Lubricating oil abrasive particle detection method and system

Abstract

The invention discloses a lubricating oil abrasive particle detection method and system, and the method comprises the steps: data collection: arranging a capacitive sensor on the outer wall of a pipeline, and obtaining the initial capacitance data C1 of a to-be-detected region through the capacitive sensor; data processing I: performing dimension reduction processing on the initial capacitance data C1 through a principal component analysis method to obtain second-order capacitance data C2; second data processing: inputting the second-order capacitance data C2 into the neural network model, and adding constraint conditions to obtain third-order capacitance data C3; image reconstruction: inputting the third-order capacitance data C3 into the finite element analysis model to obtain an image in the to-be-detected area; and evaluation and optimization: evaluating the reconstructed image, and if the image is not clear, repeating the second data processing and the image reconstruction process until the imaging quality meets the requirement. According to the method, the data is analyzed through the principal component analysis method, a simpler and more accurate neural network model can be established, the whole process consumes less time, the calculation amount is moderate, and the accuracy can also be ensured.

Description

Technical Field

[0001] The present application relates to the technical field of testing materials using electromagnetic technology, and in particular to a lubricating oil wear particle detection method and system. Background Art

[0002] Lubricating oil is a commonly used medium in the industrial field, especially in some transmission fields. It mainly plays the role of full lubrication and cooling.

[0003] During the use of lubricating oil, it is inevitable that there are some particulate impurities, which may affect the normal performance of the lubricating oil. Therefore, some related technologies usually need to monitor the lubricating oil to obtain the situation of the particles inside the lubricating oil so as to make corresponding correct countermeasures. Especially in some fields of use with safety requirements, such as aviation auxiliary power unit APU (Auxiliary power unit), the particle situation in the lubricating oil is an important factor affecting flight safety.

[0004] Currently, there are methods for offline detection or online detection. For example, there is sampling detection in offline detection, but offline detection cannot obtain the lubricating oil condition in real time, and its immediacy is poor and its reliability is insufficient. Online detection has capacitance tomography technology, which can image the medium condition in the pipeline in real time, but currently online detection still has the defects of large amount of calculation and insufficient accuracy.

[0005] Application Contents

[0006] The present application provides a lubricating oil wear particle detection method and system to solve the current technical problems of large amount of calculation and insufficient accuracy in online detection.

[0007] In the first aspect, the present application provides a method for detecting lubricating oil wear particles, including: data acquisition: arranging a capacitive sensor on the outer wall of the pipeline, and obtaining the initial capacitance data C1 of the area to be tested through the capacitive sensor; data processing one: performing dimensionality reduction processing on the initial capacitance data C1 through the principal component analysis method to obtain second-order capacitance data C2; data processing two: inputting the second-order capacitance data C2 into the neural network model, adding constraints, and obtaining third-order capacitance data C3; image reconstruction: inputting the third-order capacitance data C3 into the finite element analysis model to obtain an image in the area to be tested; evaluation and optimization: evaluating the reconstructed image. If the image is not clear, repeating the data processing two and the image reconstruction process until the imaging quality meets the requirements.

[0008] In an embodiment of the present application, the initial capacitance data C1 is subjected to dimensionality reduction processing by the principal component analysis method, which can effectively reduce the dimension of the input data, remove redundant information and retain only the most important feature set in the initial capacitance data C1. Considering that the principal component analysis method performs linear dimensionality reduction, there are some nonlinear original data in the initial capacitance data C1. Therefore, it is difficult to capture the important features of the initial capacitance data C1 using only the principal component analysis method. Therefore, it is considered to further process the reduced-dimensional second-order capacitance data C2, that is, to process it through a neural network model. The neural network model optimizes the algorithm according to the second-order data C2 under the action of constraints, and outputs the third-order capacitance data C3. Since the neural network model can capture and fit nonlinear features, a suitable neural network model can be selected or designed in the actual operation process, so that the neural network model can establish a more accurate mapping relationship between the second-order capacitance data C2 and the third-order capacitance data C3. Therefore, the neural network model can be used to capture or improve the data missed by the principal component analysis method. In this way, the second-order capacitance data C2 after principal component analysis reduces the computational complexity of the neural network model, speeds up training, reduces the risk of overfitting, and helps the neural network model to better learn the key features in the data, thereby establishing a more streamlined neural network model. With the help of the neural network model, the second-order capacitance data C2 can be processed twice to finally output accurate third-order capacitance data C3. The entire process takes less time, the amount of calculation is moderate, and the data accuracy can also be guaranteed. Finally, the finite element analysis model is used for image reconstruction. The third-order capacitance data C3 is data that has been reduced in dimension and improved by the neural model, and can maintain a certain degree of accuracy and simplicity. The image reconstruction using the finite element analysis model takes less time and the results are more accurate.

[0009] As one of the optional embodiments of the present application, the step of establishing a neural network model is also included between data processing one and data processing two: solving the positive problem: according to the collected initial capacitance data C1 and the characteristic parameters of the acquisition system, the theoretical capacitance value distribution is calculated and recorded as the target capacitance data C0; a neural network model suitable for processing capacitance data is set, and a suitable network structure and number of layers are selected, and constraints are added; the second-order capacitance data C2 is used as input data, and the target capacitance data C0 is used as output data, and the neural network model is trained through the second-order capacitance data C2 and the target capacitance data C0, so that the neural network model learns the mapping relationship between the input and output data.

[0010] In the embodiment of the present application, the target capacitance data C0 is obtained in theory by solving the positive problem. The target capacitance data C0 is a set of data that can most accurately reflect the image of the area to be tested in theory. This is obtained by a series of rigorous formula derivation and calculation. It is very mature in the existing related technology and will not be repeated. The process of establishing a traditional neural network model is: input the initial capacitance data C1 and the target capacitance data C0, and after multiple trainings, let the neural network learn the mapping relationship between C1 and C0, and then input the corresponding initial capacitance data C1 in the subsequent step to obtain the capacitance data for image reconstruction. However, there are obvious problems in this way, that is, there is a large amount of data interference in the initial capacitance data C1, and the data volume itself is large, which greatly increases the learning time and load of the neural network model, and also has a large interference with the mapping accuracy of the neural network model, which is very resistant to immediate image reconstruction. In the present application, the initial capacitance data C1 is first processed by the principal component analysis method, and the processed second-order capacitance data C2 is used as input when training the neural network. Then the neural network model only needs to learn the relationship between C2 and C0. In this way, the training process of the neural network model is greatly reduced, and because the guiding capacitance data is also the target capacitance data C0, when C2 is subsequently input, a more accurate third-order capacitance data C3 for the image reconstruction process can be obtained, and the third-order capacitance data C3 is relatively close to the target capacitance data C0.

[0011] As one of the optional embodiments of the present application, the steps of solving the direct problem include: establishing a finite element model: dividing the target object into discrete small cells, each cell having a certain conductivity; constructing a direct problem equation: constructing a direct problem equation based on the initial capacitance data C1 and the finite element model, the direct problem equation is used to indicate the relationship between the initial capacitance data C1 and the conductivity of the area to be measured; solving the direct problem: by solving the direct problem equation, the conductivity distribution of the area to be measured is obtained, recorded as the target capacitance data C0.

[0012] As one of the optional embodiments of the present application, the evaluation and optimization step also includes: if the results of multiple image reconstructions do not meet expectations, different types of neural network models are replaced, and the step of establishing the neural network model is repeated, wherein the neural network model includes a convolutional neural network, a recurrent neural network, and a feedforward neural network.

[0013] In the embodiment of the present application, since there are many types of neural network models and their adaptability varies, if the results of multiple image reconstructions do not meet expectations, it proves that the neural network model may have poor adaptability. During the evaluation and optimization process, the neural network model is changed to find the most suitable neural network model, thereby improving the accuracy of subsequent data processing.

[0014] As one of the optional embodiments of the present application, the evaluation and optimization step also includes: if the results of multiple image reconstructions do not meet expectations, then changing the combination of constraints to repeat the training, wherein the constraints include physical constraints, spatial constraints, prior knowledge constraints, etc.

[0015] In the embodiment of the present application, by trying multiple combinations of constraint conditions, the constraint conditions that best suit the current detection process can be found during the neural network training process, so as to remove noise in the capacitance data to the greatest extent, improve the smoothness of the data, and thereby improve the spatial resolution of the capacitance image, making the imaging results more accurate and detailed.

[0016] As one of the optional embodiments of the present application, before the data processing step, a data preprocessing step is also included: processing abnormal values ​​and noise in the initial capacitance data C1.

[0017] In an embodiment of the present application, in the principal component analysis method, there may be a situation where some data is not captured sufficiently. This is because some abnormal values ​​or noise are collected in the initial capacitance data C1 due to circuit noise, capacitance sensor installation position and precision control. Since the principal component analysis method is calculated based on the covariance matrix or correlation matrix of the data, the presence of abnormal values ​​or noise may cause the calculation result of the covariance matrix to deviate from the actual situation, thereby affecting the extraction and interpretation of the principal components. Therefore, preprocessing the data in advance can reduce the impact on the results of the principal component analysis process and ensure the accuracy of the results.

[0018] As one of the optional embodiments of the present application, the initial capacitance data C1 is reduced in dimension by the principal component analysis method, and the step of obtaining the second-order capacitance data C2 includes: standardizing the initial capacitance data C1 so that the mean of each feature is 0 and the variance is 1; calculating the covariance matrix of the standardized capacitance data C1; performing eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and corresponding eigenvectors; sorting the eigenvalues ​​in order from large to small, and selecting the eigenvectors corresponding to the first k eigenvalues ​​as principal components; projecting the standardized capacitance data C1 onto the selected principal component to obtain the reduced-dimensional second-order capacitance data C2.

[0019] As one of the optional embodiments of the present application, the characteristic parameters of the acquisition system include: the geometric shape of the capacitive sensor, the electrode spacing, the electrode size, etc.

[0020] As one of the optional embodiments of the present application, after the evaluation and optimization step, it also includes: if the image shows that the particle diameter D≥0.5cm, or the image shows that the solubility is continuously high within T minutes in each monitoring cycle, a warning message is issued, wherein 15≤T≤25.

[0021] On the other hand, a lubricating oil detection system is also provided, including: a capacitive sensor, which is provided in multiple numbers and is evenly fitted on the outer wall of the pipeline in a ring shape; a data acquisition module, including a driving unit and a transmission unit, the driving unit is used to apply an alternating voltage to stimulate the area to be tested; a terminal, which is used to analyze and process the collected capacitance data and reconstruct an image; wherein the transmission unit is electrically connected to the capacitive sensor to convert the current signal into a digital signal and transmit it to the terminal.

[0022] One of the above technical solutions has the following advantages or beneficial effects: reducing the dimension of the initial capacitance data C1 by the principal component analysis method can effectively reduce the dimension of the input data, remove redundant information and retain only the most important feature set in the initial capacitance data C1. Considering that the principal component analysis method performs linear dimensionality reduction, there are some nonlinear original data in the initial capacitance data C1. Therefore, it is difficult to capture the important features of the initial capacitance data C1 using only the principal component analysis method. Therefore, it is considered to further process the reduced-dimensional second-order capacitance data C2, that is, to process it through a neural network model. The neural network model optimizes the algorithm according to the second-order data C2 under the action of constraints, and outputs the third-order capacitance data C3. Since the neural network model can capture and fit nonlinear features, a suitable neural network model can be selected or designed in the actual operation process, so that the neural network model can establish a more accurate mapping relationship between the second-order capacitance data C2 and the third-order capacitance data C3. Therefore, the neural network model can be used to capture or improve the data missed by the principal component analysis method. In this way, the second-order capacitance data C2 after principal component analysis reduces the computational complexity of the neural network model, speeds up training, reduces the risk of overfitting, and helps the neural network model to better learn the key features in the data, thereby establishing a more streamlined neural network model. With the help of the neural network model, the second-order capacitance data C2 can be processed twice to finally output accurate third-order capacitance data C3. The entire process takes less time, the amount of calculation is moderate, and the data accuracy can also be guaranteed. Finally, the finite element analysis model is used for image reconstruction. The third-order capacitance data C3 is data that has been reduced in dimension and improved by the neural model, and can maintain a certain degree of accuracy and simplicity. The image reconstruction using the finite element analysis model takes less time and the results are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The technical solution and other beneficial effects of the present application will be made apparent by describing in detail the specific implementation methods of the present application in conjunction with the accompanying drawings.

[0024] Figure 1 is an overall flow chart of the lubricating oil wear particle detection method provided in an embodiment of the present application;

[0025] Figure 2 It is a flowchart of establishing a neural network model provided by an embodiment of the present application;

[0026] Figure 3 It is a logic flow chart for displaying the judgment after evaluation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0028] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally indicates that the associated objects before and after are in an "or" relationship.

[0029] Related background information.

[0030] Electrical Capacitance Tomography (ECT) is a non-invasive imaging technology that can be used to provide feedback on the internal fluid conditions of an object. The imaging principle is that each phase of the substance being measured has a different dielectric constant. When the component distribution or concentration distribution of each phase changes, the equivalent dielectric constant of the mixed fluid will change, thereby changing the capacitance value between the measuring electrode pairs.

[0031] ECT technology places multiple electrodes around an object to form a capacitive sensor array surrounding the object. These electrodes record the change in capacitance inside the object by alternating between applying voltage and measuring current. During the imaging process, a small alternating voltage, called the excitation voltage, is first applied around the area to be imaged. Subsequently, a voltage is applied by driving the electrodes to cause current to flow through the object, and a pair of measurements is made between each pair of electrodes. When current passes through the object, differences in the electrical properties inside the object (for example, conductivity) cause changes in the distribution of the current, which in turn changes the capacitance value. Through multiple measurements, a series of capacitance data is obtained, which is sent to a computer for processing and reconstruction.

[0032] During the reconstruction process, the measured capacitance data is analyzed and processed by imaging algorithms to generate images. These algorithms include finite element method, gradient method, back projection method, etc. Among them, several commonly used image reconstruction algorithms are mainly introduced:

[0033] Finite Element Method (FEM): The finite element method is a reconstruction algorithm widely used in electrical capacitance tomography. It divides the area to be imaged into many small finite element units, and models and calculates the capacitance change of each unit. By solving the direct problem of capacitance change, the measured data is fitted with the simulated data to achieve image reconstruction. The finite element method can take into account complex capacitance distribution and object shape, but the computational complexity is high.

[0034] Gradient Method: Gradient method is a reconstruction algorithm based on mathematical gradient calculation. It obtains an estimate of the capacitance distribution inside the object by calculating and optimizing the gradient of the capacitance data. Gradient method is usually an iterative algorithm, which gradually improves the reconstruction result through continuous iterative optimization process. Gradient method has faster calculation speed and lower complexity, but has certain robustness issues for weak boundaries and noise.

[0035] Back Projection Method: Back Projection Method is a simple and commonly used reconstruction algorithm. It is based on the projection principle and projects the measured capacitance data back into the area to be imaged to obtain an image. Back Projection Method has certain advantages in speed and computational complexity, but is more sensitive to noise and nonlinear effects.

[0036] The imaging principle of ECT technology is based on the electrical differences of objects and can be widely used in imaging in the fields of fluid distribution, solid particle distribution, fluid-solid interface positioning, etc. However, the imaging accuracy of ECT technology is affected by factors such as object shape, electrode arrangement, and environmental noise, resulting in insufficient clarity and accuracy of current image reconstruction results.

[0037] This application improves the clarity and accuracy of imaging by optimizing the image reconstruction process.

[0038] The following is combined with Figure 1-3 The present application is further described.

[0039] In a first aspect, the present application provides a lubricating oil wear particle detection method, comprising:

[0040] Step 1, data acquisition: arrange capacitance sensors on the outer wall of the pipeline, and obtain the initial capacitance data C1 of the area to be measured through the capacitance sensors;

[0041] Step 2, data processing 1: perform dimensionality reduction processing on the initial capacitance data C1 by principal component analysis method to obtain the second-order capacitance data C2;

[0042] Step 3, data processing 2: input the second-order capacitance data C2 into the neural network model, add constraints, and obtain the third-order capacitance data C3;

[0043] Step 4, image reconstruction: input the third-order capacitance data C3 into the finite element analysis model to obtain the image in the area to be measured;

[0044] Step 5, evaluation and optimization: Evaluate the reconstructed image. If the image is not clear, repeat the data processing 2 and image reconstruction process until the image quality meets the requirements.

[0045] In an embodiment of the present application, the initial capacitance data C1 is subjected to dimensionality reduction processing by the principal component analysis method, which can effectively reduce the dimension of the input data, remove redundant information and retain only the most important feature set in the initial capacitance data C1. Considering that the principal component analysis method performs linear dimensionality reduction, there are some nonlinear original data in the initial capacitance data C1. Therefore, it is difficult to capture the important features of the initial capacitance data C1 using only the principal component analysis method. Therefore, it is considered to further process the reduced-dimensional second-order capacitance data C2, that is, to process it through a neural network model. The neural network model optimizes the algorithm according to the second-order data C2 under the action of constraints, and outputs the third-order capacitance data C3. Since the neural network model can capture and fit nonlinear features, a suitable neural network model can be selected or designed in the actual operation process, so that the neural network model can establish a more accurate mapping relationship between the second-order capacitance data C2 and the third-order capacitance data C3. Therefore, the neural network model can be used to capture or improve the data missed by the principal component analysis method. In this way, the second-order capacitance data C2 after principal component analysis reduces the computational complexity of the neural network model, speeds up training, reduces the risk of overfitting, and helps the neural network model to better learn the key features in the data, thereby establishing a more streamlined neural network model. With the help of the neural network model, the second-order capacitance data C2 can be processed twice to finally output accurate third-order capacitance data C3. The entire process takes less time, the amount of calculation is moderate, and the data accuracy can also be guaranteed. Finally, the finite element analysis model is used for image reconstruction. The third-order capacitance data C3 is data that has been reduced in dimension and improved by the neural model, and can maintain a certain degree of accuracy and simplicity. The image reconstruction using the finite element analysis model takes less time and the results are more accurate.

[0046] In an optional embodiment of the present application, for Step 1, data collection.

[0047] Specifically, it includes: applying an excitation signal, applying the excitation signal through one electrode of the capacitance sensor, the excitation signal can be an AC signal or a pulse signal, and is used to stimulate the charge distribution in the area to be measured; measuring the capacitance value, measuring the capacitance value through another electrode of the capacitance sensor, and measuring the capacitance value can use a capacitance measurement circuit to connect the capacitance sensor to a measuring device. In an example, the measuring device can be a terminal such as a computer or a single-chip microcomputer; collecting multiple groups of data, repeating the above two steps, collecting multiple groups of capacitance data and recording them to form initial capacitance data C1.

[0048] In one of the optional embodiments of the present application, for Step 2, data processing one.

[0049] Specifically, it includes: standardizing the initial capacitance data C1 so that the mean of each feature is zero and the variance is one, which is to ensure that the importance of each feature is relatively balanced and to avoid excessive influence of some features on the principal component analysis results; calculating the covariance matrix of the standardized capacitance data C1, and the covariance matrix is ​​used to indicate the correlation between different features; performing eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and corresponding eigenvectors, wherein the eigenvalues ​​represent the importance of each feature and the eigenvectors represent the projection of each feature in the direction of the principal component; sorting the eigenvalues ​​in order from large to small, and selecting the eigenvectors corresponding to the first k eigenvalues ​​as the principal components. The principal components are selected in this step, and k is the dimension after dimensionality reduction, which can be selected as needed; projecting the standardized capacitance data C1 onto the selected principal component, that is, multiplying the initial capacitance data C1 by the selected principal component eigenvector to obtain the second-order capacitance data C2 after dimensionality reduction.

[0050] For Step 3, data processing 2.

[0051] Specifically, it includes using a trained neural network model, inputting the second-order capacitance data C2 into the model, and obtaining the predicted third-order capacitance data C3 through the forward propagation process of the model. Among them, the constraints include physical law constraints: according to the physical laws of the problem, the constraints can be established on basic physical principles. For example, in the capacitance problem mentioned in this application, the constraint capacitance value satisfies a specific mathematical formula or relationship with the dielectric constant, or the number of electrodes, current intensity, etc. For example, the capacitance value C satisfies the following formula with the dielectric constant, the electrode area A and the distance d between the electrodes: C = ε r *A / d.

[0052] In other optional examples, constraints may also include prior knowledge constraints: Constraints may be added to the model based on known prior knowledge or experience. For example, based on existing data or the knowledge of domain experts, the output of the model may be constrained to be within a certain range or to satisfy a certain distribution. Data constraints: Constraints may be added based on the characteristics and limitations of the data. For example, the range, distribution, or specific data attributes of the input data may be constrained. Parameter constraints: Constrain the parameters of the model to limit their value range or to satisfy certain conditions. For example, the weight matrix of the model may be constrained to be sparse or to satisfy certain regularization conditions.

[0053] It should be noted that between data processing 1 and data processing 2, there is also step 0: establishing a neural network model.

[0054] Specifically, it includes: Step 01, solving the direct problem: according to the collected initial capacitance data C1 and the characteristic parameters of the acquisition system, calculate the theoretical capacitance value distribution, recorded as the target capacitance data C0; Step 02, set a neural network model suitable for processing capacitance data, and select a suitable network structure and number of layers, and add constraints; Step 03, use the second-order capacitance data C2 as input data, and the target capacitance data C0 as output data, and train the neural network model through the second-order capacitance data C2 and the target capacitance data C0, so that the neural network model learns the mapping relationship between input and output data.

[0055] In the embodiment of the present application, the target capacitance data C0 is obtained in theory by solving the positive problem. The target capacitance data C0 is a set of data that can most accurately reflect the image of the area to be tested in theory. This is obtained by a series of rigorous formula derivation and calculation. It is very mature in the existing related technology and will not be repeated. The process of establishing a traditional neural network model is: input the initial capacitance data C1 and the target capacitance data C0, and after multiple trainings, let the neural network learn the mapping relationship between C1 and C0, and then input the corresponding initial capacitance data C1 in the subsequent step to obtain the capacitance data for image reconstruction. However, there are obvious problems in this way, that is, there is a large amount of data interference in the initial capacitance data C1, and the data volume itself is large, which greatly increases the learning time and load of the neural network model, and also has a large interference with the mapping accuracy of the neural network model, which is very resistant to immediate image reconstruction. In the present application, the initial capacitance data C1 is first processed by the principal component analysis method, and the processed second-order capacitance data C2 is used as input when training the neural network. Then the neural network model only needs to learn the relationship between C2 and C0. In this way, the training process of the neural network model is greatly reduced, and because the guiding capacitance data is also the target capacitance data C0, when C2 is subsequently input, a more accurate third-order capacitance data C3 for the image reconstruction process can be obtained, and the third-order capacitance data C3 is relatively close to the target capacitance data C0.

[0056] As one of the optional embodiments of the present application, the steps of solving the direct problem include: establishing a finite element model: dividing the target object into discrete small cells, each cell having a certain conductivity; constructing a direct problem equation: constructing a direct problem equation based on the initial capacitance data C1 and the finite element model, the direct problem equation is used to indicate the relationship between the initial capacitance data C1 and the conductivity of the area to be measured; solving the direct problem: by solving the direct problem equation, the conductivity distribution of the area to be measured is obtained, recorded as the target capacitance data C0.

[0057] Specifically.

[0058] Establish a finite element model: First, a finite element model of the object needs to be established, which describes the geometric shape and capacitive properties of the object. The object area is divided into many small units, such as triangles or quadrilaterals (for two-dimensional cases) or tetrahedrons or hexahedrons (for three-dimensional cases). In each unit, an appropriate basis function is selected to describe the capacitance distribution.

[0059] Determine boundary conditions: Determine boundary conditions based on actual conditions, including the position and shape of the capacitance probe, the potential or current input of the capacitance probe, etc.

[0060] Establish a direct problem model: According to the finite element model and boundary conditions, establish a direct problem model. The direct problem model describes the relationship between the capacitance value measured by the sampling probe and the capacitance distribution of the object. By solving the direct problem model, the theoretical capacitance value distribution can be obtained.

[0061] The relationship between the capacitance value C and the dielectric constant ε distribution can be expressed as the following formula:

[0062]

[0063] Where: ε(x,y) is the dielectric constant distribution function of the measured field, is the potential distribution function, V is the potential difference between the source electrode and the detection electrode, and Γ is the electrode surface. Solving the distribution of the dielectric constant matrix G by the capacitance matrix C and the sensitive field matrix S is the image reconstruction of ECT.

[0064] Meshing: Meshing the object area according to the geometric shape of the object and the finite element model. Meshing divides the object area into many small units for numerical calculation.

[0065] Solve the direct problem model: Use numerical calculation methods, such as the finite element method, to solve the direct problem model. By solving the model equation, the capacitance value distribution is calculated.

[0066] As one of the optional embodiments of the present application, the evaluation and optimization step also includes: if the results of multiple image reconstructions do not meet expectations, different types of neural network models are replaced, and the step of establishing the neural network model is repeated, wherein the neural network model includes a convolutional neural network, a recurrent neural network, and a feedforward neural network.

[0067] In the embodiment of the present application, since there are many types of neural network models and their adaptability varies, if the results of multiple image reconstructions do not meet expectations, it proves that the neural network model may have poor adaptability. During the evaluation and optimization process, the neural network model is changed to find the most suitable neural network model, thereby improving the accuracy of subsequent data processing.

[0068] As one of the optional embodiments of the present application, the evaluation and optimization step also includes: if the results of multiple image reconstructions do not meet expectations, then changing the combination of constraints to repeat the training, wherein the constraints include physical constraints, spatial constraints, prior knowledge constraints, etc.

[0069] In the embodiment of the present application, by trying multiple combinations of constraint conditions, the constraint conditions that best suit the current detection process can be found during the neural network training process, so as to remove noise in the capacitance data to the greatest extent, improve the smoothness of the data, and thereby improve the spatial resolution of the capacitance image, making the imaging results more accurate and detailed.

[0070] As one of the optional embodiments of the present application, before the data processing step, a data preprocessing step is also included: processing abnormal values ​​and noise in the initial capacitance data C1.

[0071] Specifically, noise removal: Noise is a common interference factor in capacitance measurement data, which may come from electronic equipment, environmental interference, etc. In order to improve data quality, filtering technology can be used to denoise the data. Common filtering methods include mean filtering, median filtering, Gaussian filtering, etc., and the appropriate filtering method is selected according to the actual situation. Correction: In ECT technology, capacitance measurement data may be affected by factors such as electrode spacing and electrode shape, resulting in deviations in the measurement results. In order to eliminate these deviations, correction operations are required. Correction can be performed through experimental measurement and numerical simulation, and the appropriate correction method is selected according to the actual situation. Normalization: In order to make different measurement data comparable, the capacitance measurement data can be normalized. Common normalization methods include maximum and minimum value normalization, mean variance normalization, etc. Normalization can map the data to a specific range so that the data has similar scale and distribution characteristics.

[0072] In an embodiment of the present application, in the principal component analysis method, there may be a situation where some data is not captured sufficiently. This is because some abnormal values ​​or noise are collected in the initial capacitance data C1 due to circuit noise, capacitance sensor installation position and precision control. Since the principal component analysis method is calculated based on the covariance matrix or correlation matrix of the data, the presence of abnormal values ​​or noise may cause the calculation result of the covariance matrix to deviate from the actual situation, thereby affecting the extraction and interpretation of the principal components. Therefore, preprocessing the data in advance can reduce the impact on the results of the principal component analysis process and ensure the accuracy of the results.

[0073] As one of the optional embodiments of the present application, as one of the optional embodiments of the present application, the characteristic parameters of the acquisition system include: the geometric shape of the capacitive sensor, the electrode spacing, the electrode size, etc.

[0074] As one of the optional embodiments of the present application, after the evaluation and optimization step, it also includes: if the image shows that the particle diameter D≥0.5cm, or the image shows that the solubility is continuously high within T minutes in each monitoring cycle, a warning message is issued, wherein 15≤T≤25.

[0075] On the other hand, a lubricating oil detection system is also provided, including: a plurality of capacitance sensors, which are arranged in a ring shape and are evenly fitted on the outer wall of the pipeline; a data acquisition module, including a driving unit and a transmission unit, wherein the driving unit is used to apply an alternating voltage to stimulate the area to be tested; a terminal, which is used to analyze and process the collected capacitance data and reconstruct an image; wherein the transmission unit is electrically connected to the capacitance sensor to convert the current signal into a digital signal and transmit it to the terminal

[0076] The above description is only a partial implementation method of the embodiments of the present application and does not constitute any form of limitation on the application. The protection scope of the embodiments of the present application is not limited thereto. Any simple modifications, equivalent changes and modifications that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed in the embodiments of the present application should be covered within the protection scope of the embodiments of the present application.

Claims

1. A method for detecting lubricating oil wear particles, characterized in that: include: Data acquisition: Capacitive sensors are arranged on the outer wall of the pipeline to obtain the initial capacitance data C1 of the area to be measured; Data processing 1: performing dimensionality reduction processing on the initial capacitance data C1 by a principal component analysis method to obtain second-order capacitance data C2; Data processing 2: Input the second-order capacitance data C2 into the neural network model, add constraints, and obtain the third-order capacitance data C3; Image reconstruction: Input the third-order capacitance data C3 into the finite element analysis model to obtain the image of the area to be measured; Evaluation and optimization: Evaluate the reconstructed image. If the image is not clear, repeat the data processing and image reconstruction process until the image quality meets the requirements.

2. The lubricating oil wear particle detection method according to claim 1, characterized in that: The steps of establishing a neural network model are also included between data processing one and data processing two: Solution to the direct problem: Based on the collected initial capacitance data C1 and the characteristic parameters of the acquisition system, calculate the theoretical capacitance value distribution and record it as the target capacitance data C0; Set up a neural network model suitable for processing capacitance data, select the appropriate network structure and number of layers, and add constraints; The second-order capacitance data C2 is used as input data, and the target capacitance data C0 is used as output data. The neural network model is trained by the second-order capacitance data C2 and the target capacitance data C0 so that the neural network model learns the mapping relationship between the input and output data.

3. The lubricating oil wear particle detection method according to claim 2, characterized in that: The steps of solving the positive problem include: Establish a finite element model: divide the target object into discrete small cells, each of which has a certain conductivity; Constructing a direct problem equation: constructing a direct problem equation according to the initial capacitance data C1 and the finite element model, wherein the direct problem equation is used to indicate the relationship between the initial capacitance data C1 and the conductivity of the area to be measured; Direct problem solution: By solving the direct problem equation, the conductivity distribution of the area to be measured is obtained, which is recorded as the target capacitance data C0.

4. The lubricating oil wear particle detection method according to claim 2, characterized in that: The evaluation and optimization step also includes: if the results of multiple image reconstructions do not meet expectations, different types of neural network models are replaced, and the step of establishing the neural network model is repeated, wherein the neural network model includes a convolutional neural network, a recurrent neural network, and a feedforward neural network.

5. The lubricating oil wear particle detection method according to claim 4, characterized in that: The evaluation and optimization step also includes: if the results of multiple image reconstructions do not meet expectations, then changing the combination of constraint conditions to repeat the training, wherein the constraint conditions include physical constraints, spatial constraints, prior knowledge constraints, etc.

6. The lubricating oil wear particle detection method according to claim 2 or 5, characterized in that: Before the data processing step, a data preprocessing step is also included: processing abnormal values ​​and noise in the initial capacitance data C1.

7. The lubricating oil wear particle detection method according to claim 1, characterized in that: The step of performing dimensionality reduction processing on the initial capacitance data C1 by the principal component analysis method to obtain the second-order capacitance data C2 comprises: The initial capacitance data C1 is standardized so that the mean of each feature is 0 and the variance is 1; Calculate the covariance matrix of the standardized capacitance data C1; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues ​​and corresponding eigenvectors; Sort the eigenvalues ​​in descending order, and select the eigenvectors corresponding to the first k eigenvalues ​​as the principal components; The standardized capacitance data C1 is projected onto the selected principal component to obtain the second-order capacitance data C2 after dimension reduction.

8. The lubricating oil wear particle detection method according to claim 2, characterized in that: The characteristic parameters of the acquisition system include: the geometric shape of the capacitive sensor, the electrode spacing, the electrode size, etc.

9. The lubricating oil wear particle detection method according to claim 1, characterized in that: After the evaluation and optimization step, the method further includes: if the image shows that the particle diameter D is ≥ 0.5 cm, or the image shows that the solubility is continuously high within T minutes in each monitoring cycle, a warning message is issued, wherein 15≤T≤25.

10. A lubricating oil detection system, characterized in that: include: Capacitive sensors are provided in multiple numbers and are evenly arranged in a ring shape on the outer wall of the pipeline; The data acquisition module includes a driving unit and a transmission unit, wherein the driving unit is used to apply an alternating voltage to stimulate the area to be measured; A terminal, used for analyzing and processing the collected capacitance data and reconstructing images; The transmission unit is electrically connected to the capacitive sensor to convert the current signal into a digital signal and transmit it to the terminal.

Images