Distribution network dispatching pre-order ticket issuing method and device, storage medium and computer equipment

Abstract

The invention provides a distribution network scheduling pre-order ticket issuing method and device, a storage medium and computer equipment, and relates to the technical field of power system scheduling, and the method comprises the steps: analyzing distribution network scheduling information, matching a corresponding target template in a scheduling pre-order ticket template library, filling the target template, and obtaining a scheduling pre-order ticket draft; a structured grammar rule base is used for conducting grammar verification on a scheduling pre-order ticket draft, and the structured grammar rule base is constructed on the basis of the natural language processing technology by extracting scheduling terms from distribution network scheduling standard text terms, analyzing a grammar structure and converting an operation logic relation; when the grammar verification of the scheduling pre-order ticket draft is passed, calculating the text similarity between the scheduling pre-order ticket draft and the historical scheduling pre-order ticket; and when the scheduling pre-order ticket draft similarity verification is passed, obtaining the scheduling pre-order ticket, and issuing the scheduling pre-order ticket after confirming that the scheduling pre-order ticket passes the verification. Therefore, the method effectively improves the timeliness and accuracy of the scheduling instruction.

Description

Technical Field

[0001] This application relates to the field of power system dispatching technology, and in particular to a method, apparatus, storage medium and computer equipment for issuing advance orders for distribution network dispatching. Background Technology

[0002] In power systems, power dispatching bears the crucial responsibility of organizing, commanding, guiding, and coordinating grid operation, maintenance, and accident handling. Changes in power system operation modes and adjustments to equipment operating conditions require on-site operators to execute relevant operations based on dispatch orders issued by the dispatch center. Correct understanding and efficient execution of dispatch orders help strengthen collaboration between dispatching and on-site operators, and improve operators' ability to perform primary and secondary electrical operations safely and in accordance with regulations.

[0003] With the continuous expansion of the distribution network, the number of devices requiring dispatch management has increased significantly, and the operation mode of the power grid has become increasingly complex. To improve the standardization and automation of dispatching work, the current distribution network dispatching support platform has introduced a remote command issuance mechanism based on network command technology, realizing the online generation, issuance, and circulation of dispatching commands. Under this mechanism, the dispatching center usually issues a dispatching advance order as a preparatory notification before the formal dispatching command is issued, clarifying the operation content, operators, and planned time to be executed, so as to facilitate the preparation of personnel, tools, and safety measures on site.

[0004] However, in actual operation, especially in high-voltage scenarios such as sudden faults and heavy load switching, the dispatch center experiences a lag in acquiring equipment status data, making it impossible to monitor the real-time operating conditions of field equipment. This lag in equipment status perception leads to discrepancies between the dispatch commands or advance orders generated by the dispatch system and the actual field conditions. This can result in command execution failures or the need for repeated issuance, increasing operational risks and manpower consumption, and affecting the timeliness and efficiency of dispatch response. Therefore, it is evident that the timeliness and accuracy of dispatch commands are currently insufficient in the distribution network dispatching process. Summary of the Invention

[0005] The purpose of this application is to at least address one of the aforementioned technical deficiencies, particularly the technical deficiency of insufficient timeliness and accuracy of dispatch instructions in the current power distribution network dispatching process.

[0006] Firstly, this application provides a method for issuing distribution network dispatch advance orders, the method comprising:

[0007] After parsing the obtained distribution network scheduling information, the corresponding target template is matched in the pre-established scheduling advance order template library, and the parsed distribution network scheduling information is filled into the target template to obtain the draft scheduling advance order.

[0008] The draft scheduling order is grammatically validated using a pre-built structured grammar rule base. The structured grammar rule base is based on natural language processing technology and is constructed by extracting scheduling terms from the standard text of the distribution network scheduling, parsing the grammatical structure, and transforming the operational logic relationship.

[0009] When the draft dispatch order passes the syntax check, calculate the text similarity between the draft dispatch order and the historical dispatch order, and perform a similarity check on the draft dispatch order.

[0010] If the similarity verification of the draft dispatch order passes, the draft dispatch order that passes the similarity verification will be used as the dispatch order, and it will be released through the preset channels when the dispatch order is confirmed to have passed the review.

[0011] In one embodiment, the step of parsing the acquired distribution network scheduling information includes:

[0012] Obtain distribution network scheduling information;

[0013] After removing redundant symbols from the distribution network dispatch information, word segmentation is performed, and multiple preset dispatch feature fields are extracted using regular expressions. The dispatch feature fields include operation type, voltage level, and equipment type.

[0014] In one embodiment, the step of matching the corresponding target template in a pre-established scheduling advance order template library includes:

[0015] In the dispatch advance order template library, a candidate set of dispatch advance order templates is determined based on the operation type corresponding to the distribution network dispatch information;

[0016] Within the candidate set of dispatch advance order templates, a subset of candidate dispatch advance order templates is determined based on the voltage level corresponding to the distribution network dispatch information;

[0017] Within the candidate subset of dispatch pre-order ticket templates, the target template is determined based on the equipment type corresponding to the distribution network dispatch information.

[0018] In one embodiment, the process of establishing a scheduling advance order template library includes:

[0019] Historical scheduling advance order data are categorized based on multiple preset scheduling feature fields;

[0020] Based on the scheduling feature fields corresponding to each category, generate a scheduling advance order template for each category. The scheduling advance order template is used to represent the scheduling rules.

[0021] Configure an identifier and version information for each scheduling advance ticket template, and when a change in scheduling rules is detected, locate and update the scheduling advance ticket template and its version information corresponding to the changed scheduling rules based on the changed scheduling rules and each identifier.

[0022] In one embodiment, the process of building a structured syntax rule base includes:

[0023] Using a pre-trained named entity recognition model, multiple scheduling terms in the standard text of distribution network scheduling are identified, and a dependency parsing algorithm is used to extract the syntactic dependency relationships between the scheduling terms and construct syntactic structure information.

[0024] Multiple high-frequency terms are obtained from the standard text terms of distribution network scheduling and historical scheduling records. The semantic equivalence and causal relationships between these high-frequency terms are determined, and a term mapping relationship table is constructed.

[0025] A sequence pattern mining algorithm is used to extract multiple step sequence patterns from the collected historical scheduling orders, and each step sequence pattern is converted into a corresponding logical expression.

[0026] A structured grammar rule base is formed based on grammatical structure information, term mapping relationship table, and various logical expressions.

[0027] In one embodiment, the step of determining semantic equivalence and causal relationships between various high-frequency terms includes:

[0028] Using word vector models, we can identify semantic equivalence relationships between high-frequency terms.

[0029] Temporal correlation analysis was used to determine the causal relationships between various high-frequency terms.

[0030] In one embodiment, the step of performing a similarity check on the draft scheduling advance order includes:

[0031] If the text similarity of the draft dispatch order is higher than the preset threshold, then the similarity check of the draft dispatch order is deemed to have passed.

[0032] If the text similarity of the draft dispatch order is not higher than the preset threshold, an early warning message is generated to remind manual review, and when a review confirmation instruction is received, the similarity check of the draft dispatch order is determined to be passed.

[0033] Secondly, this application provides a distribution network dispatch advance order issuing device, the device comprising:

[0034] The module for determining the draft dispatch order is used to parse the obtained distribution network dispatch information, match the corresponding target template in the pre-established dispatch order template library, and fill the parsed distribution network dispatch information into the target template to obtain the draft dispatch order.

[0035] The dispatch advance order draft syntax verification module is used to perform syntax verification on the dispatch advance order draft using a pre-built structured syntax rule base. The structured syntax rule base is based on natural language processing technology and is constructed by extracting dispatch terms from the standard text clauses of the distribution network dispatch, parsing the syntax structure and transforming the operational logic relationship.

[0036] The similarity verification module for the draft scheduling advance order is used to calculate the text similarity between the draft scheduling advance order and historical scheduling advance orders when the syntax verification of the draft scheduling advance order passes, and to perform similarity verification on the draft scheduling advance order.

[0037] The dispatch advance order release module is used to release the dispatch advance order draft that has passed the similarity verification as a dispatch advance order if the similarity verification is passed. When the dispatch advance order is confirmed to have passed the review, it is released through preset channels.

[0038] Thirdly, this application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of any of the distribution network scheduling advance order issuance methods described in the above embodiments.

[0039] Fourthly, this application provides a computer device, including: one or more processors, and a memory;

[0040] The memory stores computer-readable instructions, which, when executed by one or more processors, perform the steps of any of the distribution network dispatch advance order issuance methods described in the above embodiments.

[0041] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0042] The distribution network dispatch advance order issuance method, apparatus, storage medium, and computer equipment provided in this application parse the obtained distribution network dispatch information and fill in the corresponding templates in a pre-set dispatch advance order template library to form a draft dispatch advance order, improving the standardization and normalization of the advance order generation process. Simultaneously, a structured grammar rule library based on natural language processing technology is introduced to perform grammatical verification on the draft dispatch advance order, effectively ensuring the semantic accuracy and logical rationality of the advance order content. Furthermore, text similarity comparison with historical dispatch advance orders is used to assist in judging the rationality of the draft, enhancing the consistency between the instruction content and actual operational experience. Finally, after approval, the dispatch advance order is issued through pre-set channels, achieving efficient generation and reliable issuance of dispatch instructions. Therefore, this method can effectively alleviate the dispatch execution error problem caused by the lag in equipment status perception and improve the timeliness and accuracy of dispatch instructions. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A flowchart illustrating the distribution network dispatch advance order issuance method provided in this application embodiment;

[0045] Figure 2 This is a schematic diagram of the distribution network dispatch advance order issuing device provided in an embodiment of this application;

[0046] Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] This application provides a method for issuing advance orders for distribution network dispatching. The following embodiments illustrate this method using computer equipment as an example. It is understood that the computer equipment can be various devices with data processing capabilities, including but not limited to a single server, server cluster, personal laptop, desktop computer, etc. Figure 1 As shown, the method includes:

[0049] S101: After parsing the obtained distribution network scheduling information, match the corresponding target template in the pre-established scheduling advance ticket template library, and fill the parsed distribution network scheduling information into the target template to obtain the draft scheduling advance ticket.

[0050] Distribution network dispatch information refers to data and instruction information related to the operation and management of the distribution network, collected or generated by the dispatch center. This typically includes parameters related to dispatch instructions such as the identification information of the operating equipment, the operation content, the operation time, the personnel arrangement, and the adjustment of the operating mode. The dispatch advance order template library refers to a collection of advance order templates with a unified format and structure, pre-built and stored by computer equipment. Each template corresponds to a common dispatch scenario or operation type, used to standardize the generation of dispatch advance order content. The target template is the template in the dispatch advance order template library that corresponds to the current distribution network dispatch information; it is the selected object during the template matching process. The draft dispatch advance order refers to the preliminary advance order text generated by filling the target template with the distribution network dispatch information.

[0051] Specifically, after receiving the distribution network scheduling information transmitted from the data interface of the scheduling system, the computer equipment first performs structured parsing on the distribution network scheduling information. This structured parsing can be accomplished by running a scheduling information processing program configured with a natural language parsing module. Based on a set of rules, it can extract various scheduling parameters from the original scheduling text into structured fields, such as the unique identifier of the operating equipment, the operation type, and the predetermined time window, thereby generating a structured data object containing complete scheduling elements.

[0052] Next, the computer equipment performs template matching on the structured scheduling data objects within the constructed scheduling advance order template library. This matching process can be based on priority comparisons of key fields, such as operation type, equipment type, and scheduling method, or it can combine machine learning models to predict the most matching template, ultimately determining the target template that matches the scheduling scenario. To improve matching accuracy, each template in the template library can include adaptation condition tags. The computer equipment completes precise matching based on the correspondence between the condition tags and the structured data fields.

[0053] After the target template is determined, the computer device performs a template filling operation. The template contains pre-defined field placeholders, such as "${device number}", "${operation content}", and "${execution time}". The computer device fills the corresponding field values ​​from the parsed structured scheduling data into the corresponding placeholders, dynamically generating a draft scheduling order. This process can be implemented through string replacement, template rendering engines, etc., to ensure that the final draft scheduling order conforms to a unified format specification and is readable and executable.

[0054] To ensure the completeness of the filling, the computer equipment can also perform field integrity verification after the template filling is completed. All required fields in the draft order are checked for non-emptiness and data type. If any missing or formatting errors are found, the draft is marked as "to be completed", prompting manual supplementation or re-parsing.

[0055] In this step, by performing structured parsing, template matching, and template filling operations on distribution network dispatching information in computer equipment, the automation and standardization of dispatching advance orders generation can be significantly improved, while reducing the risk of errors and omissions in the manual writing process. Structured parsing can accurately extract dispatching elements and improve data processing efficiency; template matching ensures a high degree of adaptability between the dispatching ticket format and the current dispatching task; and template filling enables the rapid generation of advance order content, thereby accelerating the dispatching instruction preparation process and enhancing response speed and system coordination capabilities. This method can effectively solve the problems of low efficiency, inconsistent formats, and reliance on manual experience in traditional dispatching ticket compilation, improving the manageability, traceability, and execution consistency of distribution network dispatching instructions, and thus enhancing the stability and reliability of the entire distribution network dispatching system.

[0056] S102: Use a pre-built structured grammar rule base to perform grammatical verification on the draft dispatch order. The structured grammar rule base is based on natural language processing technology and is constructed by extracting dispatch terms from the standard text of the distribution network dispatch, parsing the grammatical structure and transforming the operational logic relationship.

[0057] The structured syntax rule base refers to a set of rules pre-built and stored in computer equipment that conforms to grammatical and logical norms. Based on Natural Language Processing (NLP) technology, this rule base is constructed by extracting commonly used dispatching terms, syntactic structures, and corresponding operational logic relationships from the standard text clauses of power industry distribution network dispatching, and then using structured representation. Syntax validation refers to the automated review of the text content of the draft dispatching order by computer equipment using the structured syntax rule base to examine its format, word choice, logical relationships, etc., to determine whether it conforms to dispatching business specifications and industry expression standards.

[0058] Specifically, after generating a draft scheduling order, the computer equipment initiates a grammar rule matching process for that draft. The grammar verification is performed based on a structured grammar rule base, which includes legal expressions of scheduling actions and equipment combinations, logical templates for the sequence of operation steps, common scheduling terms, and sentence structures. During the verification process, the draft scheduling order is first segmented into sentences and words, with a lexical analyzer breaking down sentences into basic words and part-of-speech tags. Next, according to the structured grammar rules, the sentence structure is parsed in a tree structure, identifying syntactic components such as subject-verb-object structures, modifiers, and conditional clauses, and comparing them one by one with the grammar structure patterns in the rule base to determine if they match existing standard grammar patterns.

[0059] Subsequently, the computer equipment performs semantic and logical consistency checks. For example, if the draft contains expressions that do not conform to operational logic, such as "performing grounding operations before the circuit breaker is fully tripped," the system can automatically identify such conflicts and mark them as errors based on the operation sequence constraint model stored in the rule base. It can also perform standardization checks on terminology usage, such as identifying the expression "disconnecting the power switch," which does not conform to the dispatch terminology specification, as something that should be replaced with the standard term "tripping the switch," and providing suggested modifications.

[0060] To enhance processing efficiency and accuracy, a contextual semantic enhancement model can be combined to judge the meaning of complex sentences, such as whether there are implicit conditions or expressions that omit the subject, so as to ensure that the syntax check covers various scheduling expressions.

[0061] Therefore, by using computer equipment to call a structured grammar rule base to perform grammatical verification on the draft dispatch order, not only can grammatical errors, non-standard terminology, and operational logic conflicts in the statements be automatically identified, but the professionalism and consistency of dispatch instructions can also be ensured. This grammar verification process uses natural language processing technology to build a standard grammar rule framework, which can effectively cover the diverse expression patterns of instructions in distribution network dispatch scenarios. This improves the accuracy of the draft dispatch order in terms of language standardization and business logic, reduces the burden of manual review, reduces the execution risks caused by semantic deviations, and further enhances the reliability of dispatch order issuance and the security of on-site execution.

[0062] S103: When the syntax check of the draft scheduling advance order passes, calculate the text similarity between the draft scheduling advance order and the historical scheduling advance order, and perform similarity check on the draft scheduling advance order.

[0063] Historical dispatch orders refer to a collection of dispatch order texts that have been successfully issued and executed by on-site personnel according to instructions, possessing proven executability and standardization. Text similarity refers to the quantitative evaluation by computer equipment, using specific algorithms, of the similarity between two text contents in terms of structure, semantics, and terminology, used to measure the consistency between the current draft and existing standardized documents. Similarity verification refers to the computer equipment judging whether the draft text reaches a preset similarity threshold based on the similarity calculation results, and determining the standardization and rationality of the draft accordingly.

[0064] Specifically, once the draft dispatch order passes the syntax check, the computer equipment compares the text of the draft with the historical dispatch orders. A set of filtered historical order sample sets can be preloaded, which are classified and managed according to operation type, equipment type, scene tag, etc., so as to make corresponding comparisons with the draft.

[0065] Subsequently, computer equipment performs text preprocessing on each historical advance order in the draft and sample, including stop word removal, standardization of terminology, standardization of unit symbols, word segmentation, and part-of-speech tagging, to enhance comparison accuracy. Then, the preprocessed text is converted into vector representations using semantic embedding models (such as Word2Vec, BERT, etc.) or sentence vector models (such as Sentence-BERT), forming an intermediate data structure that facilitates similarity calculation.

[0066] Next, the similarity between the draft document vector and each historical document vector is calculated. Cosine similarity, Euclidean distance, Jaccard similarity, etc., can be used to measure the textual closeness between the two. The calculation results are compared with a set similarity threshold, and the similarity check is determined based on the comparison results. In other embodiments, similarity verification can employ various judgment strategies to improve accuracy. For example, by setting "high similarity," "medium similarity," and "low similarity" tiered thresholds, a tiered processing approach can be implemented, allowing for automatic approval, recommendation for modification, or submission for manual review of the draft document; or multiple matching historical documents can be selected to calculate the average similarity, comprehensively evaluating the consistency of the draft document; or similarity can be calculated separately for fields such as "operation steps," "equipment name," and "safety measures," and then weighted and summarized to identify specific sources of deviation, which helps improve the flexibility and accuracy of similarity verification and enhances the practical value of judging the normative nature of the draft document.

[0067] Therefore, by using computer equipment to perform text similarity calculation and similarity verification after the draft scheduling order has passed grammatical verification, we can not only use historically successfully executed scheduling orders as normative reference templates to effectively verify the rationality of the current draft in terms of semantic expression, logical structure and terminology, but also identify potential expression deviations or inconsistencies in advance, reducing the risk of on-site misoperation caused by non-standard information expression.

[0068] S104: If the similarity verification of the draft dispatch order passes, the draft dispatch order that passes the similarity verification will be used as the dispatch order, and when the dispatch order is confirmed to have passed the review, it will be released through the preset channels.

[0069] Among them, the dispatch advance order ticket refers to the final, official ticket that can be used for dispatch execution after the draft has undergone multi-level verification and approval. The pre-set channel refers to the data transmission channel with traceability and security guarantees set up in the dispatch system, including the dispatch master station system, field terminal equipment, SMS platform, or dispatch information publishing bus, etc., used to securely and reliably deliver official instructions to the designated recipient.

[0070] Specifically, when the computer equipment detects that the draft scheduling order has passed the similarity check, it first marks the draft as "passed" and generates an official ticket number to ensure that the draft has a unique identifier.

[0071] Following this, the review preparation process begins. Computer equipment can perform a combined assessment based on preset rules, evaluating factors such as the completeness of the document content, the consistency of the scheduling operation logic, and the compliance of security terminology. Alternatively, a manual review mode can be set up, in which case the computer equipment submits the draft content to an authorized dispatcher's account through the internal scheduling control platform, awaiting confirmation. Manual confirmation can be completed via a confirmation button, electronic signature, or other means.

[0072] After approval, a preset release strategy is invoked to forward the dispatch order content to the target receiving system through a designated output interface. For example, the release interface of the SCADA or DMS dispatch system can be called to push the order to the field operation terminal; or the order content can be pushed to the dispatch monitoring dashboard and instruction archiving system through an internal message queue, achieving a closed-loop release across the entire chain. This release process can be encrypted, logged, and equipped with fault retry mechanisms to ensure the security and timeliness of dispatch instruction release.

[0073] Furthermore, after detecting that the draft dispatch order has passed the similarity check, the computer equipment first marks it as "passed" and generates a unique order number. It then enters the review process, which can be automated according to preset rules, or the draft can be sent to the dispatcher's terminal for manual review. The dispatcher annotates and modifies the draft through the interface, confirms whether it passes the review, and records a complete review log. After approval, the order is sent to the executing unit's terminal according to the set release trigger conditions (such as automatic or manual release), and relevant personnel are notified via SMS, system messages, etc., while the release time is recorded. After release, the system collects feedback results, including whether it was received and the reason, ensuring a closed-loop and traceable dispatch process.

[0074] In this step, draft dispatch orders that pass similarity verification are directly incorporated into the formal order processing flow and released through preset channels after confirmation and approval. This standardizes and automates the generation, verification, review, and release of dispatch orders, significantly improving order circulation efficiency and reducing the frequency of manual intervention. By setting clear order status markers and release control logic, risks such as mis-issuance of unreviewed orders, duplicate orders, or lost instructions can be avoided. Simultaneously, the traceability and consistency of formal order content are ensured, thereby enhancing the operational security, response timeliness, and management standardization of the dispatch system and meeting the demand for highly reliable dispatch instructions in complex distribution network scenarios.

[0075] In the above embodiments, by parsing the obtained distribution network scheduling information and matching the corresponding template in the preset scheduling advance order template library, a draft scheduling advance order is formed, improving the standardization and normalization of the advance order generation process. Simultaneously, a structured grammar rule library based on natural language processing technology is introduced to perform grammatical verification on the draft scheduling advance order, effectively ensuring the semantic accuracy and logical rationality of the advance order content. Furthermore, by comparing the text similarity with historical scheduling advance orders, the rationality of the draft is further assessed, enhancing the consistency between the instruction content and actual operational experience. Finally, after approval, the scheduling advance order is released through preset channels, achieving efficient generation and reliable release of scheduling instructions. Therefore, this method can effectively alleviate the scheduling execution error problem caused by the lag in equipment status perception and improve the timeliness and accuracy of scheduling instructions.

[0076] In one embodiment, the step of parsing the acquired distribution network scheduling information includes:

[0077] Obtain distribution network scheduling information;

[0078] After removing redundant symbols from the distribution network dispatch information, word segmentation is performed, and multiple preset dispatch feature fields are extracted using regular expressions. The dispatch feature fields include operation type, voltage level, and equipment type.

[0079] Redundant symbols refer to punctuation marks, repetitive conjunctions, or formatting symbols in the dispatch information text that do not substantially contribute to the dispatching intent, such as redundant line breaks, spaces, and separators. Dispatch feature fields refer to key content fields extracted from the structured dispatch text through analysis, typically including operation type, voltage level, and equipment type. Operation type refers to the action category of distribution network dispatching behavior, such as closing and opening. Voltage level refers to the operating voltage of the equipment involved in the electrical operation, such as 10kV and 35kV. Equipment type refers to the classification name of the specific power equipment involved in the dispatch order, such as switches, transformers, and cable branch boxes.

[0080] Specifically, a data communication module can be configured in the distribution network dispatch support platform to interface with the distribution network OMS system. This module can use communication mechanisms such as RESTful API or WebSocket to listen for and receive dispatch command information and reclosing operation data pushed from the distribution network OMS platform in real time, ensuring that dispatch text related to operation can be captured as soon as possible.

[0081] Next, the acquired raw scheduling information is preprocessed. In this stage, the processing logic includes removing redundant symbols, such as invalid newlines, tabs, duplicate spaces, and abnormal punctuation, to standardize the text format and reduce interference from subsequent word segmentation. After deduplication, a Chinese natural language processing library (such as Jieba or a custom dictionary) is called to segment the text, breaking it down into the smallest linguistic units based on word meaning.

[0082] After word segmentation, the computer equipment enters the scheduling field extraction process. In this stage, multiple regular expression rules are pre-loaded. These rules are designed based on training with a large number of scheduling statement samples and correspond to the extraction of scheduling feature fields such as operation type, voltage level, and equipment type. For example, the operation type field can match keywords such as "closing," "opening," and "switching operation"; the voltage level field can match unit identifiers such as "10kV" and "20kV"; and the equipment type field can identify equipment terms such as "ring mains unit," "switching station," and "distribution transformer." In this way, key information fields can be accurately extracted from scheduling information with varying structures and diverse content.

[0083] In this embodiment, redundancy cleaning and word segmentation are performed on distribution network scheduling information using computer equipment, and key feature fields are extracted using regular expressions. This not only enables structured recognition of scheduling text but also significantly improves the accuracy and automation of subsequent scheduling ticket generation. Especially in the context of frequent issues such as inconsistent scheduling information formats and significant differences in expression during actual distribution network operation, this processing method effectively enhances the system's ability to understand scheduling intentions and reduces the need for manual intervention. Furthermore, the extraction of structured fields serves as a preliminary data foundation for ticket template matching, semantic retrieval, and risk verification, contributing to the construction of an intelligent and standardized scheduling auxiliary decision-making system and enhancing the overall intelligence and operational security of the scheduling system.

[0084] In one embodiment, the step of matching the corresponding target template in a pre-established scheduling advance order template library includes:

[0085] In the dispatch advance order template library, a candidate set of dispatch advance order templates is determined based on the operation type corresponding to the distribution network dispatch information;

[0086] Within the candidate set of dispatch advance order templates, a subset of candidate dispatch advance order templates is determined based on the voltage level corresponding to the distribution network dispatch information;

[0087] Within the candidate subset of dispatch pre-order ticket templates, the target template is determined based on the equipment type corresponding to the distribution network dispatch information.

[0088] Specifically, computer equipment extracts operation type fields from structured distribution network scheduling information, such as "power outage / restoration," "fault isolation," or "line maintenance." Using this operation type as the first-level matching condition, a subset of templates with the corresponding operation type tag is selected from the scheduling advance order template library, forming a candidate set of scheduling advance order templates. This step can achieve fast filtering by building an inverted index to ensure template retrieval efficiency even in high-concurrency scheduling scenarios.

[0089] Next, from the template candidate set obtained from the first layer of filtering, the voltage level field is further extracted to match the distribution network dispatch information. For example, when the voltage level in the dispatch information is "10kV", templates without the "10kV" tag will be filtered out from the above operation type matching templates, and only templates that match the voltage level will be retained, thus obtaining a candidate subset of dispatch advance order templates.

[0090] Then, the computer equipment performs a final filtering from the candidate template subset based on the equipment type field contained in the distribution network dispatch information. For example, when the equipment type field is "ABC line" and belongs to the "line" category, only templates marked as "line" are retained, thus determining the final target template. Taking "10kV ABC line power outage maintenance" as an example, this information is matched with the "line power outage and restoration" category template, and the placeholders used to fill the operation task field in the template are identified. "10kV ABC line power outage maintenance" is automatically filled into the field position, forming a draft dispatch order containing complete semantics.

[0091] In this embodiment, a multi-level precise matching strategy is adopted, sequentially filtering templates based on operation type, voltage level, and equipment type. This effectively narrows the search scope and improves template matching efficiency and accuracy. Through hierarchical comparison, the template that best matches the scheduling semantics and scenario conditions can be automatically matched, reducing manual intervention and significantly improving the consistency and standardization of draft scheduling orders. Simultaneously, the standardization of the template structure and the pre-setting of field placeholders allow scheduling information to be automatically filled into the corresponding fields of the target template, ensuring semantic integrity and standard format, which is beneficial for improving the automated generation capability of invoices and the subsequent approval rate. Therefore, this hierarchical filtering strategy not only improves the accuracy and automation level of invoice matching but also enhances the preparation efficiency before issuing scheduling commands, playing a positive role in ensuring the reliability and security of distribution network scheduling operations.

[0092] In one embodiment, the process of establishing a scheduling advance order template library includes:

[0093] Historical scheduling advance order data are categorized based on multiple preset scheduling feature fields;

[0094] Based on the scheduling feature fields corresponding to each category, generate a scheduling advance order template for each category. The scheduling advance order template is used to represent the scheduling rules.

[0095] Configure an identifier and version information for each scheduling advance ticket template, and when a change in scheduling rules is detected, locate and update the scheduling advance ticket template and its version information corresponding to the changed scheduling rules based on the changed scheduling rules and each identifier.

[0096] The dispatch feature fields refer to key fields extracted from distribution network dispatch information or historical dispatch advance orders that characterize the nature of the dispatch task, including but not limited to operation type, voltage level, and equipment type. The dispatch advance order template is a standardized structural format built based on the dispatch feature fields. It is a template file used to standardize the generation of dispatch advance order text content. The template has several pre-set fixed fields, such as operation task, issuer, recipient, advance order time, operation steps, and safety measures. The template content corresponds to specific dispatch rules. The template identifier is a unique number used to identify each template. The template version information identifies the specific version number of the template at different iteration stages. Together, they achieve dynamic management and version control of the template.

[0097] Specifically, computer equipment parses historical dispatch advance order data stored locally or in a database, extracting dispatch feature fields corresponding to each dispatch record. Specifically, dispatch records can be grouped and categorized according to operation type (e.g., equipment power outage / restoration, protection device activation / deactivation), voltage level (e.g., 10kV, 35kV), and equipment type (e.g., transformer, switch, ring main unit), constructing a multi-level classification system. This system allows dispatch data to be clearly divided into multiple subsets, providing a structured basis for template generation.

[0098] Next, the computer equipment generates standardized templates for each category based on the classification system. Taking "Equipment Power Outage / Restoration + 10kV + Ring Main Unit" as an example, the computer equipment will automatically create a template for this category, pre-setting fixed fields such as "Operation Task, Issuer, Recipient, Pre-order Time, Operation Steps, and Safety Measures" as the basic structure. These fixed fields reserve positions for information to be filled, allowing new scheduling data to be automatically populated into the corresponding fields of the template. The template generation process can be achieved by calling a pre-set template constructor, ensuring a uniform template structure and standardized content.

[0099] Furthermore, to achieve unified management of the template lifecycle, computer equipment configures a unique template identifier and initial version information for each generated template. For example, a combination of "template category + number + version number" can be used, such as "SD-TDD-10kV-HWG-V1.0". Based on this, a template version control mechanism can be deployed. When an update to the scheduling rules is detected, such as a change in industry standards or optimization of operational procedures, the template change process is automatically triggered. At this time, the template that needs to be updated is quickly located based on the template identifier, and the template content is regenerated according to the new scheduling rules. Simultaneously, the version information is upgraded to the new version, such as from V1.0 to V1.1, and historical versions are retained for retrospective queries.

[0100] Therefore, structuring historical data using a classification method based on scheduling feature fields helps to accurately construct a template classification system, improving the relevance and adaptability of template generation. Furthermore, by designing standardized fixed fields for each template category, the template structure becomes uniform, ensuring the completeness and executability of generated scheduling advance orders. In addition, setting up a template version control mechanism allows template content to be updated synchronously as scheduling rules change, avoiding execution deviations caused by outdated template content. This not only improves the organizational efficiency and management flexibility of the template library but also ensures template adaptability and compliance, thus providing strong support for the high-quality generation of subsequent scheduling advance orders.

[0101] In one embodiment, the process of building a structured syntax rule base includes:

[0102] Using a pre-trained named entity recognition model, multiple scheduling terms in the standard text of distribution network scheduling are identified, and a dependency parsing algorithm is used to extract the syntactic dependency relationships between the scheduling terms and construct syntactic structure information.

[0103] Multiple high-frequency terms are obtained from the standard text terms of distribution network scheduling and historical scheduling records. The semantic equivalence and causal relationships between these high-frequency terms are determined, and a term mapping relationship table is constructed.

[0104] A sequence pattern mining algorithm is used to extract multiple step sequence patterns from the collected historical scheduling orders, and each step sequence pattern is converted into a corresponding logical expression.

[0105] A structured grammar rule base is formed based on grammatical structure information, term mapping relationship table, and various logical expressions.

[0106] Among them, the standard text clauses for distribution network dispatching refer to standardized written text documents issued by the power industry regulatory authority or dispatching unit to regulate the language expression, operation procedures, and safety requirements of distribution network dispatching instructions. Named entity recognition models are machine learning models based on natural language processing technology used to identify words with specific business meanings (such as equipment names, voltage levels, and substation names) in distribution network texts, preferably models trained using the BERT-CRF structure. Dependency parsing algorithms are algorithms that analyze the grammatical relationships between words in syntactic structures, such as identifying verb-object, subject-verb relationships between verbs and nouns. Terminology mapping tables are sets of term substitutions and associations constructed by statistically analyzing the semantic and causal relationships of terms in standard texts and historical records. Sequence pattern mining algorithms are algorithms used to identify the temporal relationships between dispatching steps, such as PrefixSpan and SPADE, used to extract the common step sequences in dispatching tasks.

[0107] Specifically, the computer equipment preprocesses the standard text clauses of the distribution network dispatching system using natural language processing technology. Specifically, the computer equipment segments the clause content based on a power industry-specific dictionary and loads a pre-trained named entity recognition model. This model, employing a BERT-CRF structure, can identify entity information such as substation names, voltage levels, equipment types, and serial numbers within the clauses. For example, in the clause "open the 10kV Dongbian 101 circuit breaker," the equipment identifies "open" as an operational verb and "10kV Dongbian 101 circuit breaker" as a combined entity, then breaks down and categorizes their structures.

[0108] Next, dependency parsing is performed on the segmented and identified text content. The dependency parsing algorithm identifies the grammatical relationships between verbs and nouns, such as verb-object relationships and subject-verb relationships. Taking "open a 10kV circuit breaker" as an example, the relationship between "open" and "circuit breaker" is marked as a verb-object relationship. Based on the electrical grammar rules, "open" is classified as a "isolation operation" verb, and "circuit breaker" is classified as an "isolated device" type, thus forming standardized grammatical structure rules.

[0109] Then, the computer equipment extracts multiple high-frequency terms from standard clauses and historical scheduling records using a high-frequency term statistical algorithm. It then uses word vector semantic matching technology to identify equivalence and causal relationships between terms, such as "disconnect" ≈ "pull apart," and "power on" being the reverse operation after "power outage." Based on this, a term mapping table is constructed to support unified parsing and standardized processing between different expressions.

[0110] Subsequently, the computer equipment collects a large number of executed tickets from the historical dispatch order database and uses a sequence pattern mining algorithm to extract the step sequence patterns. For example, when a substation device transitions from the "operating" state to the "maintenance" state, the algorithm can automatically identify the timing rule of "disconnecting the circuit breaker → opening the load-side disconnect switch → opening the power-side disconnect switch" and convert this sequence pattern into a logical expression, such as S1 -> S2 -> S3, to indicate that the above steps must be completed in sequence. Based on this, each operation step can be further abstracted into a "node," such as "disconnecting the circuit breaker" and "opening the load-side disconnect switch," which are two nodes, and arrows are used to indicate their execution order. At the same time, with the help of simple conditional logic structures, such as "if the circuit breaker is disconnected, the load-side disconnect switch can be opened," and expressed using conditional statements (such as if-else) in programming languages, the visual modeling and programmatic execution of the dispatch operation process can be realized.

[0111] Finally, the grammatical structure information generated by dependency analysis, the semantic mapping relationships between high-frequency terms, and the logical expressions of the step sequence are uniformly arranged and formatted into a structured grammar rule base. This grammar rule base can be stored in JSON, XML, or relational database formats and serves as an important foundational resource for subsequent scheduling advance order syntax verification, ticket generation, and automated approval.

[0112] After the grammatical structure information, terminology mapping relationships, and step logic expressions are structurally organized, the computer equipment further performs sentence-by-sentence verification on the automatically generated draft dispatch order. This verification process, based on the aforementioned constructed grammar rule base and terminology mapping relationship table, performs a series of structured matching and semantic consistency judgment tasks. These include checking whether the equipment name conforms to the combination format of "substation name + voltage level + equipment type + equipment number," and verifying the correctness of the dependency relationships between operation verbs and equipment nouns. If any missing equipment descriptions, incorrect ordering, or inappropriate verb-equivalent pairings (such as "connect circuit breaker") are found, errors are automatically marked, and modification suggestions are generated based on the terminology mapping relationships (e.g., suggesting "close circuit breaker"), prompting the dispatcher to check the draft dispatch order. Once all statements are not marked as errors or the dispatcher completes the suggested modifications and confirms that everything is correct, the draft is considered to have passed the grammar rule base verification, ensuring its structural standardization and terminology standardization.

[0113] Therefore, by using named entity recognition models and dependency parsing algorithms to structurally model the terms and grammatical relationships in the scheduling standard text clauses, not only can the core components of the scheduling language be accurately extracted, but the logical connections between words can also be revealed. Furthermore, by combining term mapping relationship tables and scheduling sequence logic rules, the semantics of the text and the execution logic are integrated, making the grammatical rules uniform and executable. The resulting structured grammatical rule library provides reusable, scalable, and automatically adaptable knowledge support for scheduling advance order generation, compliance verification, and language standardization, thereby significantly improving the intelligent processing capabilities of distribution network scheduling and the quality control level of scheduling instructions.

[0114] In one embodiment, the step of determining semantic equivalence and causal relationships between various high-frequency terms includes:

[0115] Using word vector models, we can identify semantic equivalence relationships between high-frequency terms.

[0116] Temporal correlation analysis was used to determine the causal relationships between various high-frequency terms.

[0117] High-frequency terms refer to key dispatching terms that appear more frequently than a preset threshold in distribution network dispatching standard clauses and historical dispatching records. These typically include equipment terms (such as "switch" and "circuit breaker") and status terms (such as "power outage operation" and "maintenance"). A word vector model is a language model that maps words to vector representations in a low-dimensional continuous vector space, preferably the Word2Vec model, used to calculate semantic similarity between words. Temporal correlation analysis is an analytical method that mines potential causal relationships between terms based on the chronological order of events in a large number of historical dispatching statements; it is suitable for causal inference of status terms.

[0118] Specifically, the computer equipment loads the standard text terms of the distribution network dispatching system and historical dispatching records. Based on word segmentation statistics, stop word filtering, and word frequency calculation, it filters terms and extracts representative equipment terms and status terms. For example, in a large amount of text, terms such as "circuit breaker," "switch," "maintenance," and "power outage operation" are identified as candidate high-frequency terms.

[0119] Next, the computer device loads a pre-trained Word2Vec word vector model, represents the extracted high-frequency terms with word vectors, calculates the cosine similarity between any two terms, and determines whether they are semantically equivalent based on a set threshold. For example, when the cosine similarity between "switch" and "circuit breaker" is greater than a set threshold (such as 0.85), they are considered to have a semantic equivalence relationship and are temporarily stored in the equivalent term candidate set.

[0120] Subsequently, the computer equipment performs temporal correlation analysis on the co-occurrence relationships between state terms. Specifically, the computer equipment traverses the sequence of operations related to state in historical scheduling tickets, such as "power outage operation → voltage testing → equipment maintenance," and uses sequence association mining algorithms such as Apriori and PrefixSpan to identify high-frequency paths between state terms. Combining this with the order of occurrence, causal pairs of "preceding term → subsequent term" are established. For example, from a large number of records, it is found that "power outage operation" is usually performed before "maintenance," thus marking it as a "causal term pair."

[0121] Next, the computer equipment organizes the identified equivalent term pairs and causal term pairs into an initial term mapping table. This initial table is then output by the computer equipment to a manual review interface, where experts with power dispatching knowledge confirm and revise the term relationships, eliminating unreasonable or highly ambiguous term combinations. After the review is completed, the computer equipment assigns a unique term ID to the approved term pairs and stores them in the term mapping table. This table will serve as an important basis for subsequent dispatching advance order generation, semantic disambiguation, and automatic matching.

[0122] Therefore, by automatically calculating the semantic similarity between high-frequency terms using word vector models, equivalent variants in language expression can be effectively discovered, improving term normalization capabilities. By identifying the sequential logical relationships between state terms through temporal correlation analysis, the actual execution order of the scheduling process can be accurately restored. This mapping table plays a role in the generation and verification of scheduling advance orders, helping to improve semantic consistency, reduce term ambiguity, and ultimately enhance the intelligence level and reliability of the automatic scheduling ticket generation system.

[0123] In one embodiment, the step of performing a similarity check on the draft scheduling advance order includes:

[0124] If the text similarity of the draft dispatch order is higher than the preset threshold, then the similarity check of the draft dispatch order is deemed to have passed.

[0125] If the text similarity of the draft dispatch order is not higher than the preset threshold, an early warning message is generated to remind manual review, and when a review confirmation instruction is received, the similarity check of the draft dispatch order is determined to be passed.

[0126] The preset threshold refers to the critical value used to determine whether the similarity meets the standard, preferably set to 70%. The warning message is a prompt message generated by the system when the text similarity is lower than the threshold, used to prompt the dispatcher to intervene for manual review.

[0127] Specifically, the computer equipment retrieves a set of dispatch advance order texts marked as "successfully executed" from the historical database as a standard reference sample set. After the draft dispatch advance order is generated, the content of the draft is compared one-to-one with the content of the historical successfully executed tickets. TF-IDF or BERT vector representation technology is preferably used to represent the draft and reference texts as vectors, and the similarity score is calculated based on the cosine similarity formula.

[0128] Next, the computer device compares the obtained text similarity score with a preset threshold (such as 70%). If the text similarity is higher than 70%, the device automatically marks the draft as "verification passed" and writes it into the scheduling advance ticket management database for subsequent review or release.

[0129] If the text similarity is not higher than the threshold, an early warning mechanism is automatically triggered, generating an early warning message containing the similarity score, difference point prompts, and the corresponding historical invoice number. This message is then pushed to the dispatcher's workbench interface, guiding the dispatcher to conduct a manual review. The dispatcher can review the original draft and similar reference invoices to check for any compilation errors or semantic deviations. Once the dispatcher confirms the draft content is correct and clicks the "Review Confirm" command, the computer receives the command, updates the draft to the "Verification Passed" status, and records the manual confirmation operation log for dispatch audit tracking.

[0130] Therefore, by comparing the text similarity of the draft dispatch order with historical dispatch orders, the standardization and rationality of the draft text can be quickly assessed using historical experience data. By setting thresholds and introducing a manual review mechanism, misjudgments caused by mismatches or fine-tuning of dispatch strategies can be effectively avoided. Automatically triggering warnings improves the efficiency of manual review and prevents drafts with low similarity from directly entering the release process. In this way, the flexibility and professionalism of manual judgment are combined with intelligent judgment, thereby achieving an effective balance between automatic verification of dispatch orders and human error tolerance.

[0131] The distribution network dispatch advance order issuing device provided in the embodiments of this application is described below. The distribution network dispatch advance order issuing device described below can be referred to in correspondence with the distribution network dispatch advance order issuing method described above. Figure 2 As shown, this application provides a distribution network dispatch advance order issuing device, the device comprising:

[0132] The dispatch advance order draft determination module 201 is used to parse the obtained distribution network dispatch information, match the corresponding target template in the pre-established dispatch advance order template library, and fill the parsed distribution network dispatch information into the target template to obtain the dispatch advance order draft.

[0133] The scheduling advance order draft syntax verification module 202 is used to perform syntax verification on the scheduling advance order draft using a pre-built structured syntax rule base. The structured syntax rule base is based on natural language processing technology and is constructed by extracting scheduling terms from the standard text clauses of the distribution network scheduling, parsing the syntax structure and transforming the operational logic relationship.

[0134] The scheduling advance order draft similarity verification module 203 is used to calculate the text similarity between the scheduling advance order draft and historical scheduling advance orders when the syntax verification of the scheduling advance order draft passes, and to perform similarity verification on the scheduling advance order draft.

[0135] The scheduling advance order release module 204 is used to release the scheduling advance order draft that has passed the similarity verification as a scheduling advance order if the similarity verification is passed, and release it through a preset channel when the scheduling advance order is confirmed to have passed the review.

[0136] In one embodiment, the scheduling advance ticket draft determination module 201 includes:

[0137] The distribution network dispatch information acquisition module is used to acquire distribution network dispatch information;

[0138] The distribution network dispatch information parsing module is used to remove redundant symbols from the distribution network dispatch information and perform word segmentation. It also extracts multiple preset dispatch feature fields using regular expressions. The dispatch feature fields include operation type, voltage level, and equipment type.

[0139] In one embodiment, the scheduling advance ticket draft determination module 201 includes:

[0140] The dispatch advance order template candidate set determination unit is used to determine the dispatch advance order template candidate set in the dispatch advance order template library according to the operation type corresponding to the distribution network dispatch information;

[0141] The dispatch advance order template candidate subset determination unit is used to determine the dispatch advance order template candidate subset from the dispatch advance order template candidate set according to the voltage level corresponding to the distribution network dispatch information;

[0142] The target template determination unit is used to determine the target template from the candidate subset of dispatch pre-order ticket templates based on the equipment type corresponding to the distribution network dispatch information.

[0143] In one embodiment, the scheduling advance ticket draft determination module 201 includes:

[0144] The historical dispatch advance order classification unit is used to classify historical dispatch advance order data based on multiple preset dispatch feature fields;

[0145] The scheduling advance order template generation unit is used to generate a scheduling advance order template for each category according to the scheduling feature fields corresponding to each category. The scheduling advance order template is used to represent the scheduling rules.

[0146] The scheduling advance order template update unit is used to configure an identifier and version information for each scheduling advance order template, and when a change in scheduling rules is detected, it locates and updates the scheduling advance order template and its version information corresponding to the changed scheduling rules based on the changed scheduling rules and each identifier.

[0147] In one embodiment, the scheduling advance order draft syntax verification module 202 includes:

[0148] The syntactic structure information construction unit is used to identify multiple scheduling terms in the standard text clauses of the distribution network scheduling using a pre-trained named entity recognition model, and to extract the syntactic dependency relationships between the scheduling terms using a dependency analysis algorithm to construct syntactic structure information.

[0149] The terminology mapping table construction unit is used to obtain multiple high-frequency terms from the standard text clauses of the distribution network scheduling and historical scheduling records, determine the semantic equivalence and causal relationships between the various high-frequency terms, and construct the terminology mapping table.

[0150] The logical expression construction unit is used to extract multiple step sequence patterns from the collected historical scheduling pre-order tickets using a sequence pattern mining algorithm, and to convert each step sequence pattern into a corresponding logical expression.

[0151] The structured syntax rule base building unit is used to form a structured syntax rule base based on syntax structure information, term mapping relationship table and various logical expressions.

[0152] In one embodiment, the term mapping table construction unit includes:

[0153] The semantic equivalence relation identification subunit is used to identify semantic equivalence relations between various high-frequency terms using a word vector model;

[0154] The causal relationship determination subunit is used to determine the causal relationships between various high-frequency terms using time-series correlation analysis.

[0155] In one embodiment, the scheduling advance order draft similarity verification module 203 includes:

[0156] The first scheduling advance order draft similarity verification unit is used to determine that the similarity verification of the scheduling advance order draft is passed if the text similarity of the scheduling advance order draft is higher than a preset threshold.

[0157] The second scheduling advance order draft similarity verification unit is used to generate a warning message to remind manual review if the text similarity of the scheduling advance order draft is not higher than a preset threshold, and to determine that the similarity verification of the scheduling advance order draft has passed when a review confirmation instruction is received.

[0158] In one embodiment, this application also provides a storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the distribution network dispatch advance order issuance method as described in any of the above embodiments.

[0159] In one embodiment, this application also provides a computer device storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the distribution network scheduling advance order issuance method as described in any of the above embodiments.

[0160] Indicatively, such as Figure 3 As shown, Figure 3 This is a schematic diagram of the internal structure of a computer device 300 provided in an embodiment of this application. The computer device 300 can be provided as a server. (Refer to...) Figure 3 The computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by memory 301 for storing instructions, such as application programs, that can be executed by the processing component 302. The application programs stored in memory 301 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 302 is configured to execute instructions to perform the distribution network scheduling advance order issuance method of any of the above embodiments.

[0161] The computer device 300 may also include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate on an operating system stored in memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0162] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0163] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this document, "a," "an," "the," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. "Multiple" refers to at least two, such as 2, 3, 5, or 8, etc. "And / or" includes any and all combinations of the related listed items.

[0164] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0165] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for issuing advance orders for distribution network dispatching, characterized in that, The method includes: After parsing the obtained distribution network scheduling information, the corresponding target template is matched in the pre-established scheduling advance order template library, and the parsed distribution network scheduling information is filled into the target template to obtain the draft scheduling advance order. The draft scheduling order is grammatically validated using a pre-built structured grammar rule base. This structured grammar rule base is constructed based on natural language processing technology by extracting scheduling terms from standard distribution network scheduling texts, parsing the grammatical structure, and transforming operational logic relationships. When the draft scheduling advance order passes the syntax check, the text similarity between the draft scheduling advance order and the historical scheduling advance order is calculated, and the similarity check is performed on the draft scheduling advance order. If the similarity verification of the draft scheduling advance order passes, the draft scheduling advance order that passes the similarity verification will be used as the scheduling advance order, and when the scheduling advance order is confirmed to have passed the review, it will be released through the preset channels.

2. The method for issuing advance orders for distribution network dispatching according to claim 1, characterized in that, The step of parsing the acquired distribution network scheduling information includes: Obtain the distribution network scheduling information; After removing redundant symbols from the distribution network scheduling information, word segmentation is performed, and multiple preset scheduling feature fields are extracted using regular expressions. The scheduling feature fields include operation type, voltage level, and equipment type.

3. The distribution network dispatch advance order issuance method according to claim 2, characterized in that, The step of matching the corresponding target template in the pre-established scheduling advance order template library includes: In the dispatch advance order template library, a candidate set of dispatch advance order templates is determined according to the operation type corresponding to the distribution network dispatch information; Within the candidate set of dispatch advance order templates, a subset of candidate dispatch advance order templates is determined based on the voltage level corresponding to the distribution network dispatch information; Within the candidate subset of the dispatch advance order template, the target template is determined based on the equipment type corresponding to the distribution network dispatch information.

4. The method for issuing distribution network dispatch advance orders according to any one of claims 1 to 3, characterized in that, The process of establishing the scheduling advance order template library includes: Historical scheduling advance order data are categorized based on multiple preset scheduling feature fields; Based on the scheduling feature fields corresponding to each category, a scheduling advance order template is generated for each category, and the scheduling advance order template is used to represent the scheduling rules. Each scheduling advance ticket template is configured with an identifier and version information. When a change in scheduling rules is detected, the scheduling advance ticket template and its version information corresponding to the changed scheduling rules are located and updated based on the changed scheduling rules and each of the identifiers.

5. The method for issuing advance orders for distribution network dispatching according to claim 1, characterized in that, The construction process of the structured syntax rule base includes: Using a pre-trained named entity recognition model, multiple scheduling terms in the standard text of the distribution network scheduling are identified, and a dependency analysis algorithm is used to extract the syntactic dependency relationships between the scheduling terms to construct syntactic structure information. Multiple high-frequency terms are obtained from the standard text terms of the distribution network scheduling and historical scheduling records, and the semantic equivalence and causal relationships between the various high-frequency terms are determined to construct a term mapping relationship table. A sequence pattern mining algorithm is used to extract multiple step sequence patterns from the collected historical scheduling advance orders, and each step sequence pattern is converted into a corresponding logical expression. The structured syntax rule base is formed based on the grammatical structure information, the terminology mapping table, and each of the logical expressions.

6. The method for issuing advance orders for distribution network dispatching according to claim 5, characterized in that, The step of determining the semantic equivalence and causal relationships among the various high-frequency terms includes: Using word vector models, semantic equivalence relationships between the high-frequency terms are identified; Temporal correlation analysis was used to determine the causal relationships between the various high-frequency terms.

7. The method for issuing advance orders for distribution network dispatching according to claim 1, characterized in that, The step of performing similarity verification on the draft scheduling order includes: If the text similarity of the draft scheduling advance order is higher than a preset threshold, then the similarity verification of the draft scheduling advance order is determined to be passed; If the text similarity of the draft scheduling order is not higher than a preset threshold, an early warning message is generated to remind manual review, and when a review confirmation instruction is received, the similarity check of the draft scheduling order is determined to be passed.

8. A distribution network dispatch advance order issuing device, characterized in that, The device includes: The dispatch advance order draft determination module is used to parse the obtained distribution network dispatch information, match the corresponding target template in the pre-established dispatch advance order template library, and fill the parsed distribution network dispatch information into the target template to obtain the dispatch advance order draft. The dispatch advance order draft syntax verification module is used to perform syntax verification on the dispatch advance order draft using a pre-built structured syntax rule base. The structured syntax rule base is based on natural language processing technology and is constructed by extracting dispatch terms from the standard text clauses of the distribution network dispatch, parsing the syntax structure and transforming the operational logic relationship. The scheduling advance order draft similarity verification module is used to calculate the text similarity between the scheduling advance order draft and historical scheduling advance orders when the syntax verification of the scheduling advance order draft passes, and to perform similarity verification on the scheduling advance order draft. The scheduling advance order release module is used to release the scheduling advance order draft that has passed the similarity verification as a scheduling advance order if the similarity verification is passed, and to release it through a preset channel when the scheduling advance order is confirmed to have passed the review.

9. A storage medium, characterized in that: The storage medium stores computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the distribution network dispatch advance order issuance method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the distribution network dispatch advance order issuance method as described in any one of claims 1 to 7.

Images