Search

CN-120812107-B - Intelligent water meter operation monitoring method and system based on Internet of things

CN120812107BCN 120812107 BCN120812107 BCN 120812107BCN-120812107-B

Abstract

The invention discloses an intelligent water meter operation monitoring method and system based on the Internet of things, which relates to the technical field of intelligent water meters, the method comprises the steps of constructing disturbance consistency scoring index Ipd and behavior pattern variation scoring Simv, and a multiparameter coupling backtracking risk level scoring function Rtrace is introduced, so that disturbance mutation intensity, period stability, flow direction consistency and historical behavior deviation degree can be comprehensively considered, and the recognition accuracy of unnatural disturbance water flow and potential abnormal water behavior is effectively improved. Especially under the condition of disturbance abnormality and behavior pattern abnormality coupling, the retrospective risk level scoring function Rtrace is subjected to secondary comparison and evaluation with a risk interval threshold value, so that more distinguishable risk judgment and risk level division can be realized, abnormal water use behaviors such as illegal water taking, backflow and backflow can be accurately locked under the condition of low misjudgment rate, and intelligent perception capability and behavior modeling reliability of the urban water supply system are enhanced.

Inventors

  • CHEN SHAOCHENG

Assignees

  • 深圳市嘉荣华科技有限公司

Dates

Publication Date
20260508
Application Date
20250725

Claims (8)

  1. 1. An intelligent water meter operation monitoring method based on the Internet of things is characterized by comprising the following steps: S1, setting an acquisition point at an intelligent water meter end, acquiring disturbance data in real time, transmitting the disturbance data to an Internet of things platform, and carrying out feature extraction and preprocessing on the disturbance data in the Internet of things platform to acquire a standardized disturbance data set; The S1 further comprises S13; s13, after the Internet of things platform receives the uploaded data message, unpacking and extracting disturbance data, and extracting features to obtain a disturbance feature set, wherein the disturbance feature set comprises a disturbance mutation gradient Dg, a disturbance fluctuation period Tp and a flow direction variation index Rrev; The disturbance abrupt change gradient Dg is characterized in that a disturbance flow velocity change sequence Vdiff is analyzed through a sliding time window, and the maximum absolute increase of a flow velocity change value in unit time is identified in the current sliding time window; the disturbance fluctuation period Tp is obtained by carrying out frequency spectrum analysis on a disturbance frequency distribution function Ff, extracting an amplitude peak value position corresponding to the dominant frequency, deducing the corresponding angular frequency according to the dominant frequency, wherein the angular frequency is calculated by multiplying 2 pi by the dominant frequency; The flow direction variation index Rrev is used for calculating the first derivative of the direction reflection offset vector Ph delta on a continuous time slice by using the time sequence of the direction reflection offset vector Ph delta as an input parameter in a set observation period T and utilizing a numerical difference method to obtain a phase offset rate change sequence, taking an absolute value of each time point in the phase offset rate change sequence, carrying out numerical integration on the absolute value sequence in the whole sliding time window to obtain the accumulated amount of reverse trend fluctuation in the sliding time window, dividing the accumulated amount of reverse trend fluctuation by the duration of the period T to obtain the average reverse offset degree in unit time, and obtaining a result, namely the flow direction variation index Rrev; Performing minimum and maximum normalization processing on the three parameters of the disturbance feature set based on the minimum value and the maximum value of each parameter in the historical sample data, eliminating physical dimension influence, and forming a standardized disturbance data set; S2, calculating a disturbance consistency score index Ipd based on a standardized disturbance data set, presetting a disturbance threshold Ith, and carrying out preliminary comparison evaluation on the disturbance consistency score index Ipd and the disturbance threshold Ith; The S2 comprises S21; S21, constructing a disturbance consistency score calculation model in an Internet of things platform, inputting a standardized disturbance data set acquired in real time into the constructed disturbance consistency score calculation model, calculating and outputting a disturbance consistency score index Ipd, and measuring unnatural disturbance characteristics of a local water flow state; the disturbance consistency score index IPd is calculated and output through the following disturbance consistency score calculation model; In a set observation time window, performing first derivative calculation on all water flow rate data in a disturbance flow rate change sequence Vdiff to obtain a flow rate change rate sequence, and performing point-to-point product processing on a sine function corresponding to a disturbance main frequency signal on the flow rate change rate sequence to form a disturbance derivative modulation sequence; square integration operation is carried out on the disturbance derivative modulation sequence, and a disturbance modulation energy integral value is obtained; Introducing a disturbance intensity suppression factor formed by adding an absolute value of a disturbance abrupt gradient Dg and a constant 1 as a denominator term on the basis of a disturbance modulation energy integral value result, and multiplying the disturbance intensity suppression factor ratio by a reciprocal value of a natural logarithm of a disturbance fluctuation period Tp to obtain a disturbance consistency score index Ipd; s3, triggering a behavior backtracking processing flow based on a preliminary comparison evaluation result, extracting a historical normal periodic behavior pattern, analyzing a current periodic behavior feature set Bnow, calling a normal periodic behavior feature set Bref in the normal periodic behavior pattern to compare, and outputting a variation pattern similarity score Simv; S4, comprehensively calculating a disturbance consistency score index Ipd and a variation map similarity score Simv, and outputting a backtracking risk score function Rtrace; S5, presetting a risk interval threshold, performing secondary comparison and evaluation on the risk interval threshold and a backtracking risk level score Rtrace, judging the risk level of the current user water behavior, and executing corresponding response control on the risk level.
  2. 2. The intelligent water meter operation monitoring method based on the Internet of things of claim 1, wherein S1 comprises S11 and S12; S11, setting two acquisition points in a water flow channel structure of the intelligent water meter, wherein the acquisition points comprise a first disturbance data acquisition point and a second disturbance data acquisition point; The first disturbance data acquisition point is arranged in the central area of the water inlet diversion cavity of the water meter main body and is used for acquiring initial disturbance information of water flow; The second disturbance data acquisition point is arranged at the downstream of the water outlet pressure stabilizing cavity and close to the water outlet valve port and is used for acquiring disturbance water flow behavior data; Each acquisition point is internally provided with a high-sensitivity electromagnetic flow velocity sensor, a piezoelectric water hammer induction sensor, an array type ultrasonic reflection sensor and a disturbance frequency capturing module, disturbance data are acquired in real time, the disturbance data comprise a disturbance flow velocity change sequence Vdiff, a direction reflection offset vector Phdelta and a disturbance frequency distribution function Ff, and then disturbance data form disturbance original data frames and are cached in an edge data control module of the intelligent water meter; S12, packaging and transmitting the disturbance original data frame through a wireless communication module configured in the intelligent water meter, wherein the wireless communication module adopts an NB-IoT protocol to transmit an uploading data message containing a time stamp, a data type identifier and a disturbance parameter value group to a remote internet of things platform.
  3. 3. The intelligent water meter operation monitoring method based on the Internet of things of claim 1, wherein S2 further comprises S22; s22, performing preliminary comparison and evaluation on the calculated disturbance consistency score index Ipd and a set disturbance threshold value Ith, and judging the consistency of current water flow disturbance, wherein the disturbance threshold value Ith is an empirical dynamic statistics threshold value, and selecting a score lower boundary with 95% confidence from disturbance consistency score index Ipd samples of historical normal disturbance behaviors as an initial threshold value benchmark based on a large number of historical disturbance consistency score indexes Ipd; when the disturbance consistency score index Ipd is greater than the disturbance threshold value Ith, the current disturbance behavior period of the water flow is consistent and the direction is stable, only the behavior log is recorded, and the routine monitoring is continued; when the disturbance consistency score index Ipd is less than or equal to the disturbance threshold value Ith, judging that the current water flow behavior has risk of disturbance consistency deficiency, and triggering a behavior backtracking processing flow at the moment.
  4. 4. The intelligent water meter operation monitoring method based on the Internet of things of claim 1, wherein S3 comprises S31; S31, after preliminary comparison and evaluation of a trigger behavior backtracking processing flow, extracting a historical normal periodic behavior pattern from a historical behavior pattern database in an internet-of-things platform, wherein the historical normal periodic behavior pattern is formed by monitoring water behavior samples of a plurality of users in different time periods, different water pressures and different living habits, and then carrying out normalization processing on sample data in the historical normal periodic behavior pattern to form a periodic behavior characteristic template with a unified structure; In the current behavior period, collecting continuous running data of the water meter, and constructing a current period behavior feature group Bnow, wherein the current period behavior feature group Bnow comprises an instantaneous water use duration, a maximum peak flow rate, a flow rate change rate slope, a daily water use frequency, water use interval distribution and a water use rhythm period; And calling a periodic behavior characteristic template matched with the current user type, the household type, the water meter type and the typical water consumption mode from the historical normal periodic behavior pattern, and extracting a normal periodic behavior characteristic group Bref.
  5. 5. The intelligent water meter operation monitoring method based on the Internet of things of claim 4, wherein S3 further comprises S32; S32, carrying out item-by-item difference processing on the values of the corresponding feature dimensions in the obtained current periodic behavior feature group Bnow and the normal periodic behavior feature group Bref to obtain a behavior deviation absolute value sequence, summing the behavior deviation absolute value sequence to obtain a behavior deviation total value, dividing the behavior deviation total value by the cumulative reference value sum of all feature dimensions in the normal periodic behavior feature group Bref to obtain a variation map similarity score Simv, and measuring the deviation degree of the current water behavior between the overall behavior feature space and the standard behavior map.
  6. 6. The intelligent water meter operation monitoring method based on the Internet of things of claim 1, wherein S4 comprises S41; S41, calculating a backtracking risk grade Rtrace based on a disturbance consistency grade index Ipd, a variation map similarity grade Simv and the accumulated times Tact of abnormal behaviors in the current detection period, and quantitatively analyzing the current abnormal behavior of the user.
  7. 7. The intelligent water meter operation monitoring method based on the Internet of things of claim 6, wherein S5 comprises S51 and S52; S51, extracting historical back-tracking risk grade scores Rtrace, performing frequency distribution analysis on all the historical back-tracking risk grade scores Rtrace, and setting the critical values of the historical back-tracking risk grade scores Rtrace of the normal users and the historical back-tracking risk grade scores Rtrace of the abnormal users as risk interval thresholds; The risk interval threshold comprises a first risk interval threshold F1 and a second risk interval threshold F2; the first risk interval threshold F1 is a historical retrospective risk level score Rtrace threshold value of a normal user, and the second risk threshold F2 is a historical retrospective risk level score Rtrace threshold value of an abnormal user; s52, performing secondary comparison and evaluation based on the backtracking risk level score Rtrace acquired in real time and a risk interval threshold value, judging the risk level of the current user water consumption behavior, and executing corresponding differential response control based on the risk level, wherein the specific evaluation content is as follows; if the retrospective risk level score Rtrace is smaller than the first risk interval threshold value F1, determining that the first-level risk behavior is performed, recording a behavior log and maintaining a normal running state; if the first risk interval threshold F1 is less than or equal to the retrospective risk level score Rtrace and is smaller than the second risk interval threshold F2, judging that the behavior is secondary risk behavior, starting a behavior trend tracking mechanism, prolonging the 50% data acquisition period and recording the behavior vector change in real time; if the backtracking risk level score Rtrace is more than or equal to the second risk interval threshold F2, determining three-level risk behaviors, calling a highest-level response scheme in a behavior triggering strategy library, and generating a processing command set comprising a water meter control command, an abnormal alarm command and a platform write-back command; The processing command set is analyzed in the intelligent water meter controller and then is executed; The water meter control instruction prohibits the reporting and settlement processing of the water behavior record of the user by locking the abnormal user, and cuts off the water supply of the current user; the abnormal alarm instruction sends early warning information to an administrator port through a remote Internet of things platform; The platform write-back instruction is used for updating the behavior map state of the user and synchronizing abnormal behavior information to the auditing database for auditing and review by an administrator.
  8. 8. An intelligent water meter operation monitoring system based on the Internet of things is applied to the intelligent water meter operation monitoring method based on the Internet of things, and is characterized by comprising a disturbance data acquisition module, a disturbance consistency analysis module, a similarity analysis module, a comprehensive analysis module and a risk level response control module; The disturbance data acquisition module acquires disturbance data in real time by setting an acquisition point at the intelligent water meter end, transmits the disturbance data to the Internet of things platform, and performs feature extraction and preprocessing on the disturbance data in the Internet of things platform to acquire a standardized disturbance data set; after the Internet of things platform receives the uploaded data message, unpacking and extracting disturbance data, and extracting features to obtain a disturbance feature set, wherein the disturbance feature set comprises a disturbance mutation gradient Dg, a disturbance fluctuation period Tp and a flow direction variation index Rrev; The disturbance abrupt change gradient Dg is characterized in that a disturbance flow velocity change sequence Vdiff is analyzed through a sliding time window, and the maximum absolute increase of a flow velocity change value in unit time is identified in the current sliding time window; the disturbance fluctuation period Tp is obtained by carrying out frequency spectrum analysis on a disturbance frequency distribution function Ff, extracting an amplitude peak value position corresponding to the dominant frequency, deducing the corresponding angular frequency according to the dominant frequency, wherein the angular frequency is calculated by multiplying 2 pi by the dominant frequency; The flow direction variation index Rrev is used for calculating the first derivative of the direction reflection offset vector Ph delta on a continuous time slice by using the time sequence of the direction reflection offset vector Ph delta as an input parameter in a set observation period T and utilizing a numerical difference method to obtain a phase offset rate change sequence, taking an absolute value of each time point in the phase offset rate change sequence, carrying out numerical integration on the absolute value sequence in the whole sliding time window to obtain the accumulated amount of reverse trend fluctuation in the sliding time window, dividing the accumulated amount of reverse trend fluctuation by the duration of the period T to obtain the average reverse offset degree in unit time, and obtaining a result, namely the flow direction variation index Rrev; Performing minimum and maximum normalization processing on the three parameters of the disturbance feature set based on the minimum value and the maximum value of each parameter in the historical sample data, eliminating physical dimension influence, and forming a standardized disturbance data set; The disturbance consistency analysis module calculates a disturbance consistency score index Ipd based on a standardized disturbance data set, presets a disturbance threshold Ith, and performs preliminary comparison evaluation on the disturbance consistency score index Ipd and the disturbance threshold Ith; a disturbance consistency score calculation model is built in an Internet of things platform, a standardized disturbance data set obtained in real time is input into the built disturbance consistency score calculation model, disturbance consistency score indexes Ipd are calculated and output, and unnatural disturbance characteristics of a local water flow state are measured; the disturbance consistency score index IPd is calculated and output through the following disturbance consistency score calculation model; In a set observation time window, performing first derivative calculation on all water flow rate data in a disturbance flow rate change sequence Vdiff to obtain a flow rate change rate sequence, and performing point-to-point product processing on a sine function corresponding to a disturbance main frequency signal on the flow rate change rate sequence to form a disturbance derivative modulation sequence; square integration operation is carried out on the disturbance derivative modulation sequence, and a disturbance modulation energy integral value is obtained; Introducing a disturbance intensity suppression factor formed by adding an absolute value of a disturbance abrupt gradient Dg and a constant 1 as a denominator term on the basis of a disturbance modulation energy integral value result, and multiplying the disturbance intensity suppression factor ratio by a reciprocal value of a natural logarithm of a disturbance fluctuation period Tp to obtain a disturbance consistency score index Ipd; The similarity analysis module triggers a behavior backtracking processing flow based on a preliminary comparison evaluation result, extracts a historical normal periodic behavior pattern, analyzes a current periodic behavior feature group Bnow, invokes a normal periodic behavior feature group Bref in the normal periodic behavior pattern to compare, and outputs a variation pattern similarity score Simv; the comprehensive analysis module outputs a backtracking risk scoring function Rtrace by comprehensively calculating a disturbance consistency scoring index Ipd and a variation map similarity score Simv; And the risk level response control module is used for carrying out secondary comparison and evaluation on the risk interval threshold value and the retrospective risk level score Rtrace through the preset risk interval threshold value, judging the risk level of the current user water consumption behavior, and executing corresponding response control on the risk level.

Description

Intelligent water meter operation monitoring method and system based on Internet of things Technical Field The invention relates to the technical field of intelligent water meters, in particular to an intelligent water meter operation monitoring method and system based on the Internet of things. Background With the wide deployment of the internet of things in urban infrastructure, the intelligent water service system is used as a core component of new generation urban water resource management, and the perception capability and the operation monitoring precision of the intelligent water service system are continuously improved. In particular, in the urban water supply network management scene, the intelligent water meter is used as edge sensing terminal equipment, the traditional mechanical meter is gradually replaced, and the fine acquisition and remote control of water consumption behavior, flow fluctuation and running state are realized. Based on the data perception capability of the existing intelligent water meter, the operation monitoring method for hydrodynamic disturbance recognition and behavior analysis is combined with the Internet of things platform. More particularly, the invention relates to an intelligent monitoring method for identifying and responding to abnormal backflow paths, illegal water taking behaviors and unauthorized branch connection in a pipe network without adding a special pressure sensor or flow direction identification hardware. The identification and treatment of illegal water taking, backflow pollution, private connection and other problems in the water supply system at the present stage mainly depend on deployment of high-cost hardware facilities such as differential pressure sensors, one-way check valves, physical flow direction calibration devices and the like. The equipment is high in installation and maintenance cost, is difficult to realize full coverage, and cannot meet the monitoring requirement of complex pipe network structure and frequent multi-point interference in a high-density area or an old community. Meanwhile, although the traditional intelligent water meter has the flow speed and total amount monitoring function, the traditional intelligent water meter lacks sensitive recognition capability for information such as micro disturbance, unstable reverse disturbance and the like, and cannot accurately judge whether reflux abnormal behavior or multi-source water path superposition problems exist, so that obvious hysteresis and blind areas exist for illegal behavior recognition; The problem is mainly caused by the fact that the current pipe network monitoring mechanism lacks sufficient multidimensional modeling capability for the hydraulic disturbance mode, particularly in practical application scenes such as late night water valley period, branch pipe network end area or node area with multi-user parallel branches, water pressure mutation, water flow direction reversal or frequency disturbance signals are easily ignored, and the traditional system cannot effectively collect and analyze the weak signals. In case of illegal tie-back pipeline, backflow and backflow or private water pump, the problems of adjacent metering errors, water backflow pollution risks and system hydraulic unbalance are easily caused, the metering accuracy and economic benefits of water supply enterprises are damaged, and serious threats are possibly formed on regional public health and pipe network stability. Disclosure of Invention Aiming at the defects of the prior art, the invention provides an intelligent water meter operation monitoring method and system based on the Internet of things, which solve the problems in the background art. In order to achieve the above purpose, the invention is realized by the following technical scheme that the method comprises the following steps: S1, setting an acquisition point at an intelligent water meter end, acquiring disturbance data in real time, transmitting the disturbance data to an Internet of things platform, and carrying out feature extraction and preprocessing on the disturbance data in the Internet of things platform to acquire a standardized disturbance data set; S2, calculating a disturbance consistency score index Ipd based on a standardized disturbance data set, presetting a disturbance threshold Ith, and carrying out preliminary comparison evaluation on the disturbance consistency score index Ipd and the disturbance threshold Ith; s3, triggering a behavior backtracking processing flow based on a preliminary comparison evaluation result, extracting a historical normal periodic behavior pattern, analyzing a current periodic behavior feature set Bnow, calling a normal periodic behavior feature set Bref in the normal periodic behavior pattern to compare, and outputting a variation pattern similarity score Simv; S4, comprehensively calculating a disturbance consistency score index Ipd and a variation map similarity score Simv, and outputti