CN-121996951-A - Hydropower station work ticket handling condition identification method and system based on multi-source information fusion technology

CN121996951ACN 121996951 ACN121996951 ACN 121996951ACN-121996951-A

Abstract

The invention discloses a hydropower station working ticket handling condition identification method and system based on a multisource information fusion technology, which are used for carrying out multisource heterogeneous information fusion analysis on the contents such as hydropower station monitoring system operation records, operation ticket execution records, operation duty logs, maintenance and delivery and the like based on natural language processing NLP, knowledge graph and rule engine technology, so that the hydropower station equipment state is automatically judged and updated in real time, decision support is provided for working ticket safety permission, the defect of the existing manual mode is overcome, the multisource heterogeneous information fusion analysis is carried out on the contents such as the hydropower station monitoring system operation records, operation ticket execution records, operation duty logs, maintenance and delivery and the like, the hydropower station equipment state is automatically judged and updated in real time, the real state of the hydropower station is intelligently inferred, and a reliable basis is provided for the working ticket safety permission, and the safety and intelligent level of hydropower station operation are improved.

Inventors

Gou Jiaji
YANG PENG
WANG CHUNYOU
XU CHANGFU
XIAO JUNYI
LIU GUOJIAN
CAO TIESHAN

Assignees

中国长江电力股份有限公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (10)

1. A hydropower station work ticket handling condition identification method based on a multisource information fusion technology is characterized by comprising the following steps: S1, constructing a hydropower station equipment state knowledge base and a rule engine, wherein the knowledge base comprises a whole plant equipment ledger model, a standard operation instruction base and a state judgment rule base; s2, multi-source heterogeneous data in a hydropower station production system are collected, wherein the multi-source heterogeneous data comprise monitoring system operation records, operation ticket execution records, operation duty logs and overhaul and exchange contents, natural language processing is carried out on the multi-source heterogeneous data, operation equipment and operation instructions are extracted, and an equipment state list is updated according to the rule engine; s3, identifying the running state interaction content of the equipment by using a natural language processing technology, performing cross verification with the equipment state list, and triggering manual confirmation when the running state interaction content is inconsistent with the equipment state list; s4, identifying the security measure requirements in the work ticket to be licensed, and extracting the state or security measure required by the target equipment; S5, carrying out special marking on the state description of which the recognition confidence is lower than a threshold value or cannot be matched with the rule, and pushing the state description to an operator for manual confirmation; s6, automatically comparing the extracted safety measure requirements with the current equipment state list to generate a compliance check result and alarm information.
2. The method for identifying the ticket handling conditions of the hydropower station based on the multi-source information fusion technology according to claim 1, wherein in the step S1, the state judgment rule base includes: Mapping rules of operation instructions and equipment states; five-prevention logic rules based on electrical latching logic; and the security measures correspond to the device state rules.
3. The method for recognizing the ticket handling conditions of the hydropower station based on the multi-source information fusion technology according to claim 1, wherein in the step S2, the natural language processing includes: cleaning, word segmentation, stop word removal and entity identification are carried out on the multi-source heterogeneous data; training the historical language features by using a convolutional neural network, and automatically extracting the language features; and sending the extracted operation equipment and operation instructions into a rule engine to drive equipment state knowledge base to update.
4. The method for identifying the ticket handling conditions of the hydropower station based on the multi-source information fusion technology according to claim 1, wherein in the step S2, when the same equipment state from different information sources is inconsistent, the system gives an alarm and prompts the operator to process.
5. The method for recognizing the ticket handling conditions of the hydropower station based on the multi-source information fusion technology according to claim 1, wherein in the step S4, the text in the ticket security measure column is automatically parsed by the natural language processing technology, and the equipment entity and the status entity are extracted.
6. A hydropower station ticket handling condition recognition system based on a multi-source information fusion technology for executing the hydropower station ticket handling condition recognition method based on the multi-source information fusion technology according to any one of claims 1 to 5, characterized in that the system comprises: the rule engine and state knowledge base module is used for storing the equipment model, the state mapping rule and the five-prevention logic rule; The multi-source data information acquisition module is used for acquiring equipment state related data from the monitoring system, the operation ticket system, the operation log and the overhaul waiting; The natural language processing engine module is used for cleaning, word segmentation and entity identification of the collected data and extracting operation equipment and state information; the device state management module is used for maintaining a device state list updated in real time; the intelligent checking and alarming module is used for executing the coincidence check of the working ticket safety measure and the equipment state and generating a checking result and alarming information; and the man-machine interaction interface module is used for displaying the state list, the verification result and the alarm information and providing a manual confirmation entry.
7. The system for recognizing the ticket handling conditions of a hydropower station based on the multi-source information fusion technology according to claim 6, wherein the rule engine and the state knowledge base module further comprises a standard operation instruction library for mapping natural language operation instructions into standard device state changing actions.
8. The system for recognizing the ticket handling conditions of the hydropower station based on the multi-source information fusion technology according to claim 6, wherein the natural language processing engine module adopts a convolutional neural network to conduct text feature extraction and entity recognition and supports continuous optimization of a model.
9. The method for recognizing the ticket handling conditions of the hydropower station based on the multi-source information fusion technology according to claim 6, wherein the intelligent checking and alarming module generates detailed alarming information containing conflicting equipment and state differences when the equipment state conflicts with the safety measure requirement.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements a hydropower station ticket handling condition recognition method based on a multisource information fusion technique as claimed in any one of claims 1 to 5.

Description

Hydropower station work ticket handling condition identification method and system based on multi-source information fusion technology Technical Field The invention belongs to the technical field of power plant work ticket management, and particularly relates to a hydropower station work ticket handling condition identification method and system based on a multi-source information fusion technology. Background In the background of energy interconnection in the current society, safe and stable operation of power generation equipment of a hydroelectric power plant plays an increasingly important role in power system interconnection and energy conservation. When the equipment is overhauled, maintenance personnel need to handle related work tickets at operators and then allow the operators to enter on-site work, and specific safety measures for overhauling, testing and other works of the equipment are defined in the work tickets, wherein the safety measures require the equipment to be in a specific state, such as operation, overhauling, power failure, brake separation, ground wire hanging and the like. At present, the judgment and management of the state of hydropower station equipment mainly depends on that an operator manually turns over a large number of paper or electronic operation tickets, work tickets and shift records, the specific state of certain equipment at a certain moment is comprehensively judged through brains, the judgment and management of the state of hydropower station equipment also depends on a manual mode, when the operator handles work ticket permission, the licensor needs to check whether related equipment is in a safe state or not by memorizing, turning over the paper records or cross checking in different systems, and the mode has low efficiency, depends on personal experience, is easy to cause misjudgment due to fatigue or negligence and has huge potential safety hazards. Once the equipment state is judged to be wrong, such as a loaded switch blade, a mistakenly-charged interval and the like, serious personal casualties and equipment damage accidents can be possibly caused, therefore, a hydropower station working ticket handling condition identification method and system based on a multi-source information fusion technology are needed to be designed to solve the problems. Disclosure of Invention The invention aims to overcome the defect of the existing manual mode, and automatically and real-time judge and update the state of hydropower station equipment, intelligently infer the true state of the equipment and provide reliable basis for the safety permission of the working ticket by carrying out multi-source heterogeneous information fusion analysis on the contents such as operation record of a hydropower station monitoring system, operation ticket execution record, operation duty log, maintenance and waiting and the like. In order to solve the technical problems, the invention adopts the following technical scheme: a hydropower station work ticket handling condition identification method based on a multisource information fusion technology comprises the following steps: S1, constructing a hydropower station equipment state knowledge base and a rule engine, wherein the knowledge base comprises a whole plant equipment ledger model, a standard operation instruction base and a state judgment rule base; s2, multi-source heterogeneous data in a hydropower station production system are collected, wherein the multi-source heterogeneous data comprise monitoring system operation records, operation ticket execution records, operation duty logs and overhaul and exchange contents, natural language processing is carried out on the multi-source heterogeneous data, operation equipment and operation instructions are extracted, and an equipment state list is updated according to the rule engine; s3, identifying the running state interaction content of the equipment by using a natural language processing technology, performing cross verification with the equipment state list, and triggering manual confirmation when the running state interaction content is inconsistent with the equipment state list; s4, identifying the security measure requirements in the work ticket to be licensed, and extracting the state or security measure required by the target equipment; S5, carrying out special marking on the state description of which the recognition confidence is lower than a threshold value or cannot be matched with the rule, and pushing the state description to an operator for manual confirmation; s6, automatically comparing the extracted safety measure requirements with the current equipment state list to generate a compliance check result and alarm information. Preferably, in the step S1, the state judgment rule base includes: Mapping rules of operation instructions and equipment states; five-prevention logic rules based on electrical latching logic; and the security measures correspond to the device state