Application in remote data acquisition and monitoring systems for water meters

Abstract

This invention discloses a remote data acquisition and monitoring system for water meters, belonging to the field of water management monitoring technology. It includes: an acquisition and preprocessing module: acquiring and purifying water meter volume increment and pressure difference data; a volume factor calculation module: calculating the volume change factor based on volume increment and pressure difference, combined with the influence of pipeline deposition; a jump degree calculation module: calculating the disturbance resistance jump degree based on the volume increment time response and pressure difference frequency domain characteristics; an anomaly scoring calculation module: constructing spatiotemporal correlation features and generating anomaly scores; an alarm generation module: triggering an alarm signal based on a dynamic alarm threshold and anomaly score comparison results; and a fault diagnosis module: diagnosing the fault type when an alarm signal is triggered. By combining pipeline deposition effects, eddy current decay spectra, and time-frequency domain characteristics to calculate the volume change factor and disturbance resistance jump degree, the alarm threshold is dynamically corrected, improving the accuracy of water meter fault early warning.

Description

Technical Field

[0001] This invention relates to the field of water monitoring technology, specifically to a remote data acquisition and monitoring system for water meters. Background Technology

[0002] With the development of smart water meters and IoT technology, remote data acquisition and monitoring of water meters has become one of the key technologies in the field of water management. Traditional water meter monitoring methods mainly rely on manual inspections, which suffer from problems such as untimely data collection, poor accuracy, and high labor costs. Modern water meter monitoring systems, through remote acquisition technology combined with advanced data analysis methods, can obtain key information such as water flow volume increments and pressure differentials in real time. Furthermore, intelligent algorithms process and analyze the water meter data to achieve real-time monitoring, fault warnings, and performance optimization.

[0003] Currently, in the process of water meter data acquisition and monitoring, the accurate measurement and processing of information such as volume increment and pressure difference have become core technical issues for improving system stability and accuracy. Traditional water meter monitoring technology has certain limitations, making it difficult to accurately identify and predict various fault types, such as leakage, backflow, and pipe bursts. To solve this problem, it is necessary to combine multiple sensor technologies with advanced signal processing algorithms to improve the processing accuracy of water meter data and effectively identify potential faults. Summary of the Invention

[0004] Based on the shortcomings of the prior art described above, the purpose of this invention is to provide a remote data acquisition and monitoring system for water meters to solve the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a remote data acquisition and monitoring system for water meters, comprising:

[0006] Acquisition and Preprocessing Module: Used to acquire and preprocess the volume increment and pressure difference data of the water meter to obtain the purified volume increment and pressure difference;

[0007] Volume factor calculation module: used to calculate the volume change factor based on volume increment and pressure difference, combined with the effect of pipeline deposition on pressure difference attenuation;

[0008] Jump degree calculation module: used to calculate the disturbance resistance jump degree based on the time response characteristics of volume increment and the frequency domain response characteristics of pressure difference;

[0009] Anomaly score calculation module: used to construct the spatiotemporal correlation features of volumetric mutation factor and perturbation resistance jump degree, and generate anomaly scores;

[0010] Alarm generation module: Used to generate dynamic alarm thresholds based on the historical average anomaly score plus periodic corrections. When the anomaly score exceeds the dynamic alarm threshold, an alarm signal is triggered.

[0011] Fault diagnosis module: When an alarm signal is triggered, it is used to diagnose the fault type according to the preset judgment logic.

[0012] The present invention is further configured such that the acquisition and preprocessing module includes:

[0013] The volume increment and pressure difference data of the water meter are collected in real time by the sensor to monitor the volume change and pressure difference on both sides of the water meter when the water flows through the water meter.

[0014] The collected volume increment data is processed by Gaussian filtering to obtain the purified volume increment.

[0015] The collected differential pressure data is used to obtain the purified differential pressure by applying a hyperbolic tangent compensation algorithm based on the instantaneous pressure change rate.

[0016] The present invention is further configured such that the volume factor calculation module includes:

[0017] Calculate the volume pressure difference coupling coefficient based on the volume increase and pressure difference after purification;

[0018] The pipe standing wave is excited by transmitting ultrasonic waves at a preset frequency, and the pipe wall standing wave resonance factor is calculated based on the phase offset of the transmitted and received signals and the acoustic impedance ratio of the water body in the pipe wall.

[0019] The volumetric abrupt change factor is calculated based on the volume pressure difference coupling coefficient and the pipe wall standing wave resonance factor.

[0020] The present invention is further configured such that the jump degree calculation module includes:

[0021] Eddy currents are induced by applying a preset constant alternating magnetic field. The eddy current decay spectrum coefficient is calculated based on the initial current and instantaneous current, combined with the magnetic field strength and water conductivity parameters.

[0022] The volume increment after purification is subjected to high-order time-domain difference processing, and the time-domain sharpness feature is obtained by inverse hyperbolic sine transform.

[0023] Extract the spectral energy of the differential pressure change rate within a preset frequency band, and use the time gradient magnitude of the eddy current decay spectrum coefficient as a suppression factor for attenuation weighting.

[0024] The perturbation resistance jump degree is calculated based on the time-domain sharpness characteristics and the attenuated weighted spectral energy.

[0025] The present invention is further configured such that the anomaly scoring calculation module includes:

[0026] One-dimensional convolution kernels are used to process the volume mutation factor and perturbation resistance jump degree to extract local temporal correlation features;

[0027] Perform element-wise product operation on the correlation features and the perturbation resistance jump degree;

[0028] The suppression denominator is constructed based on the time gradient norm of the difference between the volume mutation factor and the perturbation resistance jump degree.

[0029] Calculate the anomaly score based on the result of the element-wise product operation and the suppressed denominator.

[0030] The present invention is further configured such that the alarm generation module includes:

[0031] Perform a moving average integral on the abnormal scores within a preset time window to generate the historical baseline offset.

[0032] The historical baseline offset is superimposed on the preset initial threshold, and a periodic operating condition flag function is introduced to correct the superposition result, and a dynamic alarm threshold is output.

[0033] The real-time anomaly score is compared with the corresponding dynamic alarm threshold. When the anomaly score is greater than the dynamic alarm threshold, an alarm signal is triggered.

[0034] The present invention is further configured such that the fault diagnosis module includes:

[0035] When the volume mutation factor exceeds the preset leakage threshold and the spectral energy of the disturbance resistance jump degree in the preset frequency band is less than the preset low-frequency vibration energy threshold, it is determined to be a leakage fault.

[0036] When the negative time gradient of the volumetric mutation factor is less than the preset critical volatility and the real-time differential pressure is greater than the preset reflux differential pressure threshold, it is determined to be a reflux fault.

[0037] When the maximum value of the time-domain derivative of the disturbance resistance jump degree is greater than the preset burst tube threshold, it is determined to be a burst tube fault.

[0038] The present invention is further configured such that when the absolute deviation between the volume change factor and the disturbance resistance jump degree is greater than a preset deviation threshold, and the duration of this absolute deviation state is greater than a preset holding time, a sensor drift compensation mechanism is triggered, the compensation mechanism including:

[0039] The pipeline is brought into a static flow state by remotely controlling the solenoid valve, and the zero-point offset of the differential pressure is collected under static flow conditions.

[0040] Based on the collected differential pressure zero-point offset, combined with a preset time modulation compensation factor, the current differential pressure sensor data is drift compensated to obtain the compensated differential pressure, and the compensation time identifier is updated to the current time.

[0041] The present invention is further configured such that the system also includes a visualization module for synchronously rendering the volume increment and pressure difference after purification, as well as the time-series curve of the anomaly score.

[0042] The present invention is further configured such that the system also includes a data storage and transmission module, used to upload the collected raw water meter volume increment and pressure difference data, anomaly scores and fault diagnosis results to a remote server according to a preset cycle.

[0043] This invention provides a remote data acquisition and monitoring system for water meters. It includes: a data acquisition and preprocessing module for acquiring and preprocessing volume increment and pressure difference data from water meters to obtain purified volume increment and pressure difference; a volume factor calculation module for calculating a volume change factor based on volume increment and pressure difference, combined with the influence of pipeline deposition on pressure difference attenuation; a jump degree calculation module for calculating disturbance resistance jump degree based on the time response characteristics of volume increment and the frequency domain response characteristics of pressure difference; an anomaly score calculation module for constructing the spatiotemporal correlation characteristics of volume change factor and disturbance resistance jump degree to generate an anomaly score; an alarm generation module for generating a dynamic alarm threshold based on the historical average anomaly score with periodic correction, triggering an alarm signal when the anomaly score exceeds the dynamic alarm threshold; and a fault diagnosis module for diagnosing fault types according to preset judgment logic when an alarm signal is triggered. The beneficial effects include:

[0044] 1. Accurate calculation of volumetric abrupt change factor: Combining the pressure difference attenuation effect of pipeline deposition, the volumetric abrupt change factor is calculated. Through technologies such as pipeline standing waves and eddy current attenuation spectrum, the dynamic changes of water flow in the pipeline can be reflected, and potential fault hazards can be captured in time, thereby providing more accurate fault warnings for water meter monitoring systems.

[0045] 2. Precise calculation of disturbance resistance jump degree: By combining time domain and frequency domain characteristics, and using innovative technologies such as eddy current decay spectrum coefficient and time domain sharpness characteristics, it can efficiently extract and evaluate the disturbance resistance of water flow, monitor minute changes in the fluid state in the pipeline in real time, and provide timely and accurate state assessment to help detect potential faults.

[0046] 3. Anomaly score generation and alarm mechanism: Anomaly scores are generated based on the spatiotemporal correlation features of volume mutation factor and disturbance resistance jump degree. Combined with dynamic correction of historical data, an adaptive alarm threshold is provided. This method can avoid false alarms caused by environmental changes, improve the reliability of the system, and trigger alarms in a timely manner when real faults occur, ensuring the safe operation of the water meter.

[0047] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] In the attached diagram:

[0050] Figure 1 This is a structural diagram illustrating an exemplary embodiment of the present invention for a remote data acquisition and monitoring system for water meters. Detailed Implementation

[0051] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0052] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0053] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0054] Applications in remote data acquisition and monitoring systems for water meters, such as Figure 1 As shown, it includes:

[0055] Acquisition and Preprocessing Module: Used to acquire and preprocess the volume increment and pressure difference data of the water meter to obtain the purified volume increment and pressure difference;

[0056] Volume factor calculation module: used to calculate the volume change factor based on volume increment and pressure difference, combined with the effect of pipeline deposition on pressure difference attenuation;

[0057] Jump degree calculation module: used to calculate the disturbance resistance jump degree based on the time response characteristics of volume increment and the frequency domain response characteristics of pressure difference;

[0058] Anomaly score calculation module: used to construct the spatiotemporal correlation features of volumetric mutation factor and perturbation resistance jump degree, and generate anomaly scores;

[0059] Alarm generation module: Used to generate dynamic alarm thresholds based on the historical average anomaly score plus periodic corrections. When the anomaly score exceeds the dynamic alarm threshold, an alarm signal is triggered.

[0060] Fault diagnosis module: When an alarm signal is triggered, it is used to diagnose the fault type according to the preset judgment logic.

[0061] The present invention is further configured such that the acquisition and preprocessing module includes:

[0062] The volume increment and pressure difference data of the water meter are collected in real time by the sensor to monitor the volume change and pressure difference on both sides of the water meter when the water flows through the water meter.

[0063] The collected volume increment data is processed by Gaussian filtering to obtain the purified volume increment.

[0064] The collected differential pressure data is processed using a hyperbolic tangent compensation algorithm based on the instantaneous pressure change rate to obtain the purified differential pressure. Specifically, the volume increment data and differential pressure data of the water meter are collected in real time by sensors. During the data acquisition process, the data is affected by external factors such as environmental noise, electromagnetic interference, or temperature fluctuations, requiring preprocessing and noise suppression. To improve the reliability of the volume increment data, this embodiment uses Gaussian filtering to smooth it, thus improving the accuracy of the original volume increment data. A Gaussian filter is applied for weighted smoothing to remove random noise and transient fluctuations from the data, resulting in the purified volume increment. , This is the current time; the filter has a width parameter. Used to control the smoothness of the filter. The specific value is determined based on the data acquisition frequency, noise characteristics, and the frequency of the retained signal, with a range of [0.1, 5]. Smaller values ​​are preferred. A value suitable for scenarios with low noise and rapid changes, while a larger value... This is suitable for situations with strong noise or slow changes; Gaussian filtering enhances the accuracy and stability of volume increment data, reduces errors caused by instantaneous fluctuations, and thus reflects the stable trend of water flow; for the collected differential pressure data, this embodiment uses a hyperbolic tangent compensation algorithm based on the instantaneous pressure change rate to eliminate measurement errors caused by environmental changes, targeting the original differential pressure data. The time derivative of the pressure change rate is calculated and corrected using the hyperbolic tangent function to obtain the purified pressure difference. This compensation algorithm can remove the influence of external disturbances such as temperature changes and vibrations, making the differential pressure data more accurate and reflecting the state of water flow in the pipeline more realistically. Through the above data preprocessing and noise suppression steps, the purified volume increment data was obtained. and the pressure difference data after purification This provides more stable and reliable basic data for subsequent calculations. The purified data can better reflect the actual situation in the pipeline, especially when there are fluctuations in water flow or abnormal pressure differences, it can provide more accurate early warning of potential faults.

[0065] The present invention is further configured such that the volume factor calculation module includes:

[0066] Calculate the volume pressure difference coupling coefficient based on the volume increase and pressure difference after purification;

[0067] The pipe standing wave is excited by transmitting ultrasonic waves at a preset frequency, and the pipe wall standing wave resonance factor is calculated based on the phase offset of the transmitted and received signals and the acoustic impedance ratio of the water body in the pipe wall.

[0068] The volumetric abrupt change factor is calculated based on the volumetric pressure difference coupling coefficient and the pipe wall standing wave resonance factor. Specifically, in this embodiment, the volumetric abrupt change factor is calculated by combining the volume increment after purification, the pressure difference after purification, and the pressure difference attenuation correction caused by pipe deposits. This reflects the abrupt change characteristics of the water flow and eliminates the interference of pipe deposits on the measurement results, such as biofilm and scale, thereby enhancing the accuracy and reliability of the data. The calculation logic of the volumetric abrupt change factor is as follows: , For volume mutation factor, For adjustment coefficient, Sensitivity coefficient, For attenuation factor, The standing wave resonance factor of the pipe wall; The value is used to adjust the effect of pressure difference on the coupling coefficient of volume pressure difference, and the value range is [0.1, 5]. Used to adjust the nonlinear relationship between volume increment and pressure difference, ensuring that the interaction between the two can be reasonably adjusted under different pipeline and environmental conditions, with a value range of [1,2]. Used to control the rate of pressure differential decay caused by pipeline deposits, with a value range of [0.1, 1]; The volumetric pressure difference coupling coefficient reflects the dynamic changes in water flow by combining volume increment and pressure difference. Volume increment reflects the change in water flow rate per unit time, while pressure difference reflects the resistance or flow state of the water flow. Combining the two helps to capture the overall trend of water flow changes. To eliminate the interference of pipe deposits, in this embodiment, a preset frequency ultrasonic transmitter is used to excite standing waves in the pipe. The thickness of biofilm or scale on the pipe wall is quantified based on the phase shift of the received signal, and the standing wave resonance factor is calculated by combining it with the acoustic impedance ratio of the water in the pipe. By monitoring the response of ultrasonic standing waves to pipe deposits, the water flow conditions inside the pipe are indirectly reflected. The calculation logic of the standing wave resonance factor is as follows: , For pipe diameter, The phase difference between the transmitted and received waves, For ultrasonic wavelengths in water, The acoustic impedance ratio of the pipe wall to the water, For the density of water, For the speed of sound of water, For the density of the pipe wall material, The velocity of sound in the pipe wall material; the wavelength of ultrasound varies with temperature conditions, and the instantaneous wavelength of ultrasound in water is calculated based on the ambient temperature. The calculation logic is as follows: , The speed of sound in water after temperature adjustment Ultrasonic waves of a preset frequency; This reflects the effect of water temperature on the propagation characteristics of ultrasonic waves; the speed of sound in water after temperature adjustment. The calculation logic is as follows: , The speed of sound in water at the reference temperature For temperature change coefficient, The current temperature of the water, This is the reference temperature for the water; Used to reflect the degree of influence of temperature on the speed of sound in water, the value range is [0.017, 0.020], and the unit is per degree Celsius; by combining the joint analysis of pipe wall standing wave resonance factor, volume increment data and pressure difference data, the influence of pipeline sediments on water flow is quantified, and the accuracy and reliability of water meter monitoring system in complex environments are improved.

[0069] The present invention is further configured such that the jump degree calculation module includes:

[0070] Eddy currents are induced by applying a preset constant alternating magnetic field. The eddy current decay spectrum coefficient is calculated based on the initial current and instantaneous current, combined with the magnetic field strength and water conductivity parameters.

[0071] The volume increment after purification is subjected to high-order time-domain difference processing, and the time-domain sharpness feature is obtained by inverse hyperbolic sine transform.

[0072] Extract the spectral energy of the differential pressure change rate within a preset frequency band, and use the time gradient magnitude of the eddy current decay spectrum coefficient as a suppression factor for attenuation weighting.

[0073] Based on the time-domain sharpness characteristics and attenuation-weighted spectral energy, the disturbance resistance jump degree is calculated. Specifically, this step extracts multi-scale jump information by combining time-domain and frequency-domain features, thereby improving the response capability to sudden events. Based on the time response characteristics of volume increment and the frequency response characteristics of pressure difference, combined with the eddy current attenuation spectral coefficient, the disturbance resistance jump degree is calculated to evaluate the jump characteristics and anti-interference capability of water flow dynamics. The calculation logic of the disturbance resistance jump degree is as follows: , To resist the abrupt change in perturbation degree, For inverse hyperbolic sine function, For time-domain sharpness adjustment coefficient, For frequency domain suppression adjustment coefficient, The spectral energy of the pressure difference change rate within a preset frequency band, For the eddy current decay spectrum coefficient, For time gradient operators, The time gradient of the eddy current decay spectrum coefficient, For attenuation coefficient, For the time-domain sharpness feature term in the logic of perturbation resistance jump degree calculation, Frequency domain suppression term in the logic for calculating the perturbation resistance jump degree; inverse hyperbolic sine function. Used to enhance the response to small changes and suppress oversensitivity to large fluctuations; Used to suppress presets to High-frequency noise or interference signals within the Hz band; In this embodiment, a preset constant alternating magnetic field is used to induce eddy currents. By using real-time current and magnetic field strength, combined with water conductivity, pipe material, and sediment, the concentration of metal ions in the water is inverted, thereby improving the sensitivity to minor disturbances in water flow. Eddy current attenuation spectrum coefficient The calculation logic is as follows: , For time constant, For the initial current, For instantaneous current, Eddy effect intensity factor; Eddy effect intensity factor It is used to reflect the influence of water conductivity, magnetic field strength, and pipe surface deposits on the eddy current effect, and to describe the variation of eddy currents under different material and pipe conditions. The calculation logic is as follows: , For relative permeability, For vacuum permeability, For water conductivity, Magnetic field strength; time gradient of eddy current decay spectrum coefficient Used to describe the effect of eddy currents in water flow on water quality and flow disturbance, representing the rate of change of eddy currents, which is related to the change of metal ion concentration; Used to adjust the contribution of time-domain response characteristics to the degree of abrupt change in resistance to disturbance, and to control the response of time-domain sharpness to sudden changes in water flow, with a value range of [0.1,2]. Used to adjust the contribution of frequency domain response characteristics to the abrupt change in disturbance resistance, with a value range of [0.1, 2]. The value of the eddy current decay spectrum coefficient is used to control the effect of frequency domain suppression, and its range is [0,1]. The decay rate of eddies is used to describe the attenuation rate of eddies, reflecting the influence of sediments on eddy decay. The value range is [0.1, 10], and the unit is seconds. The time-domain sharpness feature focuses on the instantaneous change of the volume increment of the water flow. By calculating the third derivative and the inverse logarithmic function of the volume increment, it captures the instantaneous fluctuations of the water flow and reveals drastic changes in a short period of time. The frequency domain suppression reduces noise interference and enhances the recognition ability of real signals by analyzing pressure difference changes and eddy effects. The combination of the two can more accurately detect disturbances and potential faults in the water flow. Through the eddy effect and frequency domain suppression technology, efficient monitoring of water flow status can be achieved in complex water quality and pipeline environments, improving the ability to identify sudden changes in water flow and potential faults, and enhancing the robustness and stability of the system.

[0074] The present invention is further configured such that the anomaly scoring calculation module includes:

[0075] One-dimensional convolution kernels are used to process the volume mutation factor and perturbation resistance jump degree to extract local temporal correlation features;

[0076] Perform element-wise product operation on the correlation features and the perturbation resistance jump degree;

[0077] The suppression denominator is constructed based on the time gradient norm of the difference between the volume mutation factor and the perturbation resistance jump degree.

[0078] An anomaly score is calculated based on the result of the element-wise product operation and the suppression denominator. Specifically, in this embodiment, the dynamic change characteristics between the volumetric mutation factor and the disturbance resistance jump degree are analyzed, and feature extraction is performed using a convolutional neural network to obtain the anomaly score. This anomaly score is used to quantify the abnormal behavior of the water flow and provide a basis for subsequent fault early warning. The calculation logic of the anomaly score is as follows: , For abnormal scoring, For adjustment coefficient, The symbol for element-wise multiplication; Used to weight gradient differences and amplify the impact of significant fluctuations. For example, when the water flow velocity is high, a larger value needs to be selected to amplify the impact of fluctuations, while for more stable flows, a smaller value is selected to reduce the impact of noise. The value range is [0.1,2]. The differential gradient represents the time gradient difference between the volumetric abrupt change factor and the disturbance resistance abrupt change degree, which is used to reflect the difference in their dynamic behavior in the time dimension. By performing time differentiation on the difference between the volumetric abrupt change factor and the disturbance resistance abrupt change degree, the instantaneous changes in the dynamics of the water flow are captured. This difference reflects the abrupt changes and disturbances in the water flow at different time points. Gradient amplitude adjustment is used to enhance the response to significant changes, helping to amplify the magnitude of changes when water flow changes drastically, thereby accurately capturing sudden events; For convolution operations, one-dimensional convolution is used to smooth the volume change factor and the degree of resistance to perturbation, and to extract local features. Convolution operation helps to discover local fluctuation trends in time series data and integrate these feature information into a more stable result. In this embodiment, the convolution kernel size is preset to 3, mainly based on its advantages in capturing small fluctuations and instantaneous changes in water flow, avoiding excessive smoothing of large fluctuations, while maintaining a high time resolution. The activation function is used to nonlinearly adjust the perturbation resistance to abrupt changes, suppressing negative values ​​to zero, enhancing sensitivity to positive changes, and avoiding interference from negative perturbations; the anomaly score is obtained through the above calculation. It reflects the degree of abnormality of the current water flow state. The higher the abnormality score, the more obvious the abnormal characteristics of the current water flow, and the potential risk of leakage, backflow or pipe burst. By combining gradient difference and convolution operations, it can capture instantaneous abrupt changes and subtle changes in water flow dynamics, thereby enhancing the ability to identify water flow anomalies, and is especially suitable for small disturbances in complex environments.

[0079] The present invention is further configured such that the alarm generation module includes:

[0080] Perform a moving average integral on the abnormal scores within a preset time window to generate the historical baseline offset.

[0081] The historical baseline offset is superimposed on the preset initial threshold, and a periodic operating condition flag function is introduced to correct the superposition result, and a dynamic alarm threshold is output.

[0082] The real-time generated anomaly score is compared with the corresponding dynamic alarm threshold. When the anomaly score is greater than the dynamic alarm threshold, an alarm signal is triggered. Specifically, in this embodiment, the calculation of the dynamic alarm threshold adopts an adaptive adjustment mechanism by introducing a time window and a nighttime adjustment mechanism, thereby optimizing the accuracy and timeliness of water flow anomaly detection and fault early warning. The dynamic alarm threshold is adjusted by combining the weighted integral of historical anomaly scores with a nighttime adjustment coefficient to ensure that the system can adapt to environmental changes in different time periods and reduce the possibility of false alarms at night. When the real-time generated anomaly score is greater than the dynamic alarm threshold, the system triggers an alarm signal. The calculation logic of the dynamic alarm threshold is as follows: , For dynamic alarm thresholds, Preset initial threshold, For adjustment coefficient, For time window, For abnormal scoring, For threshold correction constant, For adjustment coefficient, This is a nighttime indicator function; in this embodiment, the time window... Setting it to 24 hours means that the dynamic alarm threshold is calculated using the weighted average of the anomaly scores from the past 24 hours, smoothing out recent abnormal behavior and enhancing the system's adaptability to long-term trends; Nighttime indicator function This is used to indicate whether the current time belongs to the nighttime period. In this embodiment, the nighttime period is defined as 00:00 to 05:00. When the time is within this period, A value of 1 indicates that it is nighttime; when it is other time periods... The value is 0. The function's purpose is to adjust the abnormal response at night and reduce the probability of false alarms caused by changes in the nighttime environment. The value used to determine the impact of the time-weighted average of outlier scores on dynamic threshold adjustment is [0.05, 0.3]. Used to control the influence of the weighted integral term on the threshold, with a value range of [1,3]; This is used to adjust the alarm threshold during nighttime hours to ensure that minor anomalies detected at night do not falsely trigger alarms. The value range is [0.1, 0.3]. Dynamically adjusting the alarm threshold allows the system to adaptively adjust the alarm triggering conditions under different environmental conditions, thereby improving the accuracy of alarms. By considering past anomaly scores, it can reduce false alarms caused by changes in the environment and pipeline conditions.

[0083] The present invention is further configured such that the fault diagnosis module includes:

[0084] When the volume mutation factor exceeds the preset leakage threshold and the spectral energy of the disturbance resistance jump degree in the preset frequency band is less than the preset low-frequency vibration energy threshold, it is determined to be a leakage fault.

[0085] When the negative time gradient of the volumetric mutation factor is less than the preset critical volatility and the real-time differential pressure is greater than the preset reflux differential pressure threshold, it is determined to be a reflux fault.

[0086] When the maximum value of the time-domain derivative of the disturbance resistance jump degree exceeds the preset pipe burst threshold, a pipe burst fault is determined. Specifically, after the alarm signal is triggered, the system performs fault diagnosis on the abnormal water flow through preset judgment logic to help identify potential fault types. Fault type diagnosis includes leakage judgment, backflow judgment, and pipe burst judgment. The judgment of leakage fault is based on the combination of the volume change factor and the disturbance resistance jump degree. The judgment logic for leakage fault is as follows: ,in, For volume mutation factor, Preset leakage threshold, To resist the abrupt change in the perturbation degree at the preset value to Spectral energy in the Hz band To preset the low-frequency vibration energy threshold; Leakage detection is used to determine whether a leakage fault exists in a pipeline system. When, it indicates that a leakage fault has been detected. When the time is specified, it indicates that no leakage fault was detected. The leakage detection method can identify minute leaks in the pipeline. By combining the characteristics of volume changes and low-frequency vibrations, it improves the accuracy of leakage detection and reduces the influence of background noise or external interference. The determination of backflow faults is based on the time gradient of the volume change factor and the pressure difference change. The logic for determining backflow faults is as follows: ,in, The time gradient of the volume mutation factor, To preset the critical volatility, For real-time differential pressure, This is the reflux differential pressure threshold. Leakage detection is used to determine whether backflow has occurred in a pipeline. When, it indicates that a return flow fault has been detected. When the signal is clear, it indicates that no backflow fault was detected. The backflow detection method can quickly identify and handle potential backflow problems, avoiding safety hazards caused by water backflow. The determination of pipe burst faults is based on the abrupt change rate of the disturbance resistance jump degree. The logic for determining pipe burst faults is as follows: ,in, To preset the pipe burst threshold, Let be the time-domain derivative of the perturbation resistance jump degree, and represent the mutation rate of the perturbation resistance jump degree; Pipe burst detection is used to determine whether a pipe burst has occurred. When this occurs, it indicates that a pipe burst fault has been detected. When the time is right, it indicates that no pipe burst fault has been detected. The pipe burst detection method can detect pipe bursts in the early stages, reduce the scope of pipe damage, and improve the system response capability. By implementing the above fault diagnosis steps, an efficient fault detection method is provided, which can identify abnormalities in water meter data in real time, such as leakage, backflow, and pipe burst faults, and issue timely alarms. The application of this method can improve the fault response capability of the water meter monitoring system, reduce the false alarm rate, and improve the reliability and stability of the water meter detection system.

[0087] The present invention is further configured such that when the absolute deviation between the volume change factor and the disturbance resistance jump degree is greater than a preset deviation threshold, and the duration of this absolute deviation state is greater than a preset holding time, a sensor drift compensation mechanism is triggered, the compensation mechanism including:

[0088] The pipeline is brought into a static flow state by remotely controlling the solenoid valve, and the zero-point offset of the differential pressure is collected under static flow conditions.

[0089] Based on the collected differential pressure zero-point offset, combined with a preset time modulation compensation factor, drift compensation is performed on the current differential pressure sensor data to obtain the compensated differential pressure, and the compensation time identifier is updated to the current time. Specifically, since the sensor may be affected by long-term use, environmental changes, or other factors, causing its readings to drift, the system needs to periodically perform drift calibration on the sensor to ensure measurement accuracy. In this embodiment, the calibration of the sensor includes the following steps: When the difference between the volumetric abrupt change factor and the disturbance resistance abrupt change degree is greater than a preset deviation threshold, it indicates that there is a significant change in the water flow state or the sensor has malfunctioned. When sensor drift occurs and this state persists for more than a preset duration, the compensation calibration process is triggered. When the sensor drift compensation mechanism is triggered, the remotely controlled solenoid valve puts the pipeline into a static flow state, i.e., the water flow stops. By closing the water valve, calibration is ensured in a static environment without water flow, avoiding measurement errors caused by dynamic water flow. In the static flow state, the current differential pressure data is recorded and stored as the zero-point offset. The zero-point offset represents the current reference value of the sensor when there is no water flow, providing basic data for the compensation calculation after calibration. Compensation calculations are performed based on the zero-point offset to correct the sensor's drift error. The differential pressure compensation calculation logic is as follows: , For the compensated pressure difference, The original pressure difference at the current moment, Zero offset For adjustment coefficient, The timestamp of the last compensation calibration time; The rate used to control sensor drift compensation ranges from [0.1, 1]. Using the compensation calculation formula described above, based on historical zero-point offsets and the current time, the current differential pressure data is gradually corrected to reduce sensor drift error. The timestamp is updated after each compensation calibration. The current time is used to mark the moment of this calibration and to provide a reference time for the next compensation calibration; the compensated pressure difference is output. The differential pressure data, after compensation and calibration, eliminates sensor drift error, provides more accurate measurement results, and ensures that subsequent analysis and decision-making are based on reliable data. When sensor drift occurs, calibration can adjust the measurement results in a timely manner, prevent the accumulation of errors, and avoid long-term measurement errors caused by drift.

[0090] The invention further includes a visualization module for synchronously rendering the volume increment and pressure difference after purification, as well as the time-series curves of anomaly scoring. Specifically, this embodiment provides a visualization module for a remote water meter data acquisition and monitoring system. The main function of this module is to synchronously render and display the time-series curves of the volume increment, pressure difference, and anomaly scoring after purification in real time. Volume increment and pressure difference data from the water meter are collected at fixed time intervals, such as every 5 seconds. After preprocessing, this data is transmitted to the visualization module for real-time display. Data synchronization is achieved via the WebSocket protocol, and the purified data is rendered into clear time-series curves using the D3.js library in JavaScript. The volume increment and pressure difference curves are displayed in blue and green, respectively, while the anomaly scoring is displayed in red. The system uses color-coded curves to differentiate data types. In the user interface, the timeline is located at the top, allowing users to select specific time periods by sliding the timeline. The data display area shows three curves: the volume increase after purification, the pressure difference, and the anomaly score. Users can choose to show or hide these curves. For the anomaly score, when it exceeds the dynamic alarm threshold, the system automatically marks a warning on the curve and pops up a warning box, informing the user of the anomaly type and specific fault data. Users can click the "View Details" button to further view relevant fault diagnosis information, helping them quickly identify problems and respond. In this way, users can intuitively understand the water meter's working status, monitor the system's operation in real time, and take timely measures when anomalies occur, improving the overall monitoring system's response speed and accuracy.

[0091] The invention is further configured such that the system also includes a data storage and feedback module, used to upload the collected raw water meter volume increment and pressure difference data, anomaly scores, and fault diagnosis results to a remote server according to a preset period. Specifically, the data storage and feedback module in this invention is used to realize the continuous acquisition, storage, and periodic feedback functions of water meter data. This module can collect the water meter's volume increment, pressure difference, anomaly scores, and fault diagnosis results from the sensor and computing module in real time and store them in a local cache unit. To ensure that data is not lost, the data cache unit supports temporary data storage during data transmission and provides seamless caching functionality during data upload. The core function of this module is to feedback the data to the remote server according to a preset period for subsequent data analysis and historical data tracing. Data transmission is completed through a standard communication protocol and can select different network communication methods, such as Wi-Fi, cellular networks, or Bluetooth, according to specific application requirements. To ensure data security, this module employs encryption technologies such as AES or TLS to encrypt transmitted data, guaranteeing data integrity and confidentiality during transmission. In practical applications, the data storage and transmission module automatically performs data acquisition, storage, and uploading tasks according to a preset cycle. Acquired data is pre-processed before storage, and data integrity and security are ensured during uploading. If a network interruption occurs during uploading, the data is automatically cached, and the upload request is re-initiated after network recovery, ensuring successful data transmission back to the remote server. This module boasts high reliability and flexibility, allowing adjustment of the data upload cycle and selection of suitable communication methods based on actual needs. Through caching mechanisms and encryption measures, it ensures data security, improves work efficiency, reduces manual intervention, and enhances the system's automation level. The above design supports the stable operation of the water meter data remote monitoring system and provides crucial historical data support and fault tracing functions.

[0092] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A remote data acquisition and monitoring system for water meters, characterized in that, include: Acquisition and Preprocessing Module: This module collects and preprocesses the volume increment and pressure difference data of the water meter to obtain the purified volume increment and pressure difference. This includes: real-time acquisition of the water meter's volume increment and pressure difference data via sensors, monitoring the volume change as water flows through the water meter and the pressure difference across the meter; Gaussian filtering of the collected volume increment data to obtain the purified volume increment; and applying a hyperbolic tangent compensation algorithm based on the instantaneous pressure change rate to the collected pressure difference data to obtain the purified pressure difference. Volume factor calculation module: Used to calculate the volume abrupt change factor based on volume increment and pressure difference, combined with the effect of pipeline deposition on pressure difference attenuation. This includes: calculating the volume pressure difference coupling coefficient based on the volume increment and pressure difference after purification; stimulating pipeline standing waves by transmitting ultrasonic waves at a preset frequency, and calculating the pipe wall standing wave resonance factor based on the phase offset of the transmitted and received signals and the acoustic impedance ratio of the pipe wall water; and calculating the volume abrupt change factor based on the volume pressure difference coupling coefficient and the pipe wall standing wave resonance factor. The abrupt change degree calculation module is used to calculate the disturbance resistance abrupt change degree based on the time response characteristics of the volume increment and the frequency domain response characteristics of the pressure difference. This includes: inducing eddy currents by applying a preset constant alternating magnetic field; calculating the eddy current decay spectrum coefficient based on the initial current and instantaneous current, combined with the magnetic field strength and water conductivity parameters; performing high-order time-domain differential processing on the purified volume increment; obtaining the time-domain sharpness characteristics using an inverse hyperbolic sine transform; extracting the spectral energy of the pressure difference change rate within a preset frequency band; using the time gradient magnitude of the eddy current decay spectrum coefficient as a suppression factor for attenuation weighting; and calculating the disturbance resistance abrupt change degree based on the time-domain sharpness characteristics and the attenuated weighted spectral energy. Anomaly scoring module: This module is used to construct the spatiotemporal correlation features of the volumetric mutation factor and the perturbation resistance jump degree, and generate anomaly scores. The process includes: processing the volumetric mutation factor and the perturbation resistance jump degree with a one-dimensional convolution kernel to extract local temporal correlation features; performing element-wise product operations on the correlation features and the perturbation resistance jump degree; constructing a suppression denominator based on the temporal gradient norm of the difference between the volumetric mutation factor and the perturbation resistance jump degree; and calculating the anomaly score based on the element-wise product result and the suppression denominator. Alarm generation module: This module generates a dynamic alarm threshold based on the historical average anomaly score and periodic corrections. When the anomaly score exceeds the dynamic alarm threshold, an alarm signal is triggered. The module includes: performing a moving average integral calculation on the anomaly scores within a preset time window to generate a historical baseline offset; superimposing the historical baseline offset onto a preset initial threshold, introducing a periodic operating condition flag function to correct the superposition result, and outputting the dynamic alarm threshold; comparing the real-time generated anomaly score with the corresponding dynamic alarm threshold, and triggering an alarm signal when the anomaly score exceeds the dynamic alarm threshold. Fault diagnosis module: When an alarm signal is triggered, it is used to diagnose the fault type according to preset judgment logic, including: when the volume change factor exceeds the preset leakage threshold and the spectral energy of the disturbance resistance jump degree in the preset frequency band is less than the preset low-frequency vibration energy threshold, it is judged as a leakage fault; when the negative time gradient of the volume change factor is less than the preset critical fluctuation rate and the real-time differential pressure is greater than the preset return differential pressure threshold, it is judged as a return fault; when the maximum value of the time domain derivative of the disturbance resistance jump degree is greater than the preset pipe burst threshold, it is judged as a pipe burst fault. When the absolute deviation between the volume change factor and the disturbance resistance jump degree is greater than a preset deviation threshold, and the duration of this absolute deviation state is greater than a preset holding time, a sensor drift compensation mechanism is triggered. The compensation mechanism includes: using a remotely controlled solenoid valve to put the pipeline into a static flow state, and collecting the differential pressure zero-point offset under static flow conditions; based on the collected differential pressure zero-point offset, combined with a preset time modulation compensation factor, performing drift compensation on the current differential pressure sensor data to obtain the compensated differential pressure, and updating the compensation time identifier to the current time.

2. The remote data acquisition and monitoring system for water meters according to claim 1, characterized in that, The system also includes a visualization module for synchronously rendering the volume increment and pressure difference after purification, as well as the time-series curve of the anomaly score.

3. The remote data acquisition and monitoring system for water meters according to claim 1, characterized in that, The system also includes a data storage and transmission module, which is used to upload the collected raw water meter volume increment and pressure difference data, anomaly scores and fault diagnosis results to a remote server according to a preset cycle.

Images