An in-situ ferromagnetic wear particle detection sensor and detection method

Abstract

This invention relates to an in-situ ferromagnetic wear particle detection sensor and method, comprising an LC oscillation circuit, an inductor-to-digital converter, and a microcontroller. The LC oscillation circuit includes an induction coil and a first capacitor connected to the induction coil. The detection sensor also includes a permanent magnet, with the induction coil arranged on one side of the permanent magnet. When a ferromagnetic particle enters the induction coil under the attraction of the permanent magnet, it causes a change in the inductance of the induction coil and a change in the equivalent parallel resistance between the induction coil and the particle. The inductor-to-digital converter receives the inductance data and equivalent parallel resistance data output from the LC oscillation circuit and converts them into resonant frequency and proximity output. The microcontroller receives and processes the resonant frequency and proximity data and outputs the processed data. This invention has high detection sensitivity, can distinguish particle characteristics, and provides real-time feedback on the mechanical operating status, facilitating timely maintenance by operators.

Description

Technical Field

[0001] This invention relates to the field of mechanical condition monitoring technology, specifically to an in-situ ferromagnetic wear particle detection sensor and detection method. Background Technology

[0002] During the operation of mechanical equipment, friction pairs generate metal wear particles. These particles are suspended in the lubricating oil and are an important indicator of the equipment's condition. Although traditional laboratory oil analysis technology is accurate, it has drawbacks such as detection lag, the need for manual sampling, and the inability to monitor in real time, which often leads to missed early fault warnings under dynamic operating conditions.

[0003] Among the related technologies, online monitoring mainly includes methods such as magnetic, inductive, capacitive, resistive, and charge-based methods, but these methods still have many shortcomings: limited detection sensitivity, poor anti-interference ability, and unstable real-time performance. Summary of the Invention

[0004] In view of this, the present invention provides an in-situ ferromagnetic wear particle detection sensor and detection method, thereby solving or at least alleviating one or more of the above-mentioned problems and other problems existing in the prior art.

[0005] To achieve the aforementioned objectives, the technical solution adopted by the present invention is as follows:

[0006] An in-situ ferromagnetic wear particle detection sensor, the detection sensor including an LC oscillation circuit, the LC oscillation circuit including an induction coil and a first capacitor connected to the induction coil;

[0007] The detection sensor also includes a permanent magnet, and the induction coil is arranged on one side of the permanent magnet. When the ferromagnetic particles enter the induction coil under the attraction of the permanent magnet, it causes a change in the inductance of the induction coil and a change in the equivalent parallel resistance between the induction coil and the particles.

[0008] The detection sensor also includes an inductor-to-digital converter connected to the LC oscillation circuit and a microcontroller connected to the inductor-to-digital converter. The inductor-to-digital converter is used to receive the inductance data and equivalent parallel resistance data output by the LC oscillation circuit and convert them into resonant frequency and proximity output. The microcontroller is used to receive the resonant frequency and proximity data and process them. The microcontroller is also used to output the processed data to the outside.

[0009] Optionally, in the aforementioned in-situ ferromagnetic wear particle detection sensor, the detection sensor further includes a compensation circuit. The compensation circuit includes a reference coil and a second capacitor connected to the reference coil. The reference coil is disposed on the other side of the permanent magnet, and the reference coil and the induction coil are symmetrically arranged. The inductor-to-digital converter is also connected to the compensation circuit. The inductor-to-digital converter is also used to synchronously acquire environmental interference signals, and the microcontroller is also used to perform error compensation on the acquired signals.

[0010] Optionally, the detection sensor further includes a housing and an insulating support assembly, one end of which has a blind groove extending along the length of the support assembly;

[0011] The permanent magnet is inserted into the blind slot. The permanent magnet has a detection end for adsorbing ferromagnetic particles. The induction coil and the reference coil are respectively spaced around the support assembly and the induction coil is set close to the detection end. A receiving groove for accommodating ferromagnetic particles is formed between the end faces of the induction coil and the detection end.

[0012] The outer casing covers the periphery of the induction coil, the reference coil, and the support assembly, and the receiving groove is connected to the outside. When ferromagnetic particles enter the receiving groove under the attraction of the permanent magnet, they cause changes in the inductance and equivalent parallel resistance of the induction coil.

[0013] In the aforementioned in-situ ferromagnetic wear particle detection sensor, optionally, the support assembly includes a support column with the blind groove at one end and a protrusion formed around the support column and extending along the length direction of the support column. The induction coil and the reference coil are respectively sleeved on the support column and located on both sides of the protrusion. The receiving groove is formed between the end faces of the support column and the detection end.

[0014] In the aforementioned in-situ ferromagnetic wear particle detection sensor, optionally, the support component is integrally formed, and the support component is made of plastic.

[0015] In the aforementioned in-situ ferromagnetic wear particle detection sensor, optionally, the outer surface of the housing is provided with an external thread, which is used to connect the detection sensor to a mechanical lubrication circuit.

[0016] In the aforementioned in-situ ferromagnetic wear particle detection sensor, optionally, the induction coil and the reference coil are the same coil.

[0017] The second technical solution adopted by the present invention is: a detection method based on the above-mentioned in-situ ferromagnetic wear particle detection sensor, the detection method comprising the following steps:

[0018] S1. Initialize the detection sensor;

[0019] S2. The inductor-to-digital converter detects changes in inductance and equivalent parallel resistance in real time, converts them into resonant frequency and proximity data, and sends them to the microcontroller.

[0020] S3. The microcontroller determines particle characteristics based on changes in resonant frequency and proximity.

[0021] S4. The microcontroller outputs the processed data signal to the host computer to display the particle characteristics in real time.

[0022] Optionally, in the aforementioned detection method, the detection method further includes the microcontroller analyzing the collected resonant frequency and proximity data using a fuzzy C-means clustering algorithm, setting the number of clusters to n, where n is an integer greater than or equal to 2, and the value of n corresponds to different degrees of pollution. The microcontroller then outputs the processed data signal to a host computer to display the pollution level classification in real time.

[0023] Quantitative assessment of pollution levels can provide accurate data support for preventative maintenance, helping to reduce equipment downtime losses and extend equipment lifespan.

[0024] Optionally, n is 3, corresponding to pollution levels of light, moderate and heavy pollution. When the pollution level is detected to be moderate or above, an early warning signal is triggered, prompting mechanical maintenance.

[0025] Optionally, the initialization detection sensor includes a resonant frequency reference value and a proximity reference value for recording a particle-free state;

[0026] In step S3, the determination of particle characteristics includes determining that the particle is a fine particle when the real-time resonant frequency is lower than the resonant frequency reference value and the real-time proximity is lower than the proximity reference value, wherein the fine particle size is less than 0.3 mm.

[0027] When the real-time resonant frequency is lower than the resonant frequency reference value and the real-time proximity is higher than the proximity reference value, the particle is determined to be a coarse particle, and the size of the coarse particle is greater than 0.3 mm.

[0028] Optionally, the reduction in the resonant frequency of the coarse particles is greater than the reduction in the resonant frequency of the fine particles.

[0029] Optionally, the detection method further includes triggering an early warning signal when the coarse particles are detected, prompting mechanical maintenance.

[0030] By adopting the above technical solution, this detection method indicates the mechanical wear status through multiple dimensions, including particle size, pollution level assessment, etc., providing multi-faceted early warning, which helps to detect the wear status in a timely manner and thus take corresponding measures.

[0031] This detection sensor can be used in oil-lubricated equipment such as coal mining machines, wind turbine gearboxes, ships, and aircraft engines.

[0032] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art:

[0033] The detection sensor of this invention has high detection sensitivity and can reliably detect ferromagnetic particles larger than 0.25mg. It can also distinguish particle characteristics and provide real-time feedback on the mechanical operating status, facilitating timely maintenance by operators. Attached Figure Description

[0034] The disclosure of this invention will become more apparent from the accompanying drawings. It should be understood that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention.

[0035] Figure 1 This is a schematic diagram of the structure of an in-situ ferromagnetic wear particle detection sensor according to an embodiment of the present invention;

[0036] Figure 2 for Figure 1 A side view of the detection sensor;

[0037] Figure 3 for Figure 2 Sectional view along line AA;

[0038] Figure 4 for Figure 1 An explosion diagram of the detection sensor;

[0039] Figure 5 This is a circuit system block diagram of the detection sensor according to an embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of the equivalent circuit of the induction coil and the particle when the particle enters the induction coil in an embodiment of the present invention.

[0041] Figure 7 This is a schematic diagram showing the sensitivity test results of the sensor.

[0042] Figure 8 This is a schematic diagram showing the relationship between particle size and mass on the frequency of the detection sensor.

[0043] Figure 9 This is a schematic diagram illustrating the relationship between particle size and mass on the proximity of the detection sensor.

[0044] Figure 10 A schematic diagram showing the influence of particles of different sizes on the inductance of the detection sensor;

[0045] Figure 11 A schematic diagram showing the influence of particles of different sizes on the resistance of the detection sensor;

[0046] Figure 12 A schematic diagram showing the influence of particles of different sizes on the impedance of the detection sensor;

[0047] Figure 13 A schematic diagram showing the influence of particles of different sizes on the resonant frequency and proximity of the detection sensor;

[0048] Figure 14 A schematic diagram illustrating the pollution level classification results of fuzzy C-means clustering for detecting sensors;

[0049] In the picture:

[0050] 1. Induction coil; 1a. First end face; 1b. Second end face; 2. Reference coil; 3. Permanent magnet; 4. Support assembly; 41. Support column; 42. Protrusion; 5. Housing; 6. External thread; 7. Receiving groove; 7a. Groove bottom; 8. Blind groove; 9. Inductor-to-digital converter; 10. LC oscillation circuit; 11. Compensation circuit; 12. Microcontroller; 13. Host computer; 14. Sealant. Detailed Implementation

[0051] Referring to the accompanying drawings and specific embodiments, the structure, composition, features, and advantages of the in-situ ferromagnetic wear particle sensor of the present invention will be described below by way of example; however, all descriptions should not be construed as limiting the present invention in any way.

[0052] For any single technical feature described or implied in the embodiments submitted herein, or any single technical feature shown or implied in the various drawings, the present invention still operates in any combination or deletion among these technical features or their equivalents without any technical obstacle, and thus these further embodiments according to the present invention should also be considered within the scope of the description herein.

[0053] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0054] refer to Figures 1 to 5 As shown, the in-situ ferromagnetic wear particle detection sensor includes an induction coil 1, a reference coil 2, a permanent magnet 3, a support assembly 4, and a housing 5.

[0055] The induction coil 1 and the reference coil 2 use the same coil. The reference coil 2 is used to compensate for errors caused by environmental factors such as temperature drift, power fluctuation and material aging, so as to ensure detection stability.

[0056] Support component 4 is made of insulating material and is used to fix the coil and permanent magnet. Figure 4 As can be seen, one end of the support component 4 has a blind groove 8 extending along the length direction of the support component 4. The permanent magnet 3 is inserted into the blind groove 8. The induction coil 1 and the reference coil 2 are respectively sleeved on the periphery of the support component 4. The induction coil 1 and the reference coil 2 are respectively symmetrically arranged on opposite sides of the permanent magnet 3. The permanent magnet has a detection end for adsorbing ferromagnetic particles. The induction coil 1 is arranged close to the detection end. A receiving groove 7 for accommodating ferromagnetic particles is formed between the end faces of the induction coil 1 and the detection end of the permanent magnet 3.

[0057] like Figure 3 As shown, the end face of the detection end of the permanent magnet 3 is the bottom surface 7a of the receiving groove 7; the induction coil 1 has two end faces, namely the first end face 1a facing the reference coil 2 and the second end face 1b opposite to the first end face 1a. The first end face 1a of the induction coil 1 is basically flush with the bottom surface 7a of the groove, or the bottom surface 7a of the groove is located between the first end face 1a and the second end face 1b, and the bottom surface 7a of the groove is set closer to the first end face 1a.

[0058] Specifically, after the permanent magnet 3 is inserted into the blind slot 8, all the permanent magnets are located in the blind slot 8. The receiving slot 7 is also formed between the end face of the support component 4 and the detection end of the permanent magnet. The end face of the support component 4 is basically flush with the second end face 1b of the reference coil 2.

[0059] In an optional embodiment, the support component 4 includes a support column 41 and a protrusion 42 formed around the support column 41 and extending along the length of the support column 41. The induction coil 1 and the reference coil 2 are respectively sleeved on the support column 41 and located on both sides of the protrusion 42. The protrusion 42 separates the induction coil 1 and the reference coil 2. The blind slot 8 is formed on the support column 41.

[0060] In some embodiments, the support column 41 and the protrusion 42 are integrally formed and are both made of plastic. The wall thickness of the blind groove 8 located at the support column 41 is 1 mm.

[0061] The outer casing 5 covers the induction coil 1, reference coil 2, and support assembly 4. The gap between the outer casing 5 and the induction coil 1, reference coil 2, and support assembly 4 is sealed with sealant 14. The receiving groove 7 is open to the outside. After the outer casing 5 is installed, the reference coil 2 is housed inside the outer casing 5. Ferromagnetic particles will enter the receiving groove 7 under the attraction of the permanent magnet 3, and thus enter the interior of the induction coil 1. The outer surface of the outer casing 5 is provided with external threads, which are used to connect the detection sensor to the lubrication circuit of the machine to be detected.

[0062] See also Figure 5 The detection sensor also includes an LC oscillation circuit 10, a compensation circuit 11, an inductor-to-digital converter 9, and a microcontroller 12, wherein:

[0063] The LC oscillation circuit 10 includes the aforementioned induction coil 1 and a first capacitor C1 connected to the induction coil 1. The compensation circuit includes the aforementioned reference coil 2 and a second capacitor C2 connected to the reference coil 2. The LC oscillation circuit 10 and the compensation circuit 11 are respectively connected to the inductor-to-digital converter 9, and the inductor-to-digital converter 9 is connected to the microcontroller 12.

[0064] When the ferromagnetic particles enter the receiving tank 7 under the adsorption of the permanent magnet 3, the ferromagnetic particles enter the induction coil 1, which will cause changes in the inductance of the induction coil 1 and changes in the equivalent parallel resistance of the induction coil and the particles.

[0065] Figure 6 The diagram illustrates the equivalent circuit principle of the induction coil and the particle when they enter the induction coil. When a ferromagnetic particle enters the induction coil, there is mutual inductance between the coil and the particle. Ls represents the inductance of the induction coil, Rs represents the initial resistance, and L(d) and R(d) represent the varying inductance and resistance, depending on the particle size, number, and distance. Different particles entering the induction coil cause changes in the inductance of coil 1 and the equivalent parallel resistance. The inductance-to-digital converter 9 can simultaneously detect both the inductance data and the equivalent parallel resistance data, then convert them into resonant frequency and proximity outputs. The resonant frequency and proximity are output parameters of the inductance-to-digital converter; the resonant frequency corresponds to the inductance, and the proximity corresponds to the equivalent parallel resistance.

[0066] The microcontroller 12 receives and processes the resonant frequency and proximity data, then outputs the processed data. For example, if the microcontroller 12 is connected to the host computer 13, it can transmit the processed data to the host computer 13 for real-time visualization and recording. Alternatively, the microcontroller can transmit data to the host computer via a USB interface using the Modbus protocol. The host computer can be a computer, mobile phone, or other visualization device.

[0067] The inductor-to-digital converter 9 is also used to synchronously acquire environmental interference signals, and the microcontroller 12 is also used to perform error compensation on the acquired signals, thereby improving the detection stability in complex oil environments.

[0068] In some alternative embodiments, such as the microcontroller 12 being an STM32F107, the inductor-to-digital converter 9 being an LDC1001-Q1, and the first capacitor C1 and the second capacitor C2 being 2nF C0G (or NPO) type dielectric capacitors, the circuit resonant frequency is in the range of several hundred kHz.

[0069] In some embodiments, the permanent magnet 3 is a neodymium iron boron (NdFeB) permanent magnet with dimensions of φ10mm×10mm; both the induction coil 1 and the reference coil 2 are made of enameled copper wire, with an inner diameter of 6mm, an axial length of 1.4mm, a thickness of 1.3mm, an insulation spacing of 0.164mm between adjacent coils, a diameter of 0.15mm for the enameled copper wire, and a theoretical inductance of 39μH.

[0070] Another embodiment of the present invention also provides a detection method based on the detection sensor of the embodiment, comprising the following steps:

[0071] Step 1: Installation of the detection sensor

[0072] Install the detection sensor into the mechanical lubrication circuit to ensure that the probe is in full contact with the oil.

[0073] Step 2: Initialize the detection sensor

[0074] Initialize the sensor, start the host computer and the sensor, initialize the microcontroller and inductor-to-digital converter, set the initial value of the resonant frequency, the data acquisition frequency and other parameters, and record the resonant frequency reference value and proximity reference value of the sensor in the particle-free state.

[0075] Step 3: Signal Acquisition

[0076] The inductor-to-digital converter detects changes in inductance and equivalent parallel resistance in real time, converts them into resonant frequency and proximity data, and then sends them to the microcontroller.

[0077] Step 4: Particle Feature Judgment

[0078] The microcontroller determines particle characteristics based on changes in resonant frequency and proximity, and outputs the processed data signal to the host computer for real-time display of particle characteristics.

[0079] In this step, particle characteristics are determined as follows:

[0080] When the real-time resonant frequency is lower than the resonant frequency reference value and the real-time proximity is lower than the proximity reference value, the particles are judged to be fine particles, and the size of fine particles is less than 0.3mm.

[0081] When the real-time resonant frequency is lower than the resonant frequency reference value and the real-time proximity is higher than the proximity reference value, the particles are judged to be coarse particles with a size of 0.3 mm or more.

[0082] Furthermore, the reduction in resonant frequency of coarse particles is greater than that of fine particles.

[0083] If the particles generated in the lubrication circuit are fine, it is usually due to normal wear. Their size is much smaller than the skin depth, and the eddy current effect is negligible. The main effect is magnetization, which increases coil inductance, lowers the resonant frequency, and reduces proximity. The smaller the particle mass, the smaller the decrease in resonant frequency and proximity. As described below, the smallest detectable particle mass can reach 0.25 mg. If the particles generated in the lubrication circuit are coarse, it is usually due to abnormal wear. Their size is close to or exceeds the skin depth. Magnetization and eddy current effects interact, significantly lowering the resonant frequency and greatly increasing the proximity.

[0084] Since particles of various sizes are usually present in actual lubrication circuits, in order to more comprehensively and accurately detect the mechanical condition, the particle contamination level is also assessed. This detection method also includes the following step five.

[0085] Step 5: Pollution Level Assessment

[0086] The microcontroller analyzes the collected resonant frequency and proximity data using a fuzzy C-means clustering algorithm, setting the number of clusters to n (e.g., n = 3, corresponding to light, moderate, and heavy pollution respectively). Euclidean distance is used as a similarity measure, and the objective function is iteratively optimized to calculate the sample membership degree. Based on the principle of maximum membership degree, samples with light pollution are concentrated in the high frequency and low proximity region (higher normalized frequency and lower normalized proximity), samples with moderate pollution are located in the middle region, and samples with heavy pollution are concentrated in the low frequency and high proximity region.

[0087] The pollution level is output to the host computer for real-time display. When the pollution level is detected to be moderate or above, an early warning signal is triggered, prompting mechanical maintenance.

[0088] When the resonant frequency and proximity change sharply, the particle generation rate may increase.

[0089] In practical applications, when the pollution level is moderate or higher, or when coarse particles are detected or the particle generation rate is accelerated, an early warning signal is triggered, prompting mechanical maintenance.

[0090] Performance testing of the detection sensor:

[0091] The following performance tests were conducted in a simulated lubrication circuit, with the test sensors installed on the simulated lubrication circuit.

[0092] 1. Fine particle detection test

[0093] The research subjects were ferromagnetic wear particles generated by the cast iron rocker arm gearbox of a coal mining machine. The particles were obtained by vibrating sieving and were 50 mesh (approximately 270 μm), 75 mesh (approximately 200 μm), 100 mesh (approximately 150 μm), 120 mesh (approximately 120 μm), and 150 mesh (approximately 100 μm) particle samples.

[0094] 1.1 Sensitivity Detection

[0095] The sensitivity test method involves introducing samples of different masses into the detection area and recording the changes in the resonant frequency and proximity readings of the detection sensor.

[0096] The results are as follows Figure 7 As shown in the figure, the minimum detectable mass is 0.25 mg (true positive rate, TPR > 0.95).

[0097] It is evident that this detection sensor has a stable response to micron-sized particles, and the detection of fine particles mainly depends on the resonant frequency.

[0098] 1.2 Signal Response Detection

[0099] Signal response detection was performed on particles of 50 mesh, 75 mesh, 100 mesh, 120 mesh and 150 mesh, with each size of particle being tested using three different masses (20 mg, 40 mg and 60 mg respectively).

[0100] The results are as follows Figure 8 and Figure 9 As shown in the figure, the resonant frequency decreases significantly and the proximity also decreases as the mass increases and the particle size decreases.

[0101] 2. Coarse particle detection test

[0102] The research subjects were spherical iron particles with diameters of 0.5 mm, 1 mm, 2 mm and 3 mm, which were added to the simulated lubrication circuit at a standard mass of 300 mg.

[0103] First, a KEYSIGHTE49901 impedance analyzer was used to scan iron particles of different sizes, including fine particles, and the scan also included a particle-free case. The fine particles were a mixture of 50 mesh, 75 mesh, 100 mesh, 120 mesh and 150 mesh in equal mass. The effect of particle size on key electrical parameters in the entire spectrum, including inductance, resistance and impedance, was evaluated.

[0104] The results are as follows Figures 10 to 12 As shown, by Figure 10 As can be seen, the inductance changes significantly with increasing frequency. At low frequencies (0-1000kHz), all samples maintain high inductance values; when the frequency rises to 2500kHz, the inductance of the 3mm particle sample drops below that of the particle-free sample. The curves for different material sizes show clear stratification, indicating that size plays a decisive role in high-frequency inductance behavior. Figure 11 As can be seen, the eddy current effect is significant for particles larger than 0.5 mm. The resistivity curve of fine particles is clearly different from that of other particles. Figure 12 As can be seen, magnetization and total eddy current effects are particularly pronounced in particle samples larger than 0.5 mm.

[0105] In this simulated lubrication circuit, the resonant frequencies and proximity data of particles of different sizes were also recorded in real time, and the results are as follows: Figure 13 As shown in the figure, the blue curve (proximity curve) indicates that the initial proximity reference value in the particle-free state is approximately 8640.7, with a standard deviation of 3.39 units. In the fine-particle state, it is lower than the proximity reference value; in the coarse-particle state, it is higher than the proximity reference value, reaching a peak at 3 mm. The red curve (resonant frequency curve) indicates that the initial frequency reference value in the particle-free state is approximately 570.75 kHz, with a standard deviation of 15.29 Hz. In the fine-particle state, it is lower than the frequency reference value; in the coarse-particle state, it is lower than the frequency in the fine-particle state, and the frequency increases with larger particle size, but remains lower than the frequency reference value.

[0106] 3. Fuzzy C-means Clustering (FCM) Test

[0107] The testing method is as follows: Three benchmark samples were configured, namely fine particles (50-150 mesh), coarse particles (0.5-1 mm), and metal fragments (2-3 mm), to simulate actual oil contamination scenarios. Signals were continuously collected for 2.5 minutes for each sample (a total of 150 data points). Fuzzy C-means (FCM) algorithm was used for cluster analysis, with the number of clusters set to 3 and the membership threshold set to 70%.

[0108] The results are as follows Figure 14 As shown, lightly contaminated samples (green clusters) are concentrated in high-frequency, low-proximity areas; moderately contaminated samples (red clusters) are located in the middle area; and heavily contaminated samples (blue clusters) are concentrated in low-frequency, high-proximity areas, which can effectively assess the oil contamination level.

[0109] Application Examples

[0110] The detection sensor of this invention was installed in the gearbox lubrication circuit of a 1.5MW wind turbine in a wind farm and continuously monitored for 6 months.

[0111] During the monitoring period, the particles detected by the sensors in the first four months were mainly fine particles, and the pollution level remained at a light level, which was consistent with the normal operation of the wind turbine. In the middle of the fifth month, the resonant frequency was detected to drop sharply and the proximity was increased. The FCM clustering results showed that the pollution level had risen to a moderate level, and the proportion of coarse particles (>300μm) had increased, indicating that the gearbox may have experienced early abnormal wear. Based on the early warning information, the maintenance team promptly shut down the machine for inspection and found that there was slight pitting on the gear teeth. After repair, the sensor monitoring data returned to normal, avoiding high losses caused by the expansion of the gearbox failure.

[0112] Compared to existing detection sensors, the detection sensor in this application embodiment has at least the following characteristics:

[0113] High sensitivity and accurate detection: This detection sensor can reliably detect ferromagnetic particles larger than 0.25mg, and can distinguish between fine particles smaller than 300μm and coarse particles larger than 300μm, solving the problem of insufficient sensitivity of existing technologies for micro-sized particles.

[0114] Real-time and non-invasive monitoring: The detection sensor is directly installed in the lubrication circuit, which can collect wear particle signals in real time without sampling. The data acquisition and analysis latency is low and does not affect the normal circulation of lubricating oil and mechanical operation, overcoming the shortcomings of high latency and strong invasiveness of traditional laboratory analysis.

[0115] Strong anti-interference capability: By using permanent magnets to directionally adsorb ferromagnetic particles, interference from non-target particles is reduced. At the same time, the reference coil compensates for errors caused by temperature, power supply and material aging. The multi-channel design significantly improves the detection stability in complex oil environments.

[0116] High practical value: It can be directly integrated into the lubrication system of mechanical equipment in industries such as wind power, petrochemicals, and high-speed rail. It achieves quantitative assessment of pollution level through fuzzy C-means clustering, providing accurate data support for preventive maintenance, which helps reduce equipment downtime losses and extend equipment service life.

[0117] In addition, in the description of this invention, unless otherwise specified and limited, it should be noted that the terms "installation", "connection" and "linking" should be interpreted broadly. For example, they can refer to direct connection or indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0118] The above embodiments are only used to illustrate the embodiments of the present invention, and are not intended to limit the embodiments of the present invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of the present invention. Therefore, all equivalent technical solutions also fall within the scope of the embodiments of the present invention, and the patent protection scope of the embodiments of the present invention should be defined by the claims.

Claims

1. An in-situ ferromagnetic wear particle detection sensor, characterized in that, The detection sensor includes an LC oscillation circuit, which includes an induction coil and a first capacitor connected to the induction coil. The detection sensor also includes a permanent magnet, and the induction coil is arranged on one side of the permanent magnet. When the ferromagnetic particles enter the induction coil under the attraction of the permanent magnet, it causes a change in the inductance of the induction coil and a change in the equivalent parallel resistance between the induction coil and the particles. The detection sensor also includes an inductor-to-digital converter connected to the LC oscillation circuit and a microcontroller connected to the inductor-to-digital converter. The inductor-to-digital converter is used to receive the inductance data and equivalent parallel resistance data output by the LC oscillation circuit and convert them into resonant frequency and proximity output. The microcontroller is used to receive the resonant frequency and proximity data and process them. The microcontroller is also used to output the processed data to the outside. The detection sensor also includes a compensation circuit, which includes a reference coil and a second capacitor connected to the reference coil. The reference coil is disposed on the other side of the permanent magnet, and the reference coil and the induction coil are symmetrically arranged. The inductor-to-digital converter is also connected to the compensation circuit. The inductor-to-digital converter is also used to synchronously acquire environmental interference signals. The microcontroller is also used to perform error compensation on the acquired signals. The induction coil and the reference coil are the same coil.

2. The in-situ ferromagnetic wear particle detection sensor according to claim 1, characterized in that, The detection sensor also includes a housing and an insulating support assembly, one end of which has a blind groove extending along the length of the support assembly; The permanent magnet is inserted into the blind slot. The permanent magnet has a detection end for adsorbing ferromagnetic particles. The induction coil and the reference coil are respectively spaced around the support assembly and the induction coil is set close to the detection end. A receiving groove for accommodating ferromagnetic particles is formed between the end faces of the induction coil and the detection end. The outer casing covers the periphery of the induction coil, the reference coil, and the support assembly, and the receiving groove is connected to the outside. When ferromagnetic particles enter the receiving groove under the attraction of the permanent magnet, they cause changes in the inductance and equivalent parallel resistance of the induction coil.

3. The in-situ ferromagnetic wear particle detection sensor according to claim 2, characterized in that, The support assembly includes a support column with the blind groove at one end and a protrusion formed around the support column and extending along the length of the support column. The induction coil and the reference coil are respectively sleeved on the support column and located on both sides of the protrusion. The receiving groove is formed between the end faces of the support column and the detection end; and / or, The support component is integrally formed, and the material of the support component is plastic; and / or, The outer surface of the housing is provided with external threads, which are used to connect the detection sensor to the mechanical lubrication circuit.

4. A detection method based on the in-situ ferromagnetic wear particle detection sensor according to any one of claims 1 to 3, characterized in that, The detection method includes the following steps: S1. Initialize the detection sensor; S2. The inductor-to-digital converter detects changes in inductance and equivalent parallel resistance in real time, converts them into resonant frequency and proximity data, and sends them to the microcontroller. S3. The microcontroller determines particle characteristics based on changes in resonant frequency and proximity. S4. The microcontroller outputs the processed data signal to the host computer to display the particle characteristics in real time.

5. The detection method according to claim 4, characterized in that, The detection method further includes the microcontroller analyzing the collected resonant frequency and proximity data using a fuzzy C-means clustering algorithm, setting the number of clusters to n, where n is an integer greater than 2, and the value of n corresponds to different degrees of pollution. The microcontroller then outputs the processed data signal to a host computer to display the pollution level classification in real time.

6. The detection method according to claim 5, characterized in that, The value of n is 3, corresponding to pollution levels of light, moderate, and heavy pollution. When a pollution level of moderate or higher is detected, an early warning signal is triggered, prompting mechanical maintenance.

7. The detection method according to claim 4, characterized in that, The initialization detection sensor includes a resonant frequency reference value and a proximity reference value for recording a particle-free state; In step S3, the determination of particle characteristics includes determining that the particle is a fine particle when the real-time resonant frequency is lower than the resonant frequency reference value and the real-time proximity is lower than the proximity reference value, wherein the fine particle size is less than 0.3 mm. When the real-time resonant frequency is lower than the resonant frequency reference value and the real-time proximity is higher than the proximity reference value, the particle is determined to be a coarse particle, and the size of the coarse particle is greater than 0.3 mm.

8. The detection method according to claim 7, characterized in that, The decrease in the resonant frequency of the coarse particles is greater than the decrease in the resonant frequency of the fine particles; and / or, The detection method also includes triggering an early warning signal when the coarse particles are detected, prompting mechanical maintenance.

Images