Method and system for monitoring pressure of oil inlet of servo pump set in real time

Abstract

The invention relates to the technical field of data processing, in particular to a real-time monitoring method and system for the pressure of an oil inlet of a servo pump set, and the method comprises the steps: obtaining the operation data of a servo pump, the operation data comprising a real-time pressure signal and a vibration signal of the oil inlet; segmenting the real-time pressure signal and the vibration signal to form a plurality of data windows; constructing a bandwidth adjustment coefficient of each data window based on the frequency of the vibration signal of each data window, obtaining the optimal bandwidth of each data window based on the bandwidth adjustment coefficient, and filtering the real-time pressure signal by using a band elimination filter based on the optimal bandwidth corresponding to each data window; constructing a temperature compensation curve; and correcting the filtered real-time pressure signal based on a temperature compensation curve to obtain an optimal pressure signal. The method has the effect of improving the denoising stability of the real-time pressure signal of the servo pump set.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and in particular to a method and system for real-time monitoring of the inlet pressure of a servo pump unit. Background Art

[0002] A servo pump unit is a component used to provide power in a hydraulic system. It controls the movement of an actuator, such as a hydraulic cylinder or a hoist, by controlling the oil supply. The servo pump unit mainly controls the pressure through the pressure difference between the inlet and outlet ports. The stability of the inlet pressure is directly related to the stable operation of the entire hydraulic system. Therefore, during the operation of the hydraulic system, it is of great significance to monitor the pressure at the inlet of the servo pump unit. The pressure at the inlet is generally collected based on a pressure sensor. Common sensors include: piezoresistive sensors, piezoelectric sensors, strain gauge sensors, etc. Inevitably, some noise interference will be generated during the process of collecting real-time pressure signals by such sensors. Therefore, in order to facilitate subsequent analysis and processing of real-time pressure signals, the real-time pressure signals need to be filtered and denoised after signal collection. Due to the difference in frequency between the noise signal and the normal signal, in the related art, a fixed-bandwidth filtering method is usually used to filter and denoise the originally collected real-time pressure signals, such as a band-stop filter, a notch filter, etc.

[0003] However, the frequency and intensity of the noise generated by the servo pump unit are different under different loads and different oil flow rates. When the working state or working environment of the servo pump unit changes greatly, the frequency and intensity of the noise change, resulting in a poor denoising effect of the method for filtering and denoising real-time pressure signals in the related art. In summary, the traditional filtering and denoising method has an unstable denoising effect and cannot cope with the complex working environment and diverse working states of the servo pump unit. Summary of the Invention

[0004] To solve the problem of unstable denoising effect of the filtering and denoising method in the related art, this application provides a method and system for real-time monitoring of the inlet pressure of a servo pump unit.

[0005] In the first aspect, this application provides a method for real-time monitoring of the inlet pressure of a servo pump unit, adopting the following technical solution: A real-time monitoring method for the inlet pressure of a servo pump set, comprising the steps of: obtaining the operation data of the servo pump, where the operation data includes the real-time pressure signal and vibration signal at the inlet; segmenting the real-time pressure signal and vibration signal to form a plurality of data windows; constructing a bandwidth adjustment coefficient for each data window based on the frequency of the vibration signal in each data window, obtaining the optimal bandwidth for each data window based on the bandwidth adjustment coefficient, and filtering the real-time pressure signal using a band-stop filter based on the optimal bandwidth corresponding to each data window; constructing a temperature compensation curve, and correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal. Among them, the step of constructing the bandwidth adjustment coefficient for each data window includes: obtaining the center frequency of the vibration signal in each data window, setting an adjacent frequency region based on the center frequency, and obtaining the average value of the corresponding amplitudes in the adjacent frequency region; taking the difference between the center frequency and the average amplitude in the adjacent frequency region as the first control factor, taking the difference between the center frequency and the average amplitude corresponding to all frequencies within the window as the second control factor, and taking the ratio of the first control factor to the second control factor as the bandwidth adjustment coefficient.

[0006] The beneficial effects are as follows: obtaining the operation data of the servo pump set to be monitored, dividing the operation data to form a plurality of data windows, determining the optimal bandwidth for each data window based on the vibration signal in each data window, and filtering the real-time pressure signal based on the optimal bandwidth. Different data windows in the signal to be monitored are filtered with different bandwidths, so that the bandwidth can be automatically adjusted in time when the working environment or operating state of the servo pump set changes, ensuring the filtering effect and improving the accuracy of subsequent pressure monitoring of the servo pump set.

[0007] During the calculation of the bandwidth adjustment coefficient, the first control factor is the difference between the center frequency and the average amplitude in the adjacent frequency region, which reflects the amplitude difference between the center frequency and the local part in the data window. The smaller the local amplitude difference, the greater the harmonic intensity carried by the vibration noise, and a larger bandwidth should be used for filtering. The second control factor is the difference between the center frequency and the average amplitude corresponding to all frequencies within the window, which reflects the difference between the center frequency and the amplitudes corresponding to all frequencies in the data window and embodies the global error between the center frequency and the data window. The first control factor and the second control factor jointly adjust the bandwidth adjustment coefficient, and then adaptively adjust the bandwidth. Compared with the traditional method of filtering with a fixed bandwidth, the filtering effect is not affected by the working environment and working state of the servo pump set, and the filtering effect is more stable.

[0008] Optionally, the step of obtaining the optimal bandwidth for each data window based on the bandwidth adjustment coefficient includes: taking the product of the preset initial bandwidth and the bandwidth adjustment coefficient as the adjustment value, and taking the sum of the initial bandwidth and the adjustment value as the optimal bandwidth.

[0009] The beneficial effects are as follows: calculating an adjustment value based on a preset initial bandwidth, where the adjustment value changes based on the dynamic change of a bandwidth adjustment coefficient, and adaptively adjusting the optimal bandwidth.

[0010] Optionally, the steps of constructing a temperature compensation curve and correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain an optimal pressure signal include: acquiring historical temperature signals during the historical operation of the servo pump set and segmenting them to form temperature windows; processing the historical temperature signals in the temperature windows to obtain a temperature change sequence reflecting data changes; classifying the temperature windows to construct a trend set; for each trend set, constructing a polynomial based on the pressure error between the historical pressure signal and the ideal pressure, the historical temperature signal, and the temperature change sequence in each trend set, and fitting the temperature compensation curve of each trend set; correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain an optimal pressure signal.

[0011] The beneficial effects are as follows: acquiring historical temperature signals during the historical operation of the servo pump set, segmenting the historical temperature signals to obtain multiple temperature windows. Processing the historical temperature signals in the temperature windows to obtain a temperature change sequence reflecting temperature changes. After classifying the temperature windows of the historical temperature signals, the change trends of the multiple historical temperature signals corresponding to each trend set are approximately the same, and the temperature values are close. It can also be understood that each trend set corresponds to a temperature change. Constructing a polynomial to fit the temperature compensation curve based on the corresponding historical temperature signal, pressure error, and temperature change sequence in the trend set. Subsequently, different temperature compensation curves can be selected according to the actual temperature change situation to compensate the real-time pressure signal, thereby further improving the accuracy of real-time pressure signal acquisition.

[0012] Optionally, for each temperature window, acquiring a first-order difference sequence of the historical temperature signal in the temperature window, and using the first-order difference sequence of the historical temperature signal as the temperature change sequence.

[0013] The beneficial effects are as follows: The first-order difference operation of the historical temperature signal can reflect the change of temperature at adjacent moments, and thus the change trend of the historical temperature signal in the temperature window can be represented by the first-order difference sequence of the historical temperature signal.

[0014] Optionally, the steps of classifying the temperature windows to construct a trend set include: using K-means to cluster the historical temperature signals to form multiple clustering clusters, and taking each clustering cluster as a trend set.

[0015] The beneficial effects are as follows: Using a clustering method to cluster the temperature windows of the historical temperature signals, so as to classify the historical temperature signals with similar change trends into the same clustering cluster to form a trend set. Each trend set represents a temperature change trend.

[0016] Optionally, the steps of constructing a temperature compensation curve and correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain an optimal pressure signal include: obtaining the stationary points of the historical temperature signal of the servo pump group operation, and taking the data between two adjacent stationary points as a change state interval; processing the historical temperature signal in the temperature window to obtain a temperature change sequence reflecting the data change; constructing a polynomial based on the pressure error, temperature change sequence, and historical temperature signal between the corresponding historical pressure signal and the ideal pressure signal in the change state interval, and fitting the temperature compensation curve; correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal.

[0017] The beneficial effect is that the stationary point, also known as the stable point, refers to the point where the function stops growing or stops decreasing. Therefore, the interval between two adjacent stationary points represents a monotonic change. A state change interval is formed between two stationary points, and each state change interval corresponds to a monotonic temperature change trend. The fitting construction of the temperature compensation interval in each state change interval is completed based on the data in the state change interval.

[0018] Optionally, the steps of compensating the real-time temperature signal based on the temperature compensation curve to obtain an optimal pressure signal include: obtaining the real-time temperature signal of the servo pump group, dividing the real-time temperature signal to form a real-time state interval of the same length as the temperature window, matching the temperature compensation curve based on the Euclidean distance between the real-time state interval and the temperature window in the trend set, and obtaining a pressure correction value based on the matched temperature compensation curve; taking the sum of the real-time pressure signal and the pressure correction value at the corresponding moment as the optimal pressure signal.

[0019] The beneficial effect is that the temperature data of the servo pump group to be monitored is collected in real time, the real-time state space is divided, and a suitable temperature compensation curve is matched according to the Euclidean distance between the temperature data in the real-time state space and the historical temperature signal in the trend set. In the real-time collected data, the temperature signal and the real-time pressure signal are included at the same moment. Substituting the temperature data at a certain moment into the temperature compensation curve to obtain the pressure correction value at that moment, and adding it to the real-time pressure signal at the same moment to obtain the optimal pressure signal.

[0020] Optionally, the steps of compensating the real-time temperature signal based on the temperature compensation curve to obtain an optimal pressure signal include: obtaining the real-time temperature signal of the servo pump group, dividing the real-time temperature signal based on the stationary points in the real-time temperature signal to form a real-time state space, matching the temperature compensation curve based on the DTW distance between the temperature data in the real-time state interval and the corresponding temperature data in the change state interval, and obtaining a pressure correction value based on the matched temperature compensation curve; taking the sum of the real-time pressure signal and the pressure correction value at the corresponding moment as the optimal pressure signal.

[0021] The beneficial effects are as follows: The DTW distance between the real-time temperature signal corresponding to the real-time state interval and the historical temperature signal in the variable state interval represents the similarity between the two. Based on this, it can be determined which temperature compensation curve should be used to compensate the real-time pressure signal.

[0022] Optionally, the steps of compensating the real-time pressure signal in the real-time state interval to obtain the pressure correction value include: for any moment in the real-time collected temperature signal, substituting the temperature data at this moment into the corresponding function of the temperature compensation curve to obtain the pressure correction value. In a second aspect, the present application provides a real-time monitoring system for the inlet pressure of a servo pump group, adopting the following technical solution: A real-time monitoring system for the inlet pressure of a servo pump group includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time monitoring method for the inlet pressure of a servo pump group as described above is implemented.

[0023] The beneficial effects are as follows: Generating a computer program for the real-time monitoring method for the inlet pressure of a servo pump group as described above and storing it in the memory to be loaded and executed by the processor. Thus, a system is made according to the memory and the processor, which is convenient to use.

[0024] The present application has the following technical effects: In the present application, the vibration signal and the real-time pressure signal of the servo pump group to be monitored are segmented to form data windows. Based on the vibration signal in the data window, the optimal bandwidth for noise reduction of the real-time pressure signal in each data window is determined, and different bandwidths of filtering and noise reduction are performed on the real-time pressure signal based on different vibration signals, improving the noise reduction effect on the real-time pressure signal. Moreover, the optimal bandwidth is dynamically adjusted during the process of real-time data acquisition to adapt to different operating states and different working environments of the servo pump group, enabling the filtering effect on the real-time pressure signal to be stable and efficient, and improving the accuracy of subsequent pressure monitoring. Description of the Drawings

[0025] Figure 1 is the flowchart of a real-time monitoring method for the inlet pressure of a servo pump group according to an embodiment of the present application.

[0026] Figure 2 is the construction method of the first temperature compensation curve of a real-time monitoring method for the inlet pressure of a servo pump group according to an embodiment of the present application.

[0027] Figure 3 is the construction method of the second temperature compensation curve of a real-time monitoring method for the inlet pressure of a servo pump group according to an embodiment of the present application. Detailed Embodiments

[0028] The embodiments of the present application disclose a method for real-time monitoring of the inlet pressure of a servo pump group. For the data to be monitored, which can also be understood as the operation data of the servo pump group collected in real time, the operation data is segmented to form multiple data windows. For each data window, its bandwidth adjustment coefficient is calculated, and based on the bandwidth adjustment coefficient, the optimal bandwidth for data filtering in each data window is calculated. After filtering, the temperature compensation curve constructed based on historical data is used to compensate the filtered real-time pressure signal, so as to accurately obtain the pressure information and improve the accuracy of subsequent anomaly detection. During the process of real-time data collection, the bandwidth of filtering for each data window is dynamically adjusted, so as to be able to adapt to the efficient filtering of the servo pump group in different working environments and working states and improve the accuracy of subsequent anomaly detection.

[0029] Referring to Figure 1 , a method for real-time monitoring of the inlet pressure of a servo pump group includes steps S1 - S4.

[0030] S1: Obtain the operation data of the servo pump. The operation data includes the real-time pressure signal and vibration signal at the inlet. Segment the real-time pressure signal and vibration signal to form multiple data windows.

[0031] Obtain various operation data of the servo pump group to be monitored. In this embodiment, based on the influence of the vibration analysis of the servo pump group on the real-time pressure signal, the operation data of the pump group to be monitored obtained in this embodiment includes the real-time pressure signal at the inlet collected by the sensor at the inlet and the inlet vibration signal. The real-time pressure signal can be collected by a piezoresistive sensor; the inlet vibration signal can be collected by a MEMS accelerometer. The collected real-time pressure signal and vibration signal are both stored in the form of a time series. In order to unify the dimension, in this embodiment, the collected real-time pressure signal and vibration signal are both signals after being standardized (normalized) processing.

[0032] During the process of real-time collecting the operation data to be monitored, as the data accumulates continuously, the length of the sequence increases continuously. Segment the operation data to form multiple data windows, and each data window includes the real-time pressure signal and vibration signal.

[0033] S2: Construct the bandwidth adjustment coefficient for each data window based on the frequency of the vibration signal in each data window, and obtain the optimal bandwidth for each data window based on the bandwidth adjustment coefficient.

[0034] Obtain the center frequency of the vibration signal in each data window, and set the adjacent frequency region based on the center frequency.

[0035] For the vibration signal in any one data window, use the Fourier transform method to obtain its amplitude spectrum, and take the frequency corresponding to the maximum amplitude value in the amplitude spectrum as the center frequency of the vibration signal.

[0036] Determining an adjacent frequency region based on a center frequency refers to a frequency range close to the center frequency. An exemplary adjacent frequency region can be centered on the center frequency and extend a certain frequency range on both sides. , thereby obtaining the adjacent frequency region. For example, a center frequency is 20KHz; the extended frequency range is 3KHz; then the frequency range of the adjacent frequency region is 17KHz - 23KHz. In other embodiments, the extended frequency range can be adjusted according to actual situations.

[0037] Obtain the average value of the corresponding amplitudes in the adjacent frequency region; use the difference between the center frequency and the average amplitude of the adjacent frequency region as the first control factor, use the difference between the center frequency and the average amplitude corresponding to all frequencies within the window as the second control factor, and use the ratio of the first control factor to the second control factor as the bandwidth adjustment coefficient.

[0038] Specifically, the calculation process of the bandwidth adjustment coefficient can be expressed as: ; in the formula, represents the bandwidth adjustment coefficient of this data window, represents the amplitude of the center frequency within this data window. represents the average value of the amplitudes in the adjacent frequency region; represents the average value of the amplitudes of all frequencies in the data window.

[0039] represents the difference between the amplitude of the center frequency and the amplitudes in the adjacent frequency region, reflecting the local amplitude difference of the center frequency relative to the data window; the smaller this local amplitude difference is, the greater the harmonic intensity carried by the vibration noise, and a large bandwidth should be used to suppress the noise. represents the difference between the amplitude of the center frequency and the amplitudes corresponding to all frequencies in the data window, reflecting the global amplitude difference relative to the data window. The smaller the global amplitude difference is, the smaller the harmonic intensity carried by the noise, and the smaller the impact on the real-time pressure signal acquisition. A smaller bandwidth should be used to retain the details of the real-time pressure signal.

[0040] S3: Obtain the optimal bandwidth of each data window based on the bandwidth adjustment coefficient.

[0041] The bandwidth adjustment coefficient is proportional to the optimal bandwidth. In this embodiment, the product of the preset initial bandwidth and the bandwidth adjustment coefficient is used as the adjustment value, and the sum of the initial bandwidth and the adjustment value is used as the optimal bandwidth.

[0042] Filter the real-time pressure signal using a band-stop filter based on the optimal bandwidth corresponding to each data window.

[0043] For each data window corresponding to the operating data to be monitored, an optimal bandwidth is included. Based on this optimal bandwidth, a band-stop filter is used to filter and denoise the real-time pressure signal in each window, completing the noise reduction of the real-time pressure signal.

[0044] S4: Construct a temperature compensation curve, and correct the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal.

[0045] At different temperatures, different degrees of compensation are required for the real-time pressure signal. Therefore, in this embodiment, the historical operating data of the servo pump group is obtained for analysis. Here, the historical operating data is mainly used to reference the influence of temperature change on the pressure measured by the sensor. Therefore, the historical operating data can be the data of the servo pump group operating at different temperatures in the past.

[0046] Collect the ideal pressure signal of the servo pump group operating in an environment with a constant temperature in the laboratory. In the laboratory environment, the ideal pressure signal collected by the sensor is not affected by temperature. The difference between the data in the ideal pressure signal and the historical pressure signal represents the measurement error of the sensor. Therefore, subtracting the data at the same ordinal position in the historical pressure signal from the data in the ideal pressure signal can obtain the pressure error, and the pressure error reflects the influence of temperature change on the pressure signal collected by the sensor; multiple pressure errors can form a pressure error sequence. To improve the accuracy of obtaining the historical pressure signal, the historical pressure signal here is the signal after filtering through steps S1 - S3.

[0047] Refer to Figure 2 , in one embodiment, the steps of constructing the temperature compensation curve include: step S41 - step S43.

[0048] S41: Obtain the historical temperature signal of the servo pump group running and segment it to form temperature windows, process the historical temperature signal in the temperature windows, and obtain a temperature change sequence reflecting data changes.

[0049] The historical temperature signal and the historical pressure signal are evenly segmented to form multiple temperature windows. Perform a first-order difference processing on the historical temperature signal in the temperature windows to obtain a temperature change sequence, and the data in the temperature change sequence represents the temperature change at adjacent times.

[0050] S42: Classify the temperature windows to construct a trend set.

[0051] In this embodiment, the K-means clustering method is used to cluster the temperature change sequences corresponding to each temperature window based on the Euclidean distance, obtaining multiple clustering clusters, and each clustering cluster is a trend set. The temperature change sequences in each trend set are similar, indicating that the temperature data changes in the corresponding temperature windows in each clustering cluster are similar, that is, each trend set represents a temperature change trend.

[0052] S43: For each trend set, construct a polynomial based on the pressure error between the historical pressure signal and the ideal pressure, the historical temperature signal, and the temperature change sequence in each trend set, and fit the temperature compensation curve of each trend set.

[0053] Construct a temperature compensation curve based on the corresponding historical temperature signal, temperature change sequence, and pressure error in each clustering cluster.

[0054] Specifically, the polynomial can be expressed as: ; represents the pressure error at the th moment, represents the temperature data at the th moment in the historical temperature signal, represents the data value corresponding to the th moment in the temperature change sequence; is the first parameter to be fitted; is the second parameter to be fitted; c is the third parameter to be fitted.

[0055] In the formula represents the influence of the instantaneous change of temperature on the pressure error, is the parameter to be fitted, and the least squares method is used to fit the polynomial temperature compensation curve.

[0056] Subsequently, obtain the real-time temperature signal of the servo pump group, segment the real-time temperature signal to form real-time state intervals of the same length as the temperature window, match the temperature compensation curve based on the Euclidean distance between the real-time state interval and the temperature window in the trend set, and obtain the pressure correction value based on the matched temperature compensation curve; take the sum of the real-time pressure signal and the pressure correction value at the corresponding moment as the optimal pressure signal.

[0057] For the real-time collected signal, segment it to form multiple real-time state intervals; the Euclidean distance between the data in the real-time state interval and the historical temperature signals in each trend set. Each trend set corresponds to multiple temperature windows, and each temperature window corresponds to a historical temperature signal. Therefore, here, the average value of the Euclidean distances between the real-time state region and multiple historical temperature signals is used as the distance between the real-time state interval and the trend set, and the trend set with the smallest distance is selected as the optimal matching set.

[0058] Substitute the temperature data corresponding to the real-time pressure signal at any moment into the temperature compensation curve corresponding to the optimal matching set to obtain the pressure correction value. Take the sum of the real-time pressure signal and the pressure correction value at the corresponding moment as the optimal pressure signal. Subsequently, the servo pump set can be monitored according to the optimal pressure signal to improve the accuracy of monitoring the servo pump set.

[0059] Refer to Figure 3 , in another embodiment, the steps of constructing the temperature compensation curve include: Step S51 - Step S52.

[0060] S51: Obtain the stationary points of the historical temperature signal during the operation of the servo pump set, and take the data between two adjacent stationary points as a change state interval; process the historical temperature signal in the temperature window to obtain a temperature change sequence reflecting the data change.

[0061] The stationary point refers to the point where the historical temperature signal change is zero, which can be obtained by means of the first-order difference sequence of the historical temperature signal. This method is a conventional technical means in this field and will not be elaborated here.

[0062] The historical temperature signal corresponding to two adjacent stationary points represents a monotonic temperature change trend. Therefore, the historical temperature signal between two stationary points is defined as a state change interval.

[0063] S52: Based on the pressure error between the corresponding historical pressure signal and the ideal pressure signal, the temperature change sequence, and the historical temperature signal in the change state interval, construct a polynomial to fit the temperature compensation curve.

[0064] The steps of constructing the polynomial to fit the temperature compensation curve here are the same as those in S43 and will not be elaborated here.

[0065] The optimal pressure signal is obtained by correcting the filtered real-time pressure signal based on the temperature compensation curve.

[0066] Obtain the real-time temperature signal of the servo pump set, segment the real-time temperature signal based on the stationary points in the real-time temperature signal to form a real-time state space, match the temperature compensation curve based on the DTW distance between the temperature data in the real-time state interval and the corresponding temperature data in the change state interval, and obtain the pressure correction value based on the matched temperature compensation curve.

[0067] Similarly to the acquisition of the stationary points in the historical temperature signal, the stationary points of the historical temperature signal collected in real time are acquired, and a real-time state interval is formed between two stationary points in the historical temperature signal collected in real time; the DTW distance between the real-time state interval and the historical state interval is acquired, and the historical state interval with the minimum DTW distance is used as the optimal matching interval. The temperature information collected in real time is substituted into the temperature compensation curve corresponding to the optimal matching interval to obtain a pressure correction value, and the sum of the real-time pressure signal and the correction value at the corresponding moment is used as the optimal pressure signal.

[0068] Based on the optimal pressure signal, pressure monitoring is performed on the servo pump group, thereby improving the accuracy of the final monitoring of the servo pump group.

[0069] The embodiment of the present application also discloses a real-time inlet pressure monitoring system for a servo pump group, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time inlet pressure monitoring method for a servo pump group according to the present application is implemented.

[0070] The above system further includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their settings and functions are known in the art, so they will not be described in detail here.

[0071] The above are all preferred embodiments of the present application. The protection scope of the present application is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application should be covered within the protection scope of the present application.

Claims

1. A real-time monitoring method for the inlet pressure of a servo pump set, characterized in that, Including the steps of: obtaining servo pump operation data, where the operation data includes the real-time pressure signal and vibration signal at the oil inlet; segmenting the real-time pressure signal and vibration signal to form multiple data windows; constructing a bandwidth adjustment coefficient for each data window based on the frequency of the vibration signal in each data window, obtaining the optimal bandwidth for each data window based on the bandwidth adjustment coefficient, and filtering the real-time pressure signal using a band-stop filter based on the optimal bandwidth corresponding to each data window; constructing a temperature compensation curve, and correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal. Among them, the steps of constructing the bandwidth adjustment coefficient for each data window include: obtaining the center frequency of the vibration signal in each data window, setting an adjacent frequency region based on the center frequency, and obtaining the average value of the corresponding amplitudes in the adjacent frequency region; taking the difference between the center frequency and the average amplitude of the adjacent frequency region as the first control factor, taking the difference between the center frequency and the average amplitude corresponding to all frequencies within the window as the second control factor, and taking the ratio of the first control factor to the second control factor as the bandwidth adjustment coefficient.

2. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 1, characterized in that, The steps of obtaining the optimal bandwidth for each data window based on the bandwidth adjustment coefficient include: taking the product of the preset initial bandwidth and the bandwidth adjustment coefficient as the adjustment value, and taking the sum of the initial bandwidth and the adjustment value as the optimal bandwidth.

3. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 1, characterized in that, The steps of constructing a temperature compensation curve and correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal include: obtaining the historical temperature signal of the operation of the servo pump group and segmenting it to form temperature windows; processing the historical temperature signal in the temperature windows to obtain a temperature change sequence reflecting data changes; classifying the temperature windows to construct a trend set; for each trend set, constructing a polynomial based on the pressure error between the historical pressure signal and the ideal pressure, the historical temperature signal, and the temperature change sequence in each trend set, and fitting the temperature compensation curve of each trend set; correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal.

4. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 3, characterized in that, For each temperature window, obtaining the first-order difference sequence of the historical temperature signal in the temperature window, and taking the first-order difference sequence of the historical temperature signal as the temperature change sequence.

5. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 3, characterized in that, The steps of classifying the temperature windows to construct a trend set include: using K-means to cluster the historical temperature signal to form multiple clustering clusters, and taking each clustering cluster as a trend set.

6. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 1, characterized in that, The steps of constructing a temperature compensation curve and correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal include: obtaining the stationary points of the historical temperature signal of the operation of the servo pump group, and taking the data between two adjacent stationary points as a change state interval; processing the historical temperature signal in the temperature window to obtain a temperature change sequence reflecting data changes; constructing a polynomial based on the pressure error between the corresponding historical pressure signal and the ideal pressure signal, the temperature change sequence, and the historical temperature signal in the change state interval, and fitting the temperature compensation curve; correcting the filtered real-time pressure signal based on the temperature compensation curve to obtain the optimal pressure signal.

7. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 3, characterized in that, The steps of compensating the real-time temperature signal based on the temperature compensation curve to obtain the optimal pressure signal include: acquiring the real-time temperature signal of the servo pump group, segmenting the real-time temperature signal to form real-time state intervals of the same length as the temperature window; matching the temperature compensation curve based on the Euclidean distance between the real-time state interval and the temperature window in the trend set, and obtaining the pressure correction value based on the matched temperature compensation curve; taking the sum of the real-time pressure signal and the pressure correction value at the corresponding moment as the optimal pressure signal.

8. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 6, characterized in that, The steps of compensating the real-time temperature signal based on the temperature compensation curve to obtain the optimal pressure signal include: acquiring the real-time temperature signal of the servo pump group, segmenting the real-time temperature signal based on the stationary points in the real-time temperature signal to form a real-time state space, matching the temperature compensation curve based on the DTW distance between the temperature data in the real-time state interval and the corresponding temperature data in the change state interval, and obtaining the pressure correction value based on the matched temperature compensation curve; taking the sum of the real-time pressure signal and the pressure correction value at the corresponding moment as the optimal pressure signal.

9. The real-time monitoring method for the inlet pressure of a servo pump set according to claim 8, characterized in that, The steps of compensating the real-time pressure signal in the real-time state interval to obtain the pressure correction value include: for any moment in the real-time acquired temperature signal, substituting the temperature data at that moment into the corresponding function of the temperature compensation curve to obtain the pressure correction value.

10. A real-time monitoring system for the inlet pressure of a servo pump set, characterized in that, including: a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time monitoring method for the inlet pressure of a servo pump group according to any one of claims 1-9 is implemented.

Images