Method for predicting the lifetime of an electromagnetic brake based on the voltage inflection point and the steady-state time characteristic

Abstract

The application discloses a method for predicting the service life of an electromagnetic brake based on voltage inflection points and steady-state time characteristics, and relates to the technical field of electromagnetic brakes. The method comprises the following steps: S1, acquiring a sampling voltage waveform; S2, positioning a power-on time and intercepting and analyzing the voltage waveform; S3, analyzing a smooth difference sequence; S4, determining an inflection point time and an inflection point voltage; S5, determining a steady-state time and a steady-state voltage; S6, calculating a first gap value; S7, calculating a second gap value; S8, calculating a final gap value; S9, analyzing the current health level of the brake according to the final gap value and outputting a prediction result; by detecting the key inflection point position in the voltage waveform, the voltage characteristics corresponding to the starting time of mechanical movement are extracted, and a gap estimation model is established in combination with the steady-state time characteristics, so that quantitative analysis of the wear state of the brake is realized; the inflection point time is used as the main estimation characteristic, the steady-state time is used as the auxiliary verification characteristic, double-feature redundant verification is realized, and the detection reliability is improved.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic brake technology, and in particular to a method for predicting the lifespan of electromagnetic brakes based on voltage inflection point and steady-state time characteristics. Background Technology

[0002] Electromagnetic brakes are widely used in wind turbines, elevators, industrial drive systems, and automated equipment, primarily to provide reliable braking force when equipment stops or loses power. With prolonged operation, the friction pads gradually wear down, and the air gap between the armature and the electromagnet continuously increases. When the air gap widens to a certain extent, it leads to a decrease in engagement capacity, a longer engagement time, and a deterioration in release characteristics, ultimately causing brake failure or even safety accidents. While significant progress has been made in brake lifespan monitoring technology, a persistent problem remains: reliance on additional sensors to collect relevant data signals, such as vibration and acoustic signals. Such solutions are easily affected by environmental noise and can only achieve qualitative assessments, not quantitative estimations of the brake air gap. Summary of the Invention

[0003] The purpose of this invention is to design an electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics in order to solve the above problems.

[0004] The present invention achieves the above objectives through the following technical solutions:

[0005] An electromagnetic brake lifetime prediction method based on voltage inflection point and steady-state time characteristics includes:

[0006] S1. Obtain the sampling voltage waveform by continuously sampling the voltage across the brake coil;

[0007] S2. Locate the power-on moment in the sampled voltage waveform and extract the voltage waveform after the power-on moment as the analysis voltage waveform.

[0008] S3. Smooth and perform first-order difference processing on the analyzed voltage waveform to obtain a smoothed difference sequence;

[0009] S4. Determine the inflection point time and inflection point voltage of the analyzed voltage waveform based on the smoothed differential sequence;

[0010] S5. Determine the steady-state moment and steady-state voltage of the analyzed voltage waveform based on the inflection point moment;

[0011] S6. Calculate the first gap value of the brake based on the inflection point voltage;

[0012] S7. Determine the moment of full power-on; and calculate the second gap value of the brake based on the moment of full power-on.

[0013] S8. Perform a weighted fusion of the first gap value and the second gap value to obtain the final gap value;

[0014] S9. Analyze the current health level of the brake based on the final gap value and output the prediction results.

[0015] The beneficial effects of this invention are as follows: by detecting the key inflection point position in the voltage waveform, the voltage feature corresponding to the start of mechanical motion is extracted, and an air gap estimation model is established by combining the steady-state time feature, thereby realizing quantitative analysis of brake wear state; by using the inflection point time as the main estimation feature and the steady-state time as the auxiliary verification feature, dual feature redundancy verification is realized, improving detection reliability. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics of the present invention;

[0017] Figure 2 This is a flowchart of the electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to the present invention.

[0018] Figure 3 This is a schematic diagram of the power-on waveform of the brake with 25mm and 65mm gaps provided in the embodiment;

[0019] Figure 4 This is the Vknee and gap linear regression fitting graph provided in this embodiment. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0021] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0022] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0023] In the description of this invention, it should be understood that the terms "upper," "lower," "inner," "outer," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use, or the orientation or positional relationship commonly understood by those skilled in the art. They are only used to facilitate the description of this invention and to simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0025] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, terms such as "set" and "connection" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0026] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0027] like Figure 1 , Figure 2 As shown, the electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics includes:

[0028] S1. Obtain the sampling voltage waveform of the voltage across the brake coil continuously. The continuous sampling is as follows: transient data acquisition during brake power-on. When the brake is ready to be powered on, it is in a waiting trigger state. When the brake power-on signal arrives (GPIO rising edge interrupt), the analog-to-digital converter (ADC) is immediately started to continuously sample the voltage across the brake coil at a preset sampling frequency. The sampling window length is 500ms.

[0029] S2. Locate the power-on moment in the sampled voltage waveform and extract the voltage waveform after the power-on moment as the analysis voltage waveform; specifically including:

[0030] S201. Extract the first n sampling points of the sampled voltage waveform as the baseline interval;

[0031] S202. Sort the voltage values ​​in the baseline interval, remove 10% of the voltage values ​​at each end, and obtain the baseline interval after removal.

[0032] S203. Analyze the average value of the baseline intervals after elimination as the baseline voltage. And calculate the power-on threshold. , is represented as: ;

[0033] S204. Starting from the nth sampling point, iterate through the sampled voltage waveforms. When the voltage value first continuously exceeds the power-on threshold... This moment will be taken as the power-on moment. The power-on threshold is exceeded for more than n / 2 consecutive sampling points. It can be determined that the voltage value has exceeded the power-on threshold for the first time consecutively. Established, the time of the first sampling point after more than n / 2 consecutive sampling points is taken as the power-on time. ;

[0034] S205. Capture the sampled voltage waveform at power-on time. The voltage waveform after power-on is used as the analysis voltage waveform to analyze the time characteristics of the voltage waveform. It is a relative zero point.

[0035] S3. Smooth and perform first-order difference processing on the analyzed voltage waveform to obtain a smoothed difference sequence; specifically including:

[0036] S301. A sliding window mean filter is used to smooth the voltage signal of the analyzed voltage waveform, resulting in a smoothed voltage... , is represented as: Where W is the window width; the window width W is an odd number and satisfies constraints 1 and 2. Constraint 1 is: ≤ , The original noise amplitude of the sampled voltage waveform. The noise amplitude threshold is defined as follows: Constraint 2 is: , The sampling period is The highest characteristic frequency during the duration of the transition period from a rapid voltage rise to a gradual voltage level; Constraint 1 determines the minimum window width. Constraint 2 determines the maximum window width Choose an odd number as the window width from constraints 1 and 2. (This facilitates symmetrical calculation of the window mean with the current sampling point as the center), ensuring that the window fully preserves the low-frequency characteristics of the inflection point while effectively suppressing noise. The window mean with a radius of W / 2 is used as the sampling value of the i-th sampling point. ;

[0037] S302. Calculate the first-order difference sequence of the smoothed voltage sequence, expressed as: ;

[0038] S303. Apply sliding window mean smoothing to the first-order difference sequence with a window width W to obtain a smoothed difference sequence.

[0039] S4. Determine the inflection point time and inflection point voltage of the analyzed voltage waveform based on the smoothed differential sequence. In the voltage waveform after power-on, when the armature begins to move, a back electromotive force is generated and superimposed on the coil voltage, causing the voltage growth rate to slow down significantly or experience a local decrease, forming the first local maximum of the voltage waveform as the inflection point; specifically including:

[0040] S401, From the moment of power-on Then, starting from the 10th sampling point, the process iterates backwards through the smoothed difference sequence. The search function finds the position i at which the smoothed differential voltage first changes from positive to negative.

[0041] S402. Perform amplitude verification on position i to determine whether the smooth voltage requirement is met. If so, then position i is taken as the inflection point, and the time and voltage at position i are taken as the inflection point time, respectively. and inflection point voltage Then proceed to S5; otherwise, iterate through the next window and return to S401; if the steady-state voltage is reached... If the conditions are still not met, then no inflection point was found this time, and the power-on detection is judged to have failed; false detections caused by baseline random noise are eliminated through amplitude verification.

[0042] S5. Determine the steady-state moment and steady-state voltage of the analyzed voltage waveform based on the inflection point moment; specifically including:

[0043] S501, From the inflection point Then, starting from the 20th sampling point, the smoothed difference sequence will be... Divide into sub-windows of 60 consecutive sampling points;

[0044] S502. Select all sampling points in the first sub-window whose absolute values ​​are all less than a preset threshold. The starting position is taken as the steady state time. And steady state time The voltage value is taken as the steady-state voltage. steady state time This refers to the moment when the armature reaches its final position.

[0045] If a steady-state moment is not found within a 500ms window, it is marked as a timeout state and an alarm is triggered.

[0046] S6. Calculate the first gap value of the brake based on the inflection point voltage. , is represented as: a and b are the first gap values. With inflection point voltage The regression coefficients of the linear regression model between them; mm / mV, mm.

[0047] S7. Determine the moment of full energization; and calculate the second gap value of the brake based on the moment of full energization. , is represented as: c and d are the second gap values. With full power-on time The regression coefficients of the linear regression model between them. , Although fully powered on The linear relationship with the gap is slightly weaker than that with the inflection point voltage. However, it can be used for consistency verification: when the deviation between the first gap value and the second gap value exceeds a preset threshold, the reliability of this detection is marked as low; the steady-state time... As the moment of full power-on;

[0048] S8. Perform weighted fusion on the first gap value and the second gap value to obtain the final gap value. , is represented as: ,in, and The first gap value is respectively Second gap value The weighting coefficients, , .

[0049] S9. Analyze the current health level of the brake based on the final gap value and output the prediction results; specifically including:

[0050] S901. Compare the final gap value with the preset level threshold to obtain the current health level of the brake; specifically: when The current health level is healthy; when At that time, the current health level is a sub-health warning; when Gap> At this time, the current health level is at the level of changing the warning. For health threshold, This is the danger threshold;

[0051] S902. Output the prediction result based on the current health level of the brake. The prediction result includes the final gap value, health level, and timestamp warning status. After outputting the prediction result, serialize the prediction result (gap value, health level, timestamp) and write it to Flash persistent storage. At the same time, encapsulate the result into a protocol frame and send it to the host computer / cloud via a 4G or UART adapter.

[0052] This method differs from existing technologies that use current peak values ​​for qualitative fault diagnosis. This method quantitatively inverts the brake gap based on the inflection point of the coil voltage waveform (the local maximum where the first derivative turns from positive to negative). It extracts the voltage characteristics corresponding to the start of mechanical motion by using the key inflection point position in the voltage waveform, and establishes an air gap estimation model by combining it with steady-state time characteristics, thereby realizing quantitative analysis of brake wear state. The voltage inflection point directly reflects the back electromotive force generated by the armature movement, with a clear physical meaning. This method only uses the voltage signal of the brake coil during power-on transients to complete gap estimation, requiring no additional hardware, resulting in low cost and simple installation. This method calculates the first and second gap values ​​and then weights and fuses them to obtain the final gap value, achieving continuous gap value output with micron-level accuracy within the range of healthy and failed gaps. This method proposes a power-on point detection method: using the truncated mean to calculate the baseline voltage, combined with a dynamic threshold (10mV) to automatically locate the power-on moment, eliminating the need for external synchronization signals and removing errors introduced by inconsistent power-on delays under different test / operating conditions. This method uses the inflection point voltage V... knee Principal gap estimation is performed using key features, with the full power-on time T as the starting point. full The system uses secondary features for fusion and weighted calculation, improving detection reliability; it offers three configurable health grading and early warning strategies: healthy, sub-healthy, and replacement warnings. For health threshold, The danger threshold can be adjusted by implementing data to adapt to different brake models.

[0053] Example 1: Experimental Data Verification

[0054] The experiment was conducted on a certain type of electromagnetic brake. The coil voltage waveforms at nine different gap positions (0.25mm~0.65mm, 0.05mm increments) were acquired using the CH3V channel of an oscilloscope at a sampling frequency of 100kHz. Each gap position was repeated five times. The experimental results are shown in Table 1. Figure 3 and Figure 4 As shown.

[0055] Table 1 shows the experimental results.

[0056]

[0057] Example 2: Embedded Implementation

[0058] This method was implemented on an ESP32 (512KB SRAM, 20MB Flash, 240MHz dual-core CPU): using an ADC to sample the coil voltage at 100kHz, DMA double-buffered acquisition, and firmware layer 2048 points / frame.

[0059] After each frame of data arrives, incremental window partitioning is performed (10ms / window), and the window mean is calculated.

[0060] After detecting the inflection point, the power-on time is continuously calculated, the gap value is automatically calculated and graded, and finally sent to the host computer via 4G or UART serial port.

[0061] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A method for predicting the lifespan of an electromagnetic brake based on voltage inflection point and steady-state time characteristics, characterized in that, include: S1. Obtain the sampling voltage waveform by continuously sampling the voltage across the brake coil; S2. Locate the power-on moment in the sampled voltage waveform and extract the voltage waveform after the power-on moment as the analysis voltage waveform. S3. Smooth and perform first-order difference processing on the analyzed voltage waveform to obtain a smoothed difference sequence; S4. Determine the inflection point time and inflection point voltage of the analyzed voltage waveform based on the smoothed differential sequence; S5. Determine the steady-state moment and steady-state voltage of the analyzed voltage waveform based on the inflection point moment; S6. Calculate the first gap value of the brake based on the inflection point voltage; S7. Determine the full power-on time; And calculate the second gap value of the brake based on the moment of full power-on; S8. Perform a weighted fusion of the first gap value and the second gap value to obtain the final gap value; S9. Analyze the current health level of the brake based on the final gap value and output the prediction results.

2. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, S2 includes: S201. Extract the first n sampling points of the sampled voltage waveform as the baseline interval; S202. Sort the voltage values ​​in the baseline interval, remove 10% of the voltage values ​​at each end, and obtain the baseline interval after removal. S203. Analyze the average value of the baseline intervals after elimination as the baseline voltage. And calculate the power-on threshold. , represented as: ; S204. Starting from the nth sampling point, iterate through the sampled voltage waveforms. When the voltage value first continuously exceeds the power-on threshold... This moment will be taken as the power-on moment. ; S205. Capture the sampled voltage waveform at power-on time. The voltage waveform after power-on is used as the analysis voltage waveform to analyze the time characteristics of the voltage waveform. It is a relative zero point.

3. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, S3 includes: S301. A sliding window mean filter is used to smooth the voltage signal of the analyzed voltage waveform, resulting in a smoothed voltage... , is represented as: Where W is the window width, and the average value of the window with a radius of W / 2 is used as the sample value of the i-th sampling point. ; S302. Calculate the first-order difference sequence of the smoothed voltage sequence, expressed as: ; S303. Apply sliding window mean smoothing to the first-order difference sequence with a window width W to obtain a smoothed difference sequence.

4. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 3, characterized in that, In S301, the window width W is an odd number and satisfies constraints 1 and 2. Constraint 1 is: ≤ , The original noise amplitude of the sampled voltage waveform. The noise amplitude threshold is defined as follows: Constraint 2 is: , The sampling period is It is the highest characteristic frequency during the duration of the transition period from a rapid rise in voltage to a gradual rise.

5. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, S4 includes: S401, From the moment of power-on Then, starting from the 10th sampling point, the process iterates backwards through the smoothed difference sequence. The search function finds the position i at which the smoothed differential voltage first changes from positive to negative. S402. Perform amplitude verification on position i to determine whether the smooth voltage requirement is met. If so, then position i is taken as the inflection point, and the time and voltage at position i are taken as the inflection point time, respectively. and inflection point voltage Then proceed to S5; otherwise, proceed to the next window and return to S401.

6. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, In S5 include:, S501, From the inflection point Then, starting from the 20th sampling point, the smoothed difference sequence will be... Divide into sub-windows of 60 consecutive sampling points; S502. Select all sampling points in the first sub-window whose absolute values ​​are all less than a preset threshold. The starting position is taken as the steady state time. And steady state time The voltage value is taken as the steady-state voltage. .

7. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, In S6, the first gap value Represented as: a and b are the first gap values. With inflection point voltage The regression coefficients of the linear regression model between them.

8. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, In S7, the second gap value Represented as: c and d are the second gap values. With full power-on time The regression coefficients of the linear regression model between them.

9. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, In S8, the final gap value Represented as: ,in, and The first gap value is respectively Second gap value The weighting coefficients.

10. The electromagnetic brake life prediction method based on voltage inflection point and steady-state time characteristics according to claim 1, characterized in that, S9 includes: S901. Compare the final gap value with the preset level threshold to obtain the current health level of the brake; specifically: when The current health level is healthy; when At that time, the current health level is a sub-health warning; when Gap> At this time, the current health level is at the level of changing the warning. For health threshold, This is the danger threshold; S902. Output the prediction results based on the current health level of the brake. The prediction results include the final gap value, health level, and timestamp warning status.

Images