CN-121979574-A - AMI master station-oriented data processing method, device, equipment and medium

CN121979574ACN 121979574 ACN121979574 ACN 121979574ACN-121979574-A

Abstract

The application relates to the field of electric power data management and discloses a data processing method, device, equipment and medium for an AMI (advanced mechanical interface) master station, wherein the method is used for preprocessing intention information by acquiring the intention information to obtain standardized data; the method comprises the steps of matching limit data matched with standardized data from a preset knowledge base, inputting the standardized data and the limit data into a preset large language model, outputting structured intermediate data, wherein the large language model is provided with output constraint, the output constraint is used for prohibiting the large language model from outputting a bottom communication protocol message, and generating an execution message meeting the bottom communication protocol according to the structured intermediate data, wherein the execution message is used for being sent to corresponding execution equipment for command execution. The application improves the safety, controllability and execution efficiency of the intelligent operation of the AMI master station.

Inventors

Zhan Saitao
LI KEGUANG
CHEN ZUOZHI
WANG TIANTIAN

Assignees

深圳市思达仪表有限公司

Dates

Publication Date: 20260505
Application Date: 20260409

Claims (10)

1. The data processing method facing the AMI master station is characterized in that the data processing method facing the AMI master station is applied to the AMI master station and comprises the following steps: acquiring intention information, and preprocessing the intention information to obtain standardized data; matching limit data matched with the standardized data from a preset knowledge base, wherein the knowledge base is used for storing the mapping relation between the standardized data and the limit data; Inputting the standardized data and the limiting data into a preset large language model, and outputting structured intermediate data, wherein an output constraint is arranged in the large language model, and the output constraint is used for prohibiting the large language model from outputting a bottom communication protocol message; and generating an execution message meeting the bottom communication protocol according to the structured intermediate data, wherein the execution message is used for being sent to corresponding execution equipment to execute the command.
2. The AMI master-oriented data processing method according to claim 1, wherein after inputting the standardized data and the constraint data into a preset large language model, outputting structured intermediate data, further comprises: Performing preset multi-stage verification on the structured intermediate data to obtain a verification result; and generating an execution message meeting the bottom communication protocol for the structured intermediate data with the passing verification result.
3. The AMI master station oriented data processing method according to claim 2, wherein the performing a preset multi-stage verification on the structured intermediate data to obtain a verification result comprises: Performing semantic and grammar constraint verification on the structured intermediate data based on a preset grammar specification to obtain grammar compliance data; Carrying out protocol validity and version consistency verification on the grammar compliance data based on the communication protocol type of the target execution equipment to obtain protocol validity data; performing authority approval verification on the protocol legal data based on a preset high-risk operation set to obtain authority approval passing data; and carrying out business constraint verification on the authority approval passing data based on a preset business constraint rule to obtain a verification result.
4. The AMI master-oriented data processing method according to claim 1, wherein after inputting the standardized data and the constraint data into a preset large language model, outputting structured intermediate data, further comprises: Performing simulation on the structured intermediate data to generate a simulation evaluation result; If the simulation evaluation result meets a preset admission threshold, carrying out batch gray scale execution on the structured intermediate data corresponding to the simulation evaluation result according to a preset gray scale strategy and a preset proportion, acquiring an execution result index after each batch is executed, continuing the next batch when the execution result index meets a preset condition, and otherwise triggering a preset rollback mechanism; and generating an execution message meeting the bottom communication protocol according to the structured intermediate data of which the execution result indexes meet the preset conditions.
5. The AMI-master-oriented data processing method according to claim 4, wherein the performing simulation on the structured intermediate data to generate a simulation evaluation result comprises: Generating a simulation execution instruction according to the structured intermediate data, and inputting the simulation execution instruction into a preset simulation system, wherein the simulation system at least fuses a historical data playback model, a digital twin model and a network congestion model; based on the simulation system, simulating the actual running environment of the AMI master station, executing simulation operation, predicting the operation success rate, the execution time delay and/or the system load influence data of executing the structured intermediate data, and quantitatively generating a multi-dimensional simulation evaluation result.
6. The AMI-master-station-oriented data processing method according to claim 4, wherein the executing the structured intermediate data batch gray corresponding to the simulation evaluation result according to a preset gray level strategy and a preset proportion comprises: Generating an idempotent key for the structured intermediate data, storing the idempotent key and the structured intermediate data in an associated manner, and executing the instruction of the same idempotent key only once in a preset time window; Dividing equipment to be executed into a plurality of batches according to a preset gray level strategy and a preset proportion dividing standard; Executing the structured intermediate data according to the divided batch sequence, and acquiring at least execution success rate, execution time delay and system load influence data as execution result indexes after each batch is executed; Comparing the execution result index with a preset expected threshold, if the execution result index reaches the expected threshold, continuing to execute the next batch, and if the execution result index does not reach the preset expected threshold, suspending execution and triggering a preset rollback mechanism.
7. The data processing method facing the AMI master station according to any one of claims 4 to 6, wherein the preset rollback mechanism comprises: invoking a rollback policy associated with the structured intermediate data, the rollback policy comprising at least a rollback trigger condition, rollback operation content, and rollback range; When the execution result index does not reach a preset expected threshold, judging whether the current trigger condition accords with the rollback trigger condition in the rollback strategy, and if so, sequentially cancelling all or part of operations of the executed batch according to the order reverse to the execution order; And generating a rollback record, and storing the rollback record and the full-link traceability identification of the structured intermediate data in a correlated manner as part of an audit evidence chain.
8. The utility model provides a data processing device towards AMI master station which characterized in that, data processing device towards AMI master station is applied to the AMI master station, includes: The intention acquisition module is used for acquiring intention information, preprocessing the intention information and obtaining standardized data; The data matching module is used for matching limit data matched with the standardized data from a preset knowledge base, and the knowledge base is used for storing the mapping relation between the standardized data and the limit data; the controlled generation module is used for inputting the standardized data and the limiting data into a preset large language model and outputting structured intermediate data, wherein output constraints are arranged in the large language model, and the output constraints are used for prohibiting the large language model from outputting a bottom communication protocol message; And the command execution module is used for generating an execution message meeting the bottom communication protocol according to the structured intermediate data, wherein the execution message is used for being sent to corresponding execution equipment to execute the command.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the AMI master-oriented data processing method according to any one of claims 1 to 7 when the computer program is executed by the processor.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the AMI master station oriented data processing method according to any one of claims 1 to 7.

Description

AMI master station-oriented data processing method, device, equipment and medium Technical Field The application relates to the field of power data management, in particular to a data processing method, device, equipment and medium for an AMI master station. Background The advanced measurement system (ADVANCED METERING Infrastructure, AMI) master station is used as the core of the electric power metering automation system and bears the key functions of acquisition task arrangement, alarm disposal, parameter issuing, inspection analysis and the like. Along with the development of large language model technology, the method is introduced into an AMI master station to realize automation and intellectualization, and becomes an industry exploration direction. At present, a large language model can be directly used for generating ammeter commands or messages and issuing the ammeter commands or messages, however, the content generated by the large language model lacks structural constraint and has the non-executable problems of field deletion, type mismatch and the like, the illusion characteristic of the large language model can cause the generation of non-existing protocol items or neglect of equipment portrait constraint, and the 645/DLMS protocol itself has manufacturer version difference, so that the risk of mismatching of the generated content with actual equipment is further increased, and the safe and stable operation and the operation and maintenance efficiency of an AMI master station are seriously restricted. Therefore, how to integrate a large language model in an AMI master station and accurately generate executable commands to improve operation and maintenance efficiency is a problem to be solved. Disclosure of Invention The embodiment of the application provides a data processing method, a device, computer equipment and a storage medium for an AMI master station, which are used for solving the problems of integrating a large language model in the AMI master station and accurately generating executable commands so as to improve operation and maintenance efficiency. In a first aspect, a data processing method facing an AMI master station, where the data processing method facing the AMI master station is applied to the AMI master station, includes: acquiring intention information, and preprocessing the intention information to obtain standardized data; matching limit data matched with the standardized data from a preset knowledge base, wherein the knowledge base is used for storing the mapping relation between the standardized data and the limit data; Inputting the standardized data and the limiting data into a preset large language model, and outputting structured intermediate data, wherein an output constraint is arranged in the large language model, and the output constraint is used for prohibiting the large language model from outputting a bottom communication protocol message; and generating an execution message meeting the bottom communication protocol according to the structured intermediate data, wherein the execution message is used for being sent to corresponding execution equipment to execute the command. In a second aspect, a data processing apparatus facing an AMI master station, where the data processing apparatus facing the AMI master station is applied to the AMI master station, includes: The intention acquisition module is used for acquiring intention information, preprocessing the intention information and obtaining standardized data; The data matching module is used for matching limit data matched with the standardized data from a preset knowledge base, and the knowledge base is used for storing the mapping relation between the standardized data and the limit data; the controlled generation module is used for inputting the standardized data and the limiting data into a preset large language model and outputting structured intermediate data, wherein output constraints are arranged in the large language model, and the output constraints are used for prohibiting the large language model from outputting a bottom communication protocol message; And the command execution module is used for generating an execution message meeting the bottom communication protocol according to the structured intermediate data, wherein the execution message is used for being sent to corresponding execution equipment to execute the command. In a third aspect, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the data processing method for an AMI master station according to the first aspect when the computer program is executed by the processor. In a fourth aspect, a computer readable storage medium stores a computer program, which when executed by a processor implements the data processing method for an AMI master station according to the first aspect. Compared with the prior art, the method has the bene