Operation state monitoring method and system based on intelligent water pump

Abstract

The invention discloses an operation state monitoring method and system based on an intelligent water pump, and relates to the technical field of water pump monitoring, and the method comprises the steps: obtaining the basic working condition parameters of the current operation of the water pump, and calculating the theoretical resonant frequency interval of a sealing cavity system based on the parameters and the inherent characteristics of the water pump; a target frequency scanning sequence is generated, the sequence is used for sequentially adjusting the rotating speed or flow of the water pump to a set of discrete steady-state values, and the system excitation main frequency corresponding to each steady-state value is included in the theoretical resonant frequency interval; according to the invention, dependence on macroscopic leakage signals is broken through, early micro-leakage is identified by using the characteristic distribution rule of resonance energy, the detection sensitivity is significantly improved, and early warning can be given in time at the initial stage of leakage.

Description

Technical Field

[0001] This invention belongs to the field of water pump monitoring technology, specifically a method and system for monitoring the operating status of intelligent water pumps. Background Technology

[0002] In industrial fluid transport systems, water pumps are core power equipment, and the sealing performance of their mechanical seals directly determines the stability and safety of equipment operation. Early micro-leakage of mechanical seals is difficult to detect effectively through traditional methods such as pressure monitoring and flow measurement because the leakage amount is small and the characteristic signal is weak. However, once micro-leakage develops into obvious leakage, it will not only cause medium loss and system pressure fluctuations, but may also lead to equipment shutdown, medium contamination, or even safety accidents.

[0003] In existing technologies, the detection of pump seal leaks mostly focuses on the mid-to-late stages of leakage, with common methods including monitoring the absolute value of the sealing cavity pressure and visual inspection of the leaking medium. These methods rely on macroscopic signal changes after the leakage reaches a certain scale, failing to capture the subtle physical changes caused by early micro-leakage. Some technologies attempt to detect sealing faults through vibration signal analysis, but these often focus on the mechanical wear of the seal components, neglecting to study the resonant characteristics of the sealing cavity excited by micro-leakage flow. They also lack frequency analysis methods that combine pump operating parameters with the inherent characteristics of the sealing cavity system, resulting in low sensitivity and high false alarm rates for early micro-leakage identification, making it difficult to meet the early warning requirements for pump sealing status in industrial scenarios. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for monitoring the operating status of an intelligent water pump, so as to solve the problems mentioned in the background art.

[0005] A method for monitoring the operational status of a smart water pump, used to detect early micro-leakage in the mechanical seal of a water pump, includes: Obtain the basic operating parameters of the water pump at present, and calculate the theoretical resonant frequency range of the sealing cavity system based on these parameters and the inherent characteristics of the water pump. A target frequency scanning sequence is generated, which is used to adjust the pump speed or flow rate to a set of discrete steady-state values ​​in sequence, wherein the system excitation frequency corresponding to each steady-state value is contained within the theoretical resonant frequency range; The target frequency scanning sequence is executed sequentially, and at each steady-state operating point, the pressure signal of the sealed cavity is collected, and its oscillation energy in the theoretical resonant frequency range is extracted. Analyze the distribution law of the oscillation energy as a function of the steady-state operating point in the sequence; If the distribution pattern shows an isolated energy peak that appears near a specific steady-state operating point and whose energy value exceeds a set threshold, then the energy peak is determined to correspond to a sealed cavity resonant mode excited by early micro-leakage flow, and an early micro-leakage warning is generated.

[0006] This invention, through in-depth research on the fluid-structure interaction vibration characteristics of a sealed cavity, discovers that early micro-leakage flow alters the internal fluid dynamics environment of the sealed cavity, changing its inherent resonance characteristics and thereby exciting specific resonant modes. This discovery uncovers a hidden correlation between micro-leakage faults and sealed cavity resonance, breaking through the limitations of existing technologies that analyze sealing faults solely from a mechanical structural perspective, and providing a completely new technical approach for micro-leakage detection.

[0007] As a further aspect of the present invention: the target frequency scanning sequence is generated in the following manner: Select the starting and ending frequencies within the theoretical resonant frequency range; The basic frequency step size is determined based on the pump's speed and flow control accuracy. Starting from the initial frequency and with the basic frequency step size as the interval, a series of target frequency values ​​covering the termination frequency are generated sequentially. Each target frequency value corresponds to an instruction to adjust the water pump to the corresponding steady-state operating point.

[0008] As a further aspect of the present invention, the specific process for extracting oscillation energy is as follows: The frequency domain spectrum of the collected steady-state pressure signal of a sealed cavity is obtained by performing a fast Fourier transform. In the frequency domain spectrum, locate the frequency band corresponding to the theoretical resonant frequency range; Calculate the sum of squares of the amplitudes of all spectral lines within the frequency band, and use this as the oscillation energy value at the steady-state operating point.

[0009] As a further aspect of the present invention, the specific process for analyzing the distribution pattern is as follows: Plot a scatter plot with the excitation frequency corresponding to each steady-state operating point as the horizontal axis and the corresponding oscillation energy value as the vertical axis. The scatter plot is interpolated to form a continuous energy-frequency relationship curve; On the curve, find all local maxima points where the energy value of that point is greater than the energy values ​​of its left and right adjacent interpolation points.

[0010] As a further aspect of the present invention: the condition for determining an energy peak as an isolated energy peak is: The energy value corresponding to the local maximum point exceeds a set multiple of the value corresponding to a background trend line calculated based on all oscillation energy values. Furthermore, within a fundamental frequency step to the left and right of the local maximum frequency, there are no other local maxima.

[0011] As a further aspect of the present invention: after determining the existence of an isolated energy peak, a frequency verification step is also included: The precise frequency value corresponding to the isolated energy peak is compared with the theoretical resonant frequency range; If the frequency value is within the theoretical resonant frequency range, then the early micro-leakage warning is confirmed.

[0012] As a further aspect of the present invention, the method also includes: After the first suspected isolated energy peak is identified and an early warning is generated, the target frequency scanning sequence is repeated during subsequent equipment start-up and shutdown processes; If, in subsequent consecutive executions, local maxima points with energy exceeding a set multiple of the background trend line are repeatedly identified at the same or adjacent frequency positions within the theoretical resonant frequency range, the level and certainty of early micro-leakage warning will be improved.

[0013] As a further aspect of the present invention: while generating an early warning of micro-leakage, the precise frequency value and water pump speed value corresponding to the isolated energy peak are automatically recorded and locked to form a characteristic fingerprint; During subsequent normal operation monitoring, the system continuously listens to whether there is an oscillating component in the sealing cavity pressure signal that is consistent with the frequency of the characteristic fingerprint, and monitors the historical trend of the amplitude of the component.

[0014] As a further aspect of the present invention: after determining the existence of an isolated energy peak, the method further includes a step of accurate peak shape mapping. Centered on the frequency corresponding to the isolated energy peak; Set a fine rescan step size, which is smaller than the base frequency step size; Control the pump speed or flow rate to adjust it to multiple steady-state values ​​sequentially within a narrow frequency range including the center frequency, with the fine rescan step size as the interval; At each steady-state value, the pressure signal of the sealed cavity is reacquired and its oscillation energy is calculated. Based on the higher resolution energy data obtained within the narrow band, a refined energy-frequency relationship curve is plotted to confirm the complete shape and precise frequency of the isolated energy peak.

[0015] Secondly, the present invention provides an intelligent water pump-based operation status monitoring system for detecting early micro-leakage of the water pump mechanical seal. The system includes: The acquisition module acquires the basic operating parameters of the water pump and, based on these parameters and the inherent characteristics of the water pump, calculates the theoretical resonant frequency range of the sealing cavity system. The generation module generates a target frequency scanning sequence, which is used to adjust the pump speed or flow rate to a set of discrete steady-state values ​​in sequence, wherein the system excitation frequency corresponding to each steady-state value is contained within the theoretical resonant frequency range; The execution module sequentially executes the target frequency scanning sequence and, at each steady-state operating point, acquires the pressure signal of the sealed cavity and extracts its oscillation energy within the theoretical resonant frequency range. The analysis module analyzes the distribution pattern of the oscillation energy as it changes with the steady-state operating point in the sequence; If the distribution pattern shows an isolated energy peak that appears near a specific steady-state operating point and whose energy value exceeds a set threshold, the early warning module determines that the energy peak corresponds to a sealed cavity resonant mode excited by early micro-leakage flow and generates an early micro-leakage warning.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This method calculates the theoretical resonant frequency range by combining pump operating parameters with the inherent characteristics of the sealing cavity system, and generates a targeted frequency scanning sequence to accurately capture the resonant signal of the sealing cavity excited by early micro-leakage flow. Compared with traditional detection methods, this method breaks through the dependence on macroscopic leakage signals and uses the characteristic distribution law of resonant energy to identify early micro-leakage, thus improving detection sensitivity and enabling timely early warning in the early stage of leakage. At the same time, by using a discrete steady-state value scanning method, signal interference in non-target frequency ranges is reduced, lowering the false judgment rate. This provides a reliable technical basis for preventive maintenance of pump mechanical seals, helps reduce equipment downtime due to failure, and improves the operational reliability of industrial fluid transportation systems.

[0017] 2. This invention generates a target frequency scanning sequence based on the theoretical resonant frequency range of the sealed cavity system. It actively adjusts the pump speed or flow rate to ensure the system's excitation frequency covers the theoretical resonant frequency range, thereby effectively exciting the sealed cavity resonant mode corresponding to early micro-leakage and significantly increasing the intensity of the fault characteristic signal. Furthermore, it solves the problem of weak and easily noise-overwhelmed micro-leakage characteristic signals in passive detection methods, achieving a technological breakthrough from passive sensing to active excitation, thus improving the effectiveness and reliability of early micro-leakage detection. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the method framework structure of the present invention.

[0019] Figure 2 This is a schematic diagram of the system framework structure of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be understood that research has shown that when early micro-leakage occurs in a mechanical seal, the leaking medium (liquid) experiences microscale, high-frequency periodic disturbances or vortex shearing as it passes through the extremely narrow sealing gap due to intense shearing and pressure abrupt changes. This disturbance acts as an excitation source, and when its frequency approaches a certain natural frequency of the sealing cavity system (including the fluid and structure within the cavity), it induces significant fluid-structure interaction resonance. This resonance manifests as a sharp increase in the oscillation energy of the sealing cavity pressure signal at a specific frequency, forming a prominent, isolated energy peak. The frequency characteristics of this peak are determined by the system characteristics of the sealing cavity, and its appearance and intensity indicate the presence of micro-leakage.

[0022] First aspect: Please refer to Figure 1 This invention provides a method for monitoring the operating status of an intelligent water pump, used to detect early micro-leakage in the mechanical seal of a water pump, including: Obtain the basic operating parameters of the water pump and, based on these parameters and the inherent characteristics of the water pump, calculate the theoretical resonant frequency range of the sealing cavity system.

[0023] In this embodiment, the basic operating parameters include the real-time speed of the water pump, inlet and outlet pressure, medium temperature, medium viscosity, and geometric dimensions such as the inner diameter and wall thickness of the sealing cavity. The inherent characteristics of the water pump include the stiffness coefficient of the rotating shaft, the damping characteristics of the sealing cavity material, and the contact stiffness of the dynamic and static rings. These are conventional knowledge in the field and will not be elaborated further.

[0024] Furthermore, when calculating the theoretical resonant frequency range, the fluid mechanics and structural dynamics coupling analysis logic commonly used in the industrial field is adopted. First, the fundamental resonant frequency of the medium in the sealed cavity is calculated through the fluid mechanics model. This model can be initially estimated using the simplified formula f=k×μ / d, where f is the fundamental resonant frequency of the flow and k is an empirical coefficient related to the structure of the sealed cavity. It is selected by combining the wall thickness and material characteristics of the sealing cavity of a conventional industrial water pump and is taken as 0.002. Different structural sealing cavities can be calibrated through three or more pre-experiments. The calibration method is to select three or more sealing cavity samples with different structural parameters, and under the same medium viscosity and operating pressure, measure the fundamental flow resonance frequency of each sample through experiments, and then substitute it into the formula to calculate the k value. The average value is taken as the k value of this type of structural sealing cavity. The calibration range is usually 0.001-0.003; μ is the medium viscosity, in mPa·s; d is the inner diameter of the sealing cavity, in mm.

[0025] It should be noted that this model uses the viscosity of the medium and the inner diameter of the sealed cavity as the core variables, and determines the reference value of the fundamental frequency according to the conventional law that the fundamental frequency of medium flow resonance is positively correlated with the viscosity of the medium and negatively correlated with the inner diameter of the sealed cavity. Then, the structural vibration characteristics of the sealed cavity are analyzed through a structural dynamics model. Combined with the vibration constraint effect of the shaft stiffness coefficient on the sealed cavity, the correction coefficient of the cavity structure to the fundamental frequency is calculated. This coefficient is estimated based on the linear relationship between the shaft stiffness and the wall thickness of the sealed cavity, and the typical range is 0.9-1.1. Finally, based on the changes in medium density caused by medium temperature and the deformation of the cavity caused by inlet and outlet pressure, the fundamental frequency reference value is corrected twice by multiplying by temperature correction factors of 0.95 to 1.05 and pressure correction factors of 0.98 to 1.02 respectively, and the theoretical resonant frequency range of 200Hz to 500Hz is finally obtained.

[0026] It should be further explained that the calculation of the theoretical resonant frequency range is based on the fluid-structure interaction principle of fluid mechanics and structural dynamics. It can be obtained through analytical calculation, estimation using empirical formulas, or numerical simulation using industrial fluid simulation software (such as FLUENT, CFX, etc.). Those skilled in the art can select or combine the above methods to determine the range based on the specific structural parameters of the sealed cavity, the properties of the medium, and the operating conditions; further elaboration is not required here.

[0027] Generate a target frequency scanning sequence (a speed adjustment sequence for point-by-point excitation of the sealed cavity resonance). The sequence is used to adjust the pump speed or flow rate to a set of discrete steady-state values ​​in sequence, where the system excitation frequency corresponding to each steady-state value is contained within the theoretical resonant frequency range.

[0028] The target frequency scanning sequence is generated as follows: the starting frequency and the ending frequency are selected within the theoretical resonant frequency range; the basic frequency step size is determined based on the pump speed and flow control accuracy; a series of target frequency values ​​covering the ending frequency are generated sequentially with the starting frequency as the starting point and the basic frequency step size as the interval, and each target frequency value corresponds to an instruction to adjust the pump to the corresponding steady-state operating point.

[0029] In this embodiment, considering the excitation frequency range of the water pump during normal operation, the starting frequency is selected as 220Hz and the ending frequency as 380Hz within the theoretical resonant frequency range of 200Hz to 500Hz. Next, the basic frequency step size is determined based on the pump speed and flow control accuracy. In this embodiment, the pump speed control accuracy is ±10r / min and the flow control accuracy is ±0.1m³ / h, which corresponds to a frequency control accuracy of ±5Hz. Based on this, the basic frequency step size is determined to be 40Hz. This step size not only meets the control accuracy requirements, but also avoids the reduction in detection efficiency caused by too many scanning points. Starting from the initial frequency and using the base frequency step size as the interval, a series of target frequency values ​​covering the termination frequency are generated sequentially: 220Hz, 260Hz, 300Hz, 340Hz, and 380Hz. Combining this with the linear mapping relationship between pump speed and system excitation frequency (i.e., for every 300 r / min increase in pump speed, the system excitation frequency increases by approximately 40Hz), the target frequency values ​​correspond to discrete steady-state speed values ​​of 1500 r / min, 1800 r / min, 2100 r / min, 2400 r / min, and 2700 r / min, respectively. This generates a target frequency scanning sequence containing these five steady-state speed values. The speed adjustment interval in the sequence is set to 10 seconds, taking into account the speed response characteristics of typical small and medium-power pumps, ensuring that the pump has sufficient time to reach a steady-state operating state after speed switching. This interval can be adjusted within the range of 5 to 15 seconds depending on the pump power.

[0030] Then, the target frequency scanning sequence is executed sequentially, and at each steady-state operating point, the pressure signal of the sealed cavity is acquired, and its oscillation energy within the theoretical resonant frequency range is extracted.

[0031] The specific process for extracting oscillation energy is as follows: perform a fast Fourier transform on the collected steady-state pressure signal of a sealed cavity to obtain its frequency domain spectrum; locate the frequency band corresponding to the theoretical resonant frequency range in the frequency domain spectrum; calculate the sum of squares of the amplitudes of all spectral lines in the frequency band as the oscillation energy value at the steady-state operating point.

[0032] Specifically, the water pump speed is adjusted in the order of 1500 r / min, 1800 r / min, 2100 r / min, 2400 r / min, and 2700 r / min. The water pump is kept in steady-state operation for 30 seconds at each speed. In combination with the conventional requirement of steady-state pressure signal acquisition, the pressure signal is acquired by a pressure sensor installed on the inner wall of the sealed cavity. The sampling frequency of the pressure sensor is 2000 Hz, which is a mature existing technology and will not be elaborated on here.

[0033] The acquired pressure signal is preprocessed. First, the DC component and low-frequency noise below 50Hz are removed from the signal by digital filtering. Then, the processed signal is converted to the frequency domain by fast Fourier transform. During the conversion, a Hanning window with a length of 1024 data points is used to window the signal. This length can balance the spectral resolution and computational efficiency to reduce the problem of spectral leakage. Then, the frequency band corresponding to the theoretical resonant frequency range of 200Hz to 500Hz is located in the obtained frequency domain spectrum.

[0034] When calculating the oscillation energy, effective signal data within a 30-second acquisition period is first selected for each steady-state operating point. Abnormal peak data exceeding three times the standard deviation of the signal average are removed. This can be achieved using common data analysis tools such as MATLAB and Python's Pandas library. After performing the aforementioned fast Fourier transform on the processed signal to obtain the frequency domain spectrum, the amplitudes of all spectral lines within the 200Hz to 500Hz frequency band are extracted, and the sum of squares of these amplitudes is calculated. Finally, the oscillation energies corresponding to each steady-state operating point are obtained as 0.02J, 0.03J, 0.15J, 0.04J, and 0.025J, respectively. This result matches the baseline energy under no-leakage conditions and the energy range of the micro-leakage simulation experiment.

[0035] The distribution law of oscillation energy with the steady-state operating points in the sequence is analyzed. The specific process of analyzing the distribution law is as follows: the excitation frequency corresponding to each steady-state operating point is used as the abscissa, and the corresponding oscillation energy value is used as the ordinate to draw a scatter plot; the scatter plot is interpolated to form a continuous energy-frequency relationship curve; on the curve, all local maxima points that satisfy the condition that the energy value of the point is greater than the energy values ​​of its left and right adjacent interpolation points are found.

[0036] In this embodiment, the excitation frequencies of 220Hz, 260Hz, 300Hz, 340Hz, and 380Hz corresponding to each steady-state operating point are first used as the horizontal axis, and the corresponding oscillation energy values ​​of 0.02J, 0.03J, 0.15J, 0.04J, and 0.025J are used as the vertical axis. A scatter plot is then drawn in industrial data visualization software such as MATLAB and Origin. Linear interpolation is used to interpolate the scatter plot, with data points supplemented at 5Hz intervals. This interval can balance curve smoothness and data volume, forming a continuous energy-frequency relationship curve. This interpolation method balances computational efficiency and curve smoothness. Those skilled in the art can choose other methods such as cubic spline interpolation according to the data accuracy requirements. On the generated continuous curve, the energy values ​​of each interpolation point are compared with those of the left and right adjacent interpolation points to find local maxima. Finally, the unique local maximum point is located at 300Hz, corresponding to a steady-state speed of 2100r / min.

[0037] Simultaneously, the energy change rate between adjacent steady-state operating points was calculated, which is the difference between the energy of the subsequent steady-state point and the energy of the preceding steady-state point divided by the energy of the preceding steady-state point. It was observed that the energy change rate of the preceding steady-state point at 260Hz corresponding to 300Hz was 400%, and the energy change rate of the subsequent steady-state point at 340Hz was 73.3%. The oscillation energy of the remaining steady-state operating points was at a low level, and the energy change rate was all below 50%, further verifying the significance of this local maximum point.

[0038] If the distribution pattern shows an isolated energy peak that appears near a specific steady-state operating point and whose energy value exceeds a set threshold, then the energy peak is determined to correspond to a sealed cavity resonant mode excited by early micro-leakage flow, and an early micro-leakage warning is generated.

[0039] The conditions for determining an isolated energy peak are: the energy value corresponding to the local maximum point exceeds a set multiple of the value corresponding to the background trend line calculated based on all oscillation energy values; and there are no other local maximum points within a range of one basic frequency step to the left and right of the frequency of the local maximum point.

[0040] Specifically, in this experimental method, the threshold setting follows the dual principle of historical data statistics plus simulation experiment calibration: historical data is selected from the oscillation energy data of the same type of water pump running continuously for 12 months and 24 hours a day under leak-free conditions. After removing about 5% of abnormal data under special working conditions such as equipment start-up and shutdown, medium replacement, and power grid fluctuations, the upper limit of the 95% confidence interval of 0.08J is calculated using statistical methods as the basic threshold. Three sets of simulation experiments were then used for calibration. Micro-leakage rates of 0.1 mL / min, 0.3 mL / min, and 0.5 mL / min were set in the sealed cavity, covering the common early micro-leakage range. The corresponding oscillation energy was collected, and the average value of 0.12 J was taken. The midpoint between the baseline threshold and the experimental average was used to finally set the oscillation energy threshold to 0.1 J. For different pump models or different media types, the same principle can be followed to re-collect corresponding historical data and simulation experimental data, and dynamically adjust the threshold value.

[0041] The oscillation energy of 0.15J corresponding to the steady-state operating point of 2100r / min exceeds the threshold and meets the criteria for an isolated energy peak. First, based on the oscillation energy values ​​of 0.02J, 0.03J, 0.15J, 0.04J, and 0.025J at each steady-state operating point, the measured data of frequency and energy are fitted using linear regression. The fitted equation of the background trend line is y=0.0001x+0.02 (x is the frequency and y is the energy). Substituting 300Hz, the corresponding value of the trend line is calculated to be 0.05J. The multiple is set to 2 times. Combined with the signal-to-noise ratio requirements for conventional signal recognition, 0.15J exceeds twice that of 0.05J. Secondly, in this embodiment, the fundamental frequency step size is set to 40Hz. Within the range of 40Hz around 300Hz, i.e. from 260Hz to 340Hz, there is only one local maximum point at 300Hz, and no other local maximum points. Therefore, it is determined to be an isolated energy peak. Thus, it can be determined that the energy peak is caused by the resonant mode of the sealed cavity excited by the early micro-leakage flow.

[0042] Furthermore, the technical implementation of early warning generation and push is as follows: the intelligent monitoring terminal of the water pump and the data acquisition module establish a physical connection through industrial Ethernet, use the MODBUSRTU communication protocol for data interaction, set the communication baud rate to 9600bps, which is a commonly used baud rate in industrial scenarios, with 8 data bits and 1 stop bit.

[0043] The acquisition module packages the detected isolated energy peak data, the corresponding steady-state operating point parameters, and the detection time into an early warning data packet in the format of device number plus timestamp plus data content, and transmits it to the monitoring terminal through the above communication protocol; After receiving the data packet, the terminal first checks whether the checksum of the data packet matches. If the checksum is successful, a red warning prompt box pops up on the local interface of the terminal and triggers the SMS gateway to send a warning SMS to the mobile phone number of the maintenance personnel. At the same time, through the message push interface of the industrial APP, the warning information including the water pump number, detection time, specific steady-state operating point parameters and oscillation energy value is pushed to the mobile terminal of the maintenance personnel. If maintenance personnel do not confirm the warning on the monitoring terminal within 5 minutes, the system will automatically push a reminder again, based on the usual time setting for maintenance response. At the same time, all warning data will be stored in the relational database of the monitoring system according to the timestamp. The database uses MySQL, a commonly used database in industrial scenarios. The data table fields include pump number, detection time, steady-state speed, oscillation energy, number of warning pushes, confirmation status, etc., which facilitates subsequent traceability and analysis.

[0044] While generating early warnings of micro-leakage, the intelligent monitoring terminal automatically records and locks the precise frequency value corresponding to the isolated energy peak (300Hz in this embodiment) and the water pump speed value (2100r / min in this embodiment). It associates these with information such as the water pump number and detection time to form a unique early micro-leakage feature fingerprint. In addition to being stored in the MySQL database, the feature fingerprint data is also encrypted and backed up locally on the terminal using the AES-128 encryption algorithm. This is a commonly used encryption algorithm for industrial data. The key is allocated by the monitoring system administrator and updated regularly to prevent data loss or tampering.

[0045] During subsequent normal operation and monitoring, the system will automatically switch to the feature fingerprint tracking mode. This mode can continuously monitor the characteristic frequency of early micro-leakage, continuously listen to the pressure signal of the sealing cavity at a sampling frequency of 2000Hz, and ensure data consistency with the previous acquisition frequency. Through real-time fast Fourier transform, the frequency domain spectrum is analyzed using the same Hanning window and 1024 data point parameters as the previous detection, and the oscillation component consistent with the feature fingerprint frequency of 300Hz is identified. Simultaneously, an amplitude variation curve of the oscillation component is established in the monitoring system, and amplitude data is recorded at 10-minute intervals to balance real-time monitoring and data volume. The trend is updated in real time through linear fitting. If the amplitude increases for three consecutive time intervals with an increase of more than 10%, combined with the normal rate setting for leakage development, it is determined that the micro-leakage has intensified. The system will generate a level-two warning and increase the frequency of push notifications, providing maintenance personnel with an accurate basis for judging the development trend of the leakage.

[0046] This method calculates the theoretical resonant frequency range by combining pump operating parameters with the inherent characteristics of the sealing cavity system, and generates a targeted frequency scanning sequence to accurately capture the resonant signal of the sealing cavity excited by early micro-leakage flow. Compared with traditional detection methods, this method breaks through the dependence on macroscopic leakage signals, uses the characteristic distribution law of resonant energy to identify early micro-leakage, significantly improves detection sensitivity, and can issue early warnings in the early stages of leakage. At the same time, by using a discrete steady-state value scanning method, signal interference in non-target frequency ranges is reduced, lowering the false judgment rate. This provides a reliable technical basis for preventive maintenance of pump mechanical seals, helps reduce equipment downtime due to failure, and improves the operational reliability of industrial fluid transportation systems.

[0047] In some embodiments of the present invention, after determining the existence of an isolated energy peak, a frequency verification step is also included: The first step is to compare the precise frequency value corresponding to the isolated energy peak with the theoretical resonant frequency range; In this embodiment, the theoretical resonant frequency range of 200Hz to 500Hz was calculated in the early stage by coupling analysis of fluid mechanics and structural dynamics, combined with parameters such as medium viscosity and inner diameter of the sealed cavity. The precise frequency value corresponding to the isolated energy peak was located by energy distribution law analysis, which was 300Hz. The range of 300Hz was compared with the theoretical range of 200Hz to 500Hz.

[0048] The second step is to confirm the early micro-leakage warning if the frequency value is within the theoretical resonant frequency range. In this embodiment, the 300Hz frequency corresponding to the isolated energy peak falls within the theoretical resonant frequency range of 200Hz to 500Hz, indicating that the energy peak matches the theoretical characteristics of the resonant mode of the sealed cavity. This eliminates the possibility that non-resonant factors caused the peak, thus confirming the effectiveness of the early micro-leakage warning generated earlier. If the frequency value is not within the theoretical resonant frequency range, it is determined to be a false peak, the warning is not confirmed, and a rescan detection process is triggered.

[0049] The core innovations of this solution are: First, by combining the pump's operating conditions and the characteristics of the sealing cavity to calculate the theoretical resonant frequency range, a target scanning sequence is generated to accurately capture micro-leakage resonant signals, overcoming the limitations of traditional methods that rely on macroscopic leakage signals. This significantly improves detection sensitivity and enables early warning of leaks. Second, a new frequency verification step, combined with energy distribution identification, forms a dual verification process, eliminating false peak interference and significantly reducing the false alarm rate. This provides a reliable basis for preventative maintenance of pump seals, reduces downtime due to failures, and improves system reliability.

[0050] In some embodiments of the present invention, the method further includes: After the first suspected isolated energy peak is identified and an early warning is generated, the target frequency scanning sequence is repeated during subsequent equipment start-up and shutdown processes; In this embodiment, after the suspected isolated energy peak at 300Hz is first identified and a first-level warning is generated, in the subsequent three operating cycles of normal pump shutdown and restart, after each startup is completed and the initial stable operating state is reached, the target frequency scanning sequence containing five steady-state speed values ​​of 1500r / min, 1800r / min, 2100r / min, 2400r / min, and 2700r / min is repeatedly executed. The interval between two start-stop cycles is no less than 2 hours to avoid interference superposition in a short period of time. The scanning parameters are consistent with the first time, namely, the speed adjustment interval is 10 seconds, each steady-state point runs for 30 seconds, and the pressure signal sampling frequency is 2000Hz, to ensure the consistency of scanning conditions.

[0051] If, in subsequent consecutive executions, local maxima points with energy exceeding a set multiple of the background trend line are repeatedly identified at the same or adjacent frequency positions within the theoretical resonant frequency range, the level and certainty of early micro-leakage warning will be improved. In this embodiment, the number of consecutive repeated verifications is set to 3 (based on the historical interference data of the same type of water pump, the probability of a single interference lasting less than 5%), the adjacent frequency range is ±10Hz, and the set multiple of the energy exceeding the corresponding value of the background trend line is still 2 times.

[0052] In the subsequent three repeated scans after equipment start-up and shutdown, local maxima were identified at 300Hz, 295Hz, and 305Hz, respectively. These local maxima were all located within the theoretical resonant frequency range of 200Hz to 500Hz and within a ±10Hz range, with corresponding energy values ​​of 0.16J, 0.14J, and 0.15J, respectively. These values ​​were all more than twice the values ​​at the corresponding frequency positions of the fitted background trend lines in each scan (0.052J, 0.048J, and 0.051J, respectively). Therefore, the repeated verification condition was met, and the early micro-leakage warning level was upgraded from Level 1 to Level 2, with the warning certainty updated from "suspected" to "highly certain." If the requirement for multiple consecutive repeated identifications is not met, the original warning level will be maintained, and verification will continue during subsequent start-up and shutdown processes.

[0053] This step is closely integrated with the above-mentioned scheme. The initial scan identification provides the initial verification object for this step, the frequency verification step provides the theoretical range basis, and the "continuous" verification through multiple repeated scans further filters out accidental interference factors. Furthermore, after the warning level is upgraded, the system will increase the frequency of warning pushes (from once every 5 minutes to once every 1 minute), add on-site audible and visual alarm prompts, and push an analysis report containing peak data from multiple scans to the operation and maintenance management platform, providing more reliable decision support for operation and maintenance personnel to formulate maintenance plans.

[0054] The core innovation of this step is to enhance the reliability of early warning by repeatedly scanning and verifying the leakage state after the equipment starts and stops, and to solve the problem of accidental interference and misjudgment that may occur in a single scan; and to dynamically improve the warning level and certainty based on the repeated peak characteristics of the same or adjacent frequencies, so that the warning results are more in line with the actual leakage state and provide accurate guidance for maintenance needs of different urgency levels.

[0055] In some embodiments of the present invention, after determining the existence of an isolated energy peak, a step of accurate peak shape mapping is further included: The frequency corresponding to the isolated energy peak is taken as the center. In this embodiment, the frequency corresponding to the isolated energy peak identified by the basic frequency scan in the early stage is 300Hz, so 300Hz is used as the center frequency of the fine rescan.

[0056] Set a fine rescan step size, which is smaller than the base frequency step size; In this embodiment, the base frequency step size is 40Hz. Combined with the frequency control accuracy of ±5Hz corresponding to the water pump speed control accuracy of ±10r / min, the fine rescan step size is set to 5Hz. This step size is both smaller than the base step size and matches the equipment control accuracy, which can ensure the reliability of the rescan data.

[0057] Control the pump speed or flow rate to adjust it to multiple steady-state values ​​sequentially within a narrow frequency band including the center frequency, with fine rescanning step sizes as intervals; In this embodiment, by combining the conventional distribution range of isolated energy peaks and based on the statistical distribution of the resonant peaks of the sealing cavity of the same type of water pump, the narrowband frequency range is set to the center frequency ±20Hz. This range can cover the peak value and the half-width regions on both sides, i.e., 280Hz to 320Hz.

[0058] Using a fine rescan step size of 5Hz, nine target frequency values ​​of 280Hz, 285Hz, 290Hz, 295Hz, 300Hz, 305Hz, 310Hz, 315Hz, and 320Hz are generated sequentially within this range. Based on the linear mapping relationship between pump speed and system excitation frequency, these target frequency values ​​correspond to steady-state speeds of 1950r / min, 2000r / min, 2050r / min, 2100r / min, 2150r / min, 2200r / min, 2250r / min, 2300r / min, and 2350r / min, respectively. The pump is then controlled to sequentially adjust to these steady-state speeds.

[0059] At each steady-state value, the pressure signal of the sealed cavity is reacquired and its oscillation energy is calculated. In this embodiment, the water pump is kept running for 20 seconds at each steady-state speed to ensure stable pressure signals. Pressure signals are acquired through a pressure sensor on the inner wall of the sealed cavity, with a sampling frequency remaining at 2000Hz. The signal preprocessing and oscillation energy calculation methods are consistent with the basic scan: first, DC components and low-frequency noise are removed; then, a fast Fourier transform is used to obtain the frequency domain spectrum, and the energy values ​​within the theoretical resonant frequency range are calculated. The final oscillation energies corresponding to each steady-state value are 0.04J, 0.07J, 0.11J, 0.14J, 0.15J, 0.13J, 0.09J, 0.06J, and 0.03J, respectively.

[0060] Based on higher resolution energy data acquired within a narrow band, refined energy-frequency relationship curves are plotted to confirm the complete shape and precise frequency of isolated energy peaks. In this embodiment, the nine frequencies from the fine rescan are used as the x-axis, and the corresponding oscillation energy is used as the y-axis. A refined energy-frequency relationship curve is plotted using cubic spline interpolation. This method can fit a smooth and continuous curve, accurately presenting peak details, and is superior to the coarse fitting effect of linear interpolation. The curve shows that the isolated energy peak has a symmetrical single-peak shape with a half-width at half-maximum (FWHM) of 15Hz. This value conforms to the typical characteristics of a sealed cavity resonance peak (the FWHM of interference peaks is usually greater than 30Hz). The precise frequency corresponding to the peak value is 298Hz, further correcting the rough positioning result of 300Hz in the basic scan, confirming that this peak is a typical sealed cavity resonance peak, and eliminating interference such as multi-peak superposition.

[0061] This step uses the isolated energy peaks identified in the basic scan to provide the rescan center, and the resonant interval confirmed in the frequency verification step to limit the rescan range. The peak shape features and precise frequencies obtained from the refined mapping provide core parameters for the subsequent feature fingerprint tracking mode. The system uses 298Hz as the final early micro-leakage feature fingerprint frequency and automatically synchronizes it to the monitoring parameters of the feature fingerprint tracking mode, replacing the original 300Hz of the basic scan, thus improving the accuracy of subsequent continuous monitoring.

[0062] The innovation of this invention lies in its hierarchical scanning strategy, which employs a basic coarse scan to lock in the approximate location and a fine rescan for precise mapping. By rescanning within a narrow band using a fine step size smaller than the basic step size, high-resolution energy data can be obtained, thus solving the problem of difficulty in balancing accuracy and efficiency in traditional single-step scanning. Furthermore, by plotting refined curves, the complete shape and precise frequency of the peak can be clearly defined, providing a key basis for identifying the authenticity of resonance peaks and accurately locking characteristic fingerprints, thereby further improving the accuracy of leak detection.

[0063] Secondly, please refer to Figure 2 This invention also proposes an intelligent water pump-based operation status monitoring system for detecting early micro-leakage of the water pump's mechanical seal. The system includes: The acquisition module acquires the basic operating parameters of the water pump and, based on these parameters and the inherent characteristics of the water pump, calculates the theoretical resonant frequency range of the sealing cavity system. The generation module generates a target frequency scanning sequence, which is used to adjust the pump speed or flow rate to a set of discrete steady-state values ​​in sequence, wherein the system excitation frequency corresponding to each steady-state value is contained within the theoretical resonant frequency range; The execution module sequentially executes the target frequency scanning sequence and, at each steady-state operating point, acquires the pressure signal of the sealed cavity and extracts its oscillation energy within the theoretical resonant frequency range. The analysis module analyzes the distribution pattern of oscillation energy as it changes with the steady-state operating point in the sequence; If the distribution pattern shows an isolated energy peak that appears near a specific steady-state operating point and whose energy value exceeds a set threshold, the early warning module determines that the energy peak corresponds to a sealed cavity resonant mode excited by early micro-leakage flow and generates an early micro-leakage warning.

[0064] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for monitoring the operating status of an intelligent water pump, used to detect early micro-leakage in the mechanical seal of a water pump, characterized in that, include: Obtain the basic operating parameters of the water pump at present, and calculate the theoretical resonant frequency range of the sealing cavity system based on these parameters and the inherent characteristics of the water pump. A target frequency scanning sequence is generated, which is used to adjust the pump speed or flow rate to a set of discrete steady-state values ​​in sequence, wherein the system excitation frequency corresponding to each steady-state value is contained within the theoretical resonant frequency range; The target frequency scanning sequence is executed sequentially, and at each steady-state operating point, the pressure signal of the sealed cavity is collected, and its oscillation energy in the theoretical resonant frequency range is extracted. Analyze the distribution law of the oscillation energy as a function of the steady-state operating point in the sequence; If the distribution pattern shows an isolated energy peak that appears near a specific steady-state operating point and whose energy value exceeds a set threshold, then the energy peak is determined to correspond to a sealed cavity resonant mode excited by early micro-leakage flow, and an early micro-leakage warning is generated.

2. The method for monitoring the operating status of an intelligent water pump according to claim 1, characterized in that, The target frequency scan sequence is generated in the following way: Select the starting and ending frequencies within the theoretical resonant frequency range; The basic frequency step size is determined based on the pump's speed and flow control accuracy. Starting from the initial frequency and with the basic frequency step size as the interval, a series of target frequency values ​​covering the termination frequency are generated sequentially. Each target frequency value corresponds to an instruction to adjust the water pump to the corresponding steady-state operating point.

3. The method for monitoring the operating status of an intelligent water pump according to claim 1, characterized in that, The specific process for extracting oscillation energy is as follows: The frequency domain spectrum of the collected steady-state pressure signal of a sealed cavity is obtained by performing a fast Fourier transform. In the frequency domain spectrum, locate the frequency band corresponding to the theoretical resonant frequency range; Calculate the sum of squares of the amplitudes of all spectral lines within the frequency band, and use this as the oscillation energy value at the steady-state operating point.

4. The method for monitoring the operating status of an intelligent water pump according to claim 1, characterized in that, The specific process of analyzing the distribution pattern is as follows: Plot a scatter plot with the excitation frequency corresponding to each steady-state operating point as the horizontal axis and the corresponding oscillation energy value as the vertical axis. The scatter plot is interpolated to form a continuous energy-frequency relationship curve; On the curve, find all local maxima points where the energy value of that point is greater than the energy values ​​of its left and right adjacent interpolation points.

5. The method for monitoring the operating status of an intelligent water pump according to claim 4, characterized in that, The criteria for determining an isolated energy peak are: The energy value corresponding to the local maximum point exceeds a set multiple of the value corresponding to a background trend line calculated based on all oscillation energy values. Furthermore, within a fundamental frequency step to the left and right of the local maximum frequency, there are no other local maxima.

6. The method for monitoring the operating status of an intelligent water pump according to claim 5, characterized in that: After determining the existence of an isolated energy peak, a frequency verification step is also included: The precise frequency value corresponding to the isolated energy peak is compared with the theoretical resonant frequency range; If the frequency value is within the theoretical resonant frequency range, then the early micro-leakage warning is confirmed.

7. The method for monitoring the operating status of an intelligent water pump according to claim 1, characterized in that, The method also includes: After the first suspected isolated energy peak is identified and an early warning is generated, the target frequency scanning sequence is repeated during subsequent equipment start-up and shutdown processes; If, in subsequent consecutive executions, local maxima points with energy exceeding a set multiple of the background trend line are repeatedly identified at the same or adjacent frequency positions within the theoretical resonant frequency range, the level and certainty of early micro-leakage warning will be improved.

8. The method for monitoring the operating status of an intelligent water pump according to claim 1, characterized in that, While generating early warning of micro-leakage, the system automatically records and locks the precise frequency value and water pump speed value corresponding to the isolated energy peak, forming a characteristic fingerprint. During subsequent normal operation monitoring, the system continuously listens to whether there is an oscillating component in the sealing cavity pressure signal that is consistent with the frequency of the characteristic fingerprint, and monitors the historical trend of the amplitude of the component.

9. The method for monitoring the operating status of an intelligent water pump according to claim 5, characterized in that, After determining the existence of an isolated energy peak, the process also includes precise peak shape mapping: Centered on the frequency corresponding to the isolated energy peak; Set a fine rescan step size, which is smaller than the base frequency step size; Control the pump speed or flow rate to adjust it to multiple steady-state values ​​sequentially within a narrow frequency range including the center frequency, with the fine rescan step size as the interval; At each steady-state value, the pressure signal of the sealed cavity is reacquired and its oscillation energy is calculated. Based on the higher resolution energy data obtained within the narrow band, a refined energy-frequency relationship curve is plotted to confirm the complete shape and precise frequency of the isolated energy peak.

10. A smart water pump-based operation status monitoring system, applicable to the smart water pump-based operation status monitoring method described in any one of claims 1 to 9, used for detecting early micro-leakage of the water pump mechanical seal, characterized in that, The system includes: The acquisition module acquires the basic operating parameters of the water pump and, based on these parameters and the inherent characteristics of the water pump, calculates the theoretical resonant frequency range of the sealing cavity system. The generation module generates a target frequency scanning sequence, which is used to adjust the pump speed or flow rate to a set of discrete steady-state values ​​in sequence, wherein the system excitation frequency corresponding to each steady-state value is contained within the theoretical resonant frequency range; The execution module sequentially executes the target frequency scanning sequence and, at each steady-state operating point, acquires the pressure signal of the sealed cavity and extracts its oscillation energy within the theoretical resonant frequency range. The analysis module analyzes the distribution pattern of the oscillation energy as it changes with the steady-state operating point in the sequence; If the distribution pattern shows an isolated energy peak that appears near a specific steady-state operating point and whose energy value exceeds a set threshold, the early warning module determines that the energy peak corresponds to a sealed cavity resonant mode excited by early micro-leakage flow and generates an early micro-leakage warning.

Images