Anti-freezing failure judgment method and system for anti-freezing water meter

Abstract

The invention belongs to the technical field of anti-freezing water meters, and particularly relates to an anti-freezing failure judgment method and system.The method comprises the steps that firstly, temperature, flow speed and pressure monitoring points are adaptively arranged according to a water meter installation scene, a low-temperature-resistant high-precision sensor is selected, factory calibration and field calibration are completed, and meanwhile an LoRa / Wi-Fi dual-mode encryption transmission link is built; then synchronously collecting multiple parameters and adding timestamps according to a dynamic period by means of a monitoring system, guaranteeing key data transmission through priority marking, completing data purification through abnormity elimination and fluctuation smoothing, and realizing parameter magnitude unification through normalization; on the basis, a multi-dimensional monitoring system covering key areas such as a water meter body, a water inlet and outlet pipe section and a heat preservation layer is constructed, core parameters such as the temperature, the water flow speed and the pressure are synchronously collected, coupling analysis is conducted, the problem that in the prior art, the monitoring dimension is single is effectively solved, and misjudgment caused by ignoring key influence factors is greatly reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of anti-freezing water meters, in particular to an anti-freezing failure judgment method and system for an anti-freezing water meter. BACKGROUND

[0002] As a core equipment for water supply metering, water meters are widely used in residential buildings, commercial buildings and outdoor pipe networks. In low temperature environments (especially below -5°C), the volume expansion of the frozen water inside the water meter can easily cause the shell to crack and the movement to be damaged, not only causing water leakage and metering failure, but also causing property loss and increasing maintenance costs. Currently, anti-freezing water meters mainly rely on passive protection methods such as wrapping with insulation layers and heating with heating bands, and the anti-freezing failure judgment is mainly based on a single temperature threshold trigger, which lacks systematicness, precision and linkage disposal capability, and is difficult to adapt to complex and variable installation environments (such as outdoor exposure, indoor concealment, high-rise pipe networks, etc.).

[0003] In summary, the existing anti-freezing failure judgment has the following problems:

[0004] 1. Single monitoring dimension, high misjudgment rate: Only the anti-freezing state is judged by the temperature of the water meter body, ignoring key influencing factors such as water flow rate and pipe network pressure. For example, after the insulation layer of an outdoor water meter is damaged, the environmental temperature is -5°C, but the pipe network water flow is continuous (flow rate 0.3 m / s), and the water body is not frozen but is judged as anti-freezing failure; on the contrary, the local temperature of an indoor water meter drops to -2°C due to air leakage through doors and windows, but the water flow is static (flow rate 0.01 m / s), and the water body has already shown a tendency to freeze but is not identified.

[0005] 2. Threshold fixed, poor adaptability: A uniform fixed temperature threshold (such as 0°C) is used, without considering the differences in water meter models, installation scenarios and seasonal changes. For example, the same threshold is used for a northern rural outdoor water meter (without a heating band) and a southern urban high-rise indoor water meter (with an insulation layer), resulting in frequent false alarms for the rural water meter, while the high-rise water meter is prone to missed failure risks due to local low temperature.

[0006] 3. No closed-loop response mechanism, disposal lag: Only an alarm signal is triggered, without linkage to active disposal measures. For example, after the anti-freezing failure of a water meter in a certain unit of a community, only an APP push reminder is sent, and the user does not check it in time, resulting in a shell rupture and water leakage due to water freezing at night, causing a loss of 100,000 yuan in the elevator shaft.

[0007] 4. Data not formed for feedback optimization, judgment accuracy not improved: Failure data and disposal results are not recorded in the system, and the judgment logic cannot be iteratively optimized according to actual scenarios, relying on initial set parameters for a long time, and the accuracy continues to decline. SUMMARY

[0008] To solve the above-mentioned monitoring dimension single, threshold fixed, no closed-loop response mechanism, data does not form the feedback optimization technical problem, the present application provides the following technical solutions:

[0009] A freeze-proof failure judgment method of a freeze-proof water meter, comprising the following specific steps:

[0010] S1, according to the water meter installation scene, the temperature, flow rate and pressure monitoring points are distributed, and low-temperature-resistant high-precision sensors are selected and double calibration is completed at the factory and on site, and a LoRa / Wi-Fi dual-mode encrypted transmission link is built, laying a comprehensive coverage, accurate data and stable transmission hardware foundation for subsequent data collection;

[0011] S2, relying on the monitoring system, multiple parameters are synchronously collected according to the dynamic period and are attached with time stamps, key data transmission is guaranteed through priority marking, data purification is completed through exception elimination and fluctuation smoothing, parameter magnitude unification is realized through normalization, effective data with consistent time sequence and no interference are output, and subsequent intelligent judgment is supported;

[0012] S3, according to the preprocessed data, a dynamic threshold interval is constructed in combination with the water meter type and environmental parameters, a judgment value is calculated through a multi-parameter coupling model, accidental fluctuations are excluded through an abnormal secondary verification with a shortened collection period, and a freeze-proof grade is defined according to the parameter deviation degree, precise identification and misjudgment avoidance are realized;

[0013] S4, a graded early warning is triggered according to the freeze-proof grade, and the effect is continuously monitored and verified after the corresponding treatment is performed, and the whole process data is input into a database to iterate the threshold model, forming a closed loop of judgment-response-optimization, and improving the effectiveness of freeze-proof treatment.

[0014] As a preferred scheme of the freeze-proof failure judgment method of the freeze-proof water meter, the specific steps of S1 are as follows:

[0015] S11, according to the water meter installation scene, temperature monitoring points are arranged in the water meter body, the inlet and outlet water pipe sections and the inner side of the heat preservation layer, flow rate monitoring points are arranged at the water outlet of the water meter, and pressure monitoring points are arranged at the water inlet end, so as to ensure that the key areas for freeze-proof are covered;

[0016] S12, low-temperature-resistant and high-precision sensors are selected, and factory calibration is completed through a standard thermostat and a pressure calibrator, and secondary calibration is performed by using a comparison method after installation on site, so as to eliminate installation deviation;

[0017] S13, a LoRa / Wi-Fi dual-mode transmission channel is used, LoRa is used in outdoor scenes, Wi-Fi is used in indoor scenes, an AES encrypted communication link is established with the water meter controller, and the data transmission stability is guaranteed.

[0018] As a preferred scheme of the freeze-proof failure judgment method of the freeze-proof water meter, in the S2, the specific steps are as follows:

[0019] S21, the controller synchronously collects temperature, water flow rate and pressure data of each monitoring point according to a dynamic period, and adds a time stamp to ensure time sequence consistency;

[0020] S22, according to the influence weight of parameters on freeze-proof failure, priority labels are added to the collected data to ensure high-priority data transmission and avoid key information loss when the transmission bandwidth is insufficient;

[0021] S23, 3σ criterion is used to remove out-of-range data caused by sensor failure, and sliding average method is used to smooth water flow rate instantaneous fluctuation to avoid accidental factors interference;

[0022] S24, temperature, water flow rate and pressure data are normalized according to historical normal interval, and converted into [0, 1] interval dimensionless index to unify data magnitude for subsequent analysis.

[0023] As a preferred scheme of the freeze-proof failure judgment method of the freeze-proof water meter, in the S3, the specific steps are as follows:

[0024] S31, combined with water meter type, installation environment and seasonal parameters, historical effective / failure data are trained by random forest algorithm to construct a dynamic threshold interval;

[0025] S32, a multi-parameter coupling model is constructed to calculate the judgment value, and when the temperature is lower than the dynamic threshold + water flow rate <0.05 m / s + pressure fluctuation >0.02 MPa, the failure condition is triggered, and the single parameter exceeding the standard is determined as a risk state;

[0026] S33, when the failure / risk condition is triggered for the first time, the data collection period is immediately shortened to 1 / 2 of the original period, and three groups of data are continuously collected and re-substituted into the coupling model, if two or more groups still meet the condition, the state is confirmed, otherwise it is determined as accidental fluctuation to avoid single abnormal misjudgment;

[0027] S34, according to the deviation degree of the judgment value from the threshold value, three levels of I, II and III are divided.

[0028] As a preferred scheme of the freeze-proof failure judgment method of the freeze-proof water meter, in the S4, the specific steps are as follows:

[0029] S41, the first level triggers the local indicator light to flash and the APP to remind; the second level increases the SMS notification to the property; and the third level starts the voice alarm and pushes the emergency platform of the water supply company;

[0030] S42, make the first automatic start-up of the heat tracing band / valve small amplitude open and close; the second remote close the small water inlet valve + send a single maintenance; the third immediately close the water inlet valve + generate an emergency work order;

[0031] S43, after treatment, if the parameters are restored, it is determined to be effective, if not, the level is upgraded, and the failure data and treatment process are entered into the database to iteratively optimize the dynamic threshold model.

[0032] A freeze-proof failure judgment system for a freeze-proof water meter, comprising:

[0033] A multi-dimensional monitoring system construction module is used to adapt to the temperature, flow rate and pressure monitoring points according to the water meter installation scene, and a low-temperature-resistant high-precision sensor is selected and double-calibrated at the factory and on site, and a LoRa / Wi-Fi dual-mode encrypted transmission link is built, laying a comprehensive coverage, accurate data and stable transmission hardware foundation for subsequent data collection.

[0034] A real-time data collection and preprocessing module is used to rely on the monitoring system, synchronously collect multiple parameters according to a dynamic period and add time stamps, ensure key data transmission through priority marking, complete data purification through abnormality elimination and fluctuation smoothing, realize parameter magnitude unification through normalization, output effective data with consistent time sequence and no interference, and support subsequent intelligent judgment.

[0035] An intelligent freeze-proof state judgment module is used to construct a dynamic threshold interval according to the preprocessed data, in combination with the water meter model and environmental parameters, calculate the judgment value through a multi-parameter coupling model, and assist in excluding accidental fluctuations through abnormal secondary verification with a shortened collection period, and define the freeze-proof level according to the parameter deviation degree, realize accurate identification and false judgment avoidance.

[0036] A failure response and linkage control module is used to trigger a graded early warning according to the freeze-proof level, continuously monitor and verify the effect after executing the corresponding treatment, enter the whole process data into the database to iteratively optimize the threshold model, form a closed loop of judgment-response-optimization, and improve the effectiveness of freeze-proof treatment.

[0037] As a preferred scheme of the freeze-proof failure judgment system for the freeze-proof water meter, the multi-dimensional monitoring system construction module comprises:

[0038] A layout optimization unit is used to set temperature monitoring points in the water meter body, inlet and outlet water pipe sections and the inner side of the heat preservation layer, set a water flow rate monitoring point at the water outlet of the water meter, and set a pressure monitoring point at the water inlet end, to ensure coverage of the key freeze-proof areas.

[0039] A sensor calibration unit is configured to select a low-temperature-resistant and high-precision sensor, and to complete factory calibration through a standard thermostat and a pressure calibration instrument, and to perform secondary calibration by using a comparison method after installation on site, so as to eliminate installation deviation.

[0040] A transmission adaptation unit is configured to use a LoRa / Wi-Fi dual-mode transmission channel, to use LoRa in an outdoor scene and to use Wi-Fi in an indoor scene, and to establish an AES encrypted communication link with a water meter controller, so as to ensure data transmission stability.

[0041] As a preferred scheme of the anti-freezing failure judgment system of the anti-freezing water meter, the real-time data acquisition and preprocessing module comprises:

[0042] A synchronous acquisition unit is configured to make the controller synchronously acquire temperature, water flow rate and pressure data of each monitoring point according to a dynamic period, and to add a time stamp to ensure time sequence consistency;

[0043] A data priority marking unit is configured to add a priority label to the acquired data according to an influence weight of the parameter on anti-freezing failure, so as to preferentially guarantee high-priority data transmission and avoid loss of key information when the transmission bandwidth is insufficient;

[0044] A data purification unit is configured to remove out-of-range data caused by sensor failure by using a 3σ criterion, and to smooth water flow rate instantaneous fluctuations by using a sliding average method, so as to avoid accidental factor interference;

[0045] A standard conversion unit is configured to normalize temperature, water flow rate and pressure data according to a historical normal interval, and to convert the data into a [0, 1] interval dimensionless index, so as to unify data magnitudes for subsequent analysis.

[0046] As a preferred scheme of the anti-freezing failure judgment system of the anti-freezing water meter, the anti-freezing state intelligent judgment module comprises:

[0047] A threshold dynamic adaptation unit is configured to combine a water meter model, an installation environment and seasonal parameters, to train historical effective / failure data by using a random forest algorithm, and to construct a dynamic threshold interval;

[0048] A coupling judgment unit is configured to construct a multi-parameter coupling model, to calculate a judgment value, and to trigger a failure condition when the temperature is lower than the dynamic threshold + the water flow rate < 0.05 m / s + the pressure fluctuation > 0.02 MPa, and to determine a risk state when a single parameter exceeds a standard;

[0049] An abnormal secondary verification unit is configured to shorten a data acquisition period to 1 / 2 of an original period when a failure / risk condition is triggered for the first time, to continuously acquire three groups of data and to re-input the data into the coupling model, to confirm a state if two or more groups of data still satisfy the condition, and to determine an accidental fluctuation otherwise, so as to avoid a single abnormal misjudgment.

[0050] a grade defining unit configured to divide the three grades of Grade I, Grade II and Grade III according to the deviation degree of the judgment value from the threshold value;

[0051] The failure response and linkage control module comprises:

[0052] The early warning linkage unit is configured to make Grade I trigger a local indicator light to flash and an APP to remind; Grade II adds an SMS notification to the property; and Grade III starts a voice alarm and pushes an emergency platform of a water supply company;

[0053] The disposal execution unit is configured to make Grade I automatically start a heat tracing band and a valve to open and close in a small range; Grade II remotely closes an inlet valve and dispatches a maintenance order; and Grade III immediately closes the inlet valve and generates an emergency work order;

[0054] The feedback optimization unit is configured to continuously monitor after disposal, determine effectiveness if parameters are restored, upgrade the grade if the parameters are not restored, and record failure data and disposal processes in a database to iteratively optimize a dynamic threshold model.

[0055] Compared with the prior art,

[0056] The application effectively solves the problem of single monitoring dimension in the prior art by constructing a multi-dimensional monitoring system covering key areas such as the water meter body, the inlet and outlet pipe sections and the heat preservation layer, synchronously collecting core parameters such as temperature, water flow rate and pressure and performing coupled analysis, greatly reduces the misjudgment caused by neglecting key influencing factors; relies on a dynamic threshold model combined with water meter types, installation environments and seasonal parameters to break the limitation of fixed thresholds, makes the judgment standard flexible and adaptable to different application scenarios, and avoids false alarms or missed judgments caused by poor adaptability; through the grading early warning and automatic disposal linkage mechanism, the traditional single alarm is upgraded to a closed-loop process of "early warning triggering-grading response-active disposal", effectively solves the disposal lag problem, reduces equipment damage and property loss caused by untimely intervention; at the same time, failure data, disposal processes and results are included in the database for continuous iteration and optimization of the judgment model, realizing dynamic improvement of the judgment accuracy, solving the pain point that the accuracy is difficult to improve due to lack of data feedback in the prior art, and achieving the effects of accurate identification of anti-freezing state, flexible adaptation, timely disposal and continuous optimization. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 is a schematic diagram of the overall framework of the application;

[0058] Figure 2 is a schematic diagram of the dimension monitoring system construction module framework of the application;

[0059] Figure 3 is a schematic diagram of the real-time data acquisition and preprocessing module framework of the application;

[0060] Figure 4 It is a framework schematic diagram of the intelligent judgment module for the anti-freezing state of the application.

[0061] Figure 5 It is a framework schematic diagram of the failure response and linkage control module of the application. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the application more clear, the embodiments of the application will be further described in detail below with reference to the drawings.

[0063] The application provides an anti-freezing failure judgment method for an anti-freezing water meter, which comprises the following specific steps:

[0064] S1, temperature, flow rate and pressure monitoring points are adapted to be arranged according to the water meter installation scene, low-temperature-resistant high-precision sensors are selected, and double calibration of factory and field is completed, and a LoRa / Wi-Fi dual-mode encrypted transmission link is built, thereby laying a comprehensive coverage, accurate data and stable transmission hardware foundation for subsequent data collection.

[0065] The specific steps of S1 are as follows:

[0066] S11, according to the water meter installation scene (outdoor / indoor, exposed / concealed), temperature monitoring points are arranged on the water meter body, the inlet and outlet pipe sections (50 cm before and after), and the inner side of the heat preservation layer, a water flow rate monitoring point is arranged at the water outlet of the water meter, and a pressure monitoring point is arranged at the water inlet end, so as to ensure that the key areas (positions prone to icing) are covered.

[0067] S12, low-temperature-resistant (-40℃~60℃) high-precision sensors (temperature error ±0.1℃, pressure error ±0.01MPa) are selected, and factory calibration is completed through a standard constant temperature box and a pressure calibrator, and after installation on site, secondary calibration is carried out by comparison method to eliminate installation deviation.

[0068] S13, a LoRa / Wi-Fi dual-mode transmission channel is adopted, LoRa (low power consumption, long distance) is used in outdoor scenes, Wi-Fi is used in indoor scenes, and an AES encrypted communication link is established with the water meter controller to ensure data transmission stability.

[0069] S2, relying on the monitoring system, multiple parameters are synchronously collected according to a dynamic period and time stamps are added, key data transmission is ensured through priority marking, data purification is completed through exception elimination and fluctuation smoothing, parameter magnitude unification is realized through normalization, effective data with consistent time sequence and no interference are output, and the subsequent intelligent judgment is supported.

[0070] The specific steps of S2 are as follows:

[0071] S21, make the controller press the dynamic cycle (regular 1 minute / time, 0.1 minutes / time when the ambient temperature <0 ℃), synchronously collect the temperature, water flow rate, pressure data of each monitoring point, and add time stamp to ensure the consistency of timing;

[0072] S22, according to the influence weight of antifreeze failure (temperature > water flow rate > pressure), add priority label (high / medium / low) to the collected data to ensure the transmission of high priority data (such as temperature) when the transmission bandwidth is insufficient, and avoid the loss of key information;

[0073] S23, use 3σ criterion to remove out-of-range data caused by sensor failure, and smooth the instantaneous fluctuation of water flow rate by sliding average method (window size 5) to avoid accidental factors interference;

[0074] S24, normalize the temperature (℃), water flow rate (m / s), pressure (MPa) data according to the historical normal interval, and convert it to [0, 1] interval dimensionless index to unify the data magnitude for subsequent analysis.

[0075] S3, according to the preprocessed data, combined with water meter type, environmental parameters to construct dynamic threshold interval, and through multi-parameter coupling model to calculate the judgment value, and supplemented by shortening the collection cycle of abnormal secondary verification to exclude accidental fluctuations, and according to the parameter deviation degree to define the antifreeze level, to realize accurate identification and false judgment avoidance;

[0076] The specific steps of S3 are as follows:

[0077] S31, combined with water meter type, installation environment, seasonal parameters, through random forest algorithm to train historical effective / failure data, to construct dynamic threshold interval (for example: when water flow rate > 0.2 m / s, temperature threshold is-2 ℃; when water flow rate = 0, temperature threshold is 0 ℃);

[0078] S32, construct multi-parameter coupling model, calculate judgment value, when temperature is lower than dynamic threshold + water flow rate < 0.05 m / s + pressure fluctuation > 0.02 MPa, trigger failure condition, and single parameter exceeds the standard to determine the risk state;

[0079] S33, when the first calculation triggers the failure / risk condition, immediately shorten the data collection cycle to 1 / 2 of the original cycle (such as 0.1 minutes / time adjusted to 0.05 minutes / time), and continuously collect 3 groups of data to re-enter the coupling model, if 2 or more still meet the conditions, confirm the state, otherwise, it is determined as accidental fluctuation, to avoid single abnormal false judgment;

[0080] S34, according to the deviation degree of judgment value and threshold, divide into three levels: I (warning, antifreeze ability decreased), II (mild failure, local icing), III (serious failure, high risk of freeze cracking).

[0081] S4, triggering a hierarchical early warning according to the anti-freezing level, and continuously monitoring and verifying the effect after executing the corresponding treatment, while recording the whole process data into the database to iterate the threshold model, forming a closed loop of judgment-response-optimization, and improving the effectiveness of anti-freezing treatment;

[0082] The specific steps of S4 are as follows:

[0083] S41, make the first level trigger the local indicator light to flash + APP reminder; the second level adds SMS notification to the property; and the third level starts voice alarm + pushes the emergency platform of the water supply company;

[0084] S42, make the first level automatically start the heating band / valve small amplitude opening and closing (promote water flow); the second level remotely close the water inlet valve + send maintenance; and the third level immediately close the water inlet valve + generate an emergency work order;

[0085] S43, continuously monitor for 15 minutes after treatment, if the parameters recover, it is determined to be effective, if not, the level is upgraded, and the invalid data and treatment process are recorded into the database to iterate and optimize the dynamic threshold model.

[0086] A freeze protection water meter freeze protection failure judgment system, please refer to Figure 1 , comprising:

[0087] A multi-dimensional monitoring system construction module is used to adapt the temperature, flow rate and pressure monitoring points according to the water meter installation scene, select low-temperature-resistant high-precision sensors and complete factory and on-site double calibration, and build a LoRa / Wi-Fi dual-mode encrypted transmission link to lay a hardware foundation for subsequent data collection, which is comprehensive, accurate and stable in data transmission;

[0088] A real-time data collection and preprocessing module is used to rely on the monitoring system to synchronously collect multiple parameters according to a dynamic period and add time stamps, ensure key data transmission through priority marking, complete data purification through exception elimination and fluctuation smoothing, and realize parameter magnitude unification through normalization to output effective data with consistent time sequence and no interference, supporting subsequent intelligent judgment;

[0089] An anti-freezing state intelligent judgment module is used to construct a dynamic threshold interval according to the preprocessed data, combined with the water meter model and environmental parameters, calculate the judgment value through a multi-parameter coupling model, and assist in excluding accidental fluctuations by shortening the collection period, and define the anti-freezing level according to the parameter deviation degree to realize accurate identification and false judgment avoidance;

[0090] A failure response and linkage control module is used to trigger a hierarchical early warning according to the anti-freezing level, and continuously monitor and verify the effect after executing the corresponding treatment, while recording the whole process data into the database to iterate the threshold model, forming a closed loop of judgment-response-optimization, and improving the effectiveness of anti-freezing treatment.

[0091] Referring to Figure 2 , the multi-dimensional monitoring system construction module comprises:

[0092] The layout optimization unit is configured to set temperature monitoring points in the water meter body, the water inlet and outlet pipe sections (50 cm before and after), and the inner side of the thermal insulation layer, set a water flow rate monitoring point at the water outlet of the water meter, and set a pressure monitoring point at the water inlet end, according to the water meter installation scene (outdoor / indoor, exposed / concealed), so as to ensure coverage of the key anti-freezing area (position where icing is prone to occur);

[0093] The sensor calibration unit is configured to select low-temperature-resistant (-40℃~60℃) and high-precision sensors (temperature error ±0.1℃, pressure error ±0.01MPa), complete factory calibration through a standard constant-temperature box and a pressure calibrator, and perform secondary calibration by using a comparison method after installation on site, so as to eliminate installation deviation.

[0094] The transmission adaptation unit is configured to adopt a LoRa / Wi-Fi dual-mode transmission channel, so as to adopt LoRa (low power consumption and long distance) in an outdoor scene, adopt Wi-Fi in an indoor scene, and establish an AES encrypted communication link with the water meter controller, so as to ensure data transmission stability.

[0095] Referring to Figure 3 , the real-time data acquisition and preprocessing module comprises:

[0096] The synchronous acquisition unit is configured to enable the controller to synchronously acquire temperature, water flow rate, and pressure data of each monitoring point at a dynamic cycle (regularly 1 minute / once, 0.1 minute / once when the environmental temperature is less than 0℃), and add a time stamp to ensure time sequence consistency.

[0097] The data priority marking unit is configured to add a priority label (high / medium / low) to the acquired data according to the influence weight of parameters on anti-freezing failure (temperature>water flow rate>pressure), so as to, when the transmission bandwidth is insufficient, preferentially ensure transmission of high-priority data (such as temperature) and avoid loss of key information.

[0098] The data purification unit is configured to remove out-of-range data caused by sensor failure by using the 3σ criterion, and smooth water flow rate instantaneous fluctuations by using a sliding average method (window size 5), so as to avoid interference from accidental factors.

[0099] The standard conversion unit is configured to normalize temperature (℃), water flow rate (m / s), and pressure (MPa) data according to historical normal intervals, and convert the data into [0, 1] interval dimensionless indexes, so as to unify the data magnitude for subsequent analysis.

[0100] Referring to Figure 4 , the anti-freezing state intelligent judgment module comprises:

[0101] A threshold dynamic adaptation unit is configured to combine water meter model, installation environment, seasonal parameters, train historical effective / ineffective data through a random forest algorithm, and construct a dynamic threshold interval (for example, when the water flow rate is greater than 0.2 m / s, the temperature threshold is -2℃; when the water flow rate is 0, the temperature threshold is 0℃).

[0102] A coupling judgment unit is configured to construct a multi-parameter coupling model and calculate a judgment value. When the temperature is lower than the dynamic threshold + water flow rate < 0.05 m / s + pressure fluctuation > 0.02 MPa, the failure condition is triggered, and the single parameter exceeding the standard is determined as a risk state.

[0103] An abnormal secondary verification unit is configured to shorten the data acquisition period to 1 / 2 of the original period (for example, from 0.1 minutes / time to 0.05 minutes / time) when the failure / risk condition is triggered for the first time, and continuously acquire 3 groups of data to re-enter the coupling model. If 2 or more groups still meet the condition, the state is confirmed, otherwise it is determined as accidental fluctuation, avoiding single abnormal misjudgment.

[0104] A grade definition unit is configured to divide three grades of I (early warning, reduced anti-freezing ability), II (mild failure, local icing), and III (serious failure, high risk of freeze cracking) according to the deviation of the judgment value from the threshold.

[0105] Please refer to Figure 5 , the failure response and linkage control module includes:

[0106] A warning linkage unit is configured to make the I grade trigger the local indicator light to flash + the APP to remind; the II grade to increase the SMS notification to the property; and the III grade to start the voice alarm + push the emergency platform of the water supply company.

[0107] A disposal execution unit is configured to make the I grade automatically start the heat tracing band / valve small amplitude opening and closing (promote water flow); the II grade to remotely close the water inlet valve + send a single maintenance; and the III grade to immediately close the water inlet valve + generate an emergency work order.

[0108] A feedback optimization unit is configured to continuously monitor for 15 minutes after disposal. If the parameters are restored, it is determined to be effective, if not, the grade is upgraded, and the failure data and disposal process are entered into the database for iterative optimization of the dynamic threshold model.

[0109] Although the present application has been described with reference to the embodiments above, various changes and modifications can be suggested to one skilled in the art, and it is intended that the present application encompass such changes and modifications as fall within the scope of the appended claims. Particularly, each feature disclosed in the description and / or the claims can be used in the combination with each of the features disclosed in the description and / or the claims, unless specifically stated otherwise. Therefore, the present application is not intended to be limited to the particular embodiments disclosed in the description and / or the claims.

Claims

1. A method for determining the failure of an anti-freezing function of an anti-freezing water meter, characterized by, The specific steps include the following: S1, according to the water meter installation scene, the temperature, flow rate, and pressure monitoring points are adapted and arranged, low-temperature-resistant and high-precision sensors are selected, and double calibration of factory and field is completed, and a LoRa / Wi-Fi dual-mode encrypted transmission link is built, laying a hardware foundation for subsequent data collection, which is comprehensive, accurate, and stable in transmission; S2, relying on the monitoring system, multiple parameters are collected synchronously according to a dynamic cycle and are attached with time stamps, and through priority marking, key data transmission is ensured, and data purification is completed through exception elimination and fluctuation smoothing, and parameter magnitude is unified through normalization, outputting effective data with consistent timing and no interference, supporting subsequent intelligent judgment; S3, according to the preprocessed data, a dynamic threshold interval is constructed in combination with the water meter type and environmental parameters, and a judgment value is calculated through a multi-parameter coupling model, and an abnormal secondary verification is supplemented to exclude accidental fluctuations, and a freeze protection level is defined according to the parameter deviation, realizing accurate identification and false judgment avoidance; S4, according to the freeze protection level, a graded early warning is triggered, and the effect is continuously monitored and verified after the corresponding treatment is performed, and the whole process data is entered into the database to iterate the threshold model, forming a closed loop of judgment-response-optimization, and improving the effectiveness of freeze protection treatment.

2. The freeze failure determination method for a freeze-proof water meter according to claim 1, characterized by The specific steps of S1 are as follows: S11, according to the water meter installation scene, temperature monitoring points are arranged on the water meter body, inlet and outlet water pipe sections, and the inner side of the insulation layer, flow rate monitoring points are arranged on the water outlet of the water meter, and pressure monitoring points are arranged on the water inlet end, ensuring that the key areas for freeze protection are covered; S12, low-temperature-resistant and high-precision sensors are selected, and factory calibration is completed through a standard thermostat and a pressure calibrator, and after installation on site, a second calibration is performed using the comparison method to eliminate installation deviations; S13, a LoRa / Wi-Fi dual-mode transmission channel is used, LoRa is used in outdoor scenes, Wi-Fi is used in indoor scenes, and an AES encrypted communication link is established with the water meter controller to ensure data transmission stability.

3. The freeze failure determination method for a freeze-proof water meter according to claim 1, characterized by The specific steps of S2 are as follows: S21, the controller synchronously collects temperature, water flow rate, and pressure data of each monitoring point according to a dynamic cycle, and adds time stamps to ensure timing consistency; S22, according to the influence weight of parameters on freeze failure, priority labels are added to the collected data to ensure high-priority data transmission when the transmission bandwidth is insufficient, and to avoid loss of key information; S23, 3σ criterion is used to eliminate out-of-range data caused by sensor failure, and sliding average method is used to smooth water flow rate instantaneous fluctuations to avoid accidental factors interference; S24, temperature, water flow rate, and pressure data are normalized according to historical normal intervals, and are converted to dimensionless indexes in the [0, 1] interval to unify data magnitude for subsequent analysis.

4. The freeze failure determination method for a freeze-proof water meter according to claim 1, characterized by The specific steps of S3 are as follows: S31, in combination with the water meter type, installation environment, and seasonal parameters, historical effective / inactive data are trained through a random forest algorithm to construct a dynamic threshold interval; S32, a multi-parameter coupling model is constructed to calculate a judgment value, when the temperature is lower than the dynamic threshold + water flow rate <0.05 m / s + pressure fluctuation >0.02 MPa, the failure condition is triggered, and a single parameter exceeding the standard is determined as a risk state. S33, when the failure / risk condition is triggered for the first time, immediately shorten the data acquisition period to 1 / 2 of the original period, and continuously acquire 3 groups of data to re-enter the coupling model. If 2 or more groups still meet the condition, the state is confirmed, otherwise it is determined as accidental fluctuation to avoid single abnormal misjudgment; S34, according to the deviation degree of the judgment value from the threshold value, divide it into three levels: level I, level II and level III.

5. The freeze failure determination method for a freeze-proof water meter according to claim 1, characterized by The specific steps of S4 are as follows: S41, make the level I trigger the local indicator light to flash + APP reminder; level II increases SMS notification property; level III starts voice alarm + pushes the emergency platform of the water supply company; S42, make level I automatically start the heat tracing band / valve small amplitude opening and closing; level II remotely closes the water inlet valve + sends a single maintenance; level III immediately closes the water inlet valve + generates an emergency work order; S43, after treatment, continuously monitor, if the parameters are restored, it is determined to be effective, if not, the level is upgraded, and the failure data and treatment process are recorded in the database to iteratively optimize the dynamic threshold model.

6. A freeze failure determination system for a freeze-proof water meter, characterized by comprising: It includes: A multi-dimensional monitoring system construction module is used to adapt to the installation scene of the water meter, and to set up temperature, flow rate and pressure monitoring points. Low-temperature-resistant and high-precision sensors are selected, and double calibration is completed at the factory and on site. A LoRa / Wi-Fi dual-mode encrypted transmission link is built to lay a hardware foundation for subsequent data acquisition, which is comprehensive, accurate and stable. A real-time data acquisition and preprocessing module is used to rely on the monitoring system to synchronously acquire multiple parameters with time stamps according to a dynamic period. The transmission of key data is guaranteed through priority marking. The data is purified through abnormality elimination and fluctuation smoothing, and the parameter magnitude is unified through normalization. The effective data with consistent timing and no interference is output to support subsequent intelligent judgment. An anti-freezing state intelligent judgment module is used to construct a dynamic threshold interval according to the preprocessed data, combined with the water meter model and environmental parameters. The judgment value is calculated through a multi-parameter coupling model. The abnormal secondary verification to exclude accidental fluctuations is supplemented by shortening the acquisition period. The anti-freezing level is defined according to the parameter deviation degree to achieve accurate identification and misjudgment avoidance. An invalid response and linkage control module is used to trigger graded early warning according to the anti-freezing level. The effect is verified after the corresponding treatment is performed. The whole process data is recorded in the database to iteratively optimize the threshold model, forming a closed loop of judgment-response-optimization to improve the effectiveness of anti-freezing treatment.

7. The freeze failure determination system for a freeze-proof water meter according to claim 6, wherein The multi-dimensional monitoring system construction module includes: A layout optimization unit is used to set temperature monitoring points in the water meter body, inlet and outlet water pipe segments, and the inner side of the insulation layer, set water flow rate monitoring points at the water outlet of the water meter, and set pressure monitoring points at the water inlet end to ensure coverage of key anti-freezing areas. A sensor calibration unit is used to select low-temperature-resistant and high-precision sensors, and to complete factory calibration through a standard thermostat and a pressure calibrator. The second calibration is performed on site using the comparison method to eliminate installation deviation. A transmission adaptation unit is used to adopt LoRa / Wi-Fi dual-mode transmission channels to use LoRa in outdoor scenes and Wi-Fi in indoor scenes. An AES encrypted communication link is established with the water meter controller to ensure data transmission stability.

8. The freeze failure determination system for a freeze-proof water meter according to claim 6, wherein The real-time data acquisition and preprocessing module comprises: A synchronous acquisition unit for synchronously acquiring temperature, water flow rate and pressure data of each monitoring point by the controller according to a dynamic cycle and adding a time stamp to ensure time sequence consistency; A data priority marking unit for adding a priority label to the acquired data according to the influence weight of parameters on anti-freezing failure, so as to ensure high-priority data transmission and avoid loss of key information when the transmission bandwidth is insufficient; A data purification unit for removing out-of-range data caused by sensor failure by using the 3σ criterion, and smoothing water flow rate instantaneous fluctuations by using the moving average method to avoid accidental factors interference; A standard conversion unit for normalizing temperature, water flow rate and pressure data according to historical normal intervals, and converting them into [0, 1] interval dimensionless indexes to unify data magnitudes for subsequent analysis.

9. The freeze failure determination system for a freeze-proof water meter according to claim 6, wherein The anti-freezing state intelligent judgment module comprises: A threshold dynamic adaptation unit for training historical effective / failure data by using a random forest algorithm to construct a dynamic threshold interval in combination with water meter type, installation environment and seasonal parameters; A coupling judgment unit for constructing a multi-parameter coupling model to calculate a judgment value, and triggering a failure condition when the temperature is lower than the dynamic threshold + water flow rate < 0.05 m / s + pressure fluctuation > 0.02 MPa, and a single parameter exceeding the standard is determined as a risk state; An abnormality secondary verification unit for shortening the data acquisition cycle to 1 / 2 of the original cycle and continuously acquiring 3 sets of data to re-enter the coupling model when the failure / risk condition is triggered for the first time, and confirming the state if 2 or more sets of data still meet the condition, otherwise determining it as accidental fluctuation to avoid single abnormal misjudgment; A grade definition unit for dividing the judgment value into three grades of grade I, grade II and grade III according to the deviation degree of the judgment value from the threshold value; The failure response and linkage control module comprises: A warning linkage unit for triggering a local indicator light blinking + APP reminder for grade I, adding a short message notification to the property for grade II, and starting voice alarm + pushing the emergency platform of the water supply company for grade III; A disposal execution unit for automatically starting a heating belt / valve small amplitude opening and closing for grade I, remotely closing the water inlet valve + sending a single maintenance for grade II, and immediately closing the water inlet valve + generating an emergency work order for grade III; A feedback optimization unit for continuously monitoring after disposal, determining effectiveness if the parameters are restored, and upgrading the grade if the parameters are not restored, and recording the failure data and disposal process in the database to iteratively optimize the dynamic threshold model.

Images