Operation ticket intelligent analysis and sequence control logic closed-loop verification method and system

Abstract

The invention discloses an operation ticket intelligent analysis and sequence control logic closed-loop verification method and system, and belongs to the field of power system automation technology and power transformation operation and maintenance, and the method comprises the steps: obtaining a typical interval sequence control operation ticket and an SCD file; identifying the typical interval sequence control operation order through a BERT-ERNIE multi-head text extraction model, and generating a corresponding test script; querying corresponding object path information in the SCD and completing association; constructing a closed-loop acceptance environment; simulating a local operation scene and initiating sequence control operation according to the associated test script to carry out correctness acceptance; on the basis of a similar interval operation ticket generation tool, carrying out one-key replacement on part of description information of the test script passing acceptance, generating similar interval sequential control operation tickets to be accepted, and carrying out correctness acceptance; and simulating a sequence control master station address in the closed-loop acceptance environment, issuing a sequence control instruction to a sequence control host based on the sequence control operation order passing correctness acceptance, and automatically virtualizing the test environment to perform forward and reverse closed-loop automatic acceptance.

Description

Technical Field

[0001] The present invention belongs to the fields of power system automation technology and substation operation and maintenance, and relates to a method and system for intelligent parsing of operation tickets and closed-loop verification of sequence control logic. Background Art

[0002] Substations are an important part of the smart grid. With the development of the construction of the smart grid, more and more existing substations are being retrofitted with one-key sequence control. Substation control operations are related to the safe operation of substations. As a very important operation link in substations, before a substation is put into operation, all operations need to be tested and accepted.

[0003] In the conventional acceptance mode, the on-site equipment needs to be powered off, and multiple people are relied on to cooperate to conduct tests one by one. Due to the large amount of on-site acceptance work, typical interval full-scale acceptance is mostly adopted, and other intervals are spot-checked. It is impossible to complete all-round reverse testing, which brings potential safety hazards to the safe operation of substations; the conventional test verification mode is inefficient, lacking a test system with perfect functions and a good test method to achieve comprehensive quantitative testing and positive and negative testing, and there is no good solution for the one-key sequence control logic function test, which cannot fully meet the requirements of substation one-key sequence control testing. It is urgent to improve the acceptance efficiency of substation regulation information through technical means.

[0004] The traditional substation sequence control acceptance depends on manual item-by-item verification of the consistency between the content of the operation ticket and the response information of the in-station sequence control system, which has problems such as low efficiency, easy omission, and inability to close-loop verify the consistency between the operation result and the expectation. In the prior art, staff need to manually organize the information of the typical interval sequence control operation ticket approved by the user into a test ticket text recognizable by the acceptance device; moreover, the reusability of the same type of intervals is poor, with problems such as low efficiency, large workload, and difficulty in ensuring correctness. Summary of the Invention

[0005] To solve the deficiencies in the prior art, the present invention provides a method and system for intelligent parsing of operation tickets and closed-loop verification of sequence control logic.

[0006] The present invention adopts the following technical solutions.

[0007] The first aspect of the present invention provides a method for intelligent parsing of operation tickets and closed-loop verification of sequence control logic, including the following steps:

[0008] Obtain the typical interval sequence control operation ticket and SCD file approved by the operation and maintenance personnel of the substation to be accepted, and perform preprocessing;

[0009] Identify the obtained typical interval sequence control operation ticket through the BERT-ERNIE multi-head text extraction model, extract key information and generate corresponding test scripts;

[0010] Query the corresponding object path information in the SCD according to the test script description information and complete the association to realize the mapping between the test script and the actual device object;

[0011] Use the switch to interconnect the sequence control host, intelligent anti-misoperation host and network gateway to build a closed-loop acceptance environment;

[0012] Based on the closed-loop acceptance environment, simulate a local operation scenario. In this scenario, initiate a sequence control operation according to the associated test script, and perform a correctness acceptance of the test script based on the sequence control host screen information;

[0013] Based on the pre-built similar interval operation ticket generation tool, perform a one-key replacement of the keywords in the test script description information that has passed the acceptance and the IED names in the path to generate a similar interval sequence control operation ticket to be accepted and perform a correctness acceptance;

[0014] Simulate the sequence control master station address in the closed-loop acceptance environment, send the sequence control instruction to the sequence control host based on the above sequence control operation ticket that has passed the correctness acceptance, and automatically virtualize the test environment according to different test types. By comparing the consistency between the sequence control host return result and the expected result, complete the positive and negative closed-loop automatic acceptance.

[0015] Optionally, the key information of the typical interval sequence control operation ticket includes the description information after the keyword, and the keywords include interval name, current state, target state, pre-execution check, post-execution condition and operation item.

[0016] Optionally, the recognition of the obtained typical interval sequence control operation ticket by the BERT-ERNIE multi-head text extraction model, and the extraction of key information and generation of the corresponding test script include:

[0017] Preprocess the typical interval sequence control operation ticket text, including word segmentation, tokenization and adding position encoding, and convert it into a word vector sequence;

[0018] Use the BERT model with multi-head attention mechanism to perform bidirectional encoding on the word vector sequence of the operation ticket text to obtain a feature matrix containing semantic and syntactic information in the text;

[0019] Based on the ERNIE model, perform knowledge enhancement on the text feature matrix representation output by the BERT model and fuse the knowledge graph information;

[0020] Based on the feature matrix representations output by the BERT model and the ERNIE model, extract the key information corresponding to the keywords, and organize the extracted key information according to the preset format and logic to generate a test script containing interval description, operation ticket description, operation item description, and operation condition description.

[0021] Optionally, the bidirectional encoding of the operation ticket text word vector sequence by the BERT model using the multi-head attention mechanism includes:

[0022] Input the preprocessed typical interval sequence control operation ticket text word vector sequence into the BERT model for linear transformation to obtain the Query matrix Q, the Key matrix K, and the Value matrix V:

[0023] Q = HW Q , K = HW K , V = HW V

[0024] Where, is the input word vector matrix, is the learnable parameter matrix; d model is the dimension of the word vector;

[0025] Calculating the feature matrix through the multi-head attention mechanism includes single-head attention calculation, multi-head concatenation, and linear transformation of the concatenated matrix to obtain a feature matrix containing semantic and syntactic information in the text.

[0026] Optionally, the single-head attention calculation includes:

[0027] For each attention head i ∈ [1, h], split the matrices Q, K, and V into h sub-matrices respectively:

[0028]

[0029] Then calculate the single-head attention output:

[0030]

[0031] Where, d k is the dimension of K.

[0032] Optionally, concatenate the outputs of multiple heads into a matrix:

[0033] MultiHead(Q, K, V) = Concat(head1, head2,... head h )

[0034] The dimension of the concatenated matrix is

[0035] The linear transformation of the concatenated matrix to obtain a feature matrix containing semantic and syntactic information in the text includes:

[0036] H out = MultiHead(Q, K, V)W O

[0037] Among them, is the output transformation matrix, and the finally output feature matrix is consistent with the input word vector dimension.

[0038] Optionally, a standardization template and a parameter mapping library are constructed. The construction of the standardization template includes abstracting the content to be replaced into a unified identifier in the accepted test script; the establishment of the parameter mapping library includes creating a comparison table of interval type parameters and storing it in the form of Excel;

[0039] Read the parameter mapping library, establish the IED naming rule and the path generation rule, and set the replacement strategy.

[0040] Optionally, the replacement strategy includes incremental replacement and exception avoidance. Incremental replacement is used to automatically generate consecutive IED numbers according to the serial number rule, and exception avoidance is used to skip the existing IED names.

[0041] Optionally, the one-key replacement of the keywords in the test script description information and the IED names in the path by using the pre-built operation ticket generation tool for the same type of interval includes:

[0042] Select the target interval type and the generation quantity, and the operation ticket generation tool for the same type of interval automatically executes the replacement action to generate the replaced test script;

[0043] Verify the test script generated by the replacement through syntax verification and semantic verification. The syntax verification includes checking the file format standardization through an XML validator; the semantic verification includes intelligently comparing the generated test script with the SCD file to confirm that the new IED name exists in the SCD and the generated path matches the device object hierarchy in the SCD.

[0044] The second aspect of the present invention provides an intelligent parsing and sequence control logic closed-loop verification system for an operation ticket, which is used for an intelligent parsing and sequence control logic closed-loop verification method described in the first aspect of the present invention. The system includes:

[0045] A data preprocessing module, which is used to receive the operation ticket and the SCD file, and perform format cleaning and structured processing;

[0046] A text parsing and script generation module, which is used to extract the key information of the operation ticket through the BERT-ERNIE model and generate a test script;

[0047] An SCD path association module, which is used to query the SCD file according to the content of the test script and associate the device signal path;

[0048] A closed-loop environment building module, which is used to connect the sequence control host, the anti-misoperation host, and the network gateway to build a closed-loop acceptance environment;

[0049] The sequence control operation acceptance module is used to execute test scripts in a closed-loop environment to verify the sequence control host screen and operation results;

[0050] The similar interval generation module is used to generate test scripts for similar intervals;

[0051] The closed-loop automatic test module is used to simulate the commands issued by the sequence control master station and automatically compare the actual results with the expected results;

[0052] The report generation module is used to record the test process and output test reports.

[0053] Compared with the prior art, the beneficial effects of the present invention at least include:

[0054] The present invention supports one-key sequence control automatic acceptance that does not rely on substation on-site protection and measurement and control devices, and can be applied to newly built substations and in-service intelligent substations. The automatic test script generation function is based on the typical interval sequence control operation ticket information approved by the user, and can generate test scripts required for acceptance with one key, eliminating the need for on-site acceptance personnel to manually write test scripts, saving the pre-preparation time, and ensuring the correctness of the verification benchmark; the typical interval replication function automatically replaces the description information and the IED name of the associated object path according to the configured test script information, improving the on-site acceptance efficiency, reducing uncertain human factors, and ensuring the acceptance quality. Description of the Drawings

[0055] Figure 1 It is the method flow chart provided by the embodiment of the present invention. Detailed Embodiments

[0056] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the spirit of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0057] To more clearly introduce the prominent substantive features of the present invention and the significant progress brought to the prior art, the following introduces an application example of implementing the present invention.

[0058] The following will be a detailed description of the embodiments of the present invention in conjunction with the drawings. The application example specifically includes:

[0059] The present invention provides a method for intelligent parsing of operation tickets and closed-loop verification of sequence control logic in Embodiment 1. As Figure 1 shown, it includes the following steps:

[0060] Step 1: Obtain the typical interval sequence control operation ticket and SCD file that have been reviewed and approved by the substation operation and maintenance personnel to be accepted, and perform preprocessing;

[0061] Specifically, the SCD file is provided by the monitoring background manufacturer on-site at the substation to be accepted;

[0062] It should be noted that if the dual-confirmation auxiliary criterion used in this station is the video confirmation method, the background manufacturer is also required to provide a video forwarding point table file.

[0063] Step 2: Use the BERT-ERNIE multi-head text extraction model to identify the typical interval sequence control operation ticket obtained in Step 1, extract key information, and generate corresponding test scripts;

[0064] Preferably, the key information of the typical interval sequence control operation ticket includes the descriptive information after the keywords, and the keywords include interval name, current status, target status, pre-execution check, post-execution condition, and operation item;

[0065] Preferably, Step 2 includes:

[0066] Step 2.1: Preprocess the text of the typical interval sequence control operation ticket, including word segmentation, tokenization, and adding position encoding, and convert it into a sequence of word vectors;

[0067] Step 2.2: Use the BERT model with multi-head attention mechanism to perform bidirectional encoding on the sequence of word vectors of the operation ticket text to capture semantic and syntactic information in the text;

[0068] Further preferably, Step 2.2 includes:

[0069] Input the sequence of word vectors of the text of the typical interval sequence control operation ticket obtained after preprocessing into the BERT model, perform a linear transformation, and obtain the Query matrix Q, Key matrix K, and Value matrix V:

[0070] Q = HW Q

[0071] K = HW K

[0072] V = HW V

[0073] Where, is the input word vector matrix, is the learnable parameter matrix. Further preferably, calculating the feature matrix through the multi-head attention mechanism includes:

[0074] (1) Single-head attention calculation

[0075] For each attention head \(i\in[1, h]\), split the matrices \(Q\), \(K\), and \(V\) into \(h\) sub - matrices respectively:

[0076]

[0077] Then calculate the single - head attention output:

[0078]

[0079] where \(d\) k is the dimension of \(K\);

[0080] (2) Multi - head concatenation

[0081] Concatenate the outputs of multiple heads into a single matrix:

[0082] MultiHead(Q,K,V)=Concat(head1,head2,...head h )

[0083] The dimension of the concatenated matrix is

[0084] (3) Linear transformation

[0085] Perform a linear transformation on the concatenated matrix to obtain the final feature matrix:

[0086] H out =MultiHead(Q,K,V)W O

[0087] where is the output transformation matrix, and the dimension of the finally output feature matrix is the same as the dimension of the input word vectors. The relationship between the number of heads \(h\) and dimensions \(d\) k 、d v is

[0088] Step 2.3: Based on the ERNIE model, enhance the text feature representation output by the BERT model, fuse knowledge graph information, and improve the accuracy of key information extraction;

[0089] Step 2.4: Extract the key information corresponding to the keywords based on the feature representations output by the BERT model and the ERNIE model, and organize the extracted key information according to the preset format and logic to generate a test script containing interval description, operation ticket description, operation item description, and operation condition description.

[0090] Further preferably, in the said Step 2.4, generating a test script containing interval description, operation ticket description, operation item description, and operation condition description includes:

[0091] The test script is in Excel format, with each interval being a separate sub - sheet;

[0092] The test script interval contains three columns of information: the name of the operation ticket, the operation items, and the operation conditions;

[0093] Specifically, the generation of the test script also includes:

[0094] Interval name: According to the interval name attribute description information in the key information of the sequential control operation ticket for the typical interval, a separate sub - sheet is created for each interval;

[0095] Operation ticket name: According to the description information after the initial state and target state keywords in the key information of the sequential control operation ticket for the typical interval, the name of the sequential control operation ticket is automatically generated in the format of "changed from the initial state to the target state";

[0096] Operation items: The description information after the operation item keyword in the key information of the sequential control operation ticket for the typical interval is automatically copied to the operation ticket control object attribute column;

[0097] Operation conditions: The description information after the pre - operation condition and post - operation condition keywords in the key information of the sequential control operation ticket for the typical interval is automatically copied to the operation ticket operation condition attribute column.

[0098] Specifically, after automatically generating the test script, it is also necessary to manually check the consistency between the test script and the information of the typical interval operation ticket provided by the user.

[0099] It should be noted that in view of the deficiencies of the prior art that the parsing of the sequential control operation ticket in the substation and the generation of the test script require manual parsing, with low efficiency, easy to make mistakes, and insufficient logical verification, the present invention proposes to intelligently parse the sequential control operation ticket in the substation through the BERT - ERNIE multi - head text extraction model and fully automatically generate the test script, which can improve the acceptance efficiency and reduce the errors caused by manual intervention.

[0100] Step 3: According to the description information of the test script, query the corresponding object path information in the SCD and complete the association to realize the mapping between the test script and the actual device object.

[0101] Step 4: Use the switch to interconnect the sequential control host, the intelligent anti - misoperation host, and the network gateway to build a closed - loop acceptance environment.

[0102] Preferably, in the said Step 4, the construction of the closed - loop acceptance environment is disconnected from the actual in - station environment of the substation to be accepted, avoiding the influence of external interference factors on the test results.

[0103] Step 5: Based on the closed-loop acceptance environment constructed in Step 4, simulate a local operation scenario. In this scenario, initiate a sequence control operation according to the associated test script, and perform a correctness acceptance of the test script based on the sequence control host screen information.

[0104] Preferably, in Step 5, the correctness acceptance of the test script based on the sequence control host screen information includes:

[0105] Manually check the correctness of the initial state, target state, operation items, disconnector position information in the primary wiring diagram, and the display of telemetry and remote signaling values on the one-key sequence control screen of the sequence control host to complete the correctness acceptance of the sequence control host screen.

[0106] It should be noted that the functions of Step 5 include that the sequence control host is the device for actual sequence control operations, and the correctness of the screen display information is related to subsequent safe use; ensure that the test scripts of the typical intervals with which the objects have been automatically associated are correct in terms of the description information and the paths of the telemetry, remote signaling, and remote control objects in the station, so as to ensure the correctness of the other interval information generated by subsequent copying.

[0107] Step 6: Based on the pre-constructed operation ticket generation tool for similar intervals, perform a one-key replacement of the keywords in the description information of the passed test script and the IEDname in the path to generate a sequence control operation ticket to be accepted for similar intervals and perform a correctness acceptance.

[0108] Preferably, the one-key replacement in Step 6 further includes: according to the configured interval information, complete the copying of the sequence control test script for similar intervals through one-key replacement of inconsistent information.

[0109] More preferably, the construction of the operation ticket generation tool for similar intervals includes:

[0110] Construct a standardized template and a parameter mapping library. The construction of the standardized template includes abstracting the content to be replaced into a unified identifier in the passed test script.

[0111] Exemplarily, in the passed test script, for the device name replacement point "[IED_TYPE]_[sequence number]", it is marked as "$$IED_PREFIX$$_NUM" in the template; for the path replacement point, retain the path end structure " / protection / switch position", and mark the reference path as "$$PATH_ROOT$$". This template fragment can be expressed as:

[0112] <operation instruction device = "$$IED_PREFIX$$_01" path = "$$PATH_ROOT$$ / protection / switch position" / >

[0113] The establishment of the parameter mapping library includes creating a comparison table of interval type parameters and storing it in the form of Excel;

[0114] Exemplarily, the comparison table of interval type parameters can be defined as:

[0115]

[0116] Read the parameter mapping library, establish the IED naming rule and path generation rule, and set the replacement strategy;

[0117] Exemplarily, the IED naming rule can be established as: "prefix + serial number", such as "L_Line_02";

[0118] Exemplarily, the path generation rule can be established as: "new root node + original end path", such as replacing "$$PATH_ROOT$$" with "Line / Protection / 02";

[0119] Exemplarily, the replacement strategy includes incremental replacement and exception avoidance. Incremental replacement is used to automatically generate consecutive device numbers according to the serial number rule, and exception avoidance is used to skip existing device names to prevent overwriting. Further preferably, the one - key replacement of keywords in the test script description information and IEDname in the path by using the pre - built operation ticket generation tool for the same type of intervals includes:

[0120] Select the target interval type and the generation quantity, and the system automatically executes the replacement action to generate the replaced test script;

[0121] Exemplarily, when the target interval type is selected as the line interval and the generation quantity is selected as 5, the system automatically executes the following actions:

[0122] Replace "$$IED_PREFIX$$" with "L_Line";

[0123] Fill in the numbers 02 - 06 according to the serial number rule;

[0124] Replace "$$PATH_ROOT$$" with "Line / Protection / [serial number]";

[0125] The generation result is:

[0126] <Operating instruction device = "L_Line_02" path = "Line / Protection / 02 / Protection / Switch position" / >

[0127] Further preferably, the one - key replacement of keywords in the test script description information and IEDname in the path by using the pre - built operation ticket generation tool for the same type of intervals further includes:

[0128] Verify the test script generated by one - key replacement through syntax verification and semantic verification. For the syntax verification: check the file format standardization through an XML validator; the semantic verification is to compare with the SCD file intelligently to confirm that the new IED name actually exists in the SCD and the generated path matches the device object hierarchy in the SCD.

[0129] It should be noted that in view of the problems in the prior art such as poor reusability, low efficiency, and large workload of the same - type bays, the present invention proposes a technical means for constructing a generation tool for operation tickets of the same - type bays. Through the combination of a standardized template and parameterized configuration, it realizes the automatic conversion from a single verified script to batch generation of the same - type bays; the present invention also adopts a dual - verification mechanism of syntax verification and semantic verification to ensure strict consistency between the output result and the SCD file, greatly reducing the time for users to manually compile sequence control tickets and improving the efficiency of substation sequence control acceptance.

[0130] Step 7: Simulate the address of the sequence control master station, send sequence control instructions to the sequence control host based on the sequence control operation ticket that has passed the correctness acceptance above, and automatically virtualize the test environment according to different test types. Complete the forward and reverse closed - loop automatic acceptance by comparing the consistency between the result returned by the sequence control host and the expected result.

[0131] Preferably, step 7 further includes sending sequence control instructions using the dispatching sequence control protocol.

[0132] Preferably, in step 7, the test types include verification of the correctness of remote control objects, verification of the correctness of operating conditions, and verification of the integrity of double - confirmation.

[0133] Step 8: Record the test process from step 5 to step 7 and generate a test report.

[0134] Preferably, the test report further includes a brief report and a detailed report, and the brief report or the detailed report can be selected for export according to needs. The report formats include pdf and excel.

[0135] In Embodiment 2 of the present invention, an intelligent parsing system for operation tickets and a closed - loop verification system for sequence control logic are provided. Based on the method for intelligent parsing of operation tickets and closed - loop verification of sequence control logic described in Embodiment 1, this system includes:

[0136] A data pre - processing module, which is used to receive operation tickets and SCD files and perform format cleaning and structured processing;

[0137] A text parsing and script generation module, which is used to extract key information of the operation ticket through the BERT - ERNIE model and generate a structured test script;

[0138] An SCD path association module, which is used to query the SCD file and bind the device signal path according to the content of the test script;

[0139] A closed-loop environment building module, which is used to connect the sequence control host, the anti-error host, and the network gateway to build an independent test network;

[0140] A sequence control operation acceptance module, which is used to execute test scripts in the closed-loop environment to verify the sequence control host screen and operation results;

[0141] A similar interval generation module, which is used to replace the device names in the test scripts to generate similar interval test scripts;

[0142] A closed-loop automatic test module, which is used to simulate the master station issuing instructions and automatically compare the actual results with the expected results;

[0143] A report generation module, which is used to record the test process and output a brief or detailed test report.

[0144] Preferably, the system provided by the present invention can be applied to newly built substations and existing intelligent substations.

[0145] The present disclosure can be a system, a method, and / or a computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions thereon for causing a processor to implement various aspects of the present disclosure.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement without departing from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. An intelligent parsing method for operation tickets and a closed-loop verification method for sequence control logic, characterized in that, It includes the following steps: Obtain the typical interval sequence control operation ticket and SCD file that have been reviewed and approved by the operation and maintenance personnel of the substation to be accepted, and perform preprocessing; Identify the obtained typical interval sequence control operation ticket through the BERT-ERNIE multi-head text extraction model, extract key information, and generate corresponding test scripts; Query the corresponding object path information in the SCD according to the description information of the test script and complete the association to realize the mapping between the test script and the actual device object; Use the switch to interconnect the sequence control host, intelligent anti-misoperation host, and network gateway to build a closed-loop acceptance environment; Simulate a local operation scenario based on the closed-loop acceptance environment, initiate sequence control operations according to the associated test script in this scenario, and perform correctness acceptance on the test script based on the screen information of the sequence control host; Based on the pre-built tool for generating operation tickets for similar intervals, perform one-key replacement on the keywords in the description information of the test script that has passed the acceptance and the IED names in the path, generate the sequence control operation ticket to be accepted for similar intervals, and perform correctness acceptance; Simulate the sequence control master station address in the closed-loop acceptance environment, send sequence control instructions to the sequence control host based on the sequence control operation ticket that has passed the correctness acceptance above, and automatically virtualize the test environment according to different test types. By comparing the consistency between the returned result of the sequence control host and the expected result, complete the forward and reverse closed-loop automatic acceptance.

2. According to the method for intelligent parsing of operation tickets and closed-loop verification of sequence control logic described in claim 1, wherein: The key information of the typical interval sequence control operation ticket includes the description information after the keyword, and the keywords include interval name, current status, target status, pre-execution check, post-execution condition, and operation item.

3. According to the method for intelligent parsing of operation tickets and closed-loop verification of sequence control logic described in claim 2, wherein: The identification of the obtained typical interval sequence control operation ticket through the BERT-ERNIE multi-head text extraction model, extraction of key information, and generation of corresponding test scripts include: Perform preprocessing on the text of the typical interval sequence control operation ticket, including word segmentation, tokenization, and adding position encoding, and convert it into a sequence of word vectors; Perform bidirectional encoding on the sequence of word vectors of the operation ticket text through the multi-head attention mechanism of the BERT model to obtain a feature matrix containing semantic and syntactic information in the text; Based on the ERNIE model, perform knowledge enhancement on the text feature matrix representation output by the BERT model and fuse knowledge graph information; Extract the key information corresponding to the keyword based on the feature matrix representations output by the BERT model and the ERNIE model, and organize the extracted key information according to the preset format and logic to generate a test script containing interval description, operation ticket description, operation item description, and operation condition description.

4. According to the method for intelligent parsing of operation tickets and closed-loop verification of sequence control logic described in claim 3, wherein: The bidirectional encoding of the sequence of word vectors of the operation ticket text through the multi-head attention mechanism of the BERT model includes: Input the word vector sequence of the typical interval sequence control operation ticket text obtained after preprocessing into the BERT model for linear transformation to obtain the Query matrix Q, the Key matrix K, and the Value matrix V: Q = HW Q , K = HW K , V = HW V Among them, is the input word vector matrix, W Q , W K , is the learnable parameter matrix; d model is the dimension of the word vector; Calculate the feature matrix through the multi-head attention mechanism, including single-head attention calculation, multi-head concatenation, and linear transformation of the concatenated matrix to obtain a feature matrix containing semantic and syntactic information in the text.

5. An operation ticket intelligent parsing and sequence control logic closed-loop verification method according to claim 4, characterized in that: The single-head attention calculation includes: For each attention head i ∈ [1, h], split the matrices Q, K, and V into h sub-matrices respectively: Then calculate the single-head attention output: where d k is the dimension of K.

6. The intelligent parsing and sequence control logic closed-loop verification method for operation tickets according to claim 5, wherein: The multi-head concatenation includes: Concatenate the outputs of multiple heads into one matrix: MultiHead(Q,K,V)=Concat(head1,head2,...head h ) The dimension of the spliced matrix is The linear transformation of the concatenated matrix to obtain a feature matrix containing semantic and syntactic information in the text includes: H out = MultiHead(Q, K, V)W O Among them, is the output transformation matrix, and the finally output feature matrix is consistent with the dimension of the input word vector.

7. An intelligent parsing and sequential control logic closed-loop verification method for operation tickets according to claim 6, characterized in that: The construction of the operation ticket generation tool for the same type of interval includes: Construct a standardized template and a parameter mapping library. The construction of the standardized template includes abstracting the content to be replaced into a unified identifier in the accepted test script. The establishment of the parameter mapping library includes creating a parameter comparison table for interval types and storing it in the form of Excel; Read the parameter mapping library, establish IED naming rules and path generation rules, and set replacement strategies.

8. An operation ticket intelligent parsing and sequence control logic closed-loop verification method according to claim 7, characterized in that: The replacement strategy includes incremental replacement and exception avoidance. Incremental replacement is used to automatically generate consecutive IED numbers according to the serial number rule, and exception avoidance is used to skip existing IED names.

9. An operation ticket intelligent parsing and sequence control logic closed-loop verification method according to claim 8, characterized in that: The one-key replacement of keywords in the test script description information and IED names in the path by using the pre-constructed operation ticket generation tool for the same type of interval includes: Select the target interval type and the generation quantity, and the operation ticket generation tool for the same type of interval automatically executes the replacement action to generate the replaced test script; Verify the test script generated by replacement through syntax verification and semantic verification. The syntax verification includes checking the file format standardization through an XML validator. The semantic verification includes intelligently comparing the generated test script with the SCD file to confirm that the new IED name exists in the SCD and the generated path matches the device object hierarchy in the SCD.

10. An intelligent parsing and sequence control logic closed-loop verification system for operation tickets, which is used to execute the intelligent parsing and sequence control logic closed-loop verification method for operation tickets described in any one of claims 1-9, characterized in that The system includes: A data preprocessing module for receiving operation tickets and SCD files and performing format cleaning and structured processing; A text parsing and script generation module for extracting key information of operation tickets through the BERT-ERNIE model and generating test scripts; An SCD path association module for querying the SCD file and associating device signal paths according to the content of the test script; A closed-loop environment construction module for connecting the sequence control host, the anti-misoperation host, and the network gateway to construct a closed-loop acceptance environment; A sequence control operation acceptance module for executing the test script in the closed-loop environment to verify the sequence control host screen and operation results; A same-kind interval generation module for generating same-kind interval test scripts; A closed-loop automatic test module for simulating the commands sent by the sequence control master station and automatically comparing the actual results with the expected results; A report generation module for recording the test process and outputting test reports.

Images