Abnormal Detection Method, Device, Electronic Equipment and Storage Medium for Plunger Pump

Abstract

The present application discloses a plunger pump anomaly detection method, device, electronic device and storage medium, which belongs to the field of plunger pump monitoring technology, wherein the plunger pump anomaly detection method includes: obtaining a pressure signal at the plunger pump outlet; dividing the pressure signal according to pressure pulsation to obtain multiple pressure pulsations; calculating a similarity matrix of multiple pressure pulsations, wherein the number of rows and columns of the similarity matrix are equal to the number of pressure pulsations; reconstructing the similarity matrix to obtain a target matrix, wherein the target matrix is the same size as the similarity matrix; calculating the reconstruction error between the target matrix and the similarity matrix; when the reconstruction error exceeds a threshold, determining that the plunger pump is abnormal. This method monitors the similarity of pressure pulsations in the pressure signal of one rotation of the plunger pump, and then detects anomalies of the plunger pump based on the similarity of pressure pulsations. No historical data set containing fault labels is required for model training, and the plunger pump can be effectively monitored in terms of state.

Description

Technical Field

[0001] The present application belongs to the technical field of plunger pump monitoring, and specifically relates to a plunger pump abnormality detection method, device, electronic equipment and storage medium. Background Art

[0002] Due to their high power density, compact structure, and high efficiency, plunger pumps are widely used in industrial fields such as engineering machinery, aerospace, and robotics. As a key component of hydraulic systems, their health directly affects the performance of the hydraulic system.

[0003] In complex construction machinery, accurate and effective online status detection of its components can greatly facilitate the overall maintenance of the hydraulic system, and can greatly reduce downtime for maintenance, arrange mechanical spare parts in advance, reduce the downtime maintenance cost of construction machinery, shorten the construction period, and most importantly, improve the safety and reliability of the hydraulic system.

[0004] In actual engineering applications, it is very difficult to collect a large amount of data containing plunger pump fault labels; and for hydraulic systems, the pressure of the hydraulic system will change with the load, so collecting data containing fault labels under different working conditions is even more difficult to achieve. Summary of the Invention

[0005] The purpose of this application is to provide a plunger pump abnormality detection method, device, electronic equipment and storage medium to solve the problem of complex and difficult plunger pump abnormality detection in the prior art.

[0006] According to a first aspect of an embodiment of the present application, a method for detecting an abnormality in a plunger pump is provided. The method may include:

[0007] Obtain the pressure signal at the outlet of the plunger pump;

[0008] The pressure signal is divided according to the pressure pulsation to obtain multiple pressure pulsations;

[0009] Calculate the similarity matrix of multiple pressure pulsations, where the number of rows and columns of the similarity matrix is equal to the number of pressure pulsations;

[0010] Reconstruct the similarity matrix to obtain the target matrix, which has the same size as the similarity matrix;

[0011] Calculate the reconstruction error between the target matrix and the similarity matrix;

[0012] When the reconstruction error exceeds a threshold, it is determined that the plunger pump is abnormal.

[0013] In some optional embodiments of the present application, the pressure signal is divided according to the pressure pulsation to obtain multiple pressure pulsations, specifically:

[0014] The pressure signal is divided into equal intervals according to the pressure pulsation so that each interval contains one or more complete pressure pulsations.

[0015] In some optional embodiments of the present application, the similarity matrix of multiple pressure pulsations is calculated, specifically:

[0016] The inner product of multiple pressure pulsations is calculated to obtain a similarity matrix.

[0017] In some optional embodiments of the present application, the similarity matrix is reconstructed to obtain a target matrix, specifically:

[0018] The similarity matrix is input into the autoencoder so that the autoencoder outputs a matrix of the same size as the similarity matrix to obtain the target matrix.

[0019] In some optional embodiments of the present application, the reconstruction error between the target matrix and the similarity matrix is calculated as follows:

[0020] The sum of the differences between the target matrix and the elements in the similarity matrix is calculated to obtain the reconstruction error.

[0021] In some optional embodiments of the present application, after calculating the reconstruction error between the target matrix and the similarity matrix, the plunger pump abnormality detection method further includes:

[0022] When the reconstruction error is less than or equal to the threshold, it is determined that the plunger pump is normal.

[0023] According to a second aspect of an embodiment of the present application, a plunger pump abnormality detection device is provided, which may include:

[0024] An acquisition module, used for acquiring a pressure signal at an outlet of the plunger pump;

[0025] A pressure signal division module is used to divide the pressure signal according to the pressure pulsation to obtain multiple pressure pulsations;

[0026] A similarity matrix calculation module is used to calculate the similarity matrix of multiple pressure pulsations, where the number of rows and columns of the similarity matrix is equal to the number of pressure pulsations;

[0027] The reconstruction module is used to reconstruct the similarity matrix to obtain the target matrix, which has the same size as the similarity matrix;

[0028] The reconstruction error calculation module is used to calculate the reconstruction error between the target matrix and the similarity matrix;

[0029] The determination module is used to determine that the plunger pump is abnormal when the reconstruction error exceeds a threshold.

[0030] According to a third aspect of an embodiment of the present application, an electronic device is provided, which may include:

[0031] processor;

[0032] a memory for storing processor-executable instructions;

[0033] The processor is configured to execute instructions to implement the plunger pump abnormality detection method as shown in any one of the embodiments of the first aspect.

[0034] According to the fourth aspect of the embodiments of the present application, a storage medium is provided. When the instructions in the storage medium are executed by the processor of an information processing device or a server, the information processing device or the server implements the plunger pump abnormality detection method as shown in any one of the embodiments of the first aspect.

[0035] The above technical solution of this application has the following beneficial technical effects:

[0036] The embodiment method of the present application obtains the pressure signal at the outlet of the plunger pump, divides the pressure signal according to the pressure pulsation, obtains multiple pressure pulsations, calculates the similarity matrix of the multiple pressure pulsations, the number of rows and columns of the similarity matrix are equal to the number of pressure pulsations, reconstructs the similarity matrix to obtain a target matrix, the target matrix is the same size as the similarity matrix, calculates the reconstruction error between the target matrix and the similarity matrix, and when the reconstruction error exceeds a threshold, determines that the plunger pump is abnormal. This method monitors the similarity of the pressure pulsation in the pressure signal of one rotation of the plunger pump, and then detects the abnormality of the plunger pump based on the similarity of the pressure pulsation. It does not require a historical data set containing fault labels for model training, and can effectively monitor the state of the plunger pump. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flow chart of a plunger pump abnormality detection method in an exemplary embodiment of the present application;

[0038] Figure 2 is the pressure pulsation of the plunger pump outlet pressure in a healthy state in an exemplary embodiment of the present application;

[0039] Figure 3 is a similarity matrix obtained by calculating the inner product of each pressure pulsation in an exemplary embodiment of the present application;

[0040] Figure 4 is the similarity matrix output by the autoencoder in an exemplary embodiment of the present application;

[0041] Figure 5 is the reconstruction error and threshold of the autoencoder in a healthy state in an exemplary embodiment of the present application;

[0042] Figure 6 is the reconstruction error and threshold of the autoencoder under a circumferential scratch fault in an exemplary embodiment of the present application;

[0043] Figure 7 is a schematic structural diagram of an electronic device in an exemplary embodiment of the present application;

[0044] Figure 8 It is a schematic diagram of the hardware structure of an electronic device in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0045] To make the objectives, technical solutions, and advantages of this application more clearly understood, this application is further described below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are merely illustrative and are not intended to limit the scope of this application. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion in the concepts of this application.

[0046] The accompanying drawings illustrate schematic diagrams of layer structures according to embodiments of the present application. These figures are not drawn to scale; for clarity, some details are exaggerated and some details may be omitted. The shapes of the various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in practice due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art may design regions / layers with different shapes, sizes, and relative positions as needed.

[0047] Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0048] In the description of this application, it should be noted that the terms "first", "second" and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance.

[0049] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0050] Research has found that due to the structure of the plunger pump (limited number of plungers, flow backflow, etc.), the outlet flow rate of the plunger pump is not a constant value, but will pulsate over time. After the pipeline impedance conversion, the outlet pressure of the plunger pump is also a pulsating time series, that is, a pseudo-periodic time series. In a healthy state, the health status of each component (for example, the plunger) is similar, so the pulsating waveforms of each outlet pressure of the plunger pump are also similar. However, when the plunger pump fails (for example, the friction pair of the plunger pump leaks), the flow pulsation will change. After the pipeline impedance conversion, the pressure signal will also change. Due to the change in the health status of some parts, the similarity of the various pressure pulsations generated by the plunger pump during one rotation will also change. In this way, the plunger pump can be detected for abnormalities based on this change.

[0051] The plunger pump abnormality detection method provided in the embodiment of the present application is described in detail below through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0052] like Figure 1 As shown, in a first aspect of an embodiment of the present application, a plunger pump abnormality detection method is provided, which may include:

[0053] S110: Acquire a pressure signal at the outlet of the plunger pump;

[0054] S120: Dividing the pressure signal according to the pressure pulsation to obtain multiple pressure pulsations;

[0055] S130: Calculating a similarity matrix of the plurality of pressure pulsations, where the number of rows and columns of the similarity matrix are equal to the number of pressure pulsations;

[0056] S140: Reconstruct the similarity matrix to obtain a target matrix, where the target matrix has the same size as the similarity matrix;

[0057] S150: Calculating the reconstruction error between the target matrix and the similarity matrix;

[0058] S160: When the reconstruction error exceeds a threshold, it is determined that the plunger pump is abnormal.

[0059] The above embodiment method monitors the similarity of pressure pulsations in the pressure signal of the plunger pump during one rotation, and then detects anomalies of the plunger pump based on the similarity of the pressure pulsations. No historical data set containing fault labels is required for model training, and the condition of the plunger pump can be effectively monitored.

[0060] For a clearer explanation, the following steps are introduced separately:

[0061] First, step S110: obtaining a pressure signal at the outlet of the plunger pump.

[0062] In this step, a high-frequency response pressure sensor can be installed at the outlet of the plunger pump to collect the pressure signal at the outlet of the plunger pump. The location of the collected pressure signal can be 2-3 times the pipe diameter away from the outlet of the plunger pump. After research, it is found that the pressure signal here can best reflect the health status of the plunger pump. The kinematic frequency of the plunger pump is calculated as follows:

[0063]

[0064] Where n is the plunger pump speed (unit: rpm), k is the number of plungers in the plunger pump. Preferably, a signal with a frequency 10 times the kinematic frequency of the plunger pump is collected. Considering the sampling theorem (the collection frequency is preferably more than twice the highest frequency in the signal), the sampling frequency of the plunger pump outlet pressure signal is set to more than 20 times the kinematic frequency of the plunger pump.

[0065]

[0066] Where FS is the sampling frequency.

[0067] According to the sampling frequency FS, the speed of the plunger pump, and the number of plungers in the plunger pump, it can be calculated that at the sampling frequency FS, corresponding to a kinematic frequency f, the approximate number of data points of the plunger pump outlet pressure signal segment is L1:

[0068]

[0069] The approximate number of outlet pressure points per one rotation of the plunger pump is L=L1*k

[0070]

[0071] In a healthy state, the states of the various parts in the plunger pump are similar. Since the plunger pump has a symmetrical structure, the pressure signal at the outlet is a pseudo-periodic signal, and the length of the period is approximately L1. Therefore, the time series subsequence of length L1 is called a pressure pulsation of the plunger pump outlet pressure. When the plunger pump is in a healthy state, there should be k pressure pulsations with similar waveforms in the time series signal of length L. The pressure pulsation of the plunger pump outlet pressure in a healthy state is as follows: Figure 2 shown.

[0072] Next is step S120 , dividing the pressure signal according to the pressure pulsation to obtain multiple pressure pulsations.

[0073] This step divides the collected pressure signal into equally spaced intervals according to the pressure pulsation, so that each interval contains a complete pressure pulsation. After the division, k time series with a length of L1 can be obtained. In this step, each interval can also contain multiple pressure pulsations.

[0074] Next is step S130: calculating a similarity matrix of the multiple pressure pulsations, where the number of rows and columns of the similarity matrix are both equal to the number of pressure pulsations.

[0075] In this step, the inner product of each time series with a length of L1 after division can be calculated to obtain a similarity matrix M. Each element in the similarity matrix M is m ij ∈M, its calculation formula is as follows:

[0076]

[0077] in, To represent the δth pressure data in the ith time series, i = j = 1, 2, ..., k, δ = 1, 2, ..., L1, K is the reduction coefficient used to normalize each element in the similarity matrix M, which is equal to the length L1 of the pulsating pressure. The obtained similarity matrix is as follows Figure 3 The number of rows and columns of the similarity matrix are equal to the number of pressure pulsations, that is, the number of plungers k in the plunger pump.

[0078] In this step, in addition to using the inner product method to calculate the similarity, the absolute value of the difference can also be used for calculation, that is, the pressure signal data at the same position in the two sequences are subtracted, the absolute value is taken, and then the sum is added.

[0079] Next is step S140: reconstructing the similarity matrix to obtain a target matrix, which has the same size as the similarity matrix.

[0080] In this step, the similarity matrix M obtained after the pressure signal processing mentioned above can be input into the autoencoder, and the autoencoder will output a matrix M′ of the same size, such as Figure 4 As shown. In addition, a variational autoencoder can also be used for reconstruction. The variational autoencoder is a derivative of the autoencoder. The internal network structure of the two is different. At the same time, the similarity matrix is input to it, and then a matrix of the same size is output, which is the reconstruction.

[0081] Next is step S150: calculating the reconstruction error between the target matrix and the similarity matrix.

[0082] This step can calculate the sum of the differences between the elements in the similarity matrix of the autoencoder input and output to obtain the reconstruction error of the similarity matrix:

[0083]

[0084] Among them, m′ ij ∈M′ is the element in the new matrix output by the autoencoder after inputting the similarity matrix M, and S is the normalization coefficient.

[0085] In addition, in order to compare the reconstruction error between the similarity matrix input to the autoencoder and the similarity matrix output by the autoencoder, the Frobenius norm can also be used. The Frobenius norm, abbreviated as F-norm, is a matrix norm, denoted as ||·|| F The Frobenius norm of a matrix A is defined as the root sum of the squares of the absolute values of the elements in the matrix A, that is,

[0086] A=[a ij ] m×n

[0087]

[0088] It should be noted that the calculation of reconstruction error is similar for the case where an interval contains multiple pressure pulsations. The pressure signal generated by one rotation of a plunger pump contains multiple pressure pulsations, and the number of pressure pulsations is equal to the number of plungers in the plunger pump. By dividing the pressure signal of one rotation equally according to the pressure pulsations, multiple pressure signal sequences can be obtained. These signal sequences are then inner-producted (multiplied and added together) to obtain a sum, which becomes an element in the similarity matrix. The resulting similarity matrix has the same number of rows and columns as the number of pressure pulsations. Both reconstruction and error refer to the similarity matrix. Inputting a similarity matrix into a trained autoencoder produces a matrix of the same size. This process is called reconstruction. The error between the input and output matrices is calculated by subtracting the elements at the same position in each matrix, taking the absolute value, and then summing the absolute values of all the subtracted elements to obtain a sum, which is the error.

[0089] Finally, step S160: when the reconstruction error exceeds the threshold, it is determined that the plunger pump is abnormal.

[0090] This step determines whether the reconstruction error of the similarity matrix is greater than the threshold. The threshold calculation process is as follows:

[0091] The healthy plunger pump pressure signal is divided into k time series of length L1 according to the aforementioned process. The inner products of these time series are then taken to form a similarity matrix. This matrix is then batch-fed into the autoencoder to train its reconstruction capabilities. After training, a portion of the similarity matrices not used in autoencoder training is used for testing. The sum of the differences between the elements in the similarity matrices at the autoencoder input and output is calculated as the reconstruction error. The mean of all reconstruction errors across all tests is then taken as the threshold for determining whether the plunger pump is abnormal. Figure 5This figure shows the results of testing the autoencoder after training it using the pressure signal of a healthy plunger pump. The red dashed line indicates the threshold for abnormality determination. This shows that the autoencoder can make relatively accurate judgments about healthy plunger pumps.

[0092] When the reconstruction error of the similarity matrix is greater than a threshold, the plunger pump is considered abnormal; otherwise, the plunger pump is considered normal. The outlet pressure signal of each rotation of the plunger pump contains the status information of the plunger pump for one cycle, so every L data points can determine the health information of the plunger pump. Figure 6 An example of the reconstruction error and threshold of the autoencoder for a plunger pump with a circumferential scratch fault is given.

[0093] from Figure 5 and Figure 6 It can be seen that the reconstruction error of the autoencoder is small in the healthy state, while the circumferential scratch fault affects the waveform of the pressure pulsation at the outlet of the plunger pump, resulting in a decrease in the similarity of the pressure pulsation time series within one rotation. Therefore, the proposed health factor based on the similarity of pressure pulsation can well divide the health state of the plunger pump. By setting a threshold, the function of anomaly detection can be easily realized.

[0094] In some embodiments, the pressure signal is divided according to the pressure pulsation to obtain multiple pressure pulsations, specifically:

[0095] The pressure signal is divided into equal intervals according to the pressure pulsation so that each interval contains one or more complete pressure pulsations.

[0096] In some embodiments, the similarity matrix of the plurality of pressure pulsations is calculated as follows:

[0097] The inner product of multiple pressure pulsations is calculated to obtain a similarity matrix.

[0098] In some embodiments, the similarity matrix is reconstructed to obtain a target matrix, specifically:

[0099] The similarity matrix is input into the autoencoder so that the autoencoder outputs a matrix of the same size as the similarity matrix to obtain the target matrix.

[0100] In some embodiments, the reconstruction error between the target matrix and the similarity matrix is calculated as follows:

[0101] The sum of the differences between the target matrix and the elements in the similarity matrix is calculated to obtain the reconstruction error.

[0102] In some embodiments, after calculating the reconstruction error between the target matrix and the similarity matrix, the plunger pump abnormality detection method further includes:

[0103] When the reconstruction error is less than or equal to the threshold, it is determined that the plunger pump is normal.

[0104] The high-frequency response pressure sensor in the above embodiment may include:

[0105] The electrodes include a cathode and an anode, and are used to generate plasma between the cathode and the anode under the action of a high-voltage DC power supply;

[0106] A support structure for supporting and fixing the electrodes to ensure a stable gap between the cathode and anode; and

[0107] The shielding cover is used to shield the plasma from the external gas, so that the plasma is not affected by the external air flow speed and only senses the changes in the external air pressure.

[0108] The high-frequency response pressure sensor also includes an external circuit for providing a DC voltage to the electrode and measuring and limiting the current in the circuit where the electrode is located. The external circuit includes: a current sampling resistor, connected in series with the electrode, for measuring the current in the external circuit where the electrode is located; a current limiting resistor, connected in series with the electrode, for limiting the magnitude of the current in the external circuit where the electrode is located; and a DC power supply, for providing voltage to the electrode, the current sampling resistor and the current limiting resistor.

[0109] The high-frequency response pressure sensor also includes an acquisition system for measuring and processing current and voltage data of the external circuit and the voltage sensor, including: a high-voltage probe for measuring the voltage across the electrode and the voltage across the current sampling resistor; an oscilloscope for receiving and processing the voltage and current data measured by the high-voltage probe; and / or a PXI acquisition system for receiving and processing voltage data attenuated 1000 times by the high-voltage probe.

[0110] The maximum voltage amplitude of the DC power supply is 10kV; when the pressure sensor is working, the voltage range across the electrodes is maintained at 300-450V; the current range of the external circuit is 1-10mA; and the power range of the pressure sensor is 0.3-4.5W.

[0111] The shielding cover is arranged above the supporting structure, and the bottom of the shielding cover contacts the supporting structure, covering the electrode inside the shielding cover; a small hole is opened on the top of the shielding cover to enable the plasma to sense changes in external air pressure.

[0112] Among them, the support structure includes: electrode support columns, including anode support columns and cathode support columns, which are respectively connected to the cathode electrode and anode electrode, and are used to support and fix the anode and cathode; a non-metallic support base, which is arranged under the electrode support column and is used to fix the electrode support column to ensure a stable gap between the anode and cathode.

[0113] The non-metallic support base and the shielding cover are made of non-metallic materials; the shielding cover is arranged above the non-metallic support base, and the bottom of the shielding cover is in contact with the non-metallic support base, covering the electrode and the electrode support column inside the shielding cover; the shielding cover and the non-metallic support base are separately processed and then assembled or formed as one piece.

[0114] The gap between the cathode and the anode is determined by the measured pressure range; the measured pressure range is a large gap range selected within the low pressure range, the low pressure range is an absolute pressure range of 0.03 to 0.1 MPa, and the large gap range is 200 to 300 μm; the measured pressure range is a small gap range selected within the high pressure range, the high pressure range is an absolute pressure range of 0.1 to 0.5 MPa, and the small gap range is 10 to 100 μm.

[0115] The electrode material is a rare metal or a corrosion-resistant and high-temperature-resistant alloy metal; and / or the electrode structure is cylindrical, semicircular, square, flat, symmetrical or asymmetrical; the electrode diameter ranges from millimeters to centimeters.

[0116] Before the high-frequency response pressure sensor is applied to the compressor to measure pressure, the following steps are also included: calibrating the voltage-pressure correlation of the high-frequency response pressure sensor, and the calibration includes static calibration and shock tube dynamic calibration.

[0117] This high-frequency response pressure sensor improves the sensor structure, shields the airflow velocity, studies the steady-state response and dynamic response characteristics of DC-driven glow discharge plasma to air pressure, and avoids the decoupling problem of airflow velocity and pressure.

[0118] It should be noted that the plunger pump anomaly detection method provided in the embodiments of the present application can be executed by a plunger pump anomaly detection device, or a control module in the plunger pump anomaly detection device for executing the plunger pump anomaly detection method. In the embodiments of the present application, the plunger pump anomaly detection method executed by the plunger pump anomaly detection device is used as an example to illustrate the plunger pump anomaly detection device provided in the embodiments of the present application.

[0119] In a second aspect of an embodiment of the present application, a plunger pump abnormality detection device is provided, which may include:

[0120] An acquisition module, used for acquiring a pressure signal at an outlet of the plunger pump;

[0121] A pressure signal division module is used to divide the pressure signal according to the pressure pulsation to obtain multiple pressure pulsations;

[0122] A similarity matrix calculation module is used to calculate the similarity matrix of multiple pressure pulsations, where the number of rows and columns of the similarity matrix is equal to the number of pressure pulsations;

[0123] The reconstruction module is used to reconstruct the similarity matrix to obtain the target matrix, which has the same size as the similarity matrix;

[0124] The reconstruction error calculation module is used to calculate the reconstruction error between the target matrix and the similarity matrix;

[0125] The determination module is used to determine that the plunger pump is abnormal when the reconstruction error exceeds a threshold.

[0126] The plunger pump anomaly detection device in the embodiment of the present application can be a device, or it can be a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0127] The plunger pump anomaly detection device in the embodiment of the present application can be a device having an operating system. The operating system can be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0128] The plunger pump abnormality detection device provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented in the method embodiment are not described here.

[0129] Alternatively, as Figure 7 As shown, an embodiment of the present application also provides an electronic device 700, including a processor 701, a memory 702, and a program or instruction stored in the memory 702 and executable on the processor 701. When the program or instruction is executed by the processor 701, each process of the above-mentioned plunger pump abnormality detection method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0130] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0131] Figure 8A schematic diagram of the hardware structure of an electronic device implementing an embodiment of the present application.

[0132] The electronic device 800 includes but is not limited to components such as a radio frequency unit 801 , a network module 802 , an audio output unit 803 , an input unit 804 , a sensor 805 , a display unit 806 , a user input unit 807 , an interface unit 808 , a memory 809 , and a processor 810 .

[0133] Those skilled in the art will understand that the electronic device 800 may also include a power source (such as a battery) to power each component, and the power source may be logically connected to the processor 810 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system. Figure 8 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be repeated here.

[0134] It should be understood that in the embodiment of the present application, the input unit 804 may include a graphics processing unit (GPU) 8041 and a microphone 8042. The graphics processor 8041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 806 may include a display panel 8061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit 807 includes a touch panel 8071 and other input devices 8072. The touch panel 8071 is also called a touch screen. The touch panel 8071 may include two parts: a touch detection device and a touch controller. Other input devices 8072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, power keys, etc.), a trackball, a mouse, and a joystick, which will not be described in detail here. The memory 809 can be used to store software programs and various data, including but not limited to applications and operating systems. The processor 810 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and applications, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 810.

[0135] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned plunger pump abnormality detection method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0136] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk.

[0137] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0138] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0139] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0140] In summary, the present application relates to a plunger pump abnormality detection method based on pressure pulsation similarity. The specific process is to collect the pressure signal at the outlet of the plunger pump, and then perform similarity analysis on the pressure pulsation generated by one rotation of the plunger pump in the signal. The number of pressure pulsations during the rotation of the plunger pump is equal to the number of plungers of the plunger pump. The pressure signal of the plunger pump rotation is divided into equal intervals according to the pressure pulsation, and then the inner product of each pressure pulsation is performed to obtain a similarity matrix. Each element in the matrix is the inner product between two pressure pulsations. The pressure signal of the plunger pump in a healthy state is collected, and the obtained similarity matrix is sent to the autoencoder for training, so that the autoencoder learns the characteristics of the element distribution in the similarity matrix, and then the plunger pump pressure signal collected in real time is used. Perform corresponding processing to obtain a similarity matrix, send the matrix to the autoencoder, output a matrix of the same size, and judge whether the plunger pump is abnormal based on whether the sum of the difference values of each element in the autoencoder input and output matrix is greater than the threshold. This method effectively solves the problem that the current plunger pump anomaly detection algorithm requires a large number of fault history data sets, but it is difficult to collect a large number of data sets containing fault labels in actual work, and the pressure of the hydraulic system will change with the load. It is almost difficult to collect data sets containing fault labels under different loads. Therefore, this application proposes a plunger pump anomaly detection algorithm based on the similarity of pressure pulsation. There is no need for historical data sets containing fault labels for model training, and the plunger pump can be effectively monitored.

[0141] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A plunger pump abnormality detection method, characterized in that: include: Obtain the pressure signal at the outlet of the plunger pump; The pressure signal is divided according to the pressure pulsation to obtain a plurality of pressure pulsations; each of the pressure pulsations is a time sequence with a length of L1; Calculating a similarity matrix of the plurality of pressure pulsations, wherein the number of rows and the number of columns of the similarity matrix are both equal to the number of the pressure pulsations; The calculation of the similarity matrix of the plurality of pressure pulsations includes: taking the inner product of each time series of length L1 to obtain a similarity matrix M, wherein each element in the similarity matrix M is m ij ∈M, its calculation formula is as follows: in, represents the δth pressure data in the ith time series, i = j = 1, 2, ..., k, δ = 1, 2, ...,, L1, K is the reduction coefficient; Reconstructing the similarity matrix to obtain a target matrix, wherein the target matrix has the same size as the similarity matrix; Calculating the reconstruction error between the target matrix and the similarity matrix; When the reconstruction error exceeds a threshold, it is determined that an abnormality occurs in the plunger pump.

2. The plunger pump abnormality detection method according to claim 1, characterized in that: The pressure signal is divided according to the pressure pulsation to obtain multiple pressure pulsations, specifically: The pressure signal is divided into equal intervals according to the pressure pulsation, so that each interval contains one or more complete pressure pulsations.

3. The plunger pump abnormality detection method according to claim 2, characterized in that: The similarity matrix of the plurality of pressure pulsations is calculated as follows: The inner product of the plurality of pressure pulsations is calculated to obtain the similarity matrix.

4. The plunger pump abnormality detection method according to claim 1, characterized in that: The similarity matrix is reconstructed to obtain a target matrix, specifically: The similarity matrix is input into an autoencoder so that the autoencoder outputs a matrix of the same size as the similarity matrix to obtain the target matrix.

5. The plunger pump abnormality detection method according to claim 1, characterized in that: The reconstruction error between the calculated target matrix and the similarity matrix is specifically: The sum of the differences between the target matrix and the elements in the similarity matrix is calculated to obtain a reconstruction error.

6. The plunger pump abnormality detection method according to claim 1, characterized in that: After calculating the reconstruction error between the target matrix and the similarity matrix, the plunger pump abnormality detection method further includes: When the reconstruction error is less than or equal to the threshold, it is determined that the plunger pump is normal.

7. A plunger pump abnormality detection device, characterized in that: include: An acquisition module, used for acquiring a pressure signal at an outlet of the plunger pump; A pressure signal division module, configured to divide the pressure signal according to pressure pulsation to obtain a plurality of pressure pulsations; each of the pressure pulsations is a time sequence with a length of L1; A similarity matrix calculation module, configured to calculate a similarity matrix of the plurality of pressure pulsations, wherein the number of rows and the number of columns of the similarity matrix are both equal to the number of the pressure pulsations; The calculation of the similarity matrix of the plurality of pressure pulsations includes: taking the inner product of each time series of length L1 to obtain a similarity matrix M, wherein each element in the similarity matrix M is m ij ∈M, its calculation formula is as follows: in, represents the δth pressure data in the ith time series, i = j = 1, 2, ..., k, δ = 1, 2, ...,, L1, K is the reduction coefficient; A reconstruction module, configured to reconstruct the similarity matrix to obtain a target matrix, wherein the target matrix has the same size as the similarity matrix; A reconstruction error calculation module, used to calculate the reconstruction error between the target matrix and the similarity matrix; The determination module is configured to determine that an abnormality occurs in the plunger pump when the reconstruction error exceeds a threshold.

8. An electronic device, characterized in that: include: A processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the plunger pump abnormality detection method according to any one of claims 1 to 6.

9. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the plunger pump abnormality detection method according to any one of claims 1 to 6 are implemented.

Images