CN-122017662-A - Power supply monitoring method based on Internet of things distributed architecture

CN122017662ACN 122017662 ACN122017662 ACN 122017662ACN-122017662-A

Abstract

The invention belongs to the technical field of Internet of things and power monitoring, and particularly relates to a power monitoring method based on an Internet of things distributed architecture, which is realized based on a distributed monitoring terminal, an edge computing node and a cloud monitoring platform, and comprises the steps of collecting multiple types of parameters of a power supply to be monitored and packaging the parameters into initial monitoring data; the method comprises the steps of preprocessing, fusing and calibrating environmental parameters, constructing a data set by associating terminal identification and position data, judging power supply work and battery states in a layering mode, dynamically optimizing a threshold value, generating a monitoring report, constructing a monitoring topological graph, updating the states and triggering early warning based on multidimensional conditions. The distributed power supply distributed deployment system and the distributed power supply distributed deployment method based on the distributed power supply distributed deployment system have the advantages that the data processing pressure is dispersed, the monitoring precision is improved, the distributed power supply real-time monitoring, accurate early warning and efficient operation and maintenance are realized, the problems of high delay, threshold value rigidness, inaccurate early warning and the like in the prior art are solved, and the distributed power supply distributed deployment scene is adapted.

Inventors

MA KAIYU
Zhong teng
DU LIJIAO

Assignees

四川梦腾科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260416

Claims (9)

1. The power supply monitoring method based on the Internet of things distributed architecture is characterized by comprising a plurality of distributed monitoring terminals, edge computing nodes and a cloud monitoring platform, and comprises the following steps of: S1, collecting real-time position data, real-time electric variable parameters, surrounding environment parameters and real-time battery parameters of a built-in battery of a power supply to be monitored, and packaging all data into initial monitoring data; S2, preprocessing initial monitoring data to obtain standardized monitoring data, carrying out fusion calibration on other standardized monitoring data except for environmental parameters based on the environmental parameters to obtain calibrated standardized monitoring data; s3, carrying out layered judgment on the distributed power supply monitoring data set based on a preset electric variable threshold value and a preset battery parameter threshold value to respectively obtain a working state judgment result and a built-in battery state judgment result of the power supply to be monitored, and dynamically and iteratively updating the preset electric variable threshold value and the preset battery parameter threshold value to realize self-adaptive optimization of the threshold value; S4, associating real-time position data, a working state judging result and a built-in battery state judging result of the power supplies to be monitored, generating a distributed monitoring report of each power supply to be monitored by combining the optimized threshold value output by the dynamic threshold value self-learning module, and constructing a distributed power supply monitoring topological graph based on the position data of each power supply to be monitored; And S5, updating the working state judging result and the built-in battery state judging result of the power supply to be monitored based on the calibrated standardized monitoring data and the dynamically updated preset threshold value, and generating early warning information comprising real-time position data, abnormal state type and abnormal parameters of the power supply to be monitored when the judging result meets the preset early warning condition.
2. The power supply monitoring method based on the distributed architecture of the internet of things according to claim 1, wherein the specific process of performing fusion calibration on other standardized monitoring data except for the environmental parameter based on the environmental parameter in the step S2 to obtain the calibrated standardized monitoring data is as follows: s21, marking a standardized real-time electric variable parameter as X, marking a real-time battery parameter as Y, taking the standardized real-time electric variable parameter as a core monitoring parameter, marking a standardized environment parameter as Z, and taking the standardized environment parameter as an interference calibration parameter, wherein the interference calibration parameter comprises temperature T, humidity H and electromagnetic interference intensity E; s22, respectively calculating interference weights of temperature, humidity and electromagnetic interference intensity on electric variable parameters and battery parameters; S23, carrying out self-adaptive calibration on the standardized electric variable parameters and battery parameters based on the comprehensive interference weight; and S24, associating the calibrated electric variable parameters and the battery parameters with the standardized real-time position data to obtain the calibrated standardized monitoring data.
3. The power monitoring method based on the distributed architecture of the internet of things according to claim 2, wherein the specific process of step S22 is as follows: s221, calculating temperature interference weight, wherein the specific formula is as follows: ; ; Wherein, w T,X is the single interference weight of temperature to the electrical variable parameter X, w T,Y is the single interference weight of temperature to the battery parameter Y, k T is the temperature interference coefficient, T is the standardized real-time environmental temperature parameter, T 0 is the standard working temperature of the power supply to be monitored; S222, calculating humidity interference weight, wherein the specific formula is as follows: ; ; Wherein, w H,X is the single interference weight of humidity to the electrical variable parameter X, w H,Y is the single interference weight of humidity to the battery parameter Y, k H is the humidity interference coefficient, H is the standardized real-time environment humidity parameter, H 0 is the standard working humidity of the power supply to be monitored; S223, calculating the electromagnetic interference intensity interference weight, wherein the specific formula is as follows: ; ; Wherein, w E,X is the single interference weight of electromagnetic interference intensity to the electrical variable parameter X, w E,Y is the single interference weight of electromagnetic interference intensity to the battery parameter Y, k E is the electromagnetic interference intensity interference coefficient, E is the standardized real-time electromagnetic interference intensity parameter, E 0 is the standard working electromagnetic interference intensity of the power supply to be monitored; S224, calculating comprehensive interference weight of the core monitoring parameters, wherein the specific formula is as follows: w X =α·w T,X +β·w H,X +γ·w E,X ; w Y =α·w T,Y +β·w H,Y +γ·w E,Y ; Wherein w X is the comprehensive interference weight of the electrical variable parameter, w Y is the comprehensive interference weight of the battery parameter, and alpha, beta and gamma are the weight distribution coefficients of temperature, humidity and electromagnetic interference intensity respectively.
4. The power monitoring method based on the distributed architecture of the internet of things according to claim 3, wherein the specific process of step S23 is as follows: s231, performing electrical variable parameter calibration, wherein the specific formula is as follows: X cal =X·(1-w X )+X 0 ·w X ; wherein X cal is a standardized electrical variable parameter after calibration, X is a real-time electrical variable parameter after standardization, w X is comprehensive interference weight of the electrical variable parameter, and X 0 is an electrical variable parameter standard reference value; s232, performing battery parameter calibration, wherein the specific formula is as follows: Y cal =Y·(1-w Y )+Y 0 ·w Y ; Wherein Y cal is the standardized battery parameter after calibration, Y is the real-time battery parameter after standardization, w Y is the comprehensive interference weight of the battery parameter, and Y 0 is the standard reference value of the battery parameter.
5. The power monitoring method based on the internet of things distributed architecture according to claim 3, wherein in step S2, the specific process of associating each calibrated standardized monitoring data with the corresponding terminal identifier and the real-time position data to construct the distributed power monitoring data set is as follows: S25, extracting terminal marks marked when each distributed monitoring terminal uploads initial monitoring data, marking as ID i , marking as L i after preprocessing standardized real-time position data, including longitude and latitude coordinates (Lon i ,Lat i ) and altitude Alt i , extracting calibrated standardized monitoring data, and marking as D cal,i ; s26, adopting a key value pair mapping algorithm, taking a terminal identification ID i as a unique main key, and establishing one-to-one association with calibrated standardized monitoring data D cal,i ; S27, based on the uniqueness of the terminal identification ID i , performing secondary association on the terminal identification-calibrated monitoring data association pair and the corresponding standardized real-time position data L i to form three-dimensional association data Assoc (ID i ) of the terminal identification-position data-calibrated monitoring data; And S28, summarizing three-dimensional association data Assoc (ID i ) of all terminals, sequencing according to terminal identification, supplementing a monitoring time stamp t i , and constructing a structured distributed power monitoring data set DS.
6. The power monitoring method based on the distributed architecture of the internet of things according to claim 5, wherein the specific process of performing the hierarchical judgment in the step S3 is as follows: S31, extracting corresponding calibrated standardized monitoring data one by one from a distributed power supply monitoring data set DS according to a terminal identification ID i , and simultaneously matching historical monitoring data DS his,i corresponding to the corresponding terminal, wherein the historical monitoring data DS comprises parameters and historical judging results after the calibration in the last 30 days; S32, calling an initial preset electric variable threshold set Th X ={Th X1 ,Th X2 ,...,Th Xk which is pre-stored by a cloud monitoring platform, and an initial preset battery parameter threshold set Th Y ={Th Y1 ,Th Y2 ,...,Th Ym ; S33, working state layering judgment, namely comparing each dimension parameter in X cal,i with a corresponding Th X threshold value one by one, and judging the working state of the power supply to be monitored; S34, carrying out built-in battery state layering judgment, namely comparing each dimension parameter in the Y cal,i with a corresponding Th Y threshold value one by one, and judging the built-in battery state; And S35, outputting a layering judgment result, namely summarizing the working state Status X,i and the built-in battery state Status Y,i of each power supply to be monitored, and associating the corresponding terminal identification ID i and the real-time position data L i to form associated data of the terminal identification-position-double-state judgment result.
7. The power supply monitoring method based on the distributed architecture of the internet of things according to claim 6, wherein the specific process of dynamically and iteratively updating the preset electric variable threshold and the preset battery parameter threshold in the step S3 to realize the adaptive optimization of the threshold is as follows: S36, acquiring a judging result set Statu si ={Status X,i ,Status Y,i , a calibrated parameter X cal,i 、Y cal,i in a current distributed power supply monitoring dataset DS, a historical monitoring dataset DS his , and a calibrated parameter containing nearly 30 days and a corresponding judging result; S37, adopting an abnormal data eliminating algorithm to screen calibrated parameters with normal judging results in the historical monitoring data, and eliminating parameters with abnormal judging results; S38, dynamically updating a preset electrical variable threshold Th X and a preset battery parameter threshold Th Y by adopting a weighted moving average iterative algorithm and combining historical effective data and current calibration data; and S39, performing deviation verification on the updated threshold value and the initial threshold value, if the deviation is within 5%, directly taking effect to replace the original preset threshold value, if the deviation exceeds 5%, combining rated parameters of a power supply to be monitored, correcting, taking effect after correction, and synchronizing the optimized threshold value after the effect to an edge computing node.
8. The power monitoring method based on the distributed architecture of the internet of things according to claim 1, wherein the specific process of step S4 is as follows: S41, extracting standardized real-time position Data L i , standardized electrical variable parameters X cal,i after calibration and standardized battery parameters Y cal,i after calibration from a distributed power supply monitoring dataset DS one by one according to a terminal identification ID i , extracting an i-Th dual-state judging result Status i of a power supply to be monitored and an optimized threshold Th i,new output by a dynamic threshold self-learning module, and performing four-dimensional binding on the position Data, the state judging result, the optimized threshold and the calibrated monitoring Data by taking the terminal identification ID i as a unique associated main key to form a complete associated dataset Data i of the single power supply to be monitored; S42, extracting and rectifying the Data i which passes the verification, including a terminal identification ID i , a power model to be monitored, a monitoring time stamp t i and real-time position original Data, comparing the calibrated standardized monitoring Data D cal,i in the Data i with an optimized threshold Th i,new , and calculating the deviation rate of each parameter and the threshold; S43, generating a distributed monitoring Report i of a single power supply to be monitored based on the deviation rate of each parameter and a threshold value, wherein the distributed monitoring Report comprises a basic information module, a monitoring parameter module, a state judging module, a threshold value module and an analysis suggesting module; S44, constructing a visual topological graph based on the position data, and realizing visual display of multiple power states.
9. The power monitoring method based on the distributed architecture of the internet of things according to claim 8, wherein the specific process of step S44 is as follows: s441, converting the actual position data L real,i of all the power supplies to be monitored into a plane rectangular coordinate (X map,i ,Y map,i ) required by topological graph drawing; S442, drawing topology nodes of each power supply to be monitored by taking a plane rectangular coordinate (X map,i ,Y map,i ) as a node position, and distinguishing a working state from a built-in battery state by adopting nodes with different colors and different shapes; S443, labeling a terminal identification ID i , a core monitoring parameter and a state judgment result beside each topological node, wherein a node offset labeling algorithm is adopted for labeling positions, so that labeling is ensured not to overlap.

Description

Power supply monitoring method based on Internet of things distributed architecture Technical Field The invention belongs to the technical field of Internet of things and power monitoring, and particularly relates to a power monitoring method based on an Internet of things distributed architecture. Background With the rapid development of the internet of things technology, the distributed deployment of various electronic devices and industrial devices is increasingly wide, the number of power supplies to be monitored (such as industrial control power supplies, standby power supplies of internet of things terminals, distributed energy storage power supplies and the like) matched with the distributed deployment is greatly increased, deployment scenes show decentralized and complicated characteristics, and higher requirements are provided for real-time monitoring, accurate early warning and efficient operation and maintenance of the running state of the power supplies. At present, the existing power supply monitoring technology mostly adopts a centralized monitoring architecture, namely, a single monitoring terminal is used for collecting operation parameters of a plurality of power supplies and then summarizing the operation parameters to a monitoring platform for processing. Such architecture suffers from significant drawbacks: Firstly, the monitoring range is limited, the power supply is difficult to adapt to the distributed deployment power supply scene, and when the power supply is distributed and deployed in different areas, the data transmission delay is high, the stability is poor, and the real-time monitoring cannot be realized; Secondly, the data processing precision is insufficient, the state judgment is carried out by adopting original monitoring data directly in the prior art, and the interference of environmental factors such as temperature, humidity, electromagnetic interference and the like on the acquisition precision of electric variable parameters and battery parameters is not considered, so that the state judgment error is larger, and the misjudgment and omission judgment situation easily occurs; Thirdly, the threshold value is set to be stiff, the state judgment is carried out by adopting a fixed threshold value, the threshold value cannot be dynamically optimized according to the historical data of the long-term operation of the power supply, the operation characteristics and the aging state of different power supplies are adapted, and the monitoring accuracy is further reduced; Fourthly, the data association degree is low, the position data, the monitoring parameters, the state judgment result and the threshold value data of the power supply are not effectively associated, a comprehensive monitoring report is difficult to generate, the distribution and the running state of the distributed power supply cannot be intuitively presented in a visual mode, and operation and maintenance personnel are difficult to quickly locate a fault power supply and efficiently develop operation and maintenance work; fifthly, the early warning mechanism is imperfect, early warning is triggered only based on single parameter abnormality, comprehensive early warning is carried out under the multidimensional conditions of parameter mutation, abnormal duration and the like, the early warning accuracy is low, and the specific solution of faults by operation and maintenance staff is not facilitated. In addition, in the existing distributed monitoring technology, the edge node only bears the data transmission function and does not participate in data preprocessing and fusion calibration, so that the data processing pressure of the cloud platform is overlarge, and the response speed of the monitoring system is further reduced. Meanwhile, the prior art lacks a perfect multi-source data fusion calibration mechanism, cannot effectively eliminate the influence of environmental interference on monitoring data, and is difficult to meet the requirement of high-precision power supply monitoring. Aiming at the defects of the prior art, the invention aims to provide a power supply monitoring method based on a distributed architecture of the Internet of things, which solves the technical problems of high data transmission delay, low monitoring precision, threshold value rigidness, low data association degree, inaccurate early warning and low operation and maintenance efficiency in the distributed power supply monitoring and realizes the real-time, accurate, efficient monitoring and intelligent operation and maintenance of the distributed power supply. Disclosure of Invention The invention aims to provide a power supply monitoring method based on a distributed architecture of the Internet of things, which is used for solving the technical problems of high data transmission delay, low monitoring precision, threshold value rigidification, low data association degree, inaccurate early warning and low operation and maintenance effi