CN118818273A - Single-turn magnetic encoder automated test fixture - Google Patents

Abstract

The application relates to an automatic testing tool for a single-turn magnetic encoder. It comprises the following steps: starting the upper computer; selecting an editor main board to be tested and corresponding firmware data; the upper computer communicates with the lower computer through RS-485 to send a burning instruction and the firmware data; the lower computer selects a correct encoder main board through an FPGA chip, and downloads the firmware data to an editor chip of the editor main board through an ISP download protocol; after the burning is finished, the lower computer reads the encoder data according to the multi-Moire encoder protocol to check the working state of the editor main board; collecting test data through the lower computer and sending the test data to the upper computer; and the upper computer analyzes the test data to obtain a test result. Therefore, the efficiency of processing a large amount of test data is improved, complex relations among variables can be comprehensively captured, deeper test result understanding is provided, and the test data processing is more intelligent.

Description

Single-turn magnetic encoder automated test fixture

Technical Field

The present application relates to the technical field of single-turn magnetic encoders, and in particular to an automated testing tool for single-turn magnetic encoders.

Background Art

The single-turn magnetic encoder used in servo motors is a high-precision position sensor that can provide accurate angle measurement within a single turn. It is widely used in situations that require high-precision positioning control, such as industrial robots, precision machining equipment, etc.

At present, the firmware burning and functional testing of the servo motor single-turn encoder motherboard still rely on developers to use the chip manufacturer's own tools to manually burn and test each board one by one. However, since humans can only handle a single motherboard at a time, this method cannot meet the high throughput requirements of modern manufacturing. In a high-paced production environment, this inefficient manual operation method greatly prolongs the production cycle and seriously affects the overall efficiency of the production line. In addition, inconsistencies in manual operations may result in incomplete firmware burning or omissions of key steps during the testing process. These omissions can cause unstable or unqualified product performance. For example, incomplete firmware burning may cause the encoder to malfunction under certain specific conditions, while omissions during the testing process may result in failure to detect potential problems.

Therefore, an optimized single-turn magnetic encoder test fixture is desired.

Summary of the invention

This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

In the first aspect, the present application provides a single-turn magnetic encoder automated testing tool, the tool comprising: a host computer and a slave computer, wherein the working process of the host computer and the slave computer comprises: starting the host computer; selecting an editor mainboard to be tested and corresponding firmware data; the host computer communicates with the slave computer via RS-485 to send burning instructions and the firmware data; the slave computer selects the correct encoder mainboard through an FPGA chip, and downloads the firmware data to the editor chip of the editor mainboard through an ISP download protocol; after the burning is completed, the slave computer reads the encoder data according to the Tamagawa encoder protocol to check the working status of the editor mainboard; the test data is collected by the slave computer and sent to the host computer; the host computer analyzes the test data to obtain test results.

Optionally, the host computer analyzes the test data to obtain test results, including: using a test embedding coding matrix to embed code each test result item in the test data to obtain a set of test result item embedding coding vectors; inputting the set of test result item embedding coding vectors into a global semantic association encoder between test results based on a converter structure to obtain a global semantic coding feature vector of the test results; inputting the set of test result item embedding coding vectors into a local semantic association encoder between test results to obtain a local semantic coding feature vector of the test results; inputting the test result global semantic coding feature vector and the test result local semantic coding feature vector into a significant fusion network based on feature principal component query matching to obtain a test result multi-scale sparse fusion representation vector; and obtaining the test result based on the test result multi-scale sparse fusion representation vector.

Optionally, the set of test result item embedding coding vectors is input into a local semantic association encoder between test results to obtain a local semantic coding feature vector of the test result, including: inputting the set of test result item embedding coding vectors into a local semantic association encoder between test results based on a 1D-CNN model to obtain the local semantic coding feature vector of the test result.

Optionally, the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result are input into a significant fusion network based on feature principal component query matching to obtain a multi-scale sparse fusion representation vector of the test result, including: standardizing the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result to obtain a standardized global semantic coding feature vector of the test result and a standardized local semantic coding feature vector of the test result; calculating the sample covariance matrix of the standardized global semantic coding feature vector of the test result and the standardized local semantic coding feature vector of the test result to obtain a global semantic sample covariance matrix of the test result and a local semantic sample covariance matrix of the test result; performing feature vector extraction based on matrix decomposition on the global semantic sample covariance matrix of the test result and the local semantic sample covariance matrix of the test result to obtain a global semantic principal component of the test result. The method comprises the following steps: inputting the set of global semantic principal component feature vectors of the test results and the set of local semantic principal component feature vectors of the test results into a maximum approximate query matching network to obtain a set of best matching pairs of the global semantic principal component feature vectors of the test results and the local semantic principal component feature vectors of the test results; inputting the best matching pairs of the global semantic principal component feature vectors of the test results and the local semantic principal component feature vectors of the test results in the set of best matching pairs of the global semantic principal component feature vectors of the test results and the local semantic principal component feature vectors of the test results into a semantic fine-grained gated joint module to obtain a set of global semantic principal component fusion feature vectors of the test results and local semantic principal component fusion feature vectors of the test results; cascading the set of global semantic principal component fusion feature vectors of the test results and local semantic principal component fusion feature vectors of the test results to obtain a multi-scale sparse fusion representation vector of the test result.

Optionally, the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result are standardized to obtain a standardized global semantic coding feature vector of the test result and a standardized local semantic coding feature vector of the test result, including: respectively calculating the mean and standard deviation of the global semantic coding feature vector of the test result to obtain the mean of the global semantic coding feature of the test result and the standard deviation of the global semantic coding feature of the test result; subtracting the global semantic coding feature vector of the test result from the mean of the global semantic coding feature of the test result by position, and then dividing the calculated global semantic offset vector of the test result by the standard deviation of the global semantic coding feature of the test result to obtain the standardized global semantic coding feature vector of the test result; respectively calculating the mean and standard deviation of the local semantic coding feature vector of the test result to obtain the mean of the local semantic coding feature of the test result and the standard deviation of the local semantic coding feature of the test result; subtracting the local semantic coding feature vector of the test result from the mean of the local semantic coding feature of the test result by position, and then dividing the calculated local semantic offset vector of the test result by the standard deviation of the local semantic coding feature of the test result to obtain the standardized local semantic coding feature vector of the test result.

Optionally, the sample covariance matrix of the standardized test result global semantic coding feature vector and the standardized test result local semantic coding feature vector is calculated to obtain a test result global semantic sample covariance matrix and a test result local semantic sample covariance matrix, including: multiplying the transpose vector of the standardized test result global semantic coding feature vector by the standardized test result global semantic coding feature vector, and then dividing the obtained standardized test result global semantic association matrix by a value obtained by subtracting one from the length of the standardized test result global semantic coding feature vector by position to obtain the test result global semantic sample covariance matrix; multiplying the transpose vector of the standardized test result local semantic coding feature vector by the standardized test result local semantic coding feature vector, and then dividing the obtained standardized test result local semantic association matrix by a value obtained by subtracting one from the length of the standardized test result local semantic coding feature vector by position to obtain the test result local semantic sample covariance matrix.

Optionally, the set of the test result global semantic principal component feature vectors and the set of the test result local semantic principal component feature vectors are input into a maximum approximate query matching network to obtain a set of best matching pairs of the test result global semantic principal component feature vectors and the test result local semantic principal component feature vectors, including: extracting predetermined test result global semantic principal component feature vectors from the set of test result global semantic principal component feature vectors; calculating the cosine similarity between the predetermined test result global semantic principal component feature vector and each test result local semantic principal component feature vector in the set of test result local semantic principal component feature vectors to obtain a set of matching query similarities; and taking the test result local semantic principal component feature vector corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined test result global semantic principal component feature vector as the best matching pair of the predetermined test result global semantic principal component feature vector and the test result local semantic principal component feature vector.

Optionally, each best matching pair of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result in the set of the best matching pairs of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result is input into the semantic fine-grained gating joint module to obtain a set of test result global-test result local semantic principal component fusion feature vectors, including: respectively calculating the position difference, position dot multiplication and position addition between the best matching pairs of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result to obtain the test result global-local semantic principal component. The invention discloses a method for preparing a test result global-local semantic principal component multi-dimensional fusion vector of the test result, a test result global-local semantic principal component dot product vector and a test result global-local semantic principal component sum vector; the test result global-local semantic principal component difference vector, the test result global-local semantic principal component dot product vector and the test result global-local semantic principal component sum vector are cascaded and then one-dimensionally convolutionally encoded to obtain a test result global-local semantic principal component multi-dimensional fusion vector; the test result global-local semantic principal component multi-dimensional fusion vector is subjected to a local window-based maximum pooling process to obtain a test result global-test result local semantic principal component fusion feature vector.

Optionally, obtaining the test result based on the test result multi-scale sparse fusion representation vector includes: inputting the test result multi-scale sparse fusion representation vector into a classifier-based test result generator to obtain the test result.

The above technical solution is adopted, by starting the host computer; selecting the editor mainboard to be tested and the corresponding firmware data; the host computer communicates with the lower computer through RS-485 to send the burning instruction and the firmware data; the lower computer selects the correct encoder mainboard through the FPGA chip, and downloads the firmware data to the editor chip of the editor mainboard through the ISP download protocol; after the burning is completed, the lower computer reads the encoder data according to the Tamagawa encoder protocol to check the working status of the editor mainboard; the test data is collected by the lower computer and sent to the host computer; the host computer analyzes the test data to obtain the test results. In this way, the efficiency of processing a large amount of test data is improved, and at the same time, the complex relationship between variables can be fully captured, providing a deeper understanding of the test results, making the test data processing more intelligent.

Other features and advantages of the present application will be described in detail in the subsequent detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

In the accompanying drawings: FIG. 1 is a flow chart of a single-turn magnetic encoder automated testing tool according to an exemplary embodiment.

FIG. 2 is a flow chart of step S107 of a single-turn magnetic encoder automated testing tool according to the embodiment shown in FIG. 1 .

Fig. 3 is a block diagram of an electronic device according to an exemplary embodiment.

Fig. 4 is a diagram showing an application scenario of a single-turn magnetic encoder automated testing tool according to an exemplary embodiment.

DETAILED DESCRIPTION

The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and/or performed in parallel. In addition, the method implementation may include additional steps and/or omit the steps shown. The scope of the present application is not limited in this respect.

The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

At present, the burning and testing of the servo motor single-turn encoder motherboard is still done by developers using the chip manufacturer's own tools to manually burn firmware and test each board. This method is inefficient and prone to errors, and is obviously not suitable for the production environment. Now we hope to design a new fast test tool to meet the needs of automated firmware burning and testing in the production environment. The new single-turn encoder fast test tool consists of a PC-side host computer and a slave control board program; the host computer provides a human-computer interaction interface, and the slave computer completes the functions of selecting the encoder motherboard, downloading firmware, and testing. The host and slave computers communicate via RS-485.

The specific implementation of the present application is described in detail below in conjunction with the accompanying drawings. Figure 1 is a flow chart of a single-turn magnetic encoder automated testing tooling according to an exemplary embodiment. As shown in Figure 1, the tooling includes: a host computer and a slave computer, wherein the working process of the host computer and the slave computer includes: step S101, starting the host computer.

Step S102: Select the editor mainboard to be tested and the corresponding firmware data.

Step S103: the upper computer communicates with the lower computer via RS-485 to send a burning instruction and the firmware data.

Step S104: the lower computer selects the correct encoder mainboard through the FPGA chip, and downloads the firmware data to the editor chip of the editor mainboard through the ISP download protocol.

Step S105, after the burning is completed, the lower computer reads the encoder data according to the Tamagawa encoder protocol to check the working status of the editor mainboard.

Step S106: Collect test data through the lower computer and send it to the upper computer.

Step S107: The host computer analyzes the test data to obtain a test result.

Among them, all logical processes such as burning and testing are done by the upper computer. The lower computer is only a communication transfer station. The lower computer needs to have two serial ports, one serial port connects the PC serial port and the upper computer communication, and the other serial port connects the FPGA chip (responsible for selecting the encoder motherboard). The lower computer needs to implement the modbus slave station to receive the instructions and data sent by the upper computer; the lower computer needs to implement the function of selecting the encoder motherboard by communicating with the FPGA. The lower computer needs to implement the ISP download protocol to download the firmware program transmitted from the PC upper computer to the encoder chip; the lower computer needs to read the motor encoder data according to the Tamagawa encoder protocol to test whether the encoder is working properly. The PC upper computer needs to have the serial port selection and setting function, and display the serial port communication status. The PC upper computer needs to have the function of selecting the encoder test board. The PC upper computer needs to select the firmware to be burned and automatically record it to the configuration file. The file path location of the last firmware burned can be automatically remembered when it is opened next time. The PC host computer needs to choose whether to proceed to the next test according to the firmware burning status sent back by the slave computer; the PC host computer needs to display the burning and test status of the 10 encoder mainboards to be tested. After all encoder mainboard tests are completed, the PC host computer needs to display the final test result PASS/NG.

Therefore, the single-turn magnetic encoder automated test fixture of the present application can process multiple motherboards at the same time, significantly improving production throughput. And through the coordinated work of the host computer and the slave computer, batch burning and testing can be achieved, greatly shortening the production cycle and meeting the demand for high efficiency in modern manufacturing.

It should be understood that test data analysis can help identify any anomalies or defects in the encoder motherboard during the burning and testing process, so as to ensure that each product meets the predetermined quality standards. However, traditional analysis of test data is usually completed manually or through simple analysis methods. Specifically, traditional methods usually rely on manual inspection and recording of test data, and compare standard values one by one. This practice is time-consuming and labor-intensive, especially when dealing with large amounts of data. It is very inefficient and may also lead to human errors. In addition, when the test data involves multiple variables and dimensions, it is difficult for traditional analysis methods to fully capture the complex relationships between these variables, which may lead to a lack of in-depth understanding of the test results, miss important correlations, and thus reduce the accuracy of the test results.

Therefore, in the process of using the host computer to analyze the test data to obtain the test results, the technical concept of the present application is to embed the coding of each test result item in the test data by using an artificial intelligence-based data processing and analysis algorithm, and respectively perform global semantic association and local semantic association on the embedded coded test results, so as to automatically obtain the test results according to the significant fusion features between the global semantic features and the local semantic features of the test results. In this way, the efficiency of processing a large amount of test data is improved, and the time and error rate of manual operation are reduced. At the same time, it is possible to fully capture the complex relationship between variables, provide a deeper understanding of the test results, and make the test data processing more intelligent.

Fig. 2 is a flow chart of step S107 of a single-turn magnetic encoder automated test fixture according to the embodiment shown in Fig. 1. As shown in Fig. 2, step S107, the host computer analyzes the test data to obtain the test result, including: step S1071, using the test embedding coding matrix to embed the test result items in the test data to obtain a set of test result item embedding coding vectors.

Step S1072: input the set of test result item embedding coding vectors into a test result global semantic association encoder based on a converter structure to obtain a test result global semantic coding feature vector.

Step S1073: input the set of test result item embedding coding vectors into the test result local semantic association encoder to obtain the test result local semantic coding feature vector.

Step S1074: input the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result into a saliency fusion network based on feature principal component query matching to obtain a multi-scale sparse fusion representation vector of the test result.

Step S1075: Obtain the test result based on the multi-scale sparse fusion representation vector of the test result.

Specifically, considering that each test result item in the test data usually contains a large number of data dimensions, if the data are directly processed, the computational complexity is large, and each data item contains key data item information. Based on this, in the technical solution of the present application, a test embedding coding matrix is used to embed code each test result item in the test data to obtain a set of test result item embedding coding vectors, that is, the test embedding coding matrix can map high-dimensional data to a low-dimensional space, reduce computational complexity, while retaining the intrinsic meaning and semantic information in the data, providing richer data support for subsequent data processing.

Next, considering that there are semantic correlations and influences between the embedded coding vectors of each test result item in the set of embedded coding vectors of the test result item, and the core of the converter structure is the self-attention mechanism, which can capture the semantic relationship between different elements in the input sequence, which is crucial for understanding the global semantic correlation between the test result items. Based on this, in the technical solution of the present application, the set of embedded coding vectors of the test result item is input into the global semantic correlation encoder between test results based on the converter structure to capture and mine the semantic correlation between each vector to obtain the global semantic coding feature vector of the test result.

Then, considering that the set of embedded coding vectors of the test result items also has local semantic-based correlations in the local range, and 1D-CNN can capture local features in sequence data through convolutional layers, this is very useful for identifying local semantic correlations between test result items. Therefore, in the technical solution of the present application, the set of embedded coding vectors of the test result items is input into the local semantic correlation encoder between test results based on the 1D-CNN model to obtain the local semantic coding feature vector of the test results. That is, the 1D-CNN model uses a convolutional layer to process the input set of embedded coding vectors, and the convolutional layer moves on the sequence through a sliding window (convolution kernel) to extract local semantic features in the embedded coding vector.

In one embodiment of the present application, the set of test result item embedding coding vectors is input into a local semantic association encoder between test results to obtain a local semantic coding feature vector of the test result, including: inputting the set of test result item embedding coding vectors into a local semantic association encoder between test results based on a 1D-CNN model to obtain the local semantic coding feature vector of the test result.

Further, considering that the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result respectively capture information of different levels of the test result. And there are significant semantic features between the two information. In order to more comprehensively capture the complexity and multi-dimensional information of the test data, so as to make accurate judgments on the test results, in the technical solution of the present application, the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result are input into a significant fusion network based on feature principal component query matching to obtain a multi-scale sparse fusion representation vector of the test result. In particular, the significant fusion network based on feature principal component query matching uses feature sparsification, principal component analysis and key significant feature integration technology to create a feature extraction and fusion method, aiming to construct a sparse and key association mapping between feature vectors. In detail, first, the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result are standardized to eliminate the influence of different feature vectors due to different dimensions, so that the features are comparable, and the global semantic encoding feature vector of the standardized test result and the local semantic encoding feature vector of the standardized test result are obtained. Next, the covariance matrix of the standardized test result global semantic coding feature vector and the standardized test result local semantic coding feature vector is calculated to obtain the test result global semantic sample covariance matrix and the test result local semantic sample covariance matrix. Then, the covariance matrix is subjected to feature vector extraction based on matrix decomposition to obtain a set of test result global semantic principal component feature vectors and a set of test result local semantic principal component feature vectors. In particular, the feature vector extraction based on matrix decomposition utilizes the principal component analysis algorithm to perform sparse processing on the standardized vectors, which can extract the principal component features that best represent the semantics of the test results from both the global and local levels while reducing the data dimension. Next, the two sets of vectors are subjected to maximum approximate query matching based on the feature vectors, and the query matching of the most relevant semantics is performed by calculating the cosine similarity between each feature, thereby finding the best matching pair. Then, the gating mechanism is used to perform fine-grained semantic joint processing on the best matching pairs of each group to capture and extract the multi-dimensional principal component correlation information between the global features of the test results and the data features, so that the model can understand and process the semantic information more finely, enhance the model's grasp of the semantic content, and obtain a set of test result global-test result local semantic principal component fusion feature vectors. Finally, the set of principal component fusion feature vectors is cascaded to further integrate and optimize the feature vectors to obtain a feature representation containing rich global and local semantic information of the test results, thereby generating a multi-scale sparse fusion representation vector of the test results.

In one embodiment of the present application, the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result are input into a significant fusion network based on feature principal component query matching to obtain a multi-scale sparse fusion representation vector of the test result, including: standardizing the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result to obtain a standardized global semantic coding feature vector of the test result and a standardized local semantic coding feature vector of the test result; calculating the sample covariance matrix of the standardized global semantic coding feature vector of the test result and the standardized local semantic coding feature vector of the test result to obtain a global semantic sample covariance matrix of the test result and a local semantic sample covariance matrix of the test result; performing matrix decomposition-based feature vector extraction on the global semantic sample covariance matrix of the test result and the local semantic sample covariance matrix of the test result to obtain a global semantic sample covariance matrix of the test result. A set of semantic principal component feature vectors and a set of local semantic principal component feature vectors of test results; inputting the set of global semantic principal component feature vectors of test results and the set of local semantic principal component feature vectors of test results into a maximum approximate query matching network to obtain a set of best matching pairs of global semantic principal component feature vectors of test results and local semantic principal component feature vectors of test results; inputting the best matching pairs of each global semantic principal component feature vector of test results and local semantic principal component feature vectors of test results in the set of best matching pairs of global semantic principal component feature vectors of test results and local semantic principal component feature vectors of test results into a semantic fine-grained gated joint module to obtain a set of global semantic principal component fusion feature vectors of test results-local semantic principal component fusion feature vectors of test results; cascading the set of global semantic principal component fusion feature vectors of test results-local semantic principal component fusion feature vectors of test results to obtain a multi-scale sparse fusion representation vector of the test result.

Further, in one embodiment of the present application, the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result are standardized to obtain a standardized global semantic coding feature vector of the test result and a standardized local semantic coding feature vector of the test result, including: respectively calculating the mean and standard deviation of the global semantic coding feature vector of the test result to obtain the mean of the global semantic coding feature of the test result and the standard deviation of the global semantic coding feature of the test result; after positionally subtracting the global semantic coding feature vector of the test result from the mean of the global semantic coding feature of the test result, the calculated global semantic offset vector of the test result and the standard deviation of the global semantic coding feature of the test result are divided by position to obtain the standardized global semantic coding feature vector of the test result; respectively calculating the mean and standard deviation of the local semantic coding feature vector of the test result to obtain the mean of the local semantic coding feature of the test result and the standard deviation of the local semantic coding feature of the test result; after positionally subtracting the local semantic coding feature vector of the test result from the mean of the local semantic coding feature of the test result, the calculated local semantic offset vector of the test result and the standard deviation of the local semantic coding feature of the test result are divided by position to obtain the standardized local semantic coding feature vector of the test result.

Further, in one embodiment of the present application, the sample covariance matrix of the standardized test result global semantic coding feature vector and the standardized test result local semantic coding feature vector is calculated to obtain a test result global semantic sample covariance matrix and a test result local semantic sample covariance matrix, including: multiplying the transpose vector of the standardized test result global semantic coding feature vector by the standardized test result global semantic coding feature vector, and then dividing the obtained standardized test result global semantic association matrix by the value obtained by subtracting one from the length of the standardized test result global semantic coding feature vector by position to obtain the test result global semantic sample covariance matrix; multiplying the transpose vector of the standardized test result local semantic coding feature vector by the standardized test result local semantic coding feature vector, and then dividing the obtained standardized test result local semantic association matrix by the value obtained by subtracting one from the length of the standardized test result local semantic coding feature vector by position to obtain the test result local semantic sample covariance matrix.

Furthermore, in one embodiment of the present application, the set of the test result global semantic principal component feature vectors and the set of the test result local semantic principal component feature vectors are input into a maximum approximate query matching network to obtain a set of best matching pairs of the test result global semantic principal component feature vectors and the test result local semantic principal component feature vectors, including: extracting predetermined test result global semantic principal component feature vectors from the set of the test result global semantic principal component feature vectors; calculating the cosine similarity between the predetermined test result global semantic principal component feature vector and each test result local semantic principal component feature vector in the set of the test result local semantic principal component feature vectors to obtain a set of matching query similarities; and taking the test result local semantic principal component feature vector corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined test result global semantic principal component feature vector as the best matching pair of the predetermined test result global semantic principal component feature vector and the test result local semantic principal component feature vector.

Furthermore, in one embodiment of the present application, each best matching pair of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result in the set of the best matching pairs of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result is input into the semantic fine-grained gated joint module to obtain a set of test result global-test result local semantic principal component fusion feature vectors, including: respectively calculating the positional difference, positional dot multiplication and positional addition between the best matching pairs of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result to obtain the global semantic principal component feature vector of the test result. -local semantic principal component difference vector, test result global-local semantic principal component dot product vector and test result global-local semantic principal component sum vector; concatenate the test result global-local semantic principal component difference vector, the test result global-local semantic principal component dot product vector and the test result global-local semantic principal component sum vector, and then perform one-dimensional convolution encoding to obtain a test result global-local semantic principal component multi-dimensional fusion vector; perform local window-based maximum pooling processing on the test result global-local semantic principal component multi-dimensional fusion vector to obtain the test result global-test result local semantic principal component fusion feature vector.

Specifically, the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result are input into a significant fusion network based on feature principal component query matching, and processed with the following significant fusion formula to obtain a multi-scale sparse fusion representation vector of the test result; wherein the significant fusion formula is: ;in, and represent the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result respectively, and are respectively the means of the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result, and are the standard deviations of the global semantic coding feature vector and the local semantic coding feature vector of the test result, respectively, and are respectively the global semantic encoding feature vector of the standardized test result and the local semantic encoding feature vector of the standardized test result, and They are and The transposed vector of and are the lengths of the global semantic encoding feature vector of the standardized test result and the local semantic encoding feature vector of the standardized test result, respectively, and are respectively the test result global semantic sample covariance matrix and the test result local semantic sample covariance matrix, and They are the orthogonal matrix of the global semantic principal component of the test result and the orthogonal matrix of the local semantic principal component of the test result. and They are the global semantic diagonal matrix of test results and the local semantic diagonal matrix of test results, The diagonal elements of the matrix are The test results of the global semantic diagonal matrix, are the weight values of the global semantic principal component feature vectors of each test result, The diagonal elements of the matrix are The test results of the local semantic diagonal matrix, are the weight values of the local semantic principal component feature vectors of each test result, and They are and The transposed matrix of is each test result global semantic principal component feature vector in the set of test result global semantic principal component feature vectors, is each local semantic principal component feature vector of the test result in the set of local semantic principal component feature vectors of the test result, To calculate the and The vector inner product between To calculate the one-norm of a vector, To return the maximum value value, is the maximum approximate matching value, , and They are position difference, position dot multiplication and position addition respectively. For cascade processing, is a one-dimensional convolutional coding operation, is the maximum pooling operation, is the first in the set of the test result global-test result local semantic principal component fusion feature vectors The global test result-local test result semantic principal component fusion feature vector, is the number of feature vectors in the set of the test result global-test result local semantic principal component fusion feature vectors, is the multi-scale sparse fusion representation vector of the test result.

Then, the multi-scale sparse fusion representation vector of the test result is input into the test result generator based on the classifier to obtain the test result. That is, the multi-scale sparse fusion representation vector of the test result obtained by significant fusion of the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result is used for classification processing to automatically obtain the test result. In this way, the efficiency of processing a large amount of test data is improved, and the time and error rate of manual operation are reduced. At the same time, it can fully capture the complex relationship between variables, provide a deeper understanding of the test results, and make the test data processing more intelligent.

After all the coding motherboard tests are completed, the PC host computer needs to display the final test results "PASS/NG", where PASS means that the motherboard has passed all the test standards and requirements, which means that the functions, performance and quality of the motherboard meet the design specifications and manufacturing standards. When the PC host computer displays "PASS", it means that the motherboard can enter the next production stage, such as packaging and shipping. NG means that the motherboard has not passed the test and there are some non-compliant issues, which may be due to hardware defects, software errors or other manufacturing problems. When "NG" is displayed, the motherboard needs further inspection and repair, or may need to be reworked or scrapped.

In one embodiment of the present application, obtaining the test result based on the multi-scale sparse fusion representation vector of the test result includes: inputting the multi-scale sparse fusion representation vector of the test result into a classifier-based test result generator to obtain the test result.

Preferably, considering that the global semantic encoding feature vector of the test result and the local semantic encoding feature vector of the test result respectively represent the associated features based on the embedded representation context and the local associated features of the embedded representation of each test result item, when performing significant fusion based on feature principal component query matching, the feature principal component query matching significance difference caused by the difference in the embedded representation association dimension will lead to the embedded semantic aggregation fine-grained alignment conflict, thereby causing the loss of aggregation key information and affecting the expression effect of the multi-scale sparse fusion representation vector of the test result.

Based on this, in the preferred example, when the test result multiscale sparse fusion representation vector is input into the classifier-based test result generator to obtain the test result, the test result multiscale sparse fusion representation vector is optimized. Specifically, the optimization process includes the following steps: clustering all eigenvalues of the test result multiscale sparse fusion representation vector based on the distance between eigenvalues, such as the difference absolute value distance, and determining the ratio of the number of features of cluster features to the number of features of the test result multiscale sparse fusion representation vector; arranging the cluster features into a test result multiscale sparse fusion clustering vector, and dividing the second norm of the test result multiscale sparse fusion clustering vector by the second norm of the test result multiscale sparse fusion representation vector to obtain a first test result multiscale sparse fusion clustering related weight value; dividing the first norm of the test result multiscale sparse fusion clustering vector by a first power value exponentially using the ratio value. A second power value of the scale value is used as an exponent of a norm of the test result multiscale sparse fusion representation vector to obtain a second test result multiscale sparse fusion clustering related weight value; for each eigenvalue of the test result multiscale sparse fusion clustering vector, multiply it by the inverse of the difference between the first and second test result multiscale sparse fusion clustering related weight values to obtain an optimized eigenvalue of the test result multiscale sparse fusion clustering vector; for each eigenvalue outside the cluster in the test result multiscale sparse fusion representation vector, multiply it by the inverse of the sum of the first and second test result multiscale sparse fusion clustering related weight values to obtain an optimized out-of-class eigenvalue of the test result multiscale sparse fusion representation vector; and the optimized eigenvalues of the test result multiscale sparse fusion clustering vector and the optimized out-of-class eigenvalues of the test result multiscale sparse fusion representation vector are combined to form an optimized test result multiscale sparse fusion representation vector.

That is, the optimization process is expressed as: ; is the multi-scale sparse fusion representation vector of the test result, for example, The number of features, is the multi-scale sparse fusion clustering vector of the test result, for example, The number of features, represents the clustering feature set corresponding to the multi-scale sparse fusion clustering vector of the test result, and Represents the binary norm and the uninorm of the vector respectively. Power, is the ratio value, is the multi-scale sparse fusion representation vector of the test result, is the multi-scale sparse fusion clustering vector of the test result, is the eigenvalue of each position of the multi-scale sparse fusion clustering vector of the test result, It is the eigenvalues of each position of the optimized out-of-class eigenvalue composition optimized test result multi-scale sparse fusion representation vector.

Here, in order to avoid the loss of key suffix semantic information relative to the original feature set as a whole due to aggregation conflict based on the aggregated features of the test result multiscale sparse fusion representation vector, the clustering ratio of the number of features of the test result multiscale sparse fusion clustering vector relative to the number of features of the test result multiscale sparse fusion representation vector is used as a decision function to perform an adversarial decision on the set absolute representation of the test result multiscale sparse fusion clustering vector and the one-norm of the test result multiscale sparse fusion representation vector, and positive and negative interactions are performed with the clustering intrinsic conflict representation of the two-norm of the test result multiscale sparse fusion clustering vector and the test result multiscale sparse fusion representation vector, respectively, to construct a solid alignment guardrail between the optimized test result multiscale sparse fusion representation vector and the original feature set as a whole based on the aggregated features, thereby achieving the mitigation of the intention of harmful information loss of the optimized test result multiscale sparse fusion representation vector based on the aggregation risk transferability, improving the expression effect of the optimized test result multiscale sparse fusion representation vector, and thus improving the accuracy of the test result obtained by inputting the test result multiscale sparse fusion representation vector into the test result generator based on the classifier. This improves the efficiency of processing large amounts of test data, reduces the time and error rate of manual operations, and can fully capture the complex relationships between variables, provide a deeper understanding of test results, and make test data processing more intelligent.

In summary, the above scheme is adopted to embed and encode each test result item in the test data by using an artificial intelligence-based data processing and analysis algorithm, and the embedded and encoded test results are respectively subjected to global semantic association and local semantic association, so as to automatically obtain the test results according to the significant fusion features between the global semantic features and the local semantic features of the test results. In this way, the efficiency of processing a large amount of test data is improved, and the time and error rate of manual operation are reduced. At the same time, it can fully capture the complex relationship between variables, provide a deeper understanding of the test results, and make the test data processing more intelligent.

Referring to FIG. 3 below, it shows a schematic diagram of the structure of an electronic device 600 suitable for implementing an embodiment of the present application. The terminal device in the embodiment of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The electronic device shown in FIG. 3 is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

As shown in FIG3 , the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to the bus 604.

Typically, the following devices may be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 609. The communication device 609 may allow the electronic device 600 to communicate wirelessly or wired with other devices to exchange data. Although FIG. 3 shows an electronic device 600 with various devices, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have alternatively.

In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

Computer program code for performing the operations of the present application may be written in one or more programming languages or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and/or flow chart, and the combination of the square boxes in the block diagram and/or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

Fig. 4 is an application scenario diagram of a single-turn magnetic encoder automated test fixture according to an exemplary embodiment. As shown in Fig. 4, in this application scenario, first, test data is collected by a lower computer (for example, C as shown in Fig. 4); then, the acquired test data is input into a server (for example, S as shown in Fig. 4) deployed with a single-turn magnetic encoder automated test fixture algorithm, wherein the server can process the test data based on the single-turn magnetic encoder automated test fixture algorithm to obtain the test result.

The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

Although the subject matter has been described in language specific to structural features and/or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims (9)

1. A single-turn magnetic encoder automated testing tool, characterized in that it comprises: a host computer and a slave computer, wherein the working process of the host computer and the slave computer comprises: starting the host computer; selecting an editor mainboard to be tested and corresponding firmware data; the host computer communicates with the slave computer via RS-485 to send a burning instruction and the firmware data; the slave computer selects the correct encoder mainboard via an FPGA chip, and downloads the firmware data to the editor chip of the editor mainboard via an ISP download protocol; after the burning is completed, the slave computer reads the encoder data according to the Tamagawa encoder protocol to check the working status of the editor mainboard; the test data is collected by the slave computer and sent to the host computer; the host computer analyzes the test data to obtain a test result. 2. The single-turn magnetic encoder automated testing tool according to claim 1 is characterized in that the host computer analyzes the test data to obtain test results, including: using a test embedding coding matrix to embed code each test result item in the test data to obtain a set of test result item embedding coding vectors; inputting the set of test result item embedding coding vectors into a global semantic association encoder between test results based on a converter structure to obtain a global semantic coding feature vector of the test results; inputting the set of test result item embedding coding vectors into a local semantic association encoder between test results to obtain a local semantic coding feature vector of the test results; inputting the test result global semantic coding feature vector and the test result local semantic coding feature vector into a significant fusion network based on feature principal component query matching to obtain a test result multi-scale sparse fusion representation vector; and obtaining the test result based on the test result multi-scale sparse fusion representation vector. 3. The single-turn magnetic encoder automated testing tool according to claim 2 is characterized in that the set of test result item embedding coding vectors is input into a local semantic association encoder between test results to obtain a local semantic coding feature vector of the test results, including: the set of test result item embedding coding vectors is input into a local semantic association encoder between test results based on a 1D-CNN model to obtain the local semantic coding feature vector of the test results. 4. The single-turn magnetic encoder automated testing tool according to claim 3 is characterized in that the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result are input into a significant fusion network based on feature principal component query matching to obtain a multi-scale sparse fusion representation vector of the test result, including: standardizing the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result to obtain a standardized global semantic coding feature vector of the test result and a standardized local semantic coding feature vector of the test result; calculating the sample covariance matrix of the standardized global semantic coding feature vector of the test result and the standardized local semantic coding feature vector of the test result to obtain a global semantic sample covariance matrix of the test result and a local semantic sample covariance matrix of the test result; performing feature vector extraction based on matrix decomposition on the global semantic sample covariance matrix of the test result and the local semantic sample covariance matrix of the test result. Take to obtain a set of global semantic principal component feature vectors of the test results and a set of local semantic principal component feature vectors of the test results; input the set of global semantic principal component feature vectors of the test results and the set of local semantic principal component feature vectors of the test results into a maximum approximate query matching network to obtain a set of best matching pairs of global semantic principal component feature vectors of the test results and local semantic principal component feature vectors of the test results; input each best matching pair of global semantic principal component feature vectors of the test results and local semantic principal component feature vectors of the test results in the set of best matching pairs of global semantic principal component feature vectors of the test results and local semantic principal component feature vectors of the test results into a semantic fine-grained gated joint module to obtain a set of global semantic principal component fusion feature vectors of the test results-local semantic principal component fusion feature vectors of the test results; cascade the set of global semantic principal component fusion feature vectors of the test results-local semantic principal component fusion feature vectors of the test results to obtain a multi-scale sparse fusion representation vector of the test results. 5. The single-turn magnetic encoder automated testing tool according to claim 4 is characterized in that the global semantic coding feature vector of the test result and the local semantic coding feature vector of the test result are standardized to obtain a standardized global semantic coding feature vector of the test result and a standardized local semantic coding feature vector of the test result, including: respectively calculating the mean and standard deviation of the global semantic coding feature vector of the test result to obtain a global semantic coding feature mean of the test result and a global semantic coding feature standard deviation of the test result; subtracting the global semantic coding feature vector of the test result from the global semantic coding feature mean of the test result by position, and calculating the obtained test result. The global semantic offset vector of the test result and the standard deviation of the global semantic coding feature of the test result are divided by position to obtain the standardized global semantic coding feature vector of the test result; the mean and standard deviation of the local semantic coding feature vector of the test result are calculated respectively to obtain the mean of the local semantic coding features of the test result and the standard deviation of the local semantic coding features of the test result; after subtracting the local semantic coding feature vector of the test result from the mean of the local semantic coding features of the test result by position, the calculated local semantic offset vector of the test result and the standard deviation of the local semantic coding features of the test result are divided by position to obtain the standardized local semantic coding feature vector of the test result. 6. The single-turn magnetic encoder automated testing tool according to claim 5 is characterized in that the sample covariance matrix of the standardized test result global semantic coding feature vector and the standardized test result local semantic coding feature vector are calculated to obtain the test result global semantic sample covariance matrix and the test result local semantic sample covariance matrix, including: multiplying the transpose vector of the standardized test result global semantic coding feature vector with the standardized test result global semantic coding feature vector, and then dividing the obtained standardized test result global semantic association matrix by the value obtained by subtracting one from the length of the standardized test result global semantic coding feature vector by position to obtain the test result global semantic sample covariance matrix; multiplying the transpose vector of the standardized test result local semantic coding feature vector with the standardized test result local semantic coding feature vector, and then dividing the obtained standardized test result local semantic association matrix by the value obtained by subtracting one from the length of the standardized test result local semantic coding feature vector by position to obtain the test result local semantic sample covariance matrix. 7. The single-turn magnetic encoder automated testing tool according to claim 6 is characterized in that the set of the test result global semantic principal component feature vectors and the set of the test result local semantic principal component feature vectors are input into a maximum approximate query matching network to obtain a set of best matching pairs of the test result global semantic principal component feature vectors and the test result local semantic principal component feature vectors, including: extracting a predetermined test result global semantic principal component feature vector from the set of the test result global semantic principal component feature vectors; calculating the cosine similarity between the predetermined test result global semantic principal component feature vector and each test result local semantic principal component feature vector in the set of the test result local semantic principal component feature vectors to obtain a set of matching query similarities; and taking the test result local semantic principal component feature vector corresponding to the largest matching query similarity in the set of matching query similarities and the predetermined test result global semantic principal component feature vector as the best matching pair of the predetermined test result global semantic principal component feature vector and the test result local semantic principal component feature vector. 8. The single-turn magnetic encoder automated testing tool according to claim 7 is characterized in that each best matching pair of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result in the set of the best matching pairs of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result is input into the semantic fine-grained gated joint module to obtain a set of test result global-test result local semantic principal component fusion feature vectors, including: respectively calculating the position difference, position dot multiplication and position addition between the best matching pairs of the global semantic principal component feature vector of the test result and the local semantic principal component feature vector of the test result to obtain The test result global-local semantic principal component difference vector, the test result global-local semantic principal component dot product vector and the test result global-local semantic principal component sum vector are obtained; the test result global-local semantic principal component difference vector, the test result global-local semantic principal component dot product vector and the test result global-local semantic principal component sum vector are cascaded and then one-dimensional convolutional encoded to obtain a test result global-local semantic principal component multi-dimensional fusion vector; the test result global-local semantic principal component multi-dimensional fusion vector is subjected to local window-based maximum pooling processing to obtain the test result global-test result local semantic principal component fusion feature vector. 9. The single-turn magnetic encoder automated testing tool according to claim 8 is characterized in that the test result is obtained based on the multi-scale sparse fusion representation vector of the test result, including: inputting the multi-scale sparse fusion representation vector of the test result into a classifier-based test result generator to obtain the test result.