A remote intelligent control method and system for permanent magnet micro pump station units

Abstract

This invention relates to the field of pump station unit control technology, and particularly to a remote intelligent control method and system for permanent magnet micro pump station units. The method includes: synchronously acquiring multiple signals and using bandpass filtering to separate key components, providing a reliable data foundation for subsequent analysis; effectively integrating multi-dimensional information to reflect the unit's status by calculating the entropy values ​​related to high-frequency vibration and current harmonics and defining dual-frequency entropy change characteristic values; establishing a correlation mapping between characteristic values ​​and pressure pulsation peak values ​​to determine latent instability states, thus anticipating potential unit failure risks in advance; predicting expected pressure pulsation values ​​based on historical trends of characteristic values ​​and generating feedforward compensation control quantities for early intervention to stabilize unit operation; calculating feedback correction factors based on actual values ​​and dynamically correcting control quantities to form hierarchical instructions, achieving precise and flexible real-time regulation; and evaluating the effectiveness of regulation by calculating the gradient change rate, providing a quantitative basis for subsequent optimization of control strategies.

Description

Technical Field

[0001] This invention relates to the field of pump station unit control technology, and in particular to a remote intelligent control method and system for a permanent magnet micro pump station unit. Background Technology

[0002] Permanent magnet micro pump station units, with their advantages of high efficiency, rapid response, and compact structure, are widely used in distributed water supply, precision agricultural irrigation, and industrial circulation scenarios. To achieve large-scale, unmanned operation and maintenance, existing remote monitoring and control methods generally adopt a single-parameter threshold alarm and single-loop feedback control mode.

[0003] However, while this model can maintain basic control under stable operating conditions and with minimal disturbances, its fundamental flaw lies in the lack of a collaborative perception and correlation analysis mechanism for the dynamic coupling relationships between the mechanical, electromagnetic, and other physical field states of the unit. When faced with complex operating disturbances, such as sudden changes in water sediment content, nonlinear changes in pipeline resistance, or early wear within the unit, the pumping station system often undergoes a latent instability process that evolves from a normal state to a fault state. During this process, multiple weak abnormal features, such as the broadening of the mechanical vibration spectrum, the evolution of stator current harmonic components, and the intensification of outlet pressure pulsation, will appear simultaneously and intertwine and couple with each other, but the change in any single parameter may not reach its independent alarm threshold. Existing methods cannot perform early and accurate collaborative identification of the aforementioned cross-domain coupled weak anomalies, leading to missed opportunities for intervention in the early stages of latent instability. This can cause repeated oscillations and a surge in energy consumption during the regulation process, and may also lead to unexpected shutdowns of the pumping station due to overly aggressive actions, seriously affecting the stability, economy, and reliability of the intelligent pumping station group operation. Therefore, this invention proposes a remote intelligent control method and system for permanent magnet micro pump station units. Summary of the Invention

[0004] The purpose of this invention is to solve the problems in the background art, and to propose a remote intelligent control method and system for permanent magnet micro pump station units.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of this invention provides a remote intelligent control method for a permanent magnet micro pump station unit, comprising: S1. Synchronously acquire vibration signals, stator current signals, and outlet pressure pulsation signals of the permanent magnet micro pump station unit; separate high-frequency vibration components and current harmonic components through bandpass filtering; S2. Calculate the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component respectively, and define the weighted difference between the two as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit. S3. Establish a real-time correlation mapping between dual-frequency entropy change characteristic values ​​and outlet pressure pulsation peak values, and determine whether the permanent magnet micro pump station unit has entered a latent instability state; where the outlet pressure pulsation peak value is the peak value extracted from the collected outlet pressure pulsation signal. S4. If it is determined that the state of hidden instability has been entered, the expected value of the outlet pressure pulsation is estimated based on the historical trend of the dual-frequency entropy change characteristic value, and a feedforward compensation control quantity is generated. S5. Obtain the actual value of the outlet pressure pulsation within the same control cycle, calculate the feedback correction factor by combining it with the expected value of the outlet pressure pulsation, and dynamically correct the feedforward compensation control quantity to form a hierarchical gradual control command. S6. After issuing and executing the tiered and gradual control instructions, calculate the gradient change rate used to evaluate the effectiveness of the control.

[0006] Furthermore, S1 specifically includes: Set the passband frequency range of the first bandpass filter; The first bandpass filter is used to filter the acquired vibration signal to obtain the high-frequency vibration component; Set the passband frequency range of the second bandpass filter; The acquired stator current signal is filtered using a second bandpass filter to obtain the current harmonic components.

[0007] Furthermore, S2 specifically includes: Receive the separated high-frequency vibration components; The high-frequency vibration component is decomposed into wavelet packet to the Mth layer to obtain the signals of each node in the Mth layer. Calculate the signal energy of each node, and calculate the proportion of the signal energy of each node to the total energy of the layer; Based on all the calculated proportions, the energy entropy of the high-frequency vibration components is calculated using the Shannon entropy formula. Receive the separated current harmonic components; A fast Fourier transform is performed on the harmonic component of the current to extract the amplitude of each harmonic except the fundamental wave. Calculate the square of each harmonic amplitude and its ratio to the sum of the squares of all harmonic amplitudes; Based on all the calculated ratios, the distortion entropy of the current harmonic components is calculated using the Shannon entropy formula. The calculated energy entropy and distortion entropy are weighted and the difference is calculated according to the preset weight coefficients to generate dual-frequency entropy change characteristic values.

[0008] Furthermore, S3 specifically includes: The multiple dual-frequency entropy change eigenvalues ​​obtained by continuous calculation are arranged in chronological order to form a dual-frequency entropy change eigenvalue sequence. Multiple outlet pressure pulsation peaks extracted synchronously are arranged in the same time order to form an outlet pressure pulsation peak sequence. Within the sliding time window, the dynamic correlation coefficient is obtained by dynamically calculating the sequence of dual-frequency entropy change eigenvalues ​​calculated in real time and the sequence of peak outlet pressure pulsations measured synchronously. Preset association confidence threshold; If the absolute value of the dynamic correlation coefficient calculated within the current sliding time window is lower than the preset correlation confidence threshold, the correlation is determined to be weakened, and multi-source data cross-verification diagnosis is initiated.

[0009] Furthermore, initiating multi-source data cross-verification and diagnostics includes: Infrared thermal image data of key components of permanent magnet micro pump station units are obtained, and the state of air gap magnetic flux density of permanent magnets is indirectly estimated based on motor electrical signals. When the infrared thermal image data shows that the local temperature rise rate exceeds the set temperature rise threshold, and at the same time the product of the air gap magnetic flux density fluctuation amplitude extracted from the indirect estimation result and the dual-frequency entropy change characteristic value exceeds the preset mutual verification threshold, the permanent magnet micro pump station unit is determined to have entered a latent instability state; otherwise, the permanent magnet micro pump station unit is determined not to have entered a latent instability state.

[0010] Furthermore, S4 specifically includes: The dual-frequency entropy change characteristic values ​​collected in several consecutive control cycles before the current moment are obtained to form a historical change sequence; Based on historical change sequences, the exponential smoothing algorithm is used to calculate the dual-frequency entropy change characteristic trend value for the next control cycle. Based on the historical trend of the dual-frequency entropy change characteristic trend value, the most similar reference pattern segment is matched, and the expected curve of the outlet pressure pulsation is extracted from the reference pattern segment. Calculate the corresponding expected value of the export pressure pulsation based on the time delay of the current moment relative to the starting point of the expected export pressure pulsation curve; Analyze the changing trend of the expected export pressure pulsation curve in the next control cycle to obtain the sign and amplitude of its first derivative; Simultaneously, the coupling degree of the permanent magnet synchronous motor dq axis current is calculated based on the real-time collected motor current signal. By combining the sign of the first derivative with the real-time measured values ​​of amplitude and coupling degree, the motor current vector angle adjustment is calculated as the advance adjustment amount; Calculate the difference between the expected value of the outlet pressure pulsation and the actual value of the outlet pressure pulsation in the current cycle, and multiply the difference by the preset frequency fine-tuning coefficient to obtain the differential component of the inverter output frequency.

[0011] Furthermore, the pre-stored historical fault-free operation database contains multiple reference pattern segments consisting of historical dual-frequency entropy change characteristic sequence segments and their subsequent actual outlet pressure pulsation sequence segments.

[0012] Furthermore, S5 specifically includes: At the end of the next control cycle, the actual value of the outlet pressure pulsation during that cycle is collected synchronously; Calculate the difference between the actual value of the outlet pressure pulsation and the expected value of the outlet pressure pulsation for the same control cycle; Multiply the calculated difference by the negative feedback gain coefficient to obtain the feedback correction factor; The feedback correction factor is added to the obtained advance adjustment amount to obtain the corrected final current vector angle adjustment command; The feedback correction factor is added to the obtained differential component to obtain the corrected final frequency adjustment command; The final current vector angle adjustment command and the final frequency adjustment command are integrated to form a hierarchical progressive control command.

[0013] Furthermore, S6 specifically includes: The first dual-frequency entropy change characteristic value and the first outlet pressure pulsation peak value are obtained and calculated in the current control cycle; wherein, the first outlet pressure pulsation peak value is the peak value extracted from the outlet pressure pulsation signal collected in the current control cycle. The second dual-frequency entropy change characteristic value and the second outlet pressure pulsation peak value are obtained and calculated from the immediately preceding control cycle; wherein, the second outlet pressure pulsation peak value is the peak value extracted from the outlet pressure pulsation signal collected from the previous control cycle. Calculate the absolute difference between the first dual-frequency entropy change eigenvalue and the second dual-frequency entropy change eigenvalue to obtain the change in eigenvalue; calculate the ratio of the change in eigenvalue to the second dual-frequency entropy change eigenvalue to obtain the first relative rate of change. Calculate the absolute difference between the peak value of the first outlet pressure pulsation and the peak value of the second outlet pressure pulsation to obtain the pressure pulsation change; calculate the ratio of the pressure pulsation change to the peak value of the second outlet pressure pulsation to obtain the second relative change rate. Compare the numerical values ​​of the first relative rate of change and the second relative rate of change, and select the one with the largest value as the gradient rate of change.

[0014] A second aspect of the present invention provides a remote intelligent control system for a permanent magnet micro pump station unit, comprising: Signal synchronous acquisition and filtering module: synchronously acquires vibration signals, stator current signals, and outlet pressure pulsation signals of the permanent magnet micro pump station unit; and separates high-frequency vibration components and current harmonic components through bandpass filtering; Dual-frequency entropy change characteristic calculation module: calculates the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component respectively, and defines the weighted difference between the two as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit; Latent instability state judgment module: Establish a real-time correlation mapping between dual-frequency entropy change characteristic values ​​and outlet pressure pulsation peak values, and determine whether the permanent magnet micro pump station unit has entered a latent instability state; Feedforward compensation control generation module: If it is determined that the system has entered a hidden instability state, the system will estimate the expected value of the outlet pressure pulsation based on the historical trend of the dual-frequency entropy change characteristic value and generate the feedforward compensation control quantity. Feedback correction dynamic control module: Obtain the actual value of the outlet pressure pulsation within the same control cycle, calculate the feedback correction factor by combining it with the expected value of the outlet pressure pulsation, and dynamically correct the feedforward compensation control quantity to form a hierarchical gradual control command. Regulation effectiveness assessment module: After issuing and executing the tiered and gradual regulation instructions, calculate the gradient change rate used to assess the regulation effectiveness.

[0015] Compared with existing technologies, the advantages of the remote intelligent control method and system for permanent magnet micro pump station units provided by the present invention are as follows: 1) By simultaneously acquiring multiple key signals, namely the vibration signal, stator current signal, and outlet pressure pulsation signal of the permanent magnet micro pump station unit, the operating status information of the permanent magnet micro pump station unit can be comprehensively obtained, providing a data foundation for subsequent analysis; by separating high-frequency vibration components and current harmonic components through bandpass filtering, interference signals can be effectively removed, key characteristics can be highlighted, signal quality can be improved, and subsequent analysis can be more accurate and reliable, and potential problems of the unit can be detected in a timely manner. 2) By calculating the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component separately, and defining the weighted difference as the dual-frequency entropy change characteristic value, the characteristics of the two different types of signals are organically combined to comprehensively reflect the changes in the unit's operating status, providing a more representative indicator for judging the unit's status and enhancing the ability to identify unit anomalies; by establishing a real-time correlation mapping between the dual-frequency entropy change characteristic value and the peak value of the outlet pressure pulsation, the changes in the unit's status can be monitored in real time, and based on this, it can be determined whether the unit has entered a latent instability state, and potential unstable factors of the unit can be detected in advance to avoid further deterioration of the fault, providing a basis for taking timely measures to ensure the stable operation of the unit; 3) By predicting the expected value of outlet pressure pulsation based on historical trends and generating feedforward compensation control quantities, unit operating parameters can be adjusted in advance before a fault occurs, proactively preventing faults, effectively suppressing outlet pressure pulsation, improving unit stability and reliability, and reducing downtime losses caused by faults; by obtaining actual values ​​to calculate feedback correction factors and dynamically correcting feedforward compensation control quantities, layered progressive control commands can be formed, which can adjust control strategies in real time according to actual operating conditions, improving the accuracy and adaptability of control, and ensuring stable operation of the unit under different operating conditions; by calculating the gradient change rate to evaluate the effectiveness of control, the effect of control measures can be intuitively understood, providing a reference for subsequent optimization of control strategies, and continuously improving the unit's operating efficiency and stability. Attached Figure Description

[0016] Figure 1 This is a flowchart of a remote intelligent control method for a permanent magnet micro pump station unit proposed in this invention.

[0017] Figure 2 This is a block diagram of a remote intelligent control system for a permanent magnet micro pump station unit proposed in this 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. Example 1:

[0019] Please see Figure 1 This invention provides a remote intelligent control method for a permanent magnet micro pump station unit, comprising: S1. Synchronously acquire vibration signals, stator current signals, and outlet pressure pulsation signals of the permanent magnet micro pump station unit; separate high-frequency vibration components and current harmonic components through bandpass filtering; S2. Calculate the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component respectively, and define the weighted difference between the two as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit. S3. Establish a real-time correlation mapping between dual-frequency entropy change characteristic values ​​and outlet pressure pulsation peak values, and determine whether the permanent magnet micro pump station unit has entered a latent instability state; where the outlet pressure pulsation peak value is the peak value extracted from the collected outlet pressure pulsation signal. S4. If it is determined that the state of hidden instability has been entered, the expected value of the outlet pressure pulsation is estimated based on the historical trend of the dual-frequency entropy change characteristic value, and a feedforward compensation control quantity is generated. Among them, the feedforward compensation control quantity includes the advance adjustment of the motor current vector angle and the micro-component of the inverter output frequency. S5. Obtain the actual value of the outlet pressure pulsation within the same control cycle, calculate the feedback correction factor by combining it with the expected value of the outlet pressure pulsation, and dynamically correct the feedforward compensation control quantity to form a hierarchical gradual control command. S6. After issuing and executing the tiered and gradual control instructions, calculate the gradient change rate used to evaluate the effectiveness of the control.

[0020] It should be further explained that, in the specific implementation process, high-frequency vibration components and current harmonic components are separated by bandpass filtering, including: Based on the inherent frequency of the pump body structure, the passband frequency range of the first bandpass filter is set; The first bandpass filter is used to filter the acquired vibration signal to obtain the high-frequency vibration component; Based on the characteristic harmonic frequency of the electromagnetic excitation of the motor, the passband frequency range of the second bandpass filter is set. The acquired stator current signal is filtered using a second bandpass filter to obtain the current harmonic components. Specifically, parameter calibration is performed for specific models of permanent magnet micro pump station units. For vibration signals, the first-order or natural frequency range of the pump body structure, including the impeller, bearing housing, and casing, is determined through experimental modal analysis or by consulting technical data. For example, for a specific micro pump, its natural frequency may mainly be concentrated between 500Hz and 2000Hz. Accordingly, the passband frequency range of the first bandpass filter is set to [500Hz, 2000Hz]. After the original vibration signal is passed through the first bandpass filter, low-frequency vibrations below 500Hz (which may come from the foundation or distant equipment) and noise above 2000Hz are effectively filtered out, and the output high-frequency vibration components mainly contain information related to the mechanical dynamic characteristics of the pump body itself. For the stator current signal, its harmonic components mainly originate from the electromagnetic excitation inside the permanent magnet synchronous motor, especially the characteristic frequencies related to the rotor permanent magnet magnetic field and stator cogging effect. These characteristic frequencies are integer or fractional multiples of the fundamental frequency of the motor power supply. Assuming the motor operates at a 50Hz fundamental frequency, its characteristic harmonic frequencies may be distributed around frequencies such as 250Hz (5th order) and 350Hz (7th order). The passband frequency range of the second bandpass filter is set to cover the characteristic harmonic frequencies, for example, [200Hz, 1000Hz]. After the original stator current signal passes through the second bandpass filter, the fundamental (50Hz) component is significantly attenuated, and the output current harmonic components are enhanced, which strengthens the harmonic information that reflects the abnormal electromagnetic state of the motor. These abnormal electromagnetic states include rotor eccentricity, slight short circuits in the windings, etc. Through targeted filtering, a data foundation is laid for the subsequent calculation of the characteristic entropy representing the mechanical and electromagnetic states.

[0021] It should be further explained that, in the specific implementation process, the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component are calculated separately, and the weighted difference between the two is defined as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit, including: Receive the separated high-frequency vibration components; The high-frequency vibration component is decomposed into wavelet packet to the Mth layer to obtain the signals of each node in the Mth layer. Calculate the signal energy of each node, and calculate the proportion of the signal energy of each node to the total energy of the layer; Based on all the calculated proportions, the energy entropy of the high-frequency vibration components is calculated using the Shannon entropy formula. Receive the separated current harmonic components; A fast Fourier transform is performed on the harmonic component of the current to extract the amplitude of each harmonic except the fundamental wave. Calculate the square of each harmonic amplitude and its ratio to the sum of the squares of all harmonic amplitudes; Based on all the calculated ratios, the distortion entropy of the current harmonic components is calculated using the Shannon entropy formula. The calculated energy entropy and distortion entropy are weighted and the difference is calculated according to the preset weight coefficients to generate dual-frequency entropy change characteristic values. Specifically, the high-frequency vibration components separated from the first bandpass filter are received, and the corresponding time-domain signals are obtained; wavelet packet decomposition is performed on the time-domain signals; a preset decomposition level M is used, where M is a preset positive integer, for example, M=4; a complete M-level wavelet packet decomposition of the time-domain signals is performed using a selected mother wavelet, where the mother wavelet is, for example, the Daubechies 4 wavelet; after decomposition, the M-th level will yield... There are 16 node signals, each corresponding to a specific frequency band; Calculate the energy of each node signal; define the node signal sequence of the i-th node signal in mode M as WP(M,i), and its energy is the sum of the squares of the amplitudes of all data points of that node signal, calculated by the following formula: In the formula, Let i represent the signal energy of a single node, and i denote the node signal index. Calculate the sum of the signal energies of all nodes in this layer, i.e., the total energy. : ; Calculate the proportion of each node's signal energy value to the total energy of that layer. : ;Proportion This reflects the distribution of vibrational energy across different fine frequency bands; Calculate energy entropy using the Shannon entropy formula in information theory. : , where the proportion When equal to zero, the definition is... Energy entropy The magnitude of the vibration energy directly reflects the uniformity of the distribution across frequency bands; when energy is concentrated in one or a few frequency bands, such as when resonance occurs at a single frequency, the proportion... Concentrated distribution, energy entropy Smaller; when energy is dispersed across many frequency bands (e.g., broadband shocks or chaotic oscillations occur), the proportion is smaller. Uniform distribution, energy entropy Larger; energy entropy It can effectively characterize abnormal changes in the mechanical vibration state from the perspective of energy distribution complexity; The current harmonic components separated from the second bandpass filter are received, and the corresponding time-domain signals are obtained. A fast Fourier transform is performed on the time-domain signal to convert it to the frequency domain, obtaining the spectrum. From the spectrum, the amplitudes of each harmonic, excluding the fundamental frequency component (which has already been attenuated in the filtering stage), are identified and extracted. Where h represents the harmonic order, such as the 5th, 7th, 11th, etc. Analyze the distribution characteristics of harmonic amplitudes; calculate the square of the amplitude of each harmonic. And calculate the sum of the squares of all harmonic amplitudes. : In the formula, The total harmonic order is given; calculate the proportion of the square of each harmonic amplitude to the sum of the squares of all harmonic amplitudes. : This ratio It reflects the distribution of total harmonic distortion energy at different harmonic orders; Calculate the distortion entropy using the Shannon entropy formula. : ; Distortion entropy The uniformity of harmonic energy distribution is quantified; when harmonic energy is concentrated at a specific order, such as a certain characteristic harmonic, the distortion entropy... The distortion entropy is relatively small; however, when the harmonic energy distribution is widespread and disordered, the distortion entropy is relatively small. A larger value may indicate more complex electromagnetic interference or a fault. After obtaining the energy entropy respectively With distortion entropy Then, a weighted difference calculation is performed according to the preset weight coefficients α and β to generate the dual-frequency entropy change eigenvalue DF: In the formula, α+β=1, and α and β are determined experimentally based on the contribution of vibration and current to the instability of a specific pump unit. It is understandable that the calculation formula of the dual-frequency entropy change characteristic value integrates and compares the uncertainty characteristics of the mechanical side and the electromagnetic side. An increase in the dual-frequency entropy change characteristic value DF may be more likely to indicate an anomaly where the complexity of mechanical vibration is dominant, while a decrease (or negative increase) in the dual-frequency entropy change characteristic value DF may be more likely to indicate an anomaly where the concentration of electromagnetic harmonics is dominant.

[0022] It needs further explanation that establishing a real-time correlation mapping between the dual-frequency entropy change characteristic value and the peak value of the outlet pressure pulsation, and determining whether the permanent magnet micro pump station unit has entered a latent instability state, includes: The multiple dual-frequency entropy change eigenvalues ​​obtained by continuous calculation are arranged in chronological order to form a dual-frequency entropy change eigenvalue sequence. Multiple outlet pressure pulsation peaks extracted synchronously are arranged in the same time order to form an outlet pressure pulsation peak sequence. Within the sliding time window, the dynamic correlation coefficient is obtained by dynamically calculating the sequence of dual-frequency entropy change eigenvalues ​​calculated in real time and the sequence of peak outlet pressure pulsations measured synchronously. A preset association confidence threshold is set, which serves as the dividing standard for judging the strength of the association between two sequences. If the absolute value of the dynamic correlation coefficient calculated within the current sliding time window is lower than the preset correlation confidence threshold, the correlation is determined to be weakened, and multi-source data cross-verification diagnosis is initiated. Specifically, the dual-frequency entropy change eigenvalue sequence is calculated according to a fixed control period. , This represents the total number of data points in the sequence, i.e., the endpoint number; synchronously, the peak value sequence of the outlet pressure pulsation is extracted from the outlet pressure pulsation signal for each control cycle using a peak detection algorithm (such as finding local extrema). ; A sliding time window is used, for example, containing data from the most recent 20 control cycles. At each new moment, the dynamic correlation between the dual-frequency entropy change eigenvalue sequence within the window and the outlet pressure pulsation peak sequence is calculated using the Pearson correlation coefficient formula. In the formula, t represents time. Describing covariance, This represents a subsequence containing the most recent consecutive data points, extracted from the entire dual-frequency entropy change eigenvalue sequence at the current moment. This represents a subsequence extracted from the peak output pressure pulsation sequence at the current moment, containing several recent consecutive data points. Representing a subsequence The sample standard deviation Representing a subsequence The sample standard deviation; the calculated This is the dynamic correlation coefficient. The closer its absolute value is to 1, the stronger the synchronization and the closer the correlation between the changing trends of the dual-frequency entropy change characteristic value sequence and the outlet pressure pulsation peak sequence within the most recent time window. A preset correlation confidence threshold is set, for example, 0.6. In each control cycle, if the absolute value of the currently calculated dynamic correlation coefficient is greater than or equal to the preset correlation confidence threshold, the mechanical-electrical-hydraulic state coupling relationship is determined to be normal, and routine monitoring continues. If the absolute value of the calculated dynamic correlation coefficient is less than the preset correlation confidence threshold, the correlation between the dual-frequency entropy change characteristic value sequence and the outlet pressure pulsation peak sequence is determined to be weakened. The multi-source data cross-verification diagnostic process is immediately triggered, and infrared thermography and electromagnetic estimation data are introduced to perform a joint diagnosis with higher confidence, thereby determining whether a latent instability state has officially been entered.

[0023] It should be further explained that, in the specific implementation process, multi-source data mutual verification and diagnosis is initiated, including: Infrared thermal image data of key components of permanent magnet micro pump station units are obtained, and the state of air gap magnetic flux density of permanent magnets is indirectly estimated based on motor electrical signals. When the infrared thermal image data shows that the local temperature rise rate exceeds the set temperature rise threshold, and at the same time the product of the air gap magnetic flux density fluctuation amplitude extracted from the indirect estimation result and the dual-frequency entropy change characteristic value exceeds the preset mutual verification threshold, the permanent magnet micro pump station unit is determined to have entered a latent instability state; otherwise, the permanent magnet micro pump station unit is determined not to have entered a latent instability state. Specifically, through the communication interface, images of key areas recently captured by infrared thermal imagers deployed at the pump station are retrieved, such as images of key areas captured within a 30-second period. Image analysis software is used to identify and select preset key areas, such as the output terminals of the motor stator windings, the outer surfaces of the front and rear bearing housings, and the flange surfaces connecting the pump body and the motor. The average temperature of the preset key areas is calculated, and a temperature-time curve is plotted. By analyzing the curve, the local temperature rise rate within a specific time period (such as the most recent 10 seconds) is calculated. The local temperature rise rate is compared with a preset temperature rise threshold, which is a preventative limit determined based on the long-term allowable operating temperature of the materials of the key components of the permanent magnet micro pump station unit, the thermal aging characteristics of the insulation class, and the statistical value of the maximum temperature rise during historical normal operation. For example, it is set to rise by 5 degrees Celsius per minute. If the temperature rise rate of any key area exceeds the temperature rise threshold, a thermal anomaly is marked as true. The instantaneous values ​​of the three-phase terminal voltage and three-phase current of the permanent magnet synchronous motor are acquired synchronously within the most recent time period. These values ​​are then input into a pre-established motor equivalent circuit model stored on a cloud platform. This model is constructed based on the motor nameplate parameters and basic electromagnetic relationships. Through model inversion algorithms, such as least squares estimation or observer algorithms, the real-time amplitude and phase angle of the rotor permanent magnet flux linkage, i.e., its position in space, are estimated. Combined with known parameters such as the number of motor pole pairs and winding coefficients, the estimated value of the magnetic flux density in the air gap is further calculated. Time-domain analysis is performed on the estimated magnetic flux density to filter out its steady-state DC component, obtaining the AC component reflecting magnetic field fluctuations. The amplitude of this AC component is then calculated as the amplitude of the air gap magnetic flux density fluctuation. The absolute value of the current dual-frequency entropy change characteristic value is calculated and multiplied by the extracted air gap magnetic flux density fluctuation amplitude to obtain a product. Simultaneously, historical operating data is retrieved, i.e., during the normal operation of the permanent magnet micro pump station unit after installation and commissioning, a large number of dual-frequency entropy change characteristic values ​​and corresponding air gap magnetic flux density fluctuation amplitudes are continuously recorded. Multiple sets of historical product values ​​calculated from the dual-frequency entropy change characteristic values ​​and corresponding air gap magnetic flux density fluctuation amplitudes are extracted to form a historical product value set. Statistical analysis is performed on this historical product value set to calculate the statistical mean and standard deviation of all historical product values. The statistical mean plus three times the standard deviation is set as the cross-validation threshold. This cross-validation threshold is used to determine the critical condition for joint anomalies in multi-source data. Under normal operating conditions, the probability of the product value exceeding the cross-validation threshold is extremely low, for example, less than 0.3%. If both conditions are met simultaneously, the permanent magnet micro pump station unit is determined to have entered a latent instability state. If only one condition is met, or if neither condition is met, the unit is determined not to have entered a latent instability state. The current abnormal signs may be due to transient sensor interference or a minor one-sided anomaly. Monitoring will continue without triggering advanced control.

[0024] It should be further explained that, in the specific implementation process, if it is determined that a latent instability state has been entered, the expected value of the outlet pressure pulsation is estimated based on the historical trend of the dual-frequency entropy change characteristic value, and a feedforward compensation control quantity is generated, including: The dual-frequency entropy change characteristic values ​​collected in several consecutive control cycles before the current moment are obtained to form a historical change sequence; Based on historical change sequences, the exponential smoothing algorithm is used to calculate the dual-frequency entropy change characteristic trend value for the next control cycle. Based on the historical trend of the dual-frequency entropy change characteristic trend value, the most similar reference pattern segment is matched, and the typical evolution curve is extracted from the subsequent actual outlet pressure pulsation data of the reference pattern segment as the expected curve of outlet pressure pulsation; among them, the pre-stored historical fault-free operation database contains multiple reference pattern segments composed of historical dual-frequency entropy change characteristic sequence segments and their subsequent actual outlet pressure pulsation sequence segments. On the expected export pressure pulsation curve, the corresponding expected export pressure pulsation value is calculated based on the time delay of the current moment relative to the starting point of the expected export pressure pulsation curve. Analyze the changing trend of the expected export pressure pulsation curve in the next control cycle to obtain the sign and amplitude of its first derivative; Simultaneously, the coupling degree of the permanent magnet synchronous motor dq axis current is calculated based on the real-time collected motor current signal. By combining the sign of the first derivative with the real-time measured values ​​of amplitude and coupling degree, the motor current vector angle adjustment is calculated using the decoupling compensation formula, which serves as the advance adjustment amount. Calculate the difference between the expected value of the outlet pressure pulsation and the actual value of the outlet pressure pulsation in the current cycle, and multiply the difference by the preset frequency fine-tuning coefficient to obtain the differential component of the inverter output frequency. Specifically, the system reads the continuously calculated and stored dual-frequency entropy change feature values ​​over a recent period to form a historical change sequence. For example, the dual-frequency entropy change feature values ​​calculated and stored over 50 control cycles in the past 5 seconds are used to form this historical change sequence. An exponential smoothing algorithm is then applied to this historical change sequence. The exponential smoothing algorithm gives higher weight to recent data, thereby better capturing the historical trend of the dual-frequency entropy change feature values. The exponential smoothing algorithm is then used to calculate the trend value of the dual-frequency entropy change feature values ​​for the next control cycle, for example, the trend value of the dual-frequency entropy change feature values ​​for the next 100 milliseconds. Maintain a historical fault-free operation database, which stores reference mode segments consisting of paired historical dual-frequency entropy change characteristic sequence segments and their subsequent actual outlet pressure pulsation sequence segments. The reference mode segments completely record the complete dynamic correspondence between the evolution of electromechanical state characteristics and the outlet pressure pulsation response during various typical disturbances or transitions in history; for example, a historical dual-frequency entropy change characteristic value sequence within 10 consecutive control cycles, followed by a historical outlet pressure pulsation value sequence within the next 20 control cycles. The currently predicted dual-frequency entropy change characteristic trend value and its adjacent historical change sequence are combined to form a current pattern reflecting the current change state. This current pattern is then matched with the historical dual-frequency entropy change characteristic sequence segments in all reference pattern segments in the historical fault-free operation database for similarity, for example, by using the Dynamic Time Warping (DTW) algorithm or Euclidean distance calculation. This will identify the target pattern segment with the highest similarity. Extract the paired actual outlet pressure pulsation sequence segments from the target mode fragment. These sequence segments represent the complete response process of outlet pressure pulsations that actually occurred in history and correspond to the current similar feature mode. Use the historical outlet pressure pulsation sequence segments as the current predicted outlet pressure pulsation expectation curve. Based on the time delay between the current control start time and the starting point of the export pressure pulsation expectation curve, the export pressure pulsation value at the corresponding time point on the export pressure pulsation expectation curve is determined, and this value is used as the export pressure pulsation expectation value for the next control cycle. Calculate the first derivative of the expected export pressure pulsation curve for the corresponding time interval in the next control cycle to obtain its trend and intensity. For example, a positive derivative with a large amplitude indicates that the pressure is expected to rise rapidly. The trend corresponds to the sign of the first derivative, and the intensity corresponds to the amplitude of the first derivative. Simultaneously, based on the real-time collected three-phase current of the motor, the coupling degree of the dq axis current of the permanent magnet synchronous motor is obtained through coordinate transformation and calculation. This value reflects the strength of the cross coupling of the magnetic field inside the motor and is a key parameter affecting the decoupling effect of vector control. The calculated sign and amplitude of the first derivative, along with the real-time measured value of the coupling degree, are substituted into a preset decoupling compensation formula. The decoupling compensation formula is as follows: Advance Adjustment Amount = (Base Adjustment Coefficient × First Derivative Amplitude × Sign Factor) / (1 + Real-time Coupling Degree Measurement). In this formula, the sign factor is +1 or -1 depending on the sign of the first derivative, and the base adjustment coefficient is a constant calibrated experimentally. This formula initially determines the angle adjustment requirement based on the changing trend of the pressure expectation curve. Then, based on the current real-time electromagnetic coupling degree, this requirement is compensated and corrected to offset the influence of coupling effects on the control performance. The final calculation result is the advance adjustment amount of the motor current vector angle, achieving an adaptive combination of control commands and the real-time dynamic characteristics and electromagnetic state of the permanent magnet micro-pump station unit. The difference between the expected value of the outlet pressure pulsation and the actual value of the outlet pressure pulsation measured in the current control cycle is calculated. This difference is multiplied by a preset frequency fine-tuning coefficient to obtain the differential component of the inverter output frequency. The frequency fine-tuning coefficient is a proportional coefficient obtained through experimental calibration. It is the amount of change in the inverter output frequency that needs to be adjusted per unit outlet pressure pulsation deviation and is used to establish a quantitative relationship between pressure deviation and speed compensation. Finally, a feedforward compensation control quantity containing advance adjustment and a small component is generated for execution in the next control cycle.

[0025] It should be further explained that, in the specific implementation process, the feedback correction factor is calculated based on the expected value of the export pressure pulsation, and the feedforward compensation control quantity is dynamically adjusted to form a hierarchical gradual control instruction, including: At the end of the next control cycle, the actual value of the outlet pressure pulsation during that cycle is collected synchronously; Calculate the difference between the actual value of the outlet pressure pulsation and the expected value of the outlet pressure pulsation for the same control cycle; Multiply the calculated difference by the negative feedback gain coefficient to obtain the feedback correction factor; The feedback correction factor is added to the obtained advance adjustment amount to obtain the corrected final current vector angle adjustment command; The feedback correction factor is added to the obtained differential component to obtain the corrected final frequency adjustment command; The final current vector angle adjustment command and the final frequency adjustment command are integrated to form a hierarchical gradual control command, which is then sent to the pump station's local controller for execution. Specifically, it is assumed that a feedforward compensation control quantity for time T+1 is generated at time T; during the T+1 control cycle, the pump station local controller executes a hierarchical progressive control command; at the same time, at the end of the control cycle, the actual outlet pressure pulsation signal during the T+1 control cycle is collected by the pressure sensor, and its peak value is extracted to obtain the actual value of the outlet pressure pulsation during the control cycle. The actual value of the outlet pressure pulsation is compared with the expected value of the outlet pressure pulsation predicted at time T with a control cycle of T+1. The difference between the two is calculated, where the difference reflects the combined effect of the feedforward prediction error and the unexpected disturbance. Multiply the calculated difference by a negative feedback gain coefficient, such as -0.5, to obtain the feedback correction factor. The negative sign ensures that the correction direction is opposite to the error direction. That is, if the actual value of the outlet pressure pulsation is higher than the expected value of the outlet pressure pulsation, the correction factor is negative, which will cause the control command to adjust in the direction of reducing pressure. The feedback correction factor is added to the lead adjustment in the feedforward compensation control quantity to obtain the corrected final current vector angle adjustment command; similarly, the feedback correction factor is added to the differential component in the feedforward compensation control quantity to obtain the corrected final frequency adjustment command; here, the addition is a scalar addition, and the feedback correction is to simultaneously make proportional fine adjustments to the angle and frequency commands; in more complex implementations, different feedback gain coefficients can be set for the angle and frequency. Finally, the two revised final instructions, namely the final current vector angle adjustment instruction and the final frequency adjustment instruction, are encapsulated in a data packet to form a set of hierarchical gradual control instructions that integrate feedforward prediction and feedback correction. Among them, the hierarchical gradual control instructions continue to be sent to the local controller of the pump station through the cloud platform, ready to be executed at time T+2 of the next control cycle.

[0026] It should be further explained that, in the specific implementation process, after issuing and executing the tiered and gradual control instructions, the gradient change rate used to evaluate the effectiveness of the control is calculated, including: The first dual-frequency entropy change characteristic value and the first outlet pressure pulsation peak value are obtained and calculated in the current control cycle; wherein, the first outlet pressure pulsation peak value is the peak value extracted from the outlet pressure pulsation signal collected in the current control cycle. The second dual-frequency entropy change characteristic value and the second outlet pressure pulsation peak value are obtained and calculated from the immediately preceding control cycle; wherein, the second outlet pressure pulsation peak value is the peak value extracted from the outlet pressure pulsation signal collected from the previous control cycle. Calculate the absolute difference between the first dual-frequency entropy change eigenvalue and the second dual-frequency entropy change eigenvalue to obtain the change in eigenvalue; calculate the ratio of the change in eigenvalue to the second dual-frequency entropy change eigenvalue to obtain the first relative rate of change. Calculate the absolute difference between the peak value of the first outlet pressure pulsation and the peak value of the second outlet pressure pulsation to obtain the pressure pulsation change; calculate the ratio of the pressure pulsation change to the peak value of the second outlet pressure pulsation to obtain the second relative change rate. Compare the numerical values ​​of the first relative rate of change and the second relative rate of change, and select the one with the largest value as the gradient rate of change; Understandably, when the permanent magnet micro pump station unit is in the calibrated stable operating condition, the gradient change rate is continuously collected and calculated in multiple control cycles to form a benchmark dataset. Calculate the arithmetic mean and standard deviation of all gradient rates of change in the benchmark dataset; The first gradient coefficient and the second gradient coefficient are preset, wherein the first gradient coefficient and the second gradient coefficient are real numbers greater than or equal to zero; The product of the first gradient coefficient and the standard deviation is subtracted from the arithmetic mean, and the resulting product is defined as the lower limit of the stability interval; the product of the second gradient coefficient and the standard deviation is added to the arithmetic mean, and the resulting product is defined as the upper limit of the stability interval. The calculated lower limit of the stability interval and the calculated upper limit of the stability interval together constitute the stability interval; Compare the calculated gradient rate of change with the defined stability interval; If the gradient change rate is greater than the upper limit of the stable interval or less than the lower limit of the stable interval, it is determined that the process has not converged and the operating status of the permanent magnet micro pump station unit has not returned to stability. At this time, the data of the current cycle is used as the new initial state, and the process is returned to start a new round of multi-source data mutual verification diagnosis. If the gradient change rate is within the stable range, it is determined that convergence has been achieved, and the operating status of the permanent magnet micro pump station unit has returned to stability. At this time, the current stratified gradual control command is maintained, and the routine monitoring process continues.

[0027] Example 2

[0028] Please see Figure 2 This invention provides a remote intelligent control system for a permanent magnet micro pump station unit, comprising: Signal synchronous acquisition and filtering module: synchronously acquires vibration signals, stator current signals, and outlet pressure pulsation signals of the permanent magnet micro pump station unit; and separates high-frequency vibration components and current harmonic components through bandpass filtering; Dual-frequency entropy change characteristic calculation module: calculates the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component respectively, and defines the weighted difference between the two as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit; Latent instability state judgment module: Establish a real-time correlation mapping between dual-frequency entropy change characteristic values ​​and outlet pressure pulsation peak values, and determine whether the permanent magnet micro pump station unit has entered a latent instability state; Feedforward compensation control generation module: If it is determined that the system has entered a hidden instability state, the system will estimate the expected value of the outlet pressure pulsation based on the historical trend of the dual-frequency entropy change characteristic value and generate the feedforward compensation control quantity. Feedback correction dynamic control module: Obtain the actual value of the outlet pressure pulsation within the same control cycle, calculate the feedback correction factor by combining it with the expected value of the outlet pressure pulsation, and dynamically correct the feedforward compensation control quantity to form a hierarchical gradual control command. Regulation effectiveness assessment module: After issuing and executing the tiered and gradual regulation instructions, calculate the gradient change rate used to assess the regulation effectiveness.

[0029] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. The focus of each embodiment is on its differences from other embodiments. In particular, the apparatus embodiments are described simply because they are fundamentally based on the method embodiments; relevant details can be found in the descriptions of the method embodiments.

[0030] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0031] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0032] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0033] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0034] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0035] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A remote intelligent control method for a permanent magnet micro pump station unit, characterized in that: S1. Synchronously acquire vibration signals, stator current signals, and outlet pressure pulsation signals of the permanent magnet micro pump station unit; separate high-frequency vibration components and current harmonic components through bandpass filtering; S2. Calculate the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component respectively, and define the weighted difference between the two as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit. S3. Establish a real-time correlation mapping between dual-frequency entropy change characteristic values ​​and outlet pressure pulsation peak values, and determine whether the permanent magnet micro pump station unit has entered a latent instability state; where the outlet pressure pulsation peak value is the peak value extracted from the collected outlet pressure pulsation signal. S4. If it is determined that the state of hidden instability has been entered, the expected value of the outlet pressure pulsation is estimated based on the historical trend of the dual-frequency entropy change characteristic value, and a feedforward compensation control quantity is generated. S5. Obtain the actual value of the outlet pressure pulsation within the same control cycle, calculate the feedback correction factor by combining it with the expected value of the outlet pressure pulsation, and dynamically correct the feedforward compensation control quantity to form a hierarchical gradual control command. S6. After issuing and executing the tiered and gradual control instructions, calculate the gradient change rate used to evaluate the effectiveness of the control.

2. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 1, characterized in that, S1 includes: Set the passband frequency range of the first bandpass filter; The first bandpass filter is used to filter the acquired vibration signal to obtain the high-frequency vibration component; Set the passband frequency range of the second bandpass filter; The acquired stator current signal is filtered using a second bandpass filter to obtain the current harmonic components.

3. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 2, characterized in that, S2 includes: Receive the separated high-frequency vibration components; The high-frequency vibration component is decomposed into wavelet packet to the Mth layer to obtain the signals of each node in the Mth layer. Calculate the signal energy of each node, and calculate the proportion of the signal energy of each node to the total energy of the layer; Based on all the calculated proportions, the energy entropy of the high-frequency vibration components is calculated using the Shannon entropy formula. Receive the separated current harmonic components; A fast Fourier transform is performed on the harmonic component of the current to extract the amplitude of each harmonic except the fundamental wave. Calculate the square of each harmonic amplitude and its ratio to the sum of the squares of all harmonic amplitudes; Based on all the calculated ratios, the distortion entropy of the current harmonic components is calculated using the Shannon entropy formula. The calculated energy entropy and distortion entropy are weighted and the difference is calculated according to the preset weight coefficients to generate dual-frequency entropy change characteristic values.

4. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 1, characterized in that, S3 includes: The multiple dual-frequency entropy change eigenvalues ​​obtained by continuous calculation are arranged in chronological order to form a dual-frequency entropy change eigenvalue sequence. Multiple outlet pressure pulsation peaks extracted synchronously are arranged in the same time order to form an outlet pressure pulsation peak sequence. Within the sliding time window, the dynamic correlation coefficient is obtained by dynamically calculating the sequence of dual-frequency entropy change eigenvalues ​​calculated in real time and the sequence of peak outlet pressure pulsations measured synchronously. Preset association confidence threshold; If the absolute value of the dynamic correlation coefficient calculated within the current sliding time window is lower than the preset correlation confidence threshold, the correlation is determined to be weakened, and multi-source data cross-verification diagnosis is initiated.

5. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 4, characterized in that, Initiating multi-source data cross-verification diagnostics includes: Infrared thermal image data of key components of permanent magnet micro pump station units are obtained, and the state of air gap magnetic flux density of permanent magnets is indirectly estimated based on motor electrical signals. When the infrared thermal image data shows that the local temperature rise rate exceeds the set temperature rise threshold, and at the same time the product of the air gap magnetic flux density fluctuation amplitude extracted from the indirect estimation result and the dual-frequency entropy change characteristic value exceeds the preset mutual verification threshold, the permanent magnet micro pump station unit is determined to have entered a latent instability state; otherwise, the permanent magnet micro pump station unit is determined not to have entered a latent instability state.

6. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 1, characterized in that, S4 includes: The dual-frequency entropy change characteristic values ​​collected in several consecutive control cycles before the current moment are obtained to form a historical change sequence; Based on historical change sequences, the exponential smoothing algorithm is used to calculate the dual-frequency entropy change characteristic trend value for the next control cycle. Based on the historical trend of the dual-frequency entropy change characteristic trend value, the most similar reference pattern segment is matched, and the expected curve of the outlet pressure pulsation is extracted from the reference pattern segment. Calculate the corresponding expected value of the export pressure pulsation based on the time delay of the current moment relative to the starting point of the expected export pressure pulsation curve; Analyze the changing trend of the expected export pressure pulsation curve in the next control cycle to obtain the sign and amplitude of its first derivative; Simultaneously, the coupling degree of the permanent magnet synchronous motor dq axis current is calculated based on the real-time collected motor current signal. By combining the sign of the first derivative with the real-time measured values ​​of amplitude and coupling degree, the motor current vector angle adjustment is calculated as the advance adjustment amount; Calculate the difference between the expected value of the outlet pressure pulsation and the actual value of the outlet pressure pulsation in the current cycle, and multiply the difference by the preset frequency fine-tuning coefficient to obtain the differential component of the inverter output frequency.

7. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 6, characterized in that, The pre-stored historical fault-free operation database contains multiple reference pattern segments consisting of historical dual-frequency entropy change characteristic sequence segments and their subsequent actual outlet pressure pulsation sequence segments.

8. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 1, characterized in that, S5 includes: At the end of the next control cycle, the actual value of the outlet pressure pulsation during that cycle is collected synchronously; Calculate the difference between the actual value of the outlet pressure pulsation and the expected value of the outlet pressure pulsation for the same control cycle; Multiply the calculated difference by the negative feedback gain coefficient to obtain the feedback correction factor; The feedback correction factor is added to the obtained advance adjustment amount to obtain the corrected final current vector angle adjustment command; The feedback correction factor is added to the obtained differential component to obtain the corrected final frequency adjustment command; The final current vector angle adjustment command and the final frequency adjustment command are integrated to form a hierarchical progressive control command.

9. The remote intelligent control method for a permanent magnet micro pump station unit according to claim 1, characterized in that, S6 includes: The first dual-frequency entropy change characteristic value and the first outlet pressure pulsation peak value are obtained and calculated in the current control cycle; wherein, the first outlet pressure pulsation peak value is the peak value extracted from the outlet pressure pulsation signal collected in the current control cycle. The second dual-frequency entropy change characteristic value and the second outlet pressure pulsation peak value are obtained and calculated from the immediately preceding control cycle; wherein, the second outlet pressure pulsation peak value is the peak value extracted from the outlet pressure pulsation signal collected from the previous control cycle. Calculate the absolute difference between the first dual-frequency entropy change eigenvalue and the second dual-frequency entropy change eigenvalue to obtain the change in eigenvalue; calculate the ratio of the change in eigenvalue to the second dual-frequency entropy change eigenvalue to obtain the first relative rate of change. Calculate the absolute difference between the peak value of the first outlet pressure pulsation and the peak value of the second outlet pressure pulsation to obtain the pressure pulsation change; calculate the ratio of the pressure pulsation change to the peak value of the second outlet pressure pulsation to obtain the second relative change rate. Compare the numerical values ​​of the first relative rate of change and the second relative rate of change, and select the one with the largest value as the gradient rate of change.

10. A remote intelligent control system for a permanent magnet micro pump station unit, characterized in that, A remote intelligent control method for a permanent magnet micro pump station unit as described in any one of claims 1-9, the system comprising: Signal synchronous acquisition and filtering module: synchronously acquires vibration signals, stator current signals, and outlet pressure pulsation signals of the permanent magnet micro pump station unit; and separates high-frequency vibration components and current harmonic components through bandpass filtering; Dual-frequency entropy change characteristic calculation module: calculates the energy entropy of the high-frequency vibration component and the distortion entropy of the current harmonic component respectively, and defines the weighted difference between the two as the dual-frequency entropy change characteristic value of the permanent magnet micro pump station unit; Latent instability state judgment module: Establish a real-time correlation mapping between dual-frequency entropy change characteristic values ​​and outlet pressure pulsation peak values, and determine whether the permanent magnet micro pump station unit has entered a latent instability state; Feedforward compensation control generation module: If it is determined that the system has entered a hidden instability state, the system will estimate the expected value of the outlet pressure pulsation based on the historical trend of the dual-frequency entropy change characteristic value and generate the feedforward compensation control quantity. Feedback correction dynamic control module: Obtain the actual value of the outlet pressure pulsation within the same control cycle, calculate the feedback correction factor by combining it with the expected value of the outlet pressure pulsation, and dynamically correct the feedforward compensation control quantity to form a hierarchical gradual control command. Regulation effectiveness assessment module: After issuing and executing the tiered and gradual regulation instructions, calculate the gradient change rate used to assess the regulation effectiveness.

Images