field order includes at least the frequency point location, branch location, measurement time location, link error component, product error component, and branch drift component. Thus, each record in the sequence dataset simultaneously possesses the frequency point location, branch location, and measurement time location, allowing for subsequent feature extraction by frequency point, branch, and time sequence, respectively. In practical applications, if the error component set involves three frequency points and four impedance branches, it is first arranged in the order of frequency point 1, frequency point 2, and frequency point 3, and then within each frequency point, it is arranged in ascending order of measurement time. The branch sequence is arranged as branch A, branch B, branch C, and branch D, ultimately forming a sequence dataset with a unified structure. In S4-2, the frequency mean and frequency range of the product error components in the sequence dataset are calculated according to the frequency sequence, and the time increment is obtained by calculating the difference between two adjacent product error components according to the time sequence, and the product feature set is output. Specifically, firstly, the frequency sequence is grouped by frequency point, and all product error components corresponding to the same frequency point are extracted. The frequency mean of each product error component is obtained by adding the product error components together and dividing by the number of records. Then, the maximum value and minimum value of the product error components in the group are taken to obtain the frequency range of that frequency point. Subsequently, the product error components are arranged in the order of measurement time within the same frequency point, and the difference between two adjacent product error components is calculated, which is the next product error component. Subtract the previous product error component to obtain the corresponding time increment; if there is only one product error component at a certain frequency point, the time increment of that frequency point is recorded as zero; the above frequency point mean, frequency point range, and time increment are written into the product feature set one by one according to the frequency point position; after this processing, each item in the product feature set corresponds to a specific frequency point position; in practical applications, if three product error components of 0.020, 0.024, and 0.022 are obtained sequentially at a certain frequency point, then the frequency point mean is 0.022, the frequency point range is 0.004, and the time increments are 0.004 and -0.002 respectively, and are written into the product feature set according to the position of that frequency point; In S4-3, the link error components in the sequence dataset are calculated by branch sequence and by wiring method. The branch drift components in the sequence dataset are calculated by time sequence and the mean difference between the beginning and end is calculated. The station feature set is then output. Specifically, the link error components are first grouped by branch sequence, and all link error components corresponding to the same impedance branch are extracted, summed, and divided by the number of records to obtain the branch mean of that impedance branch. Then, the link error components are grouped by wiring method, and all link error components corresponding to the first wiring method are extracted and the arithmetic mean is taken to obtain the first wiring mean. All link error components corresponding to the second wiring method are extracted and the arithmetic mean is taken to obtain the second wiring mean. Subsequently, the branch drift components are read according to the time sequence of each impedance branch, and all branch drift components are summed and divided by the number of branches. The mean of the difference before and after is obtained. If the branch drift components are stored separately by frequency point, the arithmetic mean of the branch drift components of the same impedance branch at each frequency point is first taken, and then it is used to calculate the mean of the difference before and after. The above branch mean, wiring mean, and mean of the difference before and after are written into the workstation feature set. The branch mean is stored according to the branch position, the wiring mean is stored according to the wiring method, and the mean of the difference before and after is stored as the overall drift feature of the workstation. In practical applications, if the link error components corresponding to branch A are 0.10, 0.12, and 0.11, then the branch mean of branch A is 0.11. If the link error component mean corresponding to the first wiring method is 0.09 and the link error component mean corresponding to the second wiring method is 0.13, then they are written into the workstation feature set respectively. If all branch drift components are 0.30, 0.24, and 0.36, then the mean of the difference before and after is 0.30. In S4-4, the frequency point mean, frequency point range, and timing increment of the product feature set are combined according to the frequency point location to generate product health parameters, and the branch mean, wiring mean, and front-to-back difference mean of the workstation feature set are combined according to the branch location to generate workstation health parameters. Specifically, for the product feature set, the frequency point mean, frequency point range, and timing increment corresponding to the frequency point are read sequentially according to the frequency point location and written into the same set of product health parameter records; if a frequency point corresponds to multiple timing increments, all of them are retained in the order of measurement time and written into the same frequency point location; for the workstation feature set, the branch mean corresponding to the branch is read sequentially according to the branch location, and then all wiring mean and front-to-back difference mean are read and written into the same set of workstation health parameter records; if a workstation contains multiple impedance branches, each Each impedance branch corresponds to a workstation health parameter record, and the average wiring value and the average difference between the beginning and end are copied and written to the corresponding record of each branch. After this processing, both product health parameters and workstation health parameters are stored using a fixed field structure, and S5 can directly call them according to the frequency point position and branch position. In practical applications, if the frequency point average value corresponding to frequency point 1 is 0.022, the frequency point range is 0.004, and the timing increment is 0.004 and -0.002, then these three items are combined according to the position of frequency point 1 to form a product health parameter record; if the branch average value corresponding to branch A is 0.11, the average value of the first wiring is 0.09, the average value of the second wiring is 0.13, and the average difference between the beginning and end is 0.30, then these four items are combined according to the position of branch A to form a workstation health parameter record. Through the above processing, S4 further converts the error components into product health parameters and workstation health parameters with clear location indices and fixed field structures. This eliminates the need for subsequent steps to revert to the original error component level for repeated sorting and calculation, allowing direct generation of deviation results, writing of compensation amounts, and generation of maintenance instructions. Simultaneously, by employing a process of rearranging, grouping, calculating, and recombining, the value ranges, calculation methods, and recording locations of frequency point averages, frequency point ranges, timing increments, branch averages, wiring averages, and the average of the difference between the beginning and end are clearly defined, avoiding issues such as frequency point and branch misalignment, unclear adjacent difference objects, or unclear sources of averages. In practical applications: when the error components... After the data set has output the link error components, product error components, and branch drift components, all records are first rearranged according to frequency point order and measurement time order to form a sequence dataset containing frequency point location, branch location, and measurement time location. Then, the frequency point mean, frequency point range, and time increment are calculated from the product error components under the same frequency point to form a product feature set. At the same time, the branch mean, wiring mean, and front-to-back difference mean are calculated from the link error components under the same branch and all branch drift components to form a workstation feature set. Finally, the product feature set is combined into product health parameters according to frequency point location, and the workstation feature set is combined into workstation health parameters according to branch location for direct use in subsequent steps.

[0022] S5. Assign product health parameters to impedance deviation results, assign workstation health parameters to workstation deviation results, write branch compensation amounts by branch, write wiring compensation amounts by wiring method, write maintenance instructions by workstation, and output verification conclusions and final verification results. In this specific embodiment, S5 is used to further convert the product health parameters and workstation health parameters obtained in S4 into directly outputtable impedance deviation results, workstation deviation results, branch compensation amounts, wiring compensation amounts, workstation maintenance instructions, and calibration conclusions. The processing logic of this process is as follows: First, the product health parameters are grouped into the corresponding impedance items according to frequency points to obtain impedance deviation results reflecting the impedance measurement status of the inspected product; then, the workstation health parameters are grouped into the corresponding calibration workstations according to impedance branches and wiring methods to obtain workstation deviation results reflecting the status of the calibration workstations; subsequently, based on the existing correspondence between impedance items, impedance branches, and wiring methods, the two types of deviation results are cross-decomposed to form branch compensation amounts, wiring compensation amounts, and workstation maintenance instructions, respectively. Maintenance instructions are issued; finally, based on the impedance deviation results and compensation instruction set, a verification conclusion is generated and combined to form the final verification result. To ensure the execution of this process, the following limitations are made in this specific implementation: Each product health parameter record includes at least the impedance item identifier, frequency point location, frequency point average, frequency point range, and timing increment; Each workstation health parameter record includes at least the workstation identifier, impedance branch identifier, wiring method identifier, branch average, wiring average, and mean difference between the beginning and end; the frequency point deviation aggregate value and the workstation deviation aggregate value are obtained using defined arithmetic operations; the branch compensation amount and the wiring compensation amount are written using the opposite of the corresponding deviation amount; the workstation maintenance instructions are generated according to the composition source of the workstation deviation results; This specific implementation process includes the following steps: In S5-1, firstly, product health parameters are mapped to each impedance item according to frequency points, and the aggregated value of frequency point deviation for each impedance item is calculated to obtain the impedance deviation result. Then, workstation health parameters are mapped to each calibration workstation according to impedance branches and wiring methods, and the aggregated value of workstation deviation for each calibration workstation is calculated to obtain the workstation deviation result. Specifically, each product health parameter record corresponding to all frequency points under the same impedance item is read one by one. First, a single frequency point deviation value is calculated for each frequency point. The single frequency point deviation value is the arithmetic mean of the absolute value of the frequency point mean, the frequency point range, and the absolute values ​​of all time increments for that frequency point. If a frequency point corresponds to multiple time increments, the arithmetic mean of the absolute values ​​of these multiple time increments is taken first, and then included in the calculation of the single frequency point deviation value. Subsequently, all single frequency point deviation values ​​under the same impedance item are added together and divided by the number of frequency points to obtain the aggregated value of frequency point deviation for that impedance item, and this aggregated value is written as the impedance deviation result for that impedance item. For workstation health parameters... First, group the data according to workstation identification. Then, within the same workstation, read the average value of the branch, the average value of the wiring, and the average value of the difference between the beginning and end of the circuit according to impedance branches and wiring methods. For each impedance branch, calculate the single-branch workstation deviation value. The single-branch workstation deviation value is the arithmetic mean of the absolute value of the branch's average value, the absolute value of the wiring's average value for the corresponding wiring method, and the absolute value of the difference between the beginning and end of the circuit. Then, add up all the single-branch workstation deviation values ​​for the same workstation and divide by the number of branches to obtain the aggregated workstation deviation value for that workstation. The aggregated value of the station deviation is written as the station deviation result for that station. In practical applications, if there are three frequency points under a certain impedance item, and the deviation values ​​of the single frequency points are 0.020, 0.024, and 0.022 respectively, then the impedance deviation result for that impedance item is the arithmetic mean of the three, which is 0.022. If there are three impedance branches under a certain station, and the station deviation values ​​of the single branches are 0.11, 0.09, and 0.10 respectively, then the station deviation result for that station is 0.10. In S5-2, the impedance deviation results and workstation deviation results are cross-combined, and the sources of each deviation are decomposed according to the correspondence between impedance items, impedance branches, and wiring methods. This generates the branch compensation amount corresponding to each impedance branch, the wiring compensation amount corresponding to each wiring method, and the workstation maintenance instructions corresponding to each verification workstation, outputting a compensation instruction set. Specifically, a correspondence table between impedance items, impedance branches, and wiring methods is first established. This table is directly obtained from the parameter record table generated in S1 when the verification sequence is generated, i.e., which impedance branches correspond to each impedance item and which wiring methods correspond to each impedance branch. All are known; subsequently, for each impedance item, its impedance deviation result is read, and then the workstation deviation result of each impedance branch corresponding to that impedance item is read. The impedance deviation result is then allocated to the corresponding impedance branch and wiring method according to the corresponding table. During decomposition, the branch source value is first calculated for a specific impedance branch. The branch source value is the arithmetic mean of the impedance deviation result of that impedance item and the workstation deviation result of that impedance branch. Then, the negative value of the branch source value is taken to obtain the branch compensation amount of that impedance branch. For a specific wiring method, the wiring mean value corresponding to that wiring method is first read, and then... The inverse of the wiring average is used to obtain the wiring compensation amount for this wiring method. For station maintenance instructions, the absolute values ​​of the branch average, the absolute values ​​of the wiring average for each wiring method, and the absolute values ​​of the mean difference between the two methods are compared. If the absolute value of the branch average is the largest, a branch maintenance instruction is generated; if the absolute value of the wiring average is the largest, a wiring maintenance instruction is generated; if the absolute value of the mean difference between the two methods is the largest, a drift maintenance instruction is generated; if two or three items are equally largest, the corresponding multiple maintenance instructions are written simultaneously at the same station. The resulting branch compensation amount and wiring compensation amount are calculated accordingly. The compensation instruction set is composed of the compensation amount and the workstation maintenance instruction. In practical applications, if the impedance deviation of a certain impedance item is 0.022 and the workstation deviation of the corresponding impedance branch is 0.10, then the branch source value of the impedance branch is 0.061, and the corresponding branch compensation amount is -0.061. If the wiring average value corresponding to a certain wiring method is 0.13, then the wiring compensation amount of the wiring method is -0.13. If the absolute value of the wiring average value under a certain workstation is greater than the absolute value of the branch average value and the absolute value of the difference between the beginning and end values, then a wiring maintenance instruction is written for that workstation. In S5-3, the impedance deviation results and compensation instruction set are used to generate verification conclusions according to the impedance items. The verification conclusions, branch compensation amounts, wiring compensation amounts, and workstation maintenance instructions are then combined into the final verification result. Specifically, for each impedance item, the impedance deviation result is first read, followed by the reading of all branch compensation amounts, wiring compensation amounts, and workstation maintenance instructions corresponding to that impedance item. Then, verification conclusions are generated according to fixed rules. The fixed rules are: when all workstation maintenance instructions corresponding to the impedance item are empty, the verification conclusion is written as qualified; when the impedance item corresponds to a branch compensation amount or wiring compensation amount, and there is at least one workstation maintenance instruction, the verification conclusion is written as re-inspection after compensation; when the impedance item corresponds to no compensation amount but has a workstation maintenance instruction, the verification conclusion is written as workstation repair; when the impedance item corresponds to... When the impedance deviation result is zero and all compensation amounts are zero, the verification conclusion is written as "directly qualified". After the above verification conclusion is generated, it is combined with the corresponding branch compensation amount, wiring compensation amount and workstation maintenance instruction according to the impedance items to form the final verification result. To ensure subsequent recall and traceability, the final verification result shall include at least the inspected product identification, impedance item identification, impedance deviation result, workstation deviation result, verification conclusion, branch compensation amount, wiring compensation amount and workstation maintenance instruction. In practical application, if the impedance deviation result of a certain impedance item is not zero, and at the same time, a certain impedance branch corresponding to the item has a branch compensation amount, a certain wiring method has a wiring compensation amount, and a wiring maintenance instruction is written at the corresponding workstation, then the verification conclusion of the impedance item shall be written as "re-inspection after compensation", and it shall be written together with the corresponding compensation amount and maintenance instruction into the final verification result. Through the above processing, S5 transforms product health parameters and workstation health parameters into output results with clear physical orientation and specific execution purpose, establishing a one-to-one correspondence between the impedance deviation on the inspected product side, the workstation deviation on the calibration workstation side, the compensation amount on the impedance branch side, the compensation amount on the wiring method side, and the maintenance actions on the workstation side. Simultaneously, by using fixed calculation rules such as writing the frequency deviation aggregation value, workstation deviation aggregation value, branch source value, and negative number compensation, the source of each result's value and calculation path are clarified, avoiding problems such as the inability to decompose the deviation source, the inability to match the compensation object, or the inability to trigger maintenance actions. In practical applications: after an inspected product completes S4 processing, it is first processed by frequency point... Product health parameters are aggregated to corresponding impedance items, and the impedance deviation results for each impedance item are calculated. Then, the workstation health parameters are aggregated to corresponding workstations according to impedance branches and wiring methods, and the workstation deviation results for each verification workstation are calculated. Subsequently, based on the parameter correspondence table in S1, the impedance deviation results and workstation deviation results are decomposed into specific impedance branches and specific wiring methods, and branch compensation amounts and wiring compensation amounts are written respectively. The corresponding workstation maintenance instructions are written according to the dominant source of the workstation deviation results. Finally, the impedance deviation results and compensation instruction sets are combined according to impedance items to generate verification conclusions, and the final verification results containing verification conclusions, branch compensation amounts, wiring compensation amounts, and workstation maintenance instructions are output.

[0023] Furthermore, it also includes an automated verification device for intelligent measurement products, the device comprising: The sequence arrangement module reads the impedance items, frequency points, ranges, and branch nominal values ​​of the product under test, generates a verification sequence according to the combination of frequency points and impedance branches, and the two wiring switches corresponding to the same impedance branch, and outputs excitation parameters, branch parameters, and wiring parameters. The parameter execution module controls the verification device to connect the impedance branch and switch the wiring mode according to the excitation parameters, branch parameters and wiring parameters, drives the tested product to perform impedance measurement, collects the measured value, measurement time and branch identification, and outputs the raw data. The error decoupling module pairs the raw data according to the same frequency point, the same impedance branch and different wiring methods, calculates the difference between two measurements to obtain the wiring difference value, takes the arithmetic mean of the wiring difference value according to the same frequency point to obtain the link error component, calculates the standard deviation of the difference between the measured value and the nominal value according to the same impedance branch to obtain the product error component, and obtains the branch drift component according to the difference between the first and last measurements of the same impedance branch, and outputs the error component. The health assessment module sorts error components by frequency and measurement time, calculates the arithmetic mean, range, and adjacent difference of product error components by frequency, calculates the arithmetic mean of link error components by branch and wiring method, calculates the arithmetic mean of the first and last difference of branch drift components, and outputs product health parameters and workstation health parameters. The results decision module maps product health parameters to impedance deviation results, and workstation health parameters to workstation deviation results. It writes branch compensation amounts by branch, wiring compensation amounts by wiring method, and maintenance instructions by workstation, and outputs the verification conclusion and final verification result.

[0024] Working principle: First, the impedance verification process is broken down into an executable, traceable, and decomposable link. Then, errors from different sources are extracted from this link. First, the impedance item, frequency, range, and reference branch nominal value of the product under test are read, generating a verification sequence with two wiring switches, and simultaneously generating excitation parameters, branch parameters, and wiring parameters. Then, the verification device is driven sequentially according to this sequence to complete the impedance measurement, forming raw data containing frequency, branch, wiring method, measured value, and measurement time. Next, two types of calculations are performed on the raw data: one is the difference between the same frequency, the same branch, and different wiring methods, used to extract the link error; the other... It measures the deviation between the measured value and the nominal value of the branch, as well as the changes from the beginning to the end, to extract product error and branch drift. Based on this, it calculates the mean, range, standard deviation, and beginning-to-end difference according to frequency point, branch, and timing, respectively, to form product health parameters and workstation health parameters. Finally, it merges the product health parameters into impedance deviation results and the workstation health parameters into workstation deviation results, and further generates branch compensation, wiring compensation, workstation maintenance instructions, and final verification conclusions. In other words, this solution does not just look at whether a single measurement result is out of tolerance, but transforms the verification process itself into an analytical process that can distinguish between product error, link error, and workstation drift. For example, when automatically calibrating a batch of intelligent measurement terminals with resistance, capacitance, and inductance measurement functions on a production line, the system first selects a specific impedance item, such as resistance measurement. Then, it calls different reference branches at multiple frequency points and performs two measurements on each branch with different wiring methods. If a product shows significantly different measurement results at the same frequency point, but a similar change occurs after switching to another branch, the system will determine that this is more likely a wiring link or workstation problem. If the value measured by a certain branch continuously deviates at the beginning and end of the entire calibration sequence, the system will attribute this change to the branch. Path drift; if the deviation between the measured value and the nominal value of the branch exists stably at multiple frequency points, it is more likely to be a measurement link error of the product under test itself; in this way, the system will not only give "qualified" or "unqualified", but will also tell the user: which impedance item is off, which branch needs compensation, which wiring method needs correction, and which station needs maintenance; in a real production line, this is equivalent to not only detecting the problem, but also locating whether the problem is more likely to be in the product itself, the wiring link, or the calibration station, thereby reducing re-inspection, reducing misjudgment, and facilitating subsequent maintenance and health management.

[0025] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automated verification method for intelligent measurement products, characterized in that, include: S1. Read the impedance item, frequency point, range and branch nominal value of the product under test, generate the verification sequence according to the combination of frequency point and impedance branch, and the two wiring switches corresponding to the same impedance branch, and output the excitation parameters, branch parameters and wiring parameters. S2. According to the excitation parameters, branch parameters and wiring parameters, control the verification device to connect the impedance branch and switch the wiring mode, drive the tested product to perform impedance measurement, collect the measured value, measurement time and branch identification, and output the raw data. S3. Pair the raw data according to the same frequency point, the same impedance branch and different wiring methods, calculate the difference between the two measurements to obtain the wiring difference value, take the arithmetic mean of the wiring difference value according to the same frequency point to obtain the link error component, calculate the standard deviation of the difference between the measured value and the nominal value according to the same impedance branch to obtain the product error component, and obtain the branch drift component according to the difference between the first and last measurements of the same impedance branch. Output the error component. S4. Sort the error components by frequency and measurement time, calculate the arithmetic mean, range and adjacent difference of the product error components by frequency, calculate the arithmetic mean of the link error components by branch and wiring method, calculate the arithmetic mean of the first and last difference of the branch drift components, and output the product health parameters and workstation health parameters. S5. Map the product health parameters to impedance deviation results, map the workstation health parameters to workstation deviation results, write the branch compensation amount by branch, write the wiring compensation amount by wiring method, write the maintenance instructions by workstation, and output the verification conclusion and final verification result.

2. The automated verification method for intelligent measurement products according to claim 1, characterized in that: S1 includes: S1-1: Read the nominal values ​​of each frequency point, each range and each impedance branch corresponding to the impedance item, generate the first test unit according to the pairwise combination of frequency point and impedance branch, and output the first test unit set. S1-2. Configure two wiring switches for each first test unit in the first test unit set, and generate a second test unit set by alternating different impedance branches under the same frequency point and arranging the wiring switches of the same impedance branch at intervals, and output the verification sequence. S1-3. Based on each second test unit in the verification sequence, extract the corresponding frequency point, range, impedance branch and wiring switching sequence, and generate excitation parameters, branch switching parameters and wiring switching parameters that correspond one-to-one with the verification sequence.

3. The automated verification method for intelligent measurement products according to claim 2, characterized in that: S2 includes: S2-1. According to the excitation parameters, branch switching parameters and wiring switching parameters, control the verification device to connect the corresponding impedance branches and switch the corresponding wiring methods in sequence, drive the tested product to perform impedance measurement one by one, and output the measurement value corresponding to each measurement. S2-2. While outputting each measurement value, read the branch identifier of the corresponding impedance branch and the measurement time of the corresponding measurement value, and bind the measurement value, branch identifier and measurement time according to the same impedance measurement, and output a single measurement record set. S2-3. Arrange each single measurement record set sequentially according to the execution order in the verification sequence to generate raw data.

4. The automated verification method for intelligent measurement products according to claim 3, characterized in that: S3 includes: S3-1. Pair the original data according to the same frequency point, the same impedance branch and different wiring methods, calculate the difference between the two measurements in the group to obtain the wiring difference value, and write the wiring difference value, frequency point, impedance branch, wiring method and measurement time into the difference record set. S3-2. Construct frequency point windows for the differential record set according to the same frequency point, calculate the mean, range and sign inversion number of the wiring differential values ​​in each frequency point window, use the wiring consistency checker to perform cross-checking on the mean, range and sign inversion number, output the link token and write the corresponding frequency point into the link candidate set.

5. The automated verification method for intelligent measurement products according to claim 4, characterized in that: S3 further includes: S3-3. Construct branch windows for the original data according to the same impedance branches, calculate the dispersion, timing slope and front-to-back difference of the difference between the measured value and the nominal value in each branch window, and use the branch stability checker to cross-check the dispersion, timing slope and front-to-back difference, output the branch token and write the corresponding impedance branch into the branch candidate set. S3-4. Establish a timing propagation chain for the link candidate set and tributary candidate set at the same measurement time. Calculate the differential aggregation value across impedance tributaries at the same frequency point, the drift aggregation value across frequency points of the same impedance tributary, and the collision count at the same measurement time. Then, use the collision resolver to update the link token and tributary token in the manner of prioritizing the locking of frequency point paths with the link token, prioritizing the locking of tributary paths with the tributary token, and triggering backoff reassembly when the collision count reaches the backoff condition. Write the results to the token status table.

6. The automated verification method for intelligent measurement products according to claim 5, characterized in that: S3 further includes: S3-5. The link error component is obtained by summing and averaging the wiring differential values ​​of the locked frequency path in the token status table according to the frequency point. The product error component is obtained by calculating the dispersion of the difference between the measured value and the nominal value of the locked branch path. The branch drift component is obtained by averaging the difference between the two measurements of the locked branch path. The link error component, product error component and branch drift component are written into the error component set. S3-6. Perform cross-view consistency verification on the error component set. When the link error component and the branch drift component cross the corresponding constraint interval at the same measurement time, a re-detection token is generated and the re-detection flag is written back. When the link error component, product error component and branch drift component satisfy the corresponding constraint interval, an acknowledgment token is generated and the error component is output.

7. The automated verification method for intelligent measurement products according to claim 6, characterized in that: S4 includes: S4-1. Rearrange the error components according to frequency point order and measurement time order to construct frequency point sequence, branch sequence and time sequence, and output sequence dataset; S4-2. Calculate the mean and range of frequency points for the product error components in the sequence dataset according to the frequency point sequence, and calculate the time-series increment by calculating the difference between two adjacent product error components according to the time-series sequence, and output the product feature set.

8. The automated verification method for intelligent measurement products according to claim 7, characterized in that: S4 further includes: S4-3. Calculate the branch mean value of the link error component in the sequence dataset according to the branch sequence and the wiring mean value according to the wiring method. Calculate the mean value of the difference between the branch drift component and the time sequence according to the time sequence. Output the workstation feature set. S4-4. Combine the frequency point average, frequency point range, and timing increment of the product feature set according to the frequency point location to generate product health parameters, and combine the branch average, wiring average, and front-to-back difference average of the workstation feature set according to the branch location to generate workstation health parameters.

9. The automated verification method for intelligent measurement products according to claim 8, characterized in that: S5 includes: S5-1. Map the product health parameters to each impedance item according to the frequency point, calculate the aggregate value of the frequency point deviation of each impedance item to obtain the impedance deviation result, and map the station health parameters to each verification station according to the impedance branch and wiring method, calculate the aggregate value of the station deviation of each verification station to obtain the station deviation result. S5-2. Cross-combine the impedance deviation results and the workstation deviation results, and decompose each source of deviation according to the correspondence between impedance items, impedance branches and wiring methods. Generate the branch compensation amount corresponding to each impedance branch, the wiring compensation amount corresponding to each wiring method and the workstation maintenance instruction corresponding to each calibration workstation, and output the compensation instruction set. S5-3. Generate verification conclusions by matching the impedance deviation results and compensation instruction set with the impedance items, and combine the verification conclusions, branch compensation amounts, wiring compensation amounts, and workstation maintenance instructions into the final verification result.

10. An automated verification device for intelligent measurement products, used to implement the automated verification method for intelligent measurement products according to any one of claims 1-9, characterized in that, include: The sequence arrangement module reads the impedance items, frequency points, ranges, and branch nominal values ​​of the product under test, generates a verification sequence according to the combination of frequency points and impedance branches, and the two wiring switches corresponding to the same impedance branch, and outputs excitation parameters, branch parameters, and wiring parameters. The parameter execution module controls the verification device to connect the impedance branch and switch the wiring mode according to the excitation parameters, branch parameters and wiring parameters, drives the tested product to perform impedance measurement, collects the measured value, measurement time and branch identification, and outputs the raw data. The error decoupling module pairs the raw data according to the same frequency point, the same impedance branch and different wiring methods, calculates the difference between two measurements to obtain the wiring difference value, takes the arithmetic mean of the wiring difference value according to the same frequency point to obtain the link error component, calculates the standard deviation of the difference between the measured value and the nominal value according to the same impedance branch to obtain the product error component, and obtains the branch drift component according to the difference between the first and last measurements of the same impedance branch, and outputs the error component. The health assessment module sorts error components by frequency and measurement time, calculates the arithmetic mean, range, and adjacent difference of product error components by frequency, calculates the arithmetic mean of link error components by branch and wiring method, calculates the arithmetic mean of the first and last difference of branch drift components, and outputs product health parameters and workstation health parameters. The results decision module maps product health parameters to impedance deviation results, and workstation health parameters to workstation deviation results. It writes branch compensation amounts by branch, wiring compensation amounts by wiring method, and maintenance instructions by workstation, and outputs the verification conclusion and final verification result.

Images