Online monitoring method and system for cavitation damage of aviation plunger pump valve plate based on terahertz spectrum technology

Abstract

The invention relates to an aviation plunger pump valve plate cavitation damage online monitoring method and system based on the terahertz spectrum technology, and the method comprises the following steps: S1, obtaining terahertz spectrum sample data at different positions of a valve plate through a terahertz detection module; step S2, based on the terahertz spectrum sample data in the step S1, carrying out cavitation defect thickness inversion calculation through a data acquisition and processing module, dividing a cavitation crack threshold value, and carrying out cavitation damage state division; and S3, visualizing the processed terahertz spectrum data, monitoring the dynamic change of the cavitation state on the surface of the plunger pump valve plate, and carrying out timely early warning and maintenance feedback on the cavitation defect state. The cavitation damage of the plunger pump valve plate is monitored in real time through the non-contact and high-precision terahertz spectrum technology, the operation reliability and safety of the plunger pump are improved, the service life of the plunger pump is prolonged, and the requirements of aviation equipment for high-precision and real-time monitoring are met.

Description

Technical Field

[0001] The present invention relates to the field of aviation plunger pump technology, and specifically to an online monitoring method and system for cavitation damage of an aviation plunger pump valve plate based on terahertz spectroscopy technology. The present invention uses non-contact, high-precision terahertz spectroscopy technology to monitor the cavitation damage of a plunger pump valve plate in real time, aiming to improve the reliability and safety of the plunger pump operation, extend its service life, and meet the needs of aviation equipment for high-precision, real-time monitoring. Background Art

[0002] The plunger pump is an important component of the hydraulic system of an aircraft engine. Its function is to extract hydraulic oil from the oil tank and pressurize it to supply various actuators in the system. Due to the complex working environment, high temperature and high pressure, the plunger pump will experience cavitation during long-term use. The so-called cavitation is the bursting of bubbles in the liquid under pressure changes, generating shock waves and damaging the pump body, which may cause the pump body to fail in severe cases. The valve plate of the plunger pump is a key fluid distribution component in the pump, and cavitation damage has a significant impact on it. When bubbles appear in the low-pressure area where the valve plate is located, these bubbles will collapse rapidly in the high-pressure area, generating shock waves and high-temperature local environments, causing severe mechanical wear and corrosion on the surface of the valve plate, forming pits and cracks. Long-term cavitation will not only lead to unstable flow and reduced efficiency, but may also accelerate the fatigue rupture of the valve plate, ultimately affecting the performance and reliability of the entire pump. Therefore, preventing cavitation is crucial to extending the service life of the plunger pump.

[0003] At present, cavitation monitoring usually relies on vibration sensors, pressure sensors and other equipment, which cannot detect the occurrence and damage of cavitation in real time without interfering with the normal operation of the pump. Existing monitoring methods often have problems such as low sensitivity, long response time, and inability to accurately locate the cavitation source. Therefore, a new high-precision, real-time, non-contact monitoring method is urgently needed.

[0004] It should be noted that the information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as an admission or suggestion in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The purpose of the present invention is to provide an online monitoring method and system for cavitation damage of the valve plate of an aviation plunger pump based on terahertz spectroscopy technology, so as to solve the problems of low monitoring accuracy, long response time and inability to locate cavitation damage in real time in the prior art. The present invention uses terahertz spectroscopy technology to achieve accurate detection of cavitation damage, and can perform online monitoring when the pump is operating normally, thereby improving the reliability and accuracy of the plunger pump.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides an online monitoring method for cavitation damage of a valve plate of an aviation plunger pump based on terahertz spectroscopy technology, comprising the following steps:

[0008] Step S1, terahertz spectrum data detection: obtaining terahertz spectrum data at different positions of the distribution plate through a terahertz spectrum detection module;

[0009] Step S2: Cavitation defect prediction: Establish a physical model of the interaction between terahertz spectrum data and cavitation defect data, and use the physical model to predict and output the corresponding cavitation defect data based on the terahertz spectrum data obtained in S1;

[0010] Step S3: Cavitation defect visualization: visualize the cavitation defect data predicted by S2, monitor the dynamic changes of the cavitation state on the surface of the piston pump distribution plate, and provide timely warning and maintenance feedback on the cavitation defect state.

[0011] Furthermore, the step S2 specifically includes:

[0012] S21, obtaining spectral sample data of the distribution plate: using a terahertz spectrum detection module to obtain terahertz spectrum sample data at different positions of the distribution plate;

[0013] S22, cavitation defect thickness inversion and threshold determination: based on the terahertz spectrum sample data obtained in S21, the cavitation defect thickness of the distribution plate is inverted and calculated, and the threshold of the cavitation defect thickness is divided;

[0014] S23, establishment of a cavitation defect prediction model: establishing a physical model of the interaction between terahertz spectrum data and cavitation defect thickness, using the terahertz spectrum sample data obtained in S21 and the cavitation defect thickness data calculated in S22 to train the physical model, and subsequently using the trained prediction model to predict and output the cavitation defect thickness based on the detected terahertz spectrum data;

[0015] S24, determination of the degree of cavitation damage: for the predicted cavitation defect thickness data, the degree of cavitation damage of the distribution plate is determined based on the cavitation defect thickness threshold determined in S22.

[0016] Furthermore, in step S22, the inverse calculation method for calculating the cavitation defect thickness using the terahertz spectrum sample data is as follows:

[0017] When there is a cavitation defect in the metal material of the distributor plate, the total time delay of the reflected signal is recorded as Δt total , Δt total is the time delay Δt of the metal material of the valve plate itself materialThe time delay Δt caused by cavitation defects defect sum:

[0018] Δt total =Δt material +Δt defect

[0019]

[0020] In the above formula, d material is the thickness of the metal material of the valve plate, d defect is the thickness of the cavitation defect to be measured, n defect is the refractive index of the terahertz spectrum of the cavitation defect area, n is the refractive index of the terahertz spectrum of the metal material, and c is the speed of light.

[0021]

[0022]

[0023] Furthermore, in step S23, a convolutional neural network structure is selected to establish a prediction model for cavitation defect thickness, the terahertz spectrum sample data of step S21 is divided into a training set and a test set, the training set is input into the neural network, and the loss function value is minimized through multiple trainings, and then the test set is input into the neural network to tune the hyperparameters of the neural network.

[0024] Furthermore, the activation function of the convolutional neural network structure adopts ReLU, and the loss function adopts mean square error. The training set is input into the neural network, and forward propagation and back propagation are performed. Through multiple trainings, the weights and biases are gradually adjusted to minimize the loss function value. The hyperparameters of the network are tuned through cross-validation, and the optimization is stopped when the root mean square error is <2%.

[0025] Furthermore, in step S22, the threshold of the cavitation defect thickness is determined according to the tolerance of the material to cavitation, and the threshold of the cavitation defect thickness includes the cavitation initial threshold d cav1 and cavitation damage threshold d cav2 , when the thickness of the cavitation defect is 0 <d defect <d cav1 When the measured cavitation defect thickness d cav1 <d defect <d cav2 When d cav2 <d defect When , it is determined to be the cavitation fatigue threshold.

[0026] In the second aspect, the present invention provides an online monitoring system for cavitation damage of an aviation plunger pump distribution plate based on terahertz spectroscopy, comprising a terahertz spectrum detection module, a cavitation defect prediction module and a visualization module. The terahertz spectrum detection module is used to obtain terahertz spectrum sample data at different positions of the distribution plate, the data acquisition and processing module predicts the degree of cavitation damage based on the obtained terahertz spectrum data, and the visualization module is used to visualize the cavitation defect data and issue an early warning.

[0027] Furthermore, the terahertz spectrum detection module includes N measuring points arranged on the outer wall of the plunger pump corresponding to the outer area of ​​the oil distribution plate, the N measuring points are distributed in a regular polygon, and a detection window is opened on the outer wall of the pump body corresponding to each measuring point position, each detection window is arranged with a rotating part and a terahertz probe integrated module, each of the terahertz probe integrated modules integrates multiple terahertz optical probes, and the rotating part drives the terahertz probe integrated module to rotate, thereby realizing real-time monitoring of the entire area of ​​the plunger pump distribution plate without blind spots.

[0028] Furthermore, through the coordinated rotation of the plurality of terahertz optical probes, the monitoring range of the terahertz spectrum detection module can be calculated in the following manner:

[0029] Among them: A regular N-gon is composed of N triangles of equal area, and the base l and height h of the triangle can be expressed as:

[0030]

[0031] Therefore, the monitoring area S defect It can be calculated by the following formula:

[0032]

[0033] S p =πr p 2

[0034] Where: S defect is the monitoring range of the regular polygon composed of N detection windows, r is the radius of the circumscribed circle of the regular polygon; S p is the cross-sectional area of ​​the distribution plate, r p is the radius of the distribution plate; the design parameters of the detection window and the distribution plate must meet the following requirements: p <r,S p defect .

[0035] In order to achieve sufficient coverage of the detection area, the number of detection points N ≥ 6.

[0036] ​Furthermore, in order to achieve multi-angle and full-area monitoring of the distribution plate, each of the terahertz probe integrated modules includes three groups of rotating terahertz probe arrays, the first and third groups of rotating terahertz probe arrays each include three terahertz optical detection probes arranged in a line, and the second group of rotating terahertz probe arrays includes seven terahertz optical detection probes arranged in a line, and each group of rotating terahertz probe arrays can rotate independently driven by the rotating parts.

[0037] Furthermore, the cavitation defect prediction module includes a sensor interface, an optical signal sensor, a signal converter, a data acquisition card, a wireless communication module, a power module, a digital signal processor, a central processing unit, a graphics processing unit and a data storage module; the terahertz spectral data is collected by the optical signal sensor, converted by the signal converter and then collected by the data acquisition card, and then transmitted to the digital signal processor via the sensor interface, and data processing is performed by the central processing unit and the graphics processing unit to complete the prediction calculation of the cavitation defect thickness, and finally the processed data is transmitted to the visualization module through the wireless communication module.

[0038] Furthermore, the visualization module includes an LED display screen, an indicator light and a wireless communication module. The LED display screen displays cavitation crack defect data, real-time values ​​monitored by sensors and health status information in real time; once the cavitation crack parameters exceed the set threshold, the visualization module sends an alarm signal.

[0039] The beneficial effects of the present invention are:

[0040] (1) Non-contact monitoring: Terahertz spectroscopy technology does not require contact with the object being measured, thus greatly avoiding damage to the plunger pump structure;

[0041] (2) Multi-angle real-time monitoring: The terahertz probe array is integrated and matched with the action mechanism to achieve multi-angle real-time monitoring of the entire area of ​​the piston pump distribution plate without blind spots;

[0042] (3) High-precision identification: The cavitation defect thickness inversion and threshold calculation model is established to improve the accuracy and efficiency of plunger pump cavitation crack defect identification;

[0043] (4) Visualization system: The integrated visualization system realizes the reconstruction and dynamic display of the cavitation crack morphology and degree, which helps users to intuitively understand the cavitation damage;

[0044] (5) Intelligent integration: The intelligent integration level of the plunger pump is enhanced, providing a health status monitoring platform and efficient solutions to meet the diversified market needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1According to some embodiments of the present invention, a workflow diagram of a method for detecting cavitation damage of a valve plate of an aviation plunger pump based on terahertz spectroscopy technology is shown;

[0046] Figure 2 According to some embodiments of the present invention, a schematic structural diagram of a system for detecting cavitation damage of a valve plate of an aviation plunger pump is shown;

[0047] Among them, 1-spring, 2-spring support, 3-steel ball, 4-steel ball support, 5-cylinder body, 6-rear end cover, 7-shoe support, 8-bolt, 9-distribution plate, 10-oil distribution plate, 11-front end cover, 12-pin, 13-oblique oil distribution plate, 14-compression spring, 15-oil seal, 16-intermediate pump body, 17-shaft, 18-bearing, 19-plunger, 20-plunger cavity, 21-plunger support, 22-terahertz detection module, 23-data acquisition and processing module, 24-visualization module, 25-detection window.

[0048] Figure 3 An isometric view of a terahertz spectroscopy detection module is shown according to some embodiments of the present invention;

[0049] Among them, 2201-rotating shaft, 2202-terahertz probe integrated module, 2203-terahertz optical probe, 2204-rotating shaft tube seat, 2205-connecting thread, 2206-positioning boss, 2207-external boss, and the rear end is connected to the data acquisition module and the processing module 23;

[0050] Figure 4 According to some embodiments of the present invention, a bottom view of a terahertz spectrum detection module is shown;

[0051] Figure 5 According to some embodiments of the present invention, a schematic diagram of a terahertz spectrum detection region is shown;

[0052] Figure 6 According to some embodiments of the present invention, a workflow diagram of a data acquisition and processing module is shown;

[0053] Figure 7 According to some embodiments of the present invention, a front view of a data acquisition and processing module is shown;

[0054] Among them, 2301-sensor interface, 2302-optical signal sensor, 2303-signal converter, 2304-data acquisition card, 2305-wireless communication module, 2306-power module, 2307-digital signal processor, 2308-central processing unit, 2309-graphics processing unit, 2310-data storage module, 2311-installation fixing plate, 2312-housing;

[0055] Figure 8 An isometric view of a data acquisition and processing module is shown according to some embodiments of the present invention;

[0056] Fig. 9 According to some embodiments of the present invention, a workflow diagram of a visualization module is shown;

[0057] Fig.10 According to some embodiments of the present invention, a front view of a visualization module is shown;

[0058] Among them, 2401-LED display screen, 2402-indicator light, 2403-wireless communication module, 2404-external power module, 2405-installation limit hole, 2406-wired communication interface, 2407-USB data external port, 2408-housing, 2409-rear fixing plate.

[0059] Fig.11 According to some embodiments of the present invention, a bottom view of a visualization integration module is shown. DETAILED DESCRIPTION

[0060] The technical features and advantages of the present invention are described in more detail below in conjunction with the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0061] See also Figure 1 The embodiment of the present invention provides an online monitoring method for cavitation damage of a valve plate of an aviation plunger pump based on terahertz spectroscopy. The online monitoring process steps of the method include:

[0062] Step S1, terahertz spectrum data detection: obtaining terahertz spectrum data at different positions of the distribution plate through a terahertz spectrum detection module;

[0063] Step S2: Cavitation defect prediction: Establish a physical model of the interaction between terahertz spectrum data and cavitation defect data, and use the physical model to predict and output the corresponding cavitation defect data based on the terahertz spectrum data obtained in S1;

[0064] Step S3: Cavitation defect visualization: visualize the cavitation defect data predicted by S2, monitor the dynamic changes of the cavitation state on the surface of the piston pump distribution plate, and provide timely warning and maintenance feedback on the cavitation defect state.

[0065] In the above method, the prediction model of cavitation defects can be obtained by the following steps:

[0066] S21, obtaining spectral sample data of the distribution plate: using a terahertz spectrum detection module to obtain terahertz spectrum sample data at different positions of the distribution plate;

[0067] S22, cavitation defect thickness inversion and threshold determination: based on the terahertz spectrum sample data obtained in S21, the cavitation defect thickness of the distribution plate is inverted and calculated, and the threshold of the cavitation defect thickness is divided;

[0068] S23, establishment of a cavitation defect prediction model: establishing a physical model of the interaction between terahertz spectrum data and cavitation defect thickness, using the terahertz spectrum sample data obtained in S21 and the cavitation defect thickness data calculated in S22 to train the physical model, and subsequently using the trained prediction model to predict and output the cavitation defect thickness based on the detected terahertz spectrum data;

[0069] S24, determination of the degree of cavitation damage: for the predicted cavitation defect thickness data, the degree of cavitation damage of the distribution plate is determined based on the cavitation defect thickness threshold determined in S22.

[0070] See also Figure 2 The online monitoring system for cavitation damage of the aviation plunger pump valve plate based on terahertz spectroscopy technology in the embodiment of the present invention mainly includes three modules: a terahertz spectroscopy detection module 22, a cavitation defect prediction module 23 and a visualization module 24.

[0071] The spring 1 and spring support 2 of the aviation plunger pump form a spring-support assembly for providing a predetermined displacement. The spring support 2 and the steel ball support 4 are respectively connected to the cylinder body 5 and are connected to the pump body by threads to fix the spring position, ensure the stability of the steel ball and provide a suitable working space for the ball valve. The steel ball 3 cooperates with the steel ball support 4 to form a ball valve structure for controlling the direction or pressure of the oil flow. The cylinder body 5 is fixedly connected to the intermediate pump body 16 and the front cover 11 through the pin 12. The cylinder body 5 includes a plunger cavity 20 and a plunger support 21 for the plunger 19 to reciprocate in the plunger cavity 20. The rear end cover 6 is connected to the cylinder body 5 and the intermediate pump body 16 by bolts, which plays a sealing and supporting role to prevent oil medium leakage. The sliding shoe support 7 is connected to the intermediate pump body 16 by bolts 8 for the linear movement of the plunger. The plunger support 21, the oil distribution plate 10, the distribution plate 9, etc. are installed in the intermediate pump body 16 to support and distribute the oil flow. The bearing 18 is connected to the shaft 17, and the two are connected to the intermediate pump 16 by a key to drive the movement of the plunger. The distribution plate 9 is fixed to the pump body by a latch and connected to the cylinder body 5 by bolts 8 to ensure the correct distribution and flow direction of the liquid flow. The oil further flows through the distribution plate 10 to guide the oil flow to each plunger cavity. The inclined distribution plate 13 is connected to the pump body by bolts 8 to change the direction of the oil flow so that the oil is evenly distributed in the plunger cavity. The front cover 11 is connected to the cylinder body 5 by bolts 8 to seal and protect the pump body. The oil seal 15 is connected to the front cover 11 and the rear cover 6 to prevent oil leakage and ensure the sealing of the pump.

[0072] Arrange N measuring points on the outer wall of the plunger pump corresponding to the outer area of ​​the oil distribution plate 9. The N measuring points are distributed in a regular polygon. Calculate the detection area and finally ensure that the monitoring area S defect Always larger than the cross-sectional area S of the valve plate p ; A detection window 25 is respectively opened at each measuring point position on the outer wall of the pump body, and an M16*1.5 internal thread is set at the detection window 25. A terahertz probe integrated module 2202 is arranged on each detection window 25, and multiple terahertz optical probes 2203 are integrated on each terahertz probe integrated module 2202, so as to realize real-time monitoring of the entire area of ​​the piston pump distribution plate without dead angles. The terahertz spectrum detection module 22 and the cavitation defect prediction module 23 are connected by threads and installed on the fixed plate 2311. After the terahertz spectrum detection module 22 and the cavitation defect prediction module 23 are integrated, they are connected to the M16*1.5 internal thread set at the detection window 25 on the outer wall of the pump body through the connecting thread 2205, and the model of the connecting thread 2205 is an M16*1.5 external thread. The data processed by the cavitation defect prediction module 23 is transmitted to the visualization module 24 through the wireless communication module 2305. The visualization module 24 has an embedded wireless signal receiver 2403, which is independently externally placed on the plunger pump body to realize remote terminal monitoring of the cavitation status of the plunger pump.

[0073] See also Figure 3-Figure 4 , is a schematic diagram of the structure of the terahertz spectrum detection module 22 in an embodiment of the present invention, including a rotating shaft 2201, a terahertz probe integrated module 2202, a terahertz optical probe 2203, a rotating shaft tube seat 2204, a connecting thread 2205, a positioning boss 2206, and an external boss 2207. Among them, the terahertz probe integrated module 2202 is fixed on the rotating shaft 2201, both ends of the rotating shaft 2201 are fixed to the bottom end of the rotating shaft tube seat 2204, the top of the rotating shaft tube seat 2204 is connected to the connecting thread 2205, the connecting thread 2205 is fixed on the positioning boss 2206, the rear end of the positioning boss 2206 is connected to the cavitation defect prediction module 23, and multiple terahertz optical probes 2203 are installed on the terahertz probe integrated module 2202 by threaded connection. To achieve multi-angle monitoring, the rotating shaft 2201 can rotate 60° in the left and right directions around its central axis. The rotation of the rotating shaft 2201 can drive the coordinated rotation of the terahertz probe integrated module 2202 and the terahertz optical probe 2203, thereby greatly improving the terahertz optical detection range and covering the full area detection coverage of the distribution plate 9.

[0074] See also Figure 5 , through the coordinated rotation of the terahertz probe integrated module 2202 and the terahertz optical probe 2203, the terahertz optical monitoring range of the embodiment of the present invention can be calculated in the following way:

[0075] The regular N-gon formed by N detection points is composed of N triangles of equal area. The base l and height h of the triangle can be expressed as:

[0076]

[0077]

[0078] Therefore, the monitoring area S defect It can be calculated by the following formula:

[0079]

[0080]

[0081] S p =πr p 2 (5)

[0082] Where: S defect is the monitoring range of a regular polygon composed of N detection windows 25, S p is the cross-sectional area of ​​the distribution plate, r is the circumscribed circle radius of the regular polygon, r p is the radius of the distribution plate. In order to achieve full-area monitoring of the distribution plate 9, the design parameters of the distribution plate 9 need to meet the following requirements:

[0083] r p <r (4)

[0084] S p defect (5)

[0085] The monitoring area S is guaranteed defect Always larger than the cross-sectional area S of the distribution plate p , achieving full coverage of the monitoring area. In order to achieve full coverage of the detection area, the number of detection points N≥6, in this embodiment, N=6.

[0086] See also Figure 4 ​In order to realize multi-angle full-area monitoring, in this embodiment, each terahertz probe integrated module 2202 includes three groups of parallel arranged rotating terahertz probe arrays, the first and third groups of rotating terahertz probe arrays have three terahertz optical detection probes 2203 respectively, and the second group of rotating terahertz probe arrays has seven terahertz optical detection probes 2203. Through the above arrangement, the detection of cavitation crack morphology in different height ranges at multiple levels can be realized. Each rotating terahertz probe array is fixed on a rotating shaft 2201, and the rotation of the rotating shaft 2201 can drive the rotating terahertz probe array and the terahertz optical detection probe 2203 to realize the left and right swing of the monitoring angle by 60°, thereby greatly improving the monitoring range of each detection window 25.

[0087] The terahertz spectrum detection module 22 uses the Pace THz Photoconductive Antenna probe as the terahertz optical detection probe 2203, with a specification of 2mm×2mm. The array of rotating terahertz probe arrays is arranged in a line, and the spacing between two adjacent terahertz optical detection probes 2203 is set to 6mm. The terahertz probe integration module 2202 efficiently generates, detects and processes signals in the terahertz band by combining high-frequency sensors and precision electronic components. Terahertz pulses are generated by the terahertz optical detection probe 2203, and the reflected terahertz wave signal is captured by the receiver. The functions of signal modulation, demodulation, amplification, etc. are integrated into the miniaturized unit of the terahertz probe integration module 2202 to achieve high-precision signal capture and processing.

[0088] See also Figure 7-Figure 8 The cavitation defect prediction module 23 includes: a sensor interface 2301, an optical signal sensor 2302, a signal converter 2303, a data acquisition card 2304, a wireless communication module 2305, a power module 2306, a digital signal processor 2307, a central processing unit 2308, a graphics processing unit 2309, a data storage module 2310, a mounting plate 2311 and a shell 2312. The sensor interface 2301 is connected to the terahertz spectrum detection module 22, which is used to collect terahertz spectrum data. The optical signal is received by the optical signal sensor 2302, converted by the signal converter 2303, and then collected by the data acquisition card 2304. The power module 2306 is used to power the digital signal processor 2307, the central processing unit 2308, the graphics processing unit 2309, and the data storage module 2310. The terahertz spectrum data is transmitted to the digital signal processor 2307, and the central processing unit 2308 and the graphics processing unit 2309 perform complex data processing, complete the inversion calculation of the cavitation defect thickness, and determine the threshold of the cavitation thickness and predict the cavitation defect data through deep learning analysis and image processing. The processed data is finally transmitted to the visualization module 24 through the wireless communication module 2305.

[0089] The specific workflow diagram of cavitation defect prediction module 23 is shown in Figure 6 After collecting the terahertz spectrum sample data of the material sample containing defects, the spectrum data is first pre-processed, including filtering, denoising, normalization and other processing methods, and then the thickness of the cavitation crack is preliminarily calculated by inversion using the formula. The method for inverting the cavitation defect thickness of the plunger pump valve plate 9 by terahertz spectrum sample data is as follows:

[0090] When there is cavitation defect in the metal material of the distributor plate 9, the total time delay of the reflected signal is recorded as Δt total , Δt total is the time delay Δt of the metal material of the valve plate itself material The time delay Δt caused by cavitation defects defect sum:

[0091] Δt total =Δt material +Δt defect (6)

[0092]

[0093]

[0094] Where, d material is the thickness of the metal material of the distribution plate 9, d defect is the thickness of the cavitation defect to be measured, n defect is the refractive index of the terahertz spectrum of the cavitation defect area, n is the refractive index of the terahertz spectrum of the metal material, and c is the speed of light.

[0095]

[0096]

[0097] After the cavitation crack thickness is calculated by the above method, the spectral sample data is divided into a training set and a test set. The convolutional neural network structure is selected to construct a physical model of the interaction between terahertz spectral data and cavitation crack thickness. The terahertz spectral sample data is used as input, and the calculated cavitation crack thickness is used as output. The number of input layer nodes is set to 100, the hidden layer is 3 layers, the activation function is ReLU, and the loss function is mean square error (MSE). The training set is input into the neural network model for forward propagation and back propagation. Through multiple trainings, the weights and biases are gradually adjusted to minimize the loss function value. The hyperparameters of the neural network (such as learning rate, batch size, number of layers, number of nodes, etc.) are tuned through cross-validation. When the root mean square error is <2%, it is proved that the neural network model has high prediction accuracy and reliability. For the subsequently detected terahertz spectral data, it only needs to be input into the trained neural network model, and the model will predict the output cavitation defect thickness value based on the learned relationship. Furthermore, the threshold of cavitation defect is determined according to the crack defect thickness under different cavitation degrees, which are the cavitation initial threshold d cav1 and cavitation damage threshold d cav2 , when the cavitation defect thickness predicted by the model is 0 <d defect <d cav1 When the cavitation defect thickness d is predicted by the model, it is determined to be the initial threshold of cavitation. At this time, the bubbles have just been generated and have not yet caused damage to the solid material surface of the distribution plate 9. cav1 <d defect <d cav2 When the cavitation damage threshold is reached, the solid material of the distributor plate 9 has obvious cracks or holes; when the cavitation defect thickness d cav2 <d defect , it is determined to be the cavitation fatigue threshold. At this time, the fatigue limit of the solid material of the distribution plate 9 has been reached, the cavitation crack continues to develop, and the distribution plate 9 has a slight leakage.

[0098] See also Figure 10-11The visualization module 24 includes: an LED display screen 2401, an indicator light 2402, a wireless communication module 2403, an external power supply module 2404, an installation limit hole 2405, a wired communication interface 2406, a USB data external port 2407, a shell 2408, and a rear fixing plate 2409. The LED display screen 2401 is used to display the cavitation crack detection results or related data in real time and provide clear visual feedback. The indicator light 2402 is located on the left side of the cavitation crack defect visualization integrated module 24, and is used to provide a concise visual indication to display the working status or warning information of the plunger pump. The wireless communication module 2403 can exchange data wirelessly with other devices or remote servers. The external power supply block 2404 provides a stable power supply for the entire integrated module. The installation limit hole 2405 is used to fix the cavitation crack defect visualization integrated module 24 on the rear fixing plate 2409, and the shell 2408 provides external protection for the module to prevent the internal circuits and components from being damaged by the external environment (such as dust, moisture, impact, etc.). The LED display screen 2401 includes common visual information: real-time dynamic display of cavitation crack defect data, real-time values ​​monitored by sensors, such as pressure, temperature, flow, etc., and health status displays the working status of hydraulic components in graphical or text form, such as "normal", "warning" or "fault"; trend charts show the trend of various parameters over time to help users analyze long-term system changes. Alarm prompt: Once the cavitation crack parameters exceed the set threshold, the system will flash a light alarm to indicate that inspection or maintenance is required.

[0099] The specific workflow of the visualization module 24 is shown in Fig. 9 The inlet and outlet pressure, flow rate, temperature and other data collected by the sensor are integrated and displayed on the LED display screen 2401. The cavitation crack defect thickness at six detection points is used to reconstruct the three-dimensional cavitation crack morphology. The interpolation method is used to estimate the three-dimensional morphology of the crack, and the point cloud three-dimensional coordinates (x, y, z) and crack thickness of all points on the cavitation crack surface are obtained. The point cloud data generated by interpolation is further converted into a three-dimensional crack morphology model through the Delaunay triangulation method. The data is displayed on the LED display screen 2401 in real time. The dynamic change characteristics of the cavitation crack over time and the thermal map are displayed in real time to observe the cavitation crack morphology in real time. Further, corresponding feedback is provided based on the real-time displayed cavitation crack situation. The cavitation crack situation can be divided into the cavitation initial threshold d according to the threshold. cav1 , Cavitation damage threshold d cav2 and fatigue crack threshold, different cavitation degrees correspond to different feedback and maintenance suggestions. When the valve plate 9 is at the initial cavitation threshold d cav1 Although the material of the valve plate 9 is not damaged, cavitation bubbles have already occurred, and the development of cavitation should be strengthened and continuously paid attention to. cav2When the material of the valve plate 9 begins to show obvious cracks, regular inspection and maintenance are required to repair the fracture surface. When it reaches the fatigue crack threshold, the valve plate 9 has leaked and there is a major safety hazard. The valve plate 9 needs to be replaced to avoid accidents. In addition, the visualization module 24 can further integrate a remote monitoring and management module, which regularly reminds maintenance personnel to perform calibration and maintenance, and upgrades and updates products through this module, thereby optimizing and expanding the performance of the visualization integration module.

[0100] The embodiment of the present invention proposes an online monitoring method for cavitation damage of an aviation plunger pump valve plate based on terahertz spectroscopy. The method uses terahertz spectroscopy technology to perform high-precision real-time detection of the surface of the plunger pump valve plate, and can effectively identify early signs of cavitation damage. By analyzing the changes in terahertz spectral signals, the cavitation defects of the pump valve plate are monitored in real time, and non-destructive diagnosis and early warning of cavitation damage are achieved, providing a new technical means for the maintenance and fault prediction of aviation equipment.

[0101] In the description of the embodiments of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "center", "top", "bottom", "top", "bottom", "inside", "outside", "inner side", "outer side" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the purpose of facilitating the description of the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Among them, "inside" refers to an internal or enclosed area or space. "Periphery" refers to the area surrounding a specific component or a specific area.

[0102] In the description of the embodiments of the present invention, the terms "first", "second", "third", and "fourth" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", and "fourth" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0103] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", and "assemble" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0104] In the description of the embodiments of the present invention, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0105] In the description of the embodiments of the present invention, it is to be understood that "-" and "~" represent the range between two values, and the range includes the endpoints. For example: "AB" represents a range greater than or equal to A and less than or equal to B. "A~B" represents a range greater than or equal to A and less than or equal to B.

[0106] In the description of the embodiments of the present invention, the term "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " herein generally indicates that the associated objects before and after are in an "or" relationship.

[0107] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An online monitoring method for cavitation damage of an aviation plunger pump valve plate based on terahertz spectroscopy technology, characterized in that The following steps are involved: Step S1, terahertz spectrum data detection: obtaining terahertz spectrum data at different positions of the distribution plate through a terahertz spectrum detection module; Step S2, cavitation defect prediction: establishing a physical model of the interaction between terahertz spectrum data and cavitation defect data, and based on the terahertz spectrum data obtained in S1, using the physical model, predicting and outputting the corresponding cavitation defect data; Step S3, cavitation defect visualization: visualize the cavitation defect data predicted by S2, monitor the dynamic changes of the cavitation state on the surface of the piston pump distribution plate, and provide timely early warning and maintenance feedback on the cavitation defect state.

2. The method according to claim 1, characterized in that: The cavitation defect prediction in step S2 specifically includes: S21, obtaining spectral sample data of the distribution plate: using a terahertz spectrum detection module to obtain terahertz spectrum sample data at different positions of the distribution plate; S22, cavitation defect thickness inversion and threshold determination: based on the terahertz spectrum sample data obtained in S21, the cavitation defect thickness of the distribution plate is inverted and calculated, and the threshold of the cavitation defect thickness is divided; S23, establishment of a cavitation defect prediction model: establishing a physical model of the interaction between terahertz spectrum data and cavitation defect thickness, using the terahertz spectrum sample data obtained in S21 and the cavitation defect thickness data calculated in S22 to train the physical model, and subsequently using the trained prediction model to predict and output the cavitation defect thickness based on the detected terahertz spectrum data; S24, determination of the degree of cavitation damage: for the predicted cavitation defect thickness data, the degree of cavitation damage of the distribution plate is determined based on the cavitation defect thickness threshold determined in S22.

3. The method according to claim 2, characterized in that In step S22, the inverse calculation method for calculating the cavitation defect thickness using the terahertz spectrum sample data is as follows: When there is a cavitation defect in the metal material of the distributor plate, the total time delay of the reflected signal is recorded as Δt total , Δt total is the time delay Δt of the metal material of the valve plate itself material The time delay Δt caused by cavitation defects defect sum: Δt total =Δt material +Δt defect In the above formula, d material is the thickness of the metal material of the valve plate, d defect is the thickness of the cavitation defect to be measured, n defect is the refractive index of the terahertz spectrum of the cavitation defect area, n is the refractive index of the terahertz spectrum of the metal material, and c is the speed of light.

4. The method according to claim 2, characterized in that: In step S23, a convolutional neural network structure is selected to establish a prediction model for cavitation defect thickness, the terahertz spectrum sample data of step S21 is divided into a training set and a test set, the training set is input into the neural network, and the loss function value is minimized through multiple trainings, and then the test set is input into the neural network to tune the hyperparameters of the neural network.

5. The method according to claim 4, characterized in that The activation function of the convolutional neural network structure adopts ReLU, and the loss function adopts mean square error. The training set is input into the neural network for forward propagation and back propagation. Through multiple trainings, the weights and biases are gradually adjusted to minimize the loss function value. The hyperparameters of the network are tuned through cross-validation, and the optimization is stopped when the root mean square error is <2%.

6. The method according to claim 2, characterized in that In step S22, the threshold of the cavitation defect thickness is determined according to the tolerance of the material to cavitation, and the threshold of the cavitation defect thickness includes the cavitation initial threshold d cav1 and cavitation damage threshold d cav2 , when the thickness of the cavitation defect is 0 <d defect <d cav1 When , it is determined as the initial threshold of cavitation; When the measured cavitation defect thickness d cav1 <d defect <d cav2 When d cav2 <d defect When , it is determined to be the cavitation fatigue threshold.

7. A system for online monitoring of cavitation damage of a valve plate of an aviation plunger pump using the method according to any one of claims 1 to 6, characterized in that include: A terahertz spectrum detection module, a cavitation defect prediction module and a visualization module. The terahertz spectrum detection module is used to obtain terahertz spectrum data at different positions of the distribution plate. The data acquisition and processing module predicts the degree of cavitation damage based on the obtained terahertz spectrum data. The visualization module is used to visualize the cavitation defect data and issue an early warning.

8. The system according to claim 7, characterized in that The terahertz spectrum detection module includes N measuring points arranged on the outer wall of the plunger pump corresponding to the outer area of ​​the oil distribution plate, and the N measuring points are distributed in a regular polygon. A detection window is opened on the outer wall of the pump body corresponding to each measuring point position, and each detection window is arranged with a rotating part and a terahertz probe integrated module. Each of the terahertz probe integrated modules integrates multiple terahertz optical probes. The rotating part drives the terahertz probe integrated module to rotate, so as to realize real-time monitoring of the entire area of ​​the plunger pump distribution plate without blind spots.

9. The system according to claim 8, characterized in that Through the coordinated rotation of the plurality of terahertz optical probes, the monitoring range of the terahertz spectrum detection module can be calculated in the following manner: Among them: A regular N-gon is composed of N triangles of equal area, and the base l and height h of the triangle can be expressed as: Therefore, the monitoring area S defect It can be calculated by the following formula: S p ＝πr p 2 Where: S defect is the monitoring range of the regular polygon composed of N detection windows, r is the radius of the circumscribed circle of the regular polygon; S p is the cross-sectional area of ​​the distribution plate, r p is the radius of the distribution plate; the design parameters of the detection window and the distribution plate must meet the following requirements: p <r,S p defect .​ 10. The system according to claim 9, characterized in that N≥6。 11. The system according to claim 8, characterized in that Each of the terahertz probe integrated modules includes three groups of rotating terahertz probe arrays. The first and third groups of rotating terahertz probe arrays each include three terahertz optical detection probes arranged in a line, and the second group of rotating terahertz probe arrays includes seven terahertz optical detection probes arranged in a line. Each group of rotating terahertz probe arrays can rotate independently driven by the rotating member.

Images