Search

CN-114640179-B - Networked order interaction system for distribution network

CN114640179BCN 114640179 BCN114640179 BCN 114640179BCN-114640179-B

Abstract

The application discloses a networked order issuing interaction system for a distribution network, which relates to the technical field of distribution network regulation and control, and comprises a factory end, an operation and maintenance end, a dispatching end and a network order issuing end which is respectively in communication connection with the factory end, the operation and maintenance end and the dispatching end, wherein the dispatching end is used for receiving instruction information input by a dispatcher, generating a pre-instruction and a positive instruction according to the instruction information and uploading the pre-instruction and the positive instruction to the network order issuing end, the network order issuing end is used for receiving the pre-instruction and the positive instruction generated and uploaded by the dispatching end and simultaneously issuing the pre-instruction and the positive instruction to the factory end and the operation and maintenance end, the factory end is used for executing the pre-instruction and the positive instruction issued by the network order issuing end, and the dispatching end is used for receiving the pre-instruction and the positive instruction issued by the network order issuing end. The application has the effects of realizing the construction of a networked dispatching instruction information interaction management mode, improving the real-time performance and quality of on-site operation and maintenance dispatching management and control operation interaction, promoting the close cooperation of dispatching and operation detection and improving the dispatching network regulation and control work efficiency.

Inventors

  • MA XIANG
  • CHEN DANFEI
  • WU MINMIN
  • HUANG JIANFENG
  • ZHU YONGMING
  • CHEN TIEYI
  • WU HUAHUA
  • GU WEI
  • XU YONG
  • Lv Leiyan
  • ZHENG XIANG
  • SONG XIN
  • FEI YONGPAN

Assignees

  • 国网浙江省电力有限公司双创中心
  • 国网浙江省电力有限公司双创中心
  • 国网浙江省电力有限公司金华供电公司
  • 国网浙江省电力有限公司金华供电公司

Dates

Publication Date
20260421
Application Date
20220330
Priority Date
20220330

Claims (8)

  1. 1. The networked order interaction system for the distribution network is characterized by comprising a factory end, an operation and maintenance end, a scheduling end and a network order sending end which is respectively in communication connection with the factory end, the operation and maintenance end and the scheduling end; the dispatching end is used for receiving instruction information input by a dispatcher, generating a pre-instruction and a positive instruction according to the instruction information, and uploading the pre-instruction and the positive instruction to the network command-sending end; The network command end is used for receiving the pre-command and the positive command generated and uploaded by the scheduling end and simultaneously transmitting the pre-command and the positive command to the factory end and the operation and maintenance end; the factory side is used for executing the pre-instruction and the positive instruction issued by the network issuing side; The operation and maintenance end is used for receiving the pre-instruction and the positive instruction issued by the network issuing end; the method for generating the positive instruction and the pre-instruction of the scheduling end comprises the following steps: s1, acquiring regulation and control voice information; S2, inputting the regulation and control voice information into a pre-trained regulation and control instruction generation model to obtain a regulation and control instruction to be transmitted, wherein the regulation and control instruction corresponds to the regulation and control voice information; s3, determining a target user identifier corresponding to the regulation voice information; s4, generating a target regulation instruction based on the target user identification and the regulation instruction to be sent; in S2, the building of the regulation instruction generation model includes the following steps: step 1, collecting dispatching voice corpus, carrying out noise reduction treatment and feature sequence extraction on the collected dispatching voice corpus, and using the extracted feature sequence to construct a dispatching voice corpus, wherein the dispatching voice corpus comprises a general corpus, an electric professional corpus and standard voice signals for matching degree test; Step 2, taking the dispatching voice corpus as input data, and inputting the dispatching voice corpus into an acoustic model and a language model which finish training, wherein the acoustic model and the language model finish training jointly output dispatching telephone texts, and the acoustic model and the language model finish training take the dispatching voice corpus as a training set and adopt a deep learning algorithm for training; step 3, collecting a regulation instruction corpus, carrying out vectorization processing on the regulation instruction corpus words, and constructing a regulation instruction corpus by using the obtained regulation instruction word vectors; Outputting a domain matching result by the domain recognition model, outputting an intention recognition result by the intention recognition model according to the domain matching result, and training the domain recognition model and the intention recognition model which finish training by taking a regulation instruction corpus as a training set and adopting a deep learning algorithm; Step 5, taking the intention recognition result as input data of a slot recognition model, outputting a regulation and control instruction matching result by the slot recognition model, and extracting a semantic tag from the scheduling instruction by the slot recognition model through a rule analysis algorithm; Step 6, according to the matching result of the regulation and control instruction, the instruction is definitely scheduled to prepare for carrying out power scheduling, and the running state of the power grid is determined by combining the power grid power flow calculation, the short-circuit current calculation and the switching-on and switching-off operation; and 7, inputting the matching results of the dispatching telephone text and the regulation instruction into a voice synthesis engine, converting the dispatching telephone text and the regulation instruction into dispatching telephone voice through the voice synthesis engine, and automatically responding to the dispatching telephone by using the dispatching telephone voice.
  2. 2. The networked order interaction system for the distribution network of claim 1, wherein the scheduling end comprises a pre-instruction processing module, a positive instruction processing module and an instruction signing module; the pre-instruction processing module consists of a pre-instruction generating unit and a pre-instruction issuing unit, wherein the pre-instruction generating unit is used for processing instruction information to generate pre-instruction information, and the pre-instruction issuing unit is used for sending the pre-instruction information to the instruction signing module; The positive instruction processing module consists of a positive instruction generating unit and a positive instruction issuing unit, wherein the positive instruction generating unit is used for processing instruction information to generate positive instruction information, and the positive instruction issuing unit is used for sending the positive instruction information to the instruction signing module; The instruction signing module is used for receiving the pre-instruction information and the positive instruction information, generating a pre-instruction and a positive instruction according to the pre-instruction information and the positive instruction information, and uploading the pre-instruction and the positive instruction to the network instruction sending terminal.
  3. 3. The networked order interaction system for the distribution network of claim 2, wherein the order signing module comprises an order repeating unit, an order confirming unit, a pre-order signing executing unit and a positive order signing executing unit; the instruction repeating unit is used for repeating the pre-instruction information and the positive instruction information; the instruction confirmation unit is used for confirming the pre-instruction information and the positive instruction information which are repeated by the instruction repeating unit, sending the pre-instruction information to the pre-instruction signing and executing unit and sending the positive instruction information to the positive instruction signing and executing unit; The pre-instruction signing and executing unit is used for generating a pre-instruction of the factory equipment according to the pre-instruction information and uploading the pre-instruction to the network command transmitting end; The positive instruction signing and executing unit is used for generating a positive instruction of the factory equipment according to the positive instruction information and uploading the positive instruction to the network issuing end.
  4. 4. The networked order interaction system for a distribution network of claim 1, wherein in S1, the control voice information is issued by a control center worker and includes information of a power system control instruction, and the power system control instruction is an instruction related to operation and operation of a power system.
  5. 5. The networked order interaction system for a distribution network according to claim 1, wherein the step 2 comprises: step 2.1, taking the dispatching voice corpus as input data, and inputting the dispatching voice corpus into an acoustic model and a language model which are trained; step 2.2, performing noise processing and acoustic feature extraction on the modulated voice corpus by the trained acoustic model to output an acoustic feature sequence; step 2.3, the trained language model carries out noise processing and semantic feature extraction on the dispatching voice corpus to output a semantic feature sequence; Step 2.4, the feature sequence with the highest comprehensive score of the acoustic model and the language model is used as a dispatch telephone feature sequence; And 2.5, carrying out language decoding on the dispatch telephone feature sequence based on the power professional corpus to obtain a dispatch telephone text.
  6. 6. A networked order interaction system for a distribution network according to claim 5 is characterized in that in step 2.2, standard voice signals are used as matching degree test input data and are input into a trained acoustic model, a standard acoustic feature sequence is output by the acoustic model, matching degree of the acoustic feature sequence and the standard acoustic feature sequence is calculated based on a distance algorithm, in step 2.3, the standard voice signals are used as matching degree test input data and are input into the trained language model, the standard semantic feature sequence is output by the language model, and matching degree of the semantic feature sequence and the standard semantic feature sequence is calculated based on a distance algorithm.
  7. 7. The networked order interaction system for the distribution network of claim 1, wherein step 6 comprises integrating and summarizing equipment tide, power grid events, operation modes, risk early warning, equipment maintenance, accident plans and weather information, inquiring and watching various power information in the same interface, realizing report self-customization, picture self-generation and telephone self-response functions by interacting with a regulation knowledge base, and realizing intelligent ticket formation of operation tickets, automatic execution of operation tickets and confirmation of switch positions according to a switching operation knowledge map and a switch position image recognition result.
  8. 8. The networked order interaction system for the distribution network of claim 1, wherein in S3, the target user identifier is an identifier corresponding to a user who issues the control voice, and the corresponding target user identifier is used for determining the associated information of the corresponding target user.

Description

Networked order interaction system for distribution network Technical Field The application relates to the technical field of distribution network regulation and control, in particular to a networked order interaction system for a distribution network. Background The distribution network is composed of overhead lines, cables, towers, distribution transformers, isolating switches, reactive compensators, some auxiliary facilities and the like, and plays a role in distributing electric energy in the power network. The power dispatching is an effective management means for ensuring safe and stable operation of the power grid, external reliable power supply and orderly execution of various power production works. Based on the technologies of computer network, intelligent telephone voice, mobile websocket+node and the like, the scheduling instruction interaction between the scheduling station/the distribution network can be realized directly through the modes of electronic digitalization, message reminding and the like. The station user can sign and receive pre-orders, forward and recite, apply for starting work, finish and report, abnormal equipment and report power grid fault information through the three-area network WEB application and the mobile terminal, and the dispatching user can give orders, permit, confirm and the like through a text asynchronous mode, so that intelligent reminding means such as task ringing, instant pushing and the like are realized, and the quality and efficiency of interaction between station personnel and dispatcher are improved to the greatest extent. However, due to poor real-time performance and quality of information interaction of on-site operation and maintenance scheduling management and control operation instructions, scheduling and operation and inspection are difficult to cooperate closely, and further the distribution network regulation and control working efficiency is affected, and the improvement is needed. Disclosure of Invention Therefore, the application aims to provide a networked order interaction system for a distribution network, which promotes close cooperation of scheduling and operation detection by improving the real-time performance and quality of on-site operation, maintenance, scheduling and control operation interaction, thereby achieving the purpose of improving the work efficiency of distribution network regulation. The specific scheme is as follows: A networked order interaction system for a distribution network comprises a factory end, an operation and maintenance end, a scheduling end and a network order sending end which is respectively in communication connection with the factory end, the operation and maintenance end and the scheduling end; the dispatching end is used for receiving instruction information input by a dispatcher, generating a pre-instruction and a positive instruction according to the instruction information, and uploading the pre-instruction and the positive instruction to the network command-sending end; The network command end is used for receiving the pre-command and the positive command generated and uploaded by the scheduling end and simultaneously transmitting the pre-command and the positive command to the factory end and the operation and maintenance end; the factory side is used for executing the pre-instruction and the positive instruction issued by the network issuing side; the scheduling end is used for receiving the pre-instruction and the positive instruction issued by the network issuing end. Preferably, the scheduling end comprises a pre-instruction processing module, a positive instruction processing module and an instruction signing module; the pre-instruction processing module consists of a pre-instruction generating unit and a pre-instruction issuing unit, wherein the pre-instruction generating unit is used for processing instruction information to generate pre-instruction information, and the pre-instruction issuing unit is used for sending the pre-instruction information to the instruction signing module; The positive instruction processing module consists of a positive instruction generating unit and a positive instruction issuing unit, wherein the positive instruction generating unit is used for processing instruction information to generate positive instruction information, and the positive instruction issuing unit is used for sending the positive instruction information to the instruction signing module; The instruction signing module is used for receiving the pre-instruction information and the positive instruction information, generating a pre-instruction and a positive instruction according to the pre-instruction information and the positive instruction information, and uploading the pre-instruction and the positive instruction to the network instruction sending terminal. Preferably, the instruction signing and receiving module consists of an instruction repeating unit, an instruction confirming unit,