CN-122017337-A - Data acquisition and transmission method for intelligent electric energy meter

CN122017337ACN 122017337 ACN122017337 ACN 122017337ACN-122017337-A

Abstract

The invention relates to the technical field of data acquisition and transmission and discloses a data acquisition and transmission method for an intelligent electric energy meter, which comprises the steps of analyzing metering data and running state data to obtain metering feature vectors and running feature vectors, inputting the metering feature vectors and the running feature vectors into an electric field scene recognition model to generate a real-time acquisition tag, a periodic acquisition tag and an event triggering acquisition tag; and calculating a comprehensive transmission load evaluation value, and adaptively configuring data acquisition frequency, a data packaging mode and a transmission scheduling strategy of single communication between the intelligent electric energy meter and the data concentrator or the cloud management platform according to the comprehensive transmission load evaluation value. The invention can realize the intelligent and fine management and control of the data acquisition and transmission process through the deep cooperation of multi-dimensional electric field scene recognition and dynamic transmission decision.

Inventors

ZHU YUKUN
YUAN LIHONG
Sha Yao
WANG LIN
ZHANG YIJING
WANG XIUWEI
Ou Chanxian
SU SHANGJIE
ZANG WENLONG

Assignees

浙江小牛电气科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (10)

1. The data acquisition and transmission method for the intelligent electric energy meter is characterized by comprising the following steps of: an intelligent sensor is arranged in the intelligent electric energy meter, and the intelligent sensor is adopted to meter data and running state data of the intelligent electric energy meter; Analyzing the metering data and the running state data to obtain metering feature vectors and running feature vectors, inputting the metering feature vectors and the running feature vectors into an electric field scene recognition model, and generating a real-time acquisition tag, a periodic acquisition tag and an event triggering acquisition tag; inputting the metering feature vector, the operation feature vector and the real-time acquisition tag into a data transmission decision model to generate transmission priority parameters, data integrity weights and abnormal data identifications; and calculating a comprehensive transmission load evaluation value based on the real-time acquisition tag, the periodic acquisition tag, the event triggering acquisition tag, the transmission priority parameter and the data integrity weight, and adaptively configuring the data acquisition frequency, the data packaging mode and the transmission scheduling strategy of single communication between the intelligent electric energy meter and the data concentrator or the cloud management platform according to the comprehensive transmission load evaluation value.
2. The data acquisition and transmission method for an intelligent electric energy meter according to claim 1, wherein the electric field scene recognition model is constructed based on a multi-feature fusion classification framework and trained by a joint optimization objective function, the joint optimization objective function is constructed by weighting the real-time acquisition tag based on a metering data fluctuation tolerance coefficient and the event trigger acquisition tag based on a state change sensitivity coefficient.
3. The data acquisition and transmission method for an intelligent electric energy meter according to claim 2, wherein the data transmission decision model is constructed based on a hybrid decision tree and a probabilistic statistical model.
4. The data acquisition and transmission method for an intelligent ammeter according to claim 3, wherein when generating a real-time acquisition tag, a periodic acquisition tag and an event triggered acquisition tag, comprising: Extracting data fluctuation amplitude features, data fluctuation frequency features and data change trend features in the metering feature vectors, and constructing a multi-dimensional electric field scene feature matrix by combining voltage stability indexes, current load rates and temperature drift parameters in the running state feature vectors; Inputting the multi-dimensional electric field scene feature matrix into a feature preprocessing layer of the electric field scene recognition model, introducing an attention mechanism module after feature dimension reduction processing is carried out by a principal component analysis method, giving dynamic weight to feature dimensions corresponding to the fluctuation tolerance coefficient of the metering data, and carrying out weight enhancement on the feature dimensions corresponding to the state change sensitivity coefficient; Based on the feature vectors weighted by the attention mechanism, generating real-time acquisition tags, periodic acquisition tags and event trigger acquisition tags through a classification output layer, wherein confidence thresholds corresponding to the acquisition tags are optimized and determined through a cross-validation mode based on a historical electric field scene data set.
5. The method for data collection and transmission of an intelligent ammeter according to claim 4, wherein the confidence threshold value corresponding to each collection tag is determined by cross-validation based on historical electric field scene data sets, comprising: constructing a historical electric field scene data set containing a plurality of different electric field scene types, and labeling each sample in the historical electric field scene data set with a corresponding real acquisition label; Dividing the historical electric field scene data set into a training subset and a verification subset by adopting a cross verification mode, traversing a plurality of candidate confidence threshold combinations in the training subset, and respectively calculating acquisition tag identification performance indexes corresponding to the confidence threshold combinations in the verification subset; And determining an optimal confidence coefficient threshold corresponding to each acquisition tag by taking comprehensive optimization of the acquisition tag identification performance index as an optimization target, wherein a timeliness penalty factor related to timeliness of data acquisition is introduced into a threshold optimization process for the real-time acquisition tag, and when the data acquisition delay is caused by misjudgment of the real-time acquisition tag, the identification performance index is corrected.
6. The method for data acquisition and transmission of an intelligent electric energy meter according to claim 5, wherein an timeliness penalty factor related to timeliness of data acquisition is introduced in a threshold optimization process, and when a real-time acquisition tag misjudgment results in data acquisition delay, the method for correcting the identification performance index comprises: In the cross verification process, carrying out timeliness evaluation on the identification result corresponding to the real-time acquisition tag, and determining a time delay value between the occurrence of a scene and the completion of data acquisition of the real-time acquisition tag; comparing the time delay value with a real-time acquisition aging threshold, and judging that the real-time acquisition delay event is generated when the time delay value exceeds the real-time acquisition aging threshold; And introducing the timeliness penalty factor aiming at the real-time acquisition delay event, and carrying out attenuation correction on the acquisition tag identification performance index corresponding to the real-time acquisition tag.
7. The data acquisition and transmission method for an intelligent ammeter according to claim 6, wherein the calculation formula of the timeliness penalty factor is: τ=α×(Δt/Δt0) 2 ; Wherein, τ is an aging penalty factor, α is a penalty factor, the value range is 0.5-1, Δt is an actual time delay value, Δt0 is a real-time acquisition aging threshold, and τ=0 when Δt is less than or equal to Δt0, no penalty is performed.
8. The method for data collection and transmission of an intelligent ammeter according to claim 7, wherein inputting the metering feature vector, the operation feature vector and the real-time collection tag into a data transmission decision model to generate transmission priority parameters, data integrity weights and abnormal data identifications comprises: Performing characteristic splicing on the active power deviation characteristic, the reactive power distortion characteristic and the three-phase imbalance characteristic in the metering characteristic vector and the communication module signal strength characteristic, the power supply surplus energy characteristic and the historical transmission success rate characteristic in the operation characteristic vector to form a transmission decision input matrix; Performing multipath rule matching processing on the transmission decision input matrix through a mixed decision tree processing layer in the data transmission decision model, and outputting a preliminary transmission priority parameter and a data integrity weight interval, wherein split nodes of the mixed decision tree processing layer are selected based on a combined evaluation result of an information gain ratio and a radix index; combining the confidence value of the real-time acquisition tag, dynamically adjusting the data integrity weight interval through a probability statistical model, and improving the lower limit constraint of the data integrity weight interval when the confidence of the real-time acquisition tag is higher than a preset confidence threshold; And calculating posterior probability of occurrence of abnormal data based on the probability statistical model, and generating an abnormal data identifier when the posterior probability exceeds a preset abnormal judgment threshold, wherein the abnormal data identifier comprises abnormal data type identifier information and corresponding abnormal confidence information.
9. The method for data collection and transmission of an intelligent ammeter according to claim 8, wherein calculating the integrated transmission load assessment value based on the real-time collection tag, the periodic collection tag, the event triggered collection tag, the transmission priority parameter and the data integrity weight comprises: Performing matrix operation on scene weight coefficients and transmission priority parameters corresponding to all the acquisition tags to obtain a basic load evaluation vector; Combining the data integrity weight with the single communication data volume to generate a data integrity load component, and correcting according to the compression processing efficiency of the data type; Weighting and fusing the basic load assessment vector and the data integrity load component, and superposing extra load increment related to abnormality when abnormal data identification exists; And generating a comprehensive transmission load evaluation value for guiding communication scheduling according to the weighted fusion and the abnormal load result.
10. The method for data collection and transmission of an intelligent electric energy meter according to claim 9, wherein when the data collection frequency, the data packet mode and the transmission scheduling policy of single communication are adaptively configured between the intelligent electric energy meter and the data concentrator or the cloud management platform according to the comprehensive transmission load evaluation value, the method comprises: Comparing the comprehensive transmission load evaluation value with a first comprehensive transmission load evaluation value and a second comprehensive transmission load evaluation value, and determining the data acquisition frequency, the data packet mode and the transmission scheduling strategy according to the comparison result, wherein the first comprehensive transmission load evaluation value is smaller than the second comprehensive transmission load evaluation value; When the comprehensive transmission load evaluation value is smaller than or equal to the first comprehensive transmission load evaluation value, determining that the data acquisition frequency, the data packet mode and the transmission scheduling policy are respectively the first data acquisition frequency, the first data packet mode and the priority transmission scheduling policy; When the comprehensive transmission load evaluation value is larger than the first comprehensive transmission load evaluation value and smaller than or equal to the second comprehensive transmission load evaluation value, determining that the data acquisition frequency, the data packet mode and the transmission scheduling policy are respectively the second data acquisition frequency, the second data packet mode and the balanced transmission scheduling policy; And when the comprehensive transmission load evaluation value is larger than the second comprehensive transmission load evaluation value, determining that the data acquisition frequency, the data packet mode and the transmission scheduling strategy are respectively a third data acquisition frequency, a third data packet mode and a current limiting transmission scheduling strategy.

Description

Data acquisition and transmission method for intelligent electric energy meter Technical Field The invention relates to the technical field of data acquisition and transmission, in particular to a data acquisition and transmission method for an intelligent electric energy meter. Background The traditional intelligent electric energy meter data acquisition and transmission method mostly adopts a fixed mode on data acquisition frequency setting, and is difficult to dynamically adjust according to actual electricity consumption conditions and data importance, so that transmission congestion can be caused by overlarge data amount in electricity consumption peak periods, and the problems of data acquisition redundancy and resource waste exist in electricity consumption valley periods. Meanwhile, in the data packet mode, there is often a lack of differential processing on different types of data (such as real electric quantity, instantaneous voltage and current value, fault alarm information, etc.), and important data and non-important data are mixed and synchronized, which may cause delay or loss of key information transmission. In addition, the transmission scheduling strategy is single, different demands of network load conditions, data priority and user on data timeliness are not fully considered, so that the reliability and efficiency of data transmission are difficult to be effectively guaranteed, and the problems of unstable data transmission, high packet loss rate, low response speed and the like are easy to occur particularly in a complex power network environment, and the accurate monitoring and efficient management of the intelligent power grid on the user power consumption data are influenced. Therefore, it is necessary to design a data acquisition and transmission method for an intelligent electric energy meter to solve the problems existing in the prior art. Disclosure of Invention In view of the above, the invention provides a data acquisition and transmission method for an intelligent electric energy meter, which aims to solve the problems of fixed acquisition frequency, lack of differential processing of data packets, single transmission scheduling strategy and the like in the data acquisition and transmission process of the intelligent electric energy meter in the prior art. The invention provides a data acquisition and transmission method for an intelligent electric energy meter, which comprises the following steps: an intelligent sensor is arranged in the intelligent electric energy meter, and the intelligent sensor is adopted to meter data and running state data of the intelligent electric energy meter; Analyzing the metering data and the running state data to obtain metering feature vectors and running feature vectors, inputting the metering feature vectors and the running feature vectors into an electric field scene recognition model, and generating a real-time acquisition tag, a periodic acquisition tag and an event triggering acquisition tag; inputting the metering feature vector, the operation feature vector and the real-time acquisition tag into a data transmission decision model to generate transmission priority parameters, data integrity weights and abnormal data identifications; and calculating a comprehensive transmission load evaluation value based on the real-time acquisition tag, the periodic acquisition tag, the event triggering acquisition tag, the transmission priority parameter and the data integrity weight, and adaptively configuring the data acquisition frequency, the data packaging mode and the transmission scheduling strategy of single communication between the intelligent electric energy meter and the data concentrator or the cloud management platform according to the comprehensive transmission load evaluation value. Further, the electric field scene recognition model is constructed based on a multi-feature fusion classification framework and trained through a combined optimization objective function, the combined optimization objective function weights and constructs the real-time acquisition tag based on a measurement data fluctuation tolerance coefficient, and the event trigger acquisition tag weights and constructs based on a state change sensitivity coefficient. Further, the data transmission decision model is constructed based on a hybrid decision tree and a probability statistical model. Further, when generating the real-time acquisition tag, the periodic acquisition tag and the event trigger acquisition tag, the method includes: Extracting data fluctuation amplitude features, data fluctuation frequency features and data change trend features in the metering feature vectors, and constructing a multi-dimensional electric field scene feature matrix by combining voltage stability indexes, current load rates and temperature drift parameters in the running state feature vectors; Inputting the multi-dimensional electric field scene feature matrix into a feature preproce