Water feeder magnetic attraction transmission part state monitoring method and system

Abstract

The invention relates to the technical field of fault prediction and health management, and discloses a water feeder magnetic attraction transmission part state monitoring method and system, and the method comprises the steps: obtaining a part vibration signal, and obtaining signal component data through hierarchical separation filtering; calculating an energy value of each component, if the energy value exceeds an energy threshold value, performing denoising processing, and performing standardization to obtain a noiseless data set; separating high-frequency and low-frequency signals through time-frequency transformation, and extracting energy characteristics of a fatigue associated frequency band; establishing an energy distribution mode through weighting, and judging that the energy distribution mode is abnormal if the energy distribution mode exceeds a quantization threshold value; a comprehensive early warning index is established, and fault grades are divided if a threshold value degree is exceeded; and generating an early warning data stream, and adjusting the operation parameters to obtain optimized configuration data. According to the method, accurate state monitoring and early fault early warning of the water feeder magnetic attraction transmission part can be realized.

Description

Technical Field

[0001] This invention relates to the field of fault prediction and health management technology, and in particular to a method and system for monitoring the status of a magnetic drive component of a water supply device. Background Technology

[0002] Currently, in the field of fault prediction and health management technology, with the continuous expansion of water supply system application scenarios and the increasing complexity of operating conditions, the magnetic drive component, as the core power component of the water supply system, directly affects the overall efficiency and service life of the equipment due to its operational stability. Monitoring the condition of the magnetic drive component is a crucial step in preventing fatigue wear and avoiding equipment downtime, and it is necessary to achieve early warning by accurately capturing fault characteristics in vibration signals.

[0003] Existing methods for monitoring the condition of magnetic drive components in water supply systems primarily rely on single-signal analysis or simple threshold judgments. However, this approach is clearly inadequate in complex operating environments. The original signal is easily affected by environmental noise, masking key fault characteristics and making it difficult to identify early, weak fault signals. Furthermore, single-band analysis cannot comprehensively capture multi-band energy changes corresponding to different fault types. Simultaneously, the lack of hierarchical signal processing and comprehensive feature fusion makes it difficult to accurately quantify fault severity, especially under specific load fluctuation scenarios, thus failing to provide reliable decision-making support for operation and maintenance.

[0004] In summary, existing technologies are hampered by environmental noise masking fault signals and the lack of integration of multi-band characteristics, making it difficult to quantify faults. This makes it challenging to achieve accurate status monitoring and early fault warning for the magnetic drive components of water supply units, and thus fails to meet the stability and safety requirements of equipment operation and maintenance under complex working conditions. Summary of the Invention

[0005] This invention provides a method and system for monitoring the status of magnetic drive components of a water supply device, so as to achieve accurate status monitoring and early fault warning of the magnetic drive components of the water supply device, and meet the stability and safety requirements of equipment operation and maintenance under complex working conditions.

[0006] Firstly, in order to solve the above-mentioned technical problems, the present invention provides a method for monitoring the status of a magnetic drive component of a water supply device, comprising: The vibration signal of the magnetic drive component of the water supply device is acquired and subjected to hierarchical separation and filtering correction to obtain the signal component data of each level; Based on the signal component data, the energy value of each component is calculated. If the energy value of a certain component exceeds the preset energy threshold, noise is removed by filtering, and then the format is standardized to obtain a dataset without noise components. For the dataset, high and low frequency signals are separated, the energy value of each frequency band signal is calculated, abnormal frequency bands are screened based on the energy value of each frequency band signal, and frequency band energy characteristics reflecting the fatigue characteristics of the component are extracted from them; Based on the frequency band energy characteristics, a comprehensive energy distribution pattern is constructed, and energy quantification indicators are extracted from the energy distribution pattern. If the quantification indicators exceed the preset normal quantization threshold, an anomaly is determined, and an anomaly determination result is obtained. The quantification indicators include: energy fluctuation amplitude, average energy value, and energy change rate. Based on the energy distribution pattern and the anomaly determination result, a comprehensive early warning score is calculated, and the fault level is classified according to the comprehensive early warning score. By integrating the fault level, the abnormal frequency band, and the comprehensive early warning score, a structured early warning information data stream is generated. The operating parameters are then adjusted based on the early warning information data stream to obtain an optimized operating parameter configuration.

[0007] In one feasible approach, acquiring the vibration signal of the magnetic drive component of the water supply device and performing hierarchical separation and filtering correction to obtain signal component data for each level includes: Vibration signals of the magnetic drive component of the water supply unit under specific loads are collected by sensors. The vibration signal is separated into layers to obtain the initial signal components of each layer; The initial signal component is compared with a preset component feature threshold. If the initial signal component at a certain level exceeds the component feature threshold, filtering correction is performed to obtain the corrected signal component. The corrected signal component is integrated with the initial signal component, and the integrated signal component is subjected to format standardization processing to obtain signal component data at each level.

[0008] In one feasible approach, the step of calculating the energy value of each component based on the signal component data, and if the energy value of a certain component exceeds a preset energy threshold, removing noise through filtering, and then performing format standardization processing to obtain a dataset of noise-free components, includes: Calculate the energy value of each signal component in the signal component data. If the energy value of a certain component exceeds a preset energy threshold, it is marked as an abnormal signal component. The abnormal signal components are filtered and denoised to obtain preliminary noise-free components; The noise-free components are format-normalized and integrated to obtain the final noise-free component dataset.

[0009] In one feasible approach, the step of separating high- and low-frequency signals from the dataset, calculating the energy value of each frequency band signal, filtering abnormal frequency bands based on the energy values ​​of each frequency band signal, and extracting frequency band energy features reflecting the fatigue characteristics of the component includes: Perform time-frequency transformation on the dataset to obtain time-frequency distribution data; Based on a preset frequency band division rule, the low-frequency signal and high-frequency signal in the time-frequency distribution data are separated to obtain segmented signals; For the segmented signal, each segmented signal corresponds to a frequency band. The energy value of the segmented signal in the time domain is calculated and accumulated to obtain the total energy of the corresponding frequency band signal. The total energy is compared with a preset fatigue judgment threshold, and frequency band signals whose total energy exceeds the fatigue judgment threshold are filtered out. Frequency band energy features reflecting the fatigue characteristics of the component are then extracted from these signals.

[0010] In one feasible approach, a comprehensive energy distribution pattern is constructed based on the frequency band energy characteristics, and quantitative indicators of energy are extracted from the energy distribution pattern. If the quantitative indicators exceed a preset normal quantization threshold, an anomaly is determined, and an anomaly determination result is obtained. The quantitative indicators include: energy fluctuation amplitude, average energy value, and energy change rate, including: The energy characteristics of each frequency band are weighted and calculated to obtain a comprehensive energy value; By integrating the comprehensive energy values ​​according to the time series, a comprehensive energy distribution pattern is constructed. Quantitative indicators of energy over time are extracted from the comprehensive energy distribution pattern. The key quantitative indicators include energy fluctuation amplitude, average energy value, and energy change rate. The quantitative indicators are compared with preset normal quantitative thresholds. If any of the quantitative indicators exceeds the normal quantitative threshold, an anomaly is determined, and an anomaly determination result is obtained.

[0011] In one feasible approach, the step of calculating a comprehensive early warning score based on the energy distribution pattern and the anomaly determination result, and classifying the fault level based on the comprehensive early warning score, includes: Anomaly features are extracted from the energy distribution pattern, including the energy peak location and fluctuation amplitude. By combining the abnormal location features with the abnormality determination results, a comprehensive early warning score is calculated. The comprehensive early warning score is compared with a preset early warning score threshold. If the comprehensive early warning score exceeds the early warning score threshold, the fault level is classified according to the extent of the exceedance, and the fault level classification result is determined.

[0012] In one feasible approach, the integration of the fault level, the abnormal frequency band, and the comprehensive early warning score to generate a structured early warning information data stream, and the adjustment of operating parameters based on the early warning information data stream to obtain an optimized operating parameter configuration, includes: Based on the classification results, the fault level, the abnormal frequency band, and the comprehensive early warning score are integrated to generate a structured early warning information data stream; The warning information data stream is transmitted according to a preset output format, and a transmission confirmation message is obtained. If the transmission confirmation information indicates successful transmission, the equipment operating parameters are adjusted for the specific load to obtain the adjusted equipment operating parameters, which include operating frequency and torque output. The adjusted equipment operating parameters are updated to the equipment operating configuration to obtain the optimized operating parameter configuration.

[0013] Secondly, the present invention provides a status monitoring system for a magnetic drive component of a water supply device, comprising: The signal component acquisition module acquires the vibration signal of the magnetic drive component of the water supply unit and performs hierarchical separation and filtering correction to obtain the signal component data of each level. The energy denoising module calculates the energy value of each component based on the signal component data. If the energy value of a component exceeds a preset energy threshold, noise is removed by filtering, and then the format is standardized to obtain a dataset without noise components. The fatigue sign determination module separates high and low frequency signals from the dataset, calculates the energy value of each frequency band signal, filters abnormal frequency bands based on the energy value of each frequency band signal, and extracts frequency band energy features that reflect the fatigue characteristics of the component. The anomaly detection module constructs a comprehensive energy distribution pattern based on the frequency band energy characteristics and extracts energy quantification indicators from the energy distribution pattern. If the quantification indicators exceed a preset normal quantization threshold, an anomaly is detected, and an anomaly detection result is obtained. The quantification indicators include: energy fluctuation amplitude, average energy value, and energy change rate. The fault level determination module calculates a comprehensive early warning score based on the energy distribution pattern and the anomaly judgment result, and classifies the fault level based on the comprehensive early warning score; The configuration optimization module integrates the fault level, the abnormal frequency band, and the comprehensive early warning score to generate a structured early warning information data stream. Based on the early warning information data stream, the operating parameters are adjusted to obtain the optimized operating parameter configuration.

[0014] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention collects the vibration signal of the magnetic drive component of the water supply device under a specific load in real time through a sensor, and uses multi-scale decomposition technology to perform hierarchical separation. The signal components exceeding the component characteristic threshold are filtered and corrected twice, and after integration and standardization, the signal component data of each level are obtained. Then, the energy value of each component is calculated and compared with the energy threshold. Abnormal components are filtered and denoised to obtain noise-free components. This invention breaks through the limitations of traditional single signal analysis, which is easily affected by noise and key features are masked. It fully explores the hierarchical features of vibration signals, eliminates the influence of environmental noise and abnormal fluctuations, provides high-precision and pure data support for fault feature extraction, effectively improves the detection rate of early weak fault signals, and solves the problem of missing early faults in the existing technology.

[0015] (2) This invention performs time-frequency transformation on the noise-free component to obtain time-frequency distribution data, sets frequency band division rules to separate high and low frequency signals in combination with equipment operating characteristics, calculates the total energy of each frequency band and compares it with the fatigue judgment threshold, selects the frequency band energy characteristics that reflect the fatigue characteristics of the component, sets weighting coefficients according to the influence degree of each frequency band, and integrates the weighted data to construct a comprehensive energy distribution model. This breaks through the limitation of traditional single frequency band analysis that cannot fully capture the energy changes of multiple frequency bands, accurately captures the frequency band energy characteristics corresponding to different fault types, provides multi-dimensional basis for judging fatigue and wear signs, significantly improves the accuracy of fault identification under complex working conditions, and makes up for the shortcomings of existing technologies in integrating multiple frequency band features.

[0016] (3) This invention integrates the comprehensive energy distribution pattern and potential fault symptom characteristics to construct a comprehensive early warning index. The intensity of the early warning signal is quantified by intensity assessment and classification. If the threshold is exceeded, the fault level is determined. A structured early warning information data stream is generated and transmitted to the equipment control system. The operating frequency, torque output and other parameters are adjusted to achieve optimized configuration. This invention breaks through the limitations of traditional monitoring that lacks quantitative classification and is difficult to provide support for operation and maintenance. It provides operation and maintenance personnel with accurate fault level and parameter adjustment basis, solves the problems of ambiguous fault warning and delayed operation and maintenance response, takes into account the accuracy of monitoring and the stability of equipment operation, and meets the operation and maintenance safety requirements of the magnetic drive component of the water supply unit under complex working conditions. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of a method for monitoring the status of a magnetic drive component of a water supply device according to the first embodiment of the present invention; Figure 2 This is a schematic diagram of a status monitoring system for a magnetic drive component of a water supply device provided in the second embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0019] Reference Figure 1 The first embodiment of the present invention provides a method for monitoring the status of a magnetic drive component of a water supply device, comprising the following steps: S101, acquire the vibration signal of the magnetic drive component of the water supply device and perform hierarchical separation and filtering correction to obtain the signal component data of each level; S102, calculate the energy value of each component based on the signal component data. If the energy value of a certain component exceeds the preset energy threshold, remove noise by filtering and then perform format standardization to obtain a dataset without noise components. S103, for the dataset, separate high and low frequency signals, calculate the energy value of each frequency band signal, filter abnormal frequency bands according to the energy value of each frequency band signal, and extract the frequency band energy characteristics that reflect the fatigue characteristics of the component. S104. Based on the frequency band energy characteristics, a comprehensive energy distribution pattern is constructed, and energy quantification indicators are extracted from the energy distribution pattern. If the quantification indicators exceed the preset normal quantization threshold, an anomaly is determined, and an anomaly determination result is obtained. The quantification indicators include: energy fluctuation amplitude, average energy value, and energy change rate. S105, Calculate a comprehensive early warning score based on the energy distribution pattern and the anomaly determination result, and classify the fault level based on the comprehensive early warning score; S106, integrate the fault level, the abnormal frequency band and the comprehensive early warning score to generate a structured early warning information data stream, and adjust the operating parameters according to the early warning information data stream to obtain the optimized operating parameter configuration.

[0020] In step S101, the vibration signal of the magnetic drive component of the water supply device is acquired and subjected to hierarchical separation and filtering correction to obtain signal component data for each level, including: Vibration signals of the magnetic drive component of the water supply unit under specific loads are collected by sensors. The vibration signal is separated into layers to obtain the initial signal components of each layer; The initial signal component is compared with a preset component feature threshold. If the initial signal component at a certain level exceeds the component feature threshold, filtering correction is performed to obtain the corrected signal component. The corrected signal component is integrated with the initial signal component, and the integrated signal component is subjected to format standardization processing to obtain signal component data at each level.

[0021] It should be noted that, firstly, when collecting vibration signals of the magnetic drive component of the water supply unit under a specific load using sensors, a high-precision piezoelectric accelerometer is selected and fixed at a critical stress location on the component, such as the contact edge between the magnetic rotor and the stator. The sensor's sampling frequency is determined based on the component's operating speed, typically more than twice the frequency corresponding to the speed to satisfy the sampling theorem. During the acquisition process, the sampling time and the current load value are recorded simultaneously to ensure the correlation between the data and the operating conditions. For example, if the magnetic drive component of a water supply unit operates at a speed of 2400 rpm (corresponding to a frequency of 40 Hz), the sensor sampling frequency is set to 100 Hz, and the vibration signal of this component under a specific load of 60 N is continuously collected.

[0022] When performing hierarchical separation of the vibration signal, the db4 wavelet basis function is selected to decompose the original vibration signal into three levels according to the frequency from low to high. During the decomposition process, the Mallat algorithm is used to achieve rapid decomposition of the signal. Each level corresponds to the initial signal components in different frequency ranges. The low-frequency level (0-50Hz) reflects the overall structural vibration of the component, the mid-frequency level (50-200Hz) corresponds to the dynamic changes of the magnetic gap, and the high-frequency level (above 200Hz) captures the vibration generated by local friction or minor wear, resulting in three types of initial signal components: low-frequency level 0-50Hz, mid-frequency level 50-200Hz, and high-frequency level above 200Hz.

[0023] When comparing the initial signal components with preset component characteristic thresholds, the component characteristic thresholds are set based on the statistical data of normal operation signals of similar water supply magnetic drive components over the past year and the minimum amplitude of abnormal signals in historical fault cases. The setting uses a percentile statistical method to select the upper limit of the amplitude of each level of signal under 95% of normal operation scenarios as the initial threshold. This threshold is then verified through at least 60,000 normal operation signal data points per quarter to ensure a confidence level of over 96%. Adjustment based on load is also supported: the threshold is increased by 12% under heavy load (80-100N) conditions and decreased by 15% under light load (20-40N) conditions. If the initial signal component of a certain level exceeds the threshold, an adaptive Kalman filter algorithm is used for correction. This algorithm updates the filter gain in real time to specifically remove interference components of that level, such as environmental vibration noise, while retaining effective vibration characteristics, resulting in a corrected signal component. For example, if the characteristic threshold of the intermediate frequency layer component is set to 0.9 mm / s, and the initial signal amplitude of this layer is 1.3 mm / s at a certain moment, which exceeds the threshold, the amplitude is reduced to 0.8 mm / s after adaptive Kalman filtering correction, thus obtaining the corrected signal component.

[0024] When integrating the corrected signal components with the initial signal components, the time axes of all signals are aligned in hierarchical order to ensure a one-to-one correspondence between signals at each level at the same time point. The integrated signal components are then standardized in format: first, the amplitude unit of all signal components is unified to millimeters per second (mm / s); then, the data storage format is standardized to CSV; finally, the sampling time interval of each level of signal is verified. If differences exist, linear interpolation is performed according to a reference interval (e.g., 0.01 seconds) to eliminate data format differences between different levels of signals, ultimately resulting in well-structured signal component data for each level. For example, after aligning the low-frequency initial signal, mid-frequency corrected signal, and high-frequency initial signal on the time axis, the amplitude unit is unified to millimeters per second, the sampling interval is 0.01 seconds, and the storage format is CSV, forming signal component data for each level that can be directly used for subsequent analysis.

[0025] In step S102, based on the signal component data, the energy value of each component is calculated. If the energy value of a certain component exceeds a preset energy threshold, noise is removed by filtering, and then format normalization is performed to obtain a dataset of noise-free components, including: Calculate the energy value of each signal component in the signal component data. If the energy value of a certain component exceeds a preset energy threshold, it is marked as an abnormal signal component. The abnormal signal components are filtered and denoised to obtain preliminary noise-free components; The noise-free components are format-normalized and integrated to obtain the final noise-free component dataset.

[0026] It should be noted that, firstly, when calculating the energy value of each signal component in the signal component data, the calculation method is to sum the product of the square of the vibration velocity amplitude in the time domain and the sampling time. The vibration velocity amplitude of each component at all sampling points within the acquisition period is squared one by one. Then, combined with the preset equivalent mass of the water supply unit's magnetic drive component, the energy is calculated based on the component design drawings and material density. For example, if the rotor's equivalent mass is 0.5 kg, the instantaneous energy at each sampling point is calculated using the kinetic energy formula. Finally, the instantaneous energies of all sampling points within the acquisition period are directly summed to obtain the corresponding energy value of that component (unit: μJ). The preset energy threshold is set based on the energy statistics of each signal component during the normal operation of similar water supply units' magnetic drive components over the past year, and the energy differentiation threshold between environmental noise and normal equipment vibration in historical monitoring. The system employs a percentile statistical method to select the upper limit of energy values ​​for each component under 95% of normal operating scenarios as the initial threshold. This threshold is then validated using at least 50,000 normal operating energy data points quarterly to ensure a confidence level of over 95%. It also supports load-based fine-tuning: the threshold is increased by 8% under heavy load (80-100N) conditions and decreased by 10% under light load (20-40N) conditions to adapt to the energy characteristics of different operating conditions. The calculated energy values ​​are compared one by one with this threshold. If the energy value of a component exceeds the threshold, it is marked as an abnormal signal component. For example, if a high-frequency signal component has a calculated energy value of 120μJ and the preset energy threshold is 100μJ, this component is marked as an abnormal signal component.

[0027] Next, when filtering and denoising the anomalous signal components, wavelet thresholding denoising technology is employed, specifically using the sym8 wavelet basis function. The anomalous components are decomposed into wavelet coefficients of different scales through three-level wavelet decomposition. Noisy high-frequency wavelet coefficients are processed using soft thresholding (the threshold is determined according to the Birgé-Massart strategy), retaining low-frequency wavelet coefficients reflecting the vibration characteristics of the component and effective high-frequency coefficients. Wavelet reconstruction then yields a preliminary noise-free component. This technique can specifically remove random noise from anomalous components while avoiding the loss of effective vibration characteristics. For example, after sym8 wavelet thresholding denoising, the original high-frequency noise of the high-frequency signal component marked as anomalous is removed, while the effective vibration signal generated by local friction of the component is retained, resulting in a preliminary noise-free component.

[0028] When standardizing the format of the initial noise-free components, the energy unit of all noise-free components is first unified to microjoules (μJ). Then, the data storage format is standardized to CSV format, with each row containing a timestamp and the corresponding energy value. Finally, the sampling time interval of each component is verified. If discrepancies exist, linear interpolation is performed at a reference interval (e.g., 0.01 seconds) to ensure that all noise-free components are synchronized in the time dimension. For example, low-frequency components originally in millijoules and mid-frequency components stored in TXT format are standardized to μJ units and CSV storage format after standardization, and the sampling time interval is adjusted to 0.01 seconds to obtain the final noise-free components.

[0029] In step S103, high and low frequency signals are separated from the dataset, the energy value of each frequency band signal is calculated, abnormal frequency bands are screened based on the energy value of each frequency band signal, and frequency band energy features reflecting the fatigue characteristics of the component are extracted from them, including: Perform time-frequency transformation on the dataset to obtain time-frequency distribution data; Based on a preset frequency band division rule, the low-frequency signal and high-frequency signal in the time-frequency distribution data are separated to obtain segmented signals; For the segmented signal, each segmented signal corresponds to a frequency band. The energy value of the segmented signal in the time domain is calculated and accumulated to obtain the total energy of the corresponding frequency band signal. The total energy is compared with a preset fatigue judgment threshold, and frequency band signals whose total energy exceeds the fatigue judgment threshold are filtered out. Frequency band energy features reflecting the fatigue characteristics of the component are then extracted from these signals.

[0030] It should be noted that, firstly, when performing time-frequency transformation on the dataset, the Short Time Fourier Transform (STFT) technique was employed, using a Hanning window as the window function. The window length was set to 256 sampling points, with an overlap rate of 50%. This transformed the noise-free component dataset in the time domain into time-frequency distribution data that simultaneously contains both time and frequency information. This data can completely preserve the amplitude changes of each frequency component at different times, providing a precise basis for subsequent frequency band analysis. For example, after STFT processing, the noise-free dataset of a magnetic drive component of a water supply device clearly showed that the signal amplitude in the 200-250Hz frequency band was significantly higher than in other time periods after 8 minutes of operation, clearly identifying the frequency range that needed to be focused on.

[0031] The preset frequency band division rules are set based on the equipment's operating characteristics, referencing the design speed of the magnetic drive component of the water supply unit (e.g., 3000 rpm corresponds to a fundamental frequency of 50 Hz), the typical frequency range of component material fatigue vibration (obtained from the equipment design manual), and frequency band data related to fatigue in historical failure cases. The time-frequency distribution data is divided into three frequency bands: a low-frequency band (0-60 Hz) corresponding to overall component structural vibration; a mid-frequency band (60-250 Hz) corresponding to dynamic changes in the contact of the magnetic drive component; and a high-frequency band (above 250 Hz) corresponding to localized minor wear vibration. Based on these rules, low-frequency and high-frequency signals are separated, resulting in independent segmented signals for each frequency band. For example, processing the time-frequency distribution data according to the above rules assigns the 0-60 Hz signal to the low-frequency segment and the 60-250 Hz signal to the mid-frequency segment, achieving effective separation of high and low frequency signals.

[0032] When calculating the total energy of the corresponding frequency band signals for segmented signals, all vibration velocity time-domain amplitude points of the segmented signals within the acquisition period are extracted segment by segment. The vibration velocity amplitude is obtained by time-domain integration of the raw signals acquired by the sensors, ensuring that the dimensions conform to energy calculation. After squaring the vibration velocity amplitude of each component at all sampling points within the acquisition period, the instantaneous energy of each sampling point is calculated according to the kinetic energy formula. Then, the instantaneous energies of all sampling points within the acquisition period are directly summed to obtain the energy value (unit: μJ) corresponding to that component. This value can intuitively quantify the intensity of vibration signals in each frequency band. For example, after the above calculation, the total energy of the vibration velocity amplitude points of a low-frequency segmented signal within a 5-minute acquisition period is 180μJ, while the total energy of the mid-frequency segmented signal calculated simultaneously is 110μJ, clearly reflecting the energy differences between different frequency bands.

[0033] When comparing the total energy with the preset fatigue judgment threshold, the fatigue judgment threshold is based on fatigue test data of similar water supply magnetic drive components over the past two years. It statistically analyzes the energy critical values ​​for each frequency band at different fatigue stages and sets the threshold based on the 95% distribution range of energy in each frequency band during normal operation. The threshold is set using a percentile statistical method, taking the upper limit of energy in each frequency band under 95% normal operation scenarios as the initial threshold. After verification with no less than 80,000 fatigue tests and normal operation data points per quarter, the confidence level of this threshold reaches over 96%. It also supports fine-tuning based on load: the threshold is increased by 10% under heavy load (80-100N) conditions and decreased by 8% under light load (20-40N) conditions. If the total energy in a certain frequency band exceeds the corresponding threshold, features are extracted by combining the correlation attributes between that frequency band and component fatigue (e.g., low-frequency band related to structural fatigue, mid-frequency band related to magnetic component contact fatigue) to obtain frequency band energy characteristics that reflect the component's fatigue characteristics. For example, if the fatigue judgment threshold for the low-frequency band is set to 150μJ, and the current total energy in the low-frequency band is 180μJ, which exceeds the threshold by 30μJ, the frequency band energy feature of "low-frequency band energy exceeding the threshold by 30μJ" can be extracted by combining the fatigue attributes of the low-frequency band associated structures. This feature can directly reflect the potential fatigue state of the component structure.

[0034] In step S104, a comprehensive energy distribution pattern is constructed based on the frequency band energy characteristics, and energy quantization indicators are extracted from the energy distribution pattern. If the quantization indicators exceed a preset normal quantization threshold, an anomaly is determined, and an anomaly determination result is obtained. The quantization indicators include: energy fluctuation amplitude, average energy value, and energy change rate, including: The energy characteristics of each frequency band are weighted and calculated to obtain a comprehensive energy value; By integrating the comprehensive energy values ​​according to the time series, a comprehensive energy distribution pattern is constructed. Quantitative indicators of energy over time are extracted from the comprehensive energy distribution pattern. The key quantitative indicators include energy fluctuation amplitude, average energy value, and energy change rate. The quantitative indicators are compared with preset normal quantitative thresholds. If any of the quantitative indicators exceeds the normal quantitative threshold, an anomaly is determined, and an anomaly determination result is obtained.

[0035] It should be noted that, firstly, when calculating the comprehensive energy value by weighting the energy characteristics of each frequency band, a linear weighting method is used. The weights are set according to the correlation between each frequency band and the fatigue of the magnetic drive components of the water supply unit: low-frequency band related to structural fatigue has a weight of 0.4, mid-frequency band related to magnetic contact fatigue has a weight of 0.35, and high-frequency band related to local wear has a weight of 0.25. The weight settings are based on the statistical data of similar component failures over the past two years, the abnormal proportion of each frequency band in the failure cases, and are verified by no less than 50,000 operating data points per quarter, with a confidence level of over 95%. At the same time, it can be finely adjusted according to the load: under heavy load (80-100N) conditions, the weight of the high-frequency band is increased by 0.05, and under light load (20-40N) conditions, the weight of the low-frequency band is increased by 0.05. To ensure the total weight of each frequency band remains constant at 1 and to avoid numerical deviations in the weighted result, 0.05 is simultaneously subtracted from the weight of the mid-frequency band. Since the correlation between the mid-frequency band and component fatigue under heavy load conditions is relatively lower than that of the high-frequency band, the weights are adjusted to 0.4 for the low-frequency band, 0.3 for the mid-frequency band, and 0.3 for the high-frequency band. Under light load (20-40N) conditions, the weight of the low-frequency band is increased by 0.05, again adhering to the principle of a total weight of 1, while 0.05 is simultaneously subtracted from the weight of the mid-frequency band. Since the correlation between the mid-frequency band and component fatigue under light load conditions is relatively lower than that of the low-frequency band, the weights are adjusted to 0.45 for the low-frequency band, 0.3 for the mid-frequency band, and 0.25 for the high-frequency band. During calculation, the energy characteristic value of each frequency band is multiplied by its corresponding weight and then summed to obtain the comprehensive energy value. For example, the energy characteristic value is 150μJ in the low-frequency band, 120μJ in the mid-frequency band, and 90μJ in the high-frequency band. The total energy value is 150×0.4+120×0.35+90×0.25=124.5μJ.

[0036] When constructing a comprehensive energy distribution model by integrating comprehensive energy values ​​over time, linear interpolation is used to unify the time interval of each comprehensive energy value. The baseline interval is set to 1 minute. The discrete comprehensive energy values ​​are arranged in chronological order of acquisition time, and combined with the time axis to generate a continuous energy change curve. This curve is the comprehensive energy distribution model. For example, the comprehensive energy values ​​(120 μJ, 124.5 μJ, 128 μJ) collected every 5 minutes are supplemented with data for each minute through linear interpolation to form an energy change curve with a 1-minute interval, thus constructing the comprehensive energy distribution model.

[0037] When extracting key quantitative indicators from a comprehensive energy distribution model over a time series, the energy fluctuation amplitude is calculated as the difference between the maximum and minimum comprehensive energy values ​​within a certain time period; the average energy value is calculated as the arithmetic mean of all comprehensive energy values ​​within that time period; and the energy change rate is calculated by dividing the difference in comprehensive energy values ​​between two adjacent time points by the time interval (1 minute). For example, within a 3-minute time period, the comprehensive energy values ​​are 120 μJ, 124.5 μJ, and 128 μJ, with an energy fluctuation amplitude of 8 μJ, an average energy value of 124.17 μJ, and energy change rates of 4.5 μJ / minute and 3.5 μJ / minute, respectively.

[0038] When comparing key quantitative indicators with preset normal quantitative thresholds, the normal quantitative thresholds are set based on statistical data of key indicators for similar components operating normally over the past two years. The initial thresholds are determined using a percentile statistical method, taking the upper limit of the indicator for 95% of normal operating scenarios. The fluctuation amplitude threshold is 10 μJ, the average energy value threshold is 130 μJ, and the energy change rate threshold is 5 μJ / minute. After verification with no less than 60,000 normal operating data points per quarter, the threshold confidence level reaches over 96%. Fine-tuning based on equipment usage time is also supported: the threshold is increased by 8% for equipment used for more than two years, and decreased by 5% for new equipment (used for less than one year). If any key indicator exceeds the corresponding threshold, an anomaly is determined, and an anomaly determination result is obtained. For example, if the energy fluctuation amplitude is 12 μJ over a certain period, exceeding the normal quantitative threshold of 10 μJ, an anomaly is determined, and an anomaly determination result is obtained.

[0039] In step S105, a comprehensive early warning score is calculated based on the energy distribution pattern and the anomaly determination result. The fault level is then classified based on the comprehensive early warning score, including: Anomaly features are extracted from the energy distribution pattern, including the energy peak location and fluctuation amplitude. By combining the abnormal location features with the abnormality determination results, a comprehensive early warning score is calculated. The comprehensive early warning score is compared with a preset early warning score threshold. If the comprehensive early warning score exceeds the early warning score threshold, the fault level is classified according to the extent of the exceedance, and the fault level classification result is determined.

[0040] It should be noted that, firstly, when extracting outlier features from the distribution pattern, a sliding window peak detection algorithm is used. The window size is set to 100 sampling points, and the window sliding step size is 20 sampling points. The time series data of the comprehensive energy distribution pattern is scanned segment by segment to identify the peak positions where the energy value is higher than the surrounding sampling points and record the corresponding timestamps. At the same time, the difference between the peak position and the adjacent normal energy value is calculated to obtain the fluctuation amplitude. For example, when scanning the comprehensive energy distribution pattern of a magnetic drive component of a water supply device, an energy peak was detected after 12 minutes of operation, corresponding to the timestamp 12:00, with a fluctuation amplitude of 22 μJ. These two outlier features were extracted.

[0041] Next, when calculating the comprehensive early warning score by integrating the abnormal location characteristics and the anomaly judgment results, a weighted summation method is used. The weights are set based on the correlation between each feature and the fault: energy peak location has a weight of 0.4, fluctuation amplitude has a weight of 0.3, and the anomaly judgment result has a weight of 0.3. First, each feature is standardized to the range of 0-100. The energy peak location is scored according to the degree of deviation from the normal time period, and the fluctuation amplitude is scored according to the proportion exceeding the normal range. An anomaly judgment result of "yes" is scored as 100 points and "no" as 0 points. Then, the weights are summed to obtain the comprehensive early warning score. For example, if the standardized score for the energy peak location is 80, the standardized score for the fluctuation amplitude is 75, and the anomaly judgment result is "yes" (score 100), the comprehensive early warning score is 80×0.4+75×0.3+100×0.3=81.5.

[0042] Subsequently, when comparing the comprehensive early warning score with the preset early warning score threshold, the threshold is set based on nearly two years of failure case data for similar water supply magnetic drive components and the statistical range of early warning scores during normal operation. The setting adopts a percentile statistical method, taking the upper limit of the early warning score for 95% of normal operating scenarios as the initial threshold. After verification with no less than 80,000 operating data points per quarter, the threshold confidence level reaches over 96%. Simultaneously, it supports fine-tuning according to load: under heavy load (80-100N) conditions, the threshold for each level is lowered by 5 points, and under light load (20-40N) conditions, it is raised by 5 points. The rules for classifying fault levels are as follows: exceeding the mild threshold (50 points) by 0-10 points is a Level 1 fault; exceeding the moderate threshold (80 points) by 0-15 points is a Level 2 fault; and exceeding the severe threshold (100 points) or the moderate threshold by more than 15 points is a Level 3 fault. For example, if the comprehensive early warning score is 81.5, exceeding the moderate threshold of 80 points by 1.5 points, the fault level is classified as Level 2 based on the magnitude of the exceedance, thus determining the fault level classification result.

[0043] In step S106, the fault level, the abnormal frequency band, and the comprehensive early warning score are integrated to generate a structured early warning information data stream. Operating parameters are adjusted based on this data stream to obtain an optimized operating parameter configuration, including: Based on the classification results, the fault level, the abnormal frequency band, and the comprehensive early warning score are integrated to generate a structured early warning information data stream; The warning information data stream is transmitted according to a preset output format, and a transmission confirmation message is obtained. If the transmission confirmation information indicates successful transmission, the equipment operating parameters are adjusted for the specific load to obtain the adjusted equipment operating parameters, which include operating frequency and torque output. The adjusted equipment operating parameters are updated to the equipment operating configuration to obtain the optimized operating parameter configuration.

[0044] It should be noted that, firstly, when generating a structured warning information data stream by integrating fault levels, abnormal frequency bands, and warning signal strength values ​​based on the classification results, the information is integrated in a fixed field order: fault level, abnormal frequency band, and warning signal strength value. Fault levels are encoded using numbers from 1 to 3, with level 1 representing minor faults, level 2 representing moderate faults, and level 3 representing severe faults. Abnormal frequency bands are represented by specific frequency ranges, and warning signal strength values ​​are represented by a combination of numerical values ​​and level descriptions. This ensures a unified data stream structure that can be directly parsed by the equipment system. For example, if the classification result is a level 2 fault, an abnormal frequency band of 0-50Hz, and a warning signal strength value of 75 (moderate), the generated data stream will contain the structured content "Fault Level 2, Abnormal Frequency Band 0-50Hz, Warning Strength 75, Moderate".

[0045] When transmitting early warning information data streams according to a preset output format, the preset output format is JSON, and the transmission technology uses the MQTT protocol. During transmission, the data transmission status is monitored in real time, and a transmission confirmation message is received after the receiving end (equipment control system) provides feedback on the data reception result. For example, an early warning data stream in JSON format is sent to the control system via the MQTT protocol, and the control system returns a "transmission successful" confirmation message upon receipt.

[0046] If the transmission confirmation message indicates successful transmission, when adjusting the equipment operating parameters for a specific load, the adjustment benchmark is based on fault handling data of similar water supply units' magnetic drive components over the past two years. The stable operating rate of the equipment after parameter adjustments under different fault levels and loads is statistically set, and the adjustment range is determined using a percentile statistical method. For a Level 1 fault, the operating frequency is reduced by 15% and the torque output by 12%; for a Level 2 fault, the operating frequency is reduced by 10% and the torque output by 10%; for a Level 3 fault, the operating frequency is reduced by 5% and the torque output by 5%. This adjustment benchmark has been validated through at least 60,000 parameter adjustment cases per quarter, achieving a confidence level of over 95%. Simultaneously, fine-tuning can be performed according to the load: a 2% reduction in adjustment range under heavy load (80-100N) conditions and a 2% increase in adjustment range under light load (20-40N) conditions. For example, if the current specific load is 50N, the fault level is Level 2, the original operating frequency is 50Hz, and the torque output is 80N·m. After adjustment, the operating frequency is reduced to 45Hz, and the torque output is reduced to 72N·m, resulting in the adjusted equipment operating parameters.

[0047] When updating the adjusted equipment operating parameters to the equipment operating configuration, a parameter overwrite update method is used. This directly replaces the corresponding field values ​​for "operating frequency" and "torque output" in the equipment configuration file. After the update, the equipment's built-in parameter verification module verifies whether the parameters meet the equipment's hardware requirements, such as whether the operating frequency is within the equipment's allowed range of 20-60Hz. Once the verification is successful, the saved new configuration data becomes the optimized operating parameter configuration. For example, overwriting the original configuration with an operating frequency of 45Hz and a torque output of 72N·m, after the parameter verification module confirms compatibility, saves it as the optimized operating configuration data.

[0048] In summary, this invention discloses a method for monitoring the condition of a magnetic drive component of a water supply device, comprising: acquiring vibration signals of the magnetic drive component of the water supply device and performing hierarchical separation and filtering correction to obtain signal component data at each level; calculating the energy value of each component based on the signal component data; if the energy value of a component exceeds a preset energy threshold, removing noise through filtering and then performing format standardization to obtain a dataset of noise-free components; separating high and low frequency signals from the dataset, calculating the energy value of each frequency band signal, filtering abnormal frequency bands based on the energy values ​​of each frequency band signal, and extracting frequency band energy features reflecting the fatigue characteristics of the component; and based on the frequency band energy... The system features a comprehensive energy distribution pattern, from which quantitative energy indicators are extracted. If these indicators exceed a preset normal threshold, an anomaly is identified, resulting in an anomaly assessment. These quantitative indicators include energy fluctuation amplitude, average energy value, and energy change rate. Based on the energy distribution pattern and the anomaly assessment result, a comprehensive early warning score is calculated, and fault levels are categorized according to the score. The fault levels, anomaly frequency bands, and comprehensive early warning scores are integrated to generate a structured early warning information data stream. Operating parameters are adjusted based on this data stream to obtain optimized operating configuration data. This system enables precise status monitoring and early fault warning of the magnetic drive components of the water supply unit, meeting the stability and safety requirements of equipment operation and maintenance under complex operating conditions.

[0049] Reference Figure 2 The second embodiment of the present invention provides a status monitoring system for a magnetic drive component of a water supply device, comprising: The signal component acquisition module acquires the vibration signal of the magnetic drive component of the water supply unit and performs hierarchical separation and filtering correction to obtain the signal component data of each level. The energy denoising module calculates the energy value of each component based on the signal component data. If the energy value of a component exceeds a preset energy threshold, noise is removed by filtering, and then the format is standardized to obtain a dataset without noise components. The fatigue sign determination module separates high and low frequency signals from the dataset, calculates the energy value of each frequency band signal, filters abnormal frequency bands based on the energy value of each frequency band signal, and extracts frequency band energy features that reflect the fatigue characteristics of the component. The anomaly detection module constructs a comprehensive energy distribution pattern based on the frequency band energy characteristics and extracts energy quantification indicators from the energy distribution pattern. If the quantification indicators exceed a preset normal quantization threshold, an anomaly is detected, and an anomaly detection result is obtained. The quantification indicators include: energy fluctuation amplitude, average energy value, and energy change rate. The fault level determination module calculates a comprehensive early warning score based on the energy distribution pattern and the anomaly judgment result, and classifies the fault level based on the comprehensive early warning score; The configuration optimization module integrates the fault level, the abnormal frequency band, and the comprehensive early warning score to generate a structured early warning information data stream. Based on the early warning information data stream, the operating parameters are adjusted to obtain the optimized operating parameter configuration.

[0050] It should be noted that the water supply magnetic drive component status monitoring system provided in this embodiment of the invention is used to execute all the process steps of the water supply magnetic drive component status monitoring method in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0051] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an energy spectrum processing module program. When the processor executes the computer program, it implements the steps in the above-described embodiments of the magnetic drive component state method for water supply devices, for example... Figure 1 The step S101 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments, such as the fault level determination module.

[0052] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0053] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0054] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0055] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0056] If the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0057] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0058] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for monitoring the status of a magnetic drive component in a water supply device, characterized in that, include: The vibration signal of the magnetic drive component of the water supply device is acquired and subjected to hierarchical separation and filtering correction to obtain the signal component data of each level; Based on the signal component data, the energy value of each component is calculated. If the energy value of a certain component exceeds the preset energy threshold, noise is removed by filtering, and then the format is standardized to obtain a dataset without noise components. For the dataset, high and low frequency signals are separated, the energy value of each frequency band signal is calculated, abnormal frequency bands are screened based on the energy value of each frequency band signal, and frequency band energy characteristics reflecting the fatigue characteristics of the component are extracted from them; Based on the frequency band energy characteristics, a comprehensive energy distribution pattern is constructed, and a quantitative index of energy is extracted from the energy distribution pattern. If the quantitative index exceeds the preset normal quantitative threshold, an anomaly is determined, and an anomaly determination result is obtained. The quantitative indicators mentioned above include: energy fluctuation amplitude, average energy value, and energy change rate; Based on the energy distribution pattern and the anomaly determination result, a comprehensive early warning score is calculated, and the fault level is classified according to the comprehensive early warning score. By integrating the fault level, the abnormal frequency band, and the comprehensive early warning score, a structured early warning information data stream is generated. The operating parameters are then adjusted based on the early warning information data stream to obtain an optimized operating parameter configuration.

2. The method for monitoring the status of the magnetic drive component of the water supply device according to claim 1, characterized in that, The process of acquiring the vibration signal of the magnetic drive component of the water supply device and performing hierarchical separation and filtering correction to obtain signal component data for each level includes: Vibration signals of the magnetic drive component of the water supply unit under specific loads are collected by sensors. The vibration signal is separated into layers to obtain the initial signal components of each layer; The initial signal component is compared with a preset component feature threshold. If the initial signal component at a certain level exceeds the component feature threshold, filtering correction is performed to obtain the corrected signal component. The corrected signal component is integrated with the initial signal component, and the integrated signal component is subjected to format standardization processing to obtain signal component data at each level.

3. The method for monitoring the status of the magnetic drive component of the water supply device according to claim 1, characterized in that, The step of calculating the energy value of each component based on the signal component data, and if the energy value of a certain component exceeds a preset energy threshold, removing noise through filtering, and then performing format standardization processing to obtain a dataset with noise-free components, includes: Calculate the energy value of each signal component in the signal component data. If the energy value of a certain component exceeds a preset energy threshold, it is marked as an abnormal signal component. The abnormal signal components are filtered and denoised to obtain preliminary noise-free components; The noise-free components are format-normalized and integrated to obtain the final noise-free component dataset.

4. The method for monitoring the status of the magnetic drive component of the water supply device according to claim 1, characterized in that, The process of separating high- and low-frequency signals from the dataset, calculating the energy value of each frequency band, filtering abnormal frequency bands based on the energy values ​​of each frequency band, and extracting frequency band energy features reflecting the fatigue characteristics of the component includes: Perform time-frequency transformation on the dataset to obtain time-frequency distribution data; Based on a preset frequency band division rule, the low-frequency signal and high-frequency signal in the time-frequency distribution data are separated to obtain segmented signals; For the segmented signal, each segmented signal corresponds to a frequency band. The energy value of the segmented signal in the time domain is calculated and accumulated to obtain the total energy of the corresponding frequency band signal. The total energy is compared with a preset fatigue judgment threshold, and frequency band signals whose total energy exceeds the fatigue judgment threshold are filtered out. Frequency band energy features reflecting the fatigue characteristics of the component are then extracted from these signals.

5. The method for monitoring the status of the magnetic drive component of the water supply device according to claim 1, characterized in that, The process involves constructing a comprehensive energy distribution pattern based on the frequency band energy characteristics, extracting energy quantization indicators from the energy distribution pattern, and determining an anomaly if the quantization indicators exceed a preset normal quantization threshold. The quantitative indicators mentioned above include: energy fluctuation amplitude, average energy value, and energy change rate, including: The energy characteristics of each frequency band are weighted and calculated to obtain a comprehensive energy value; By integrating the comprehensive energy values ​​according to the time series, a comprehensive energy distribution pattern is constructed. Quantitative indicators of energy in the time series are extracted from the energy distribution pattern. The key quantitative indicators include energy fluctuation amplitude, average energy value, and energy change rate. The quantitative indicators are compared with preset normal quantitative thresholds. If any of the quantitative indicators exceeds the normal quantitative thresholds, an anomaly is determined, and an anomaly determination result is obtained.

6. The method for monitoring the status of the magnetic drive component of the water supply device according to claim 1, characterized in that, The step of calculating a comprehensive early warning score based on the energy distribution pattern and the anomaly determination result, and classifying the fault level based on the comprehensive early warning score, includes: Anomaly features are extracted from the energy distribution pattern, including the energy peak location and fluctuation amplitude. By combining the abnormal location characteristics with the abnormality determination results, a comprehensive early warning score is calculated. The comprehensive early warning score is compared with a preset early warning score threshold. If the comprehensive early warning score exceeds the early warning score threshold, the fault level is classified according to the extent of the exceedance, and the fault level classification result is determined.

7. The method for monitoring the status of the magnetic drive component of the water supply device according to claim 6, characterized in that, The process involves integrating the fault level, the abnormal frequency band, and the comprehensive early warning score to generate a structured early warning information data stream. Based on this data stream, operating parameters are adjusted to obtain an optimized operating parameter configuration, including: Based on the classification results, the fault level, the abnormal frequency band, and the comprehensive early warning score are integrated to generate a structured early warning information data stream; The warning information data stream is transmitted according to a preset output format, and a transmission confirmation message is obtained. If the transmission confirmation information indicates successful transmission, the equipment operating parameters are adjusted for the specific load to obtain the adjusted equipment operating parameters, which include operating frequency and torque output. The adjusted equipment operating parameters are updated to the equipment operating configuration to obtain the optimized operating parameter configuration.

8. A status monitoring system for a magnetic drive component of a water supply device, characterized in that, include: The signal component acquisition module acquires the vibration signal of the magnetic drive component of the water supply unit and performs hierarchical separation and filtering correction to obtain the signal component data of each level. The energy denoising module calculates the energy value of each component based on the signal component data. If the energy value of a component exceeds a preset energy threshold, noise is removed by filtering, and then the format is standardized to obtain a dataset without noise components. The fatigue sign determination module separates high and low frequency signals from the dataset, calculates the energy value of each frequency band signal, filters abnormal frequency bands based on the energy value of each frequency band signal, and extracts frequency band energy features that reflect the fatigue characteristics of the component. The anomaly detection module constructs a comprehensive energy distribution pattern based on the frequency band energy characteristics and extracts energy quantization indicators from the energy distribution pattern. If the quantization indicators exceed a preset normal quantization threshold, an anomaly is detected, and an anomaly detection result is obtained. The quantitative indicators mentioned above include: energy fluctuation amplitude, average energy value, and energy change rate; The fault level determination module calculates a comprehensive early warning score based on the energy distribution pattern and the anomaly judgment result, and classifies the fault level based on the comprehensive early warning score; The configuration optimization module integrates the fault level, the abnormal frequency band, and the comprehensive early warning score to generate a structured early warning information data stream. Based on the early warning information data stream, the operating parameters are adjusted to obtain the optimized operating parameter configuration.

Images