Acoustic emission condition monitoring system and method for a main bearing of a wind turbine

Abstract

The application discloses a kind of acoustic emission state monitoring system and method of wind power generator main bearing, comprising: establishing calibration parameter library including amplitude threshold, count abnormal threshold, coupling state threshold and time coincidence degree threshold under healthy state;Periodically send standard diagnostic pulse to acoustic emission sensor, calculate current coupling state coefficient and determine whether sensor is in effective coupling state;When effective coupling, acoustic emission and ground brush current waveform sequence are synchronously collected and total number of events is extracted;The time coincidence degree of two types of events is calculated;According to the comparison result of coupling state, time coincidence degree and total number of events and corresponding count abnormal threshold, it is classified and determined as sensor maintenance alarm, shaft current electrocorrosion alarm, mechanical spalling alarm, ground brush loop abnormal alarm, composite abnormal alarm or normal according to priority;Output differentiated maintenance instruction.The application effectively distinguishes multiple faults through sensor self-diagnosis and acoustic-electric synchronous identification, and reduces false alarm rate.

Description

Technical Field

[0001] This invention relates to the field of wind turbine condition monitoring technology, and in particular to an acoustic emission condition monitoring system and method for the main bearing of a wind turbine. Background Technology

[0002] The main bearing of a wind turbine is a core load-bearing component of the transmission system. It endures alternating heavy loads and complex environmental stresses over long periods, making it prone to early failures such as raceway spalling, wear, and shaft current galvanic corrosion. Acoustic emission monitoring technology, with its high sensitivity to microcrack propagation and transient discharges in materials, has become an important means of detecting early failures in main bearings.

[0003] However, existing acoustic emission monitoring schemes suffer from the following technical bottlenecks: Current solutions assume the sensor installation state is constant. However, during long-term operation, the magnetic base, adhesive layer, or clamping structure can deteriorate due to vibration and temperature drift, leading to signal amplitude attenuation. In this case, the system has difficulty distinguishing between sensor failure and normal bearing operation, which can easily cause false alarms or missed alarms.

[0004] Both raceway spalling and shaft current discharge are transient impact-type acoustic emission events, and a single acoustic emission amplitude or counting characteristic is insufficient to reveal their root cause. Although some existing technologies have introduced current detection, they often lack a time synchronization correlation mechanism between acoustic emission and current pulses. The two types of signals are separated on the time axis, making it difficult to determine whether they originate from the same discharge process, thus failing to achieve reliable root cause classification.

[0005] Most existing methods rely on fixed thresholds or historical statistical baselines, and their parameters are difficult to adjust according to the physical characteristics of the object being measured (such as sensor installation transfer function and brush discharge characteristics), making it difficult for the system to achieve both high sensitivity and low false alarm rate.

[0006] In summary, there is an urgent need in this field for an anti-interference monitoring scheme that can ensure the reliability of the acquisition link at the source, decouple the root cause of the fault through acoustic-electric time correlation analysis, and quantitatively bind the criterion system to the physical characteristics of the object. Summary of the Invention

[0007] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0008] Therefore, to solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for monitoring the acoustic emission status of a wind turbine main bearing, comprising the following steps: S1. Establish a calibration parameter library, which includes at least the acoustic emission event extraction amplitude threshold, the grounding brush current pulse event extraction amplitude threshold, the acoustic emission event count anomaly threshold, the current pulse event count anomaly threshold, the coupling state threshold, and the time overlap threshold. S2. Periodically send standard diagnostic pulses to the acoustic emission sensor, calculate the current coupling state coefficient and compare it with the coupling state threshold. If it is determined to be a coupling failure, generate a coupling failure flag; if it is determined to be a valid coupling state, generate a valid channel flag. S3. When the acoustic emission sensor is in an effective coupling state, the original waveform sequence of acoustic emission and the original waveform sequence of grounding brush current are acquired synchronously. After time stamp alignment, the time sequence of acoustic emission events and the time sequence of grounding brush current pulse events are extracted respectively, and the total number of acoustic emission events and the total number of grounding brush current pulse events are counted. S4. Calculate the time overlap between the acoustic emission event time series and the grounding brush current pulse event time series; S5. Based on the comparison results of the coupling failure mark, time overlap, total number of acoustic emission events and total number of grounding brush current pulse events with the corresponding counting abnormal threshold, it is determined to be one of the following: sensor maintenance alarm, shaft current electrical corrosion alarm, mechanical peeling alarm, grounding brush circuit abnormal alarm, composite abnormal alarm or normal state. S6. Output the corresponding differentiated maintenance instructions to the wind turbine main control system based on the judgment result.

[0009] As a preferred embodiment of the acoustic emission status monitoring method for the main bearing of a wind turbine according to the present invention, in step S1, the acoustic emission event extraction amplitude threshold and the grounding brush current pulse event extraction amplitude threshold are used to extract a single abnormal event; the acoustic emission event count abnormal threshold and the grounding brush current pulse event count abnormal threshold are used to determine whether the number of abnormal events within a preset time window exceeds the limit; the coupling state threshold is a preset ratio threshold obtained based on the standard diagnostic pulse response amplitude calibration; and the time overlap threshold is a synchronization judgment threshold obtained based on the calibration of known electro-corrosion samples.

[0010] In a preferred embodiment of the acoustic emission status monitoring method for the main bearing of a wind turbine according to the present invention, in step S5, a uniqueness classification determination is performed according to the following priority: If the aforementioned coupling failure marker exists, it is determined to be a sensor maintenance alarm; If the time overlap is greater than or equal to the time overlap threshold, and the total number of acoustic emission events is greater than or equal to the acoustic emission event count abnormal threshold, and the total number of grounding brush current pulse events is greater than or equal to the grounding brush current pulse event count abnormal threshold, then it is determined to be an shaft current electrical corrosion alarm. If the aforementioned conditions are not met, and the total number of acoustic emission events is greater than or equal to the abnormal threshold for acoustic emission event counting, and the total number of grounding brush current pulse events is less than the abnormal threshold for grounding brush current pulse event counting, then it is determined to be a mechanical peeling alarm. If the aforementioned conditions are not met, and the total number of acoustic emission events is less than the abnormal threshold for acoustic emission event counting, and the total number of grounding brush current pulse events is greater than or equal to the abnormal threshold for grounding brush current pulse event counting, then it is determined to be an abnormal alarm of the grounding brush circuit. If the aforementioned conditions are not met, and the total number of acoustic emission events is greater than or equal to the abnormal threshold for acoustic emission event counting, and the total number of grounding brush current pulse events is greater than or equal to the abnormal threshold for grounding brush current pulse event counting, but the time overlap is less than the time overlap threshold, then it is determined to be a composite abnormal alarm. If none of the above conditions are met, it is considered normal.

[0011] As a preferred embodiment of the acoustic emission status monitoring method for the main bearing of a wind turbine according to the present invention, in step S6, the following differentiated maintenance instructions are output based on different judgment results: When a sensor maintenance alarm is triggered, the output will check the coupling layer of the acoustic emission sensor. When a shaft current electrical corrosion alarm is triggered, the output will check the grounding brush and the shaft current bypass circuit. When a mechanical spalling alarm is triggered, an output will be sent to check the raceway and rolling elements of the main bearing. When a grounding brush circuit malfunction alarm is triggered, the output will check the grounding brush resistance and contact status. When a compound abnormality alarm occurs, the output synchronously checks the main bearing and grounding brush circuit.

[0012] As a preferred embodiment of the acoustic emission status monitoring method for the main bearing of a wind turbine according to the present invention, the standard diagnostic pulse is generated by a diagnostic pulse generating unit located near the acoustic emission sensor and coupled to the acoustic emission sensor through a signal line or waveguide structure.

[0013] As a preferred embodiment of the acoustic emission status monitoring method for the main bearing of a wind turbine according to the present invention, the acoustic emission sensor is installed at a radial position on the outer side of the main bearing housing near the raceway projection area, and is coupled to the main bearing housing using an acoustic coupling agent.

[0014] The present invention also provides an acoustic emission status monitoring system for a wind turbine main bearing, for performing an acoustic emission status monitoring method for a wind turbine main bearing as described in any of the above claims. The system includes an embedded processing unit, an acoustic emission acquisition unit, a grounding brush current acquisition unit, a standard diagnostic pulse generation unit, and an alarm output unit. The embedded processing unit is communicatively connected to the acoustic emission acquisition unit, the grounding brush current acquisition unit, the standard diagnostic pulse generation unit, and the alarm output unit, and stores a calibration parameter library. The embedded processing unit is used to perform calibration parameter generation, coupling state determination, event extraction, time overlap calculation, fault classification determination, and alarm output control. The acoustic emission acquisition unit is used to acquire the acoustic emission waveform during the operation of the main bearing; The grounding brush current acquisition unit is used to acquire high-frequency pulse current signals in the grounding brush circuit. The standard diagnostic pulse generation unit is used to generate standard diagnostic pulses to perform online verification of the coupling state of the acoustic emission sensor; The alarm output unit is used to output corresponding maintenance commands to the main control system.

[0015] As a preferred embodiment of the acoustic emission status monitoring system for the main bearing of a wind turbine generator according to the present invention, the acoustic emission acquisition unit includes an acoustic emission sensor and a preamplifier circuit, a filter circuit, and an A / D conversion circuit connected in sequence; the grounding brush current acquisition unit includes a current sensor, a shaping circuit, and an A / D conversion circuit.

[0016] As a preferred embodiment of the acoustic emission status monitoring system for the main bearing of a wind turbine according to the present invention, the acoustic emission acquisition unit and the analog front end of the grounding brush current acquisition unit are powered by isolated power supplies; the communication interface between the embedded processing unit and each acquisition unit is connected in an isolated manner; each signal input terminal is equipped with a shielding structure and adopts a single-point grounding structure.

[0017] As a preferred embodiment of the acoustic emission status monitoring method for the main bearing of a wind turbine according to the present invention, wherein: when the communication link is interrupted, the embedded processing unit caches the fault classification judgment results and related data in the local storage area, and retransmits them to the wind turbine main control system after the communication is restored; the related data includes the current time window number, time overlap, total number of acoustic emission events, total number of current pulse events, and coupling state coefficient.

[0018] The beneficial effects of this invention are: 1. This invention establishes a calibration parameter library under healthy conditions and generates various thresholds based on calibration data. Furthermore, each threshold parameter can be established according to the actual operating state, thereby improving the system's adaptability to different operating conditions and reducing false alarms and missed alarms caused by fixed empirical thresholds.

[0019] 2. This invention achieves online detection of the coupling state of acoustic emission sensors by comparing standard diagnostic pulse excitation and response amplitude; when the sensor magnetic base, adhesive layer or clamping structure deteriorates, a coupling failure alarm can be generated, ensuring the reliability of the acquisition link from the source, avoiding misjudging sensor failure as main bearing abnormality, and reducing the occurrence of false alarms and missed alarms.

[0020] 3. This invention synchronously collects acoustic emission events and grounding brush current pulse events and aligns them with a unified time reference, and calculates the event time overlap. This mechanism can distinguish between mechanical spalling faults and shaft current electrical corrosion faults: when acoustic emission increases significantly but current pulses do not increase, it is determined to be mechanical spalling; when acoustic emission and current pulses increase significantly at the same time and have a high degree of overlap, it is determined to be shaft current electrical corrosion. This time-correlation-based analysis overcomes the limitations of existing technologies where the two types of signals are separated and it is difficult to determine the root cause of the fault.

[0021] 4. This invention establishes threshold values ​​for acoustic emission and current events, event count thresholds, and time overlap thresholds using health status calibration data. This allows the judgment criteria to be adjusted based on physical characteristics such as sensor installation transfer function and brush discharge characteristics, thereby reducing the false alarm rate while ensuring high sensitivity. This strategy solves the technical problem that existing fixed thresholds or historical statistical baselines cannot balance accuracy and reliability.

[0022] 5. This invention is based on a hierarchical classification and judgment mechanism for coupling failure, mechanical peeling, shaft current electrical corrosion, grounding brush abnormality and composite abnormality, which enables different abnormal states to be identified under a unified logical framework, improving the systematicness and interpretability of fault classification results; at the same time, it can also output maintenance suggestions to realize the linkage between fault identification and operation and maintenance decision-making, improve on-site handling efficiency and wind turbine operation reliability.

[0023] 6. The entire discrimination system designed in this invention realizes a linear closed loop of calibration, diagnosis, classification, and decision-making, achieves synergistic effects among modules, balances interpretability and feasibility, and is suitable for long-term online operation environments. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart illustrating the overall workflow of the present invention.

[0025] Figure 2 This is a sub-flowchart of step S2 in Embodiment 1 of the present invention.

[0026] Figure 3 This is a sub-flowchart of step S3 in Embodiment 1 of the present invention.

[0027] Figure 4 This is a sub-flowchart of step S5 in Embodiment 1 of the present invention.

[0028] Figure 5 This is a system architecture diagram of the present invention. Detailed Implementation

[0029] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0030] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0031] Secondly, an embodiment or embodiment referred to herein means a specific feature, structure or characteristic that may be included in at least one implementation of the present invention; an embodiment appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0032] Example 1 Reference Figure 1 The first embodiment of the present invention provides a method for monitoring the acoustic emission status of a wind turbine main bearing. The method is used to identify mechanical spalling, electrical corrosion and grounding circuit abnormalities generated during the operation of the wind turbine main bearing online, and to perform synchronous self-diagnosis of the coupling status of the acoustic emission sensor.

[0033] The method includes the following steps: S1, Calibration parameter generation; After the wind turbine has passed maintenance and the main bearing and grounding brush are confirmed to be in good condition, perform initialization calibration to establish the baseline parameters under good condition.

[0034] The health status refers to the operating condition of a wind turbine that has been inspected and confirmed to be free from bearing peeling, abnormal brush contact, abnormal discharge, and obvious structural loosening.

[0035] S11. Collect calibration data; The standard diagnostic pulse generation unit is controlled to send standard diagnostic pulses to a broadband piezoelectric acoustic emission sensor mounted on the surface of the main bearing housing, and its response waveform is received.

[0036] Preferably, the acoustic emission sensor is installed radially on the outside of the main bearing, near the raceway projection area, and the mounting surface is preferably located in the high-rigidity area of ​​the main bearing housing to improve the consistency of acoustic emission propagation. High-temperature silicone grease or a special acoustic coupling agent can also be used as the coupling medium between the acoustic emission sensor and the main bearing housing, and it can be fixed by a magnetic base, bolt plate or weather-resistant adhesive layer. When the site structure allows, a stainless steel waveguide rod can also be arranged between the acoustic emission sensor and the main bearing housing to enhance the signal transmission stability under long-distance attenuation environment.

[0037] Preferably, the amplitude of the standard diagnostic pulse is 5V, the pulse width is 100μs, and the transmission period is 24h; the 24h period is preferably set to cover one complete load change process and one temperature change process, so as to obtain representative reference response data under different environmental conditions.

[0038] The acoustic emission acquisition unit and the grounding brush current acquisition unit continuously and synchronously acquire data under stable operating conditions without faults or abnormal arcs, obtaining the acoustic emission background waveform and the grounding brush current background waveform. Preferably, the continuous data collection time is no less than 24 hours; during the calibration period, it is preferable to ensure that the wind turbine undergoes at least one start-up and shutdown process and changes in different power ranges. Different power ranges may include 0%~30% of rated power, 30%~70% of rated power, and 70%~100% of rated power to cover typical operating conditions.

[0039] During synchronous acquisition, the acoustic emission background waveform and the grounding brush current background waveform are preferably stored in a ring buffer first, and then written in batches to non-volatile memory or external storage medium according to time windows, so as to avoid buffer overflow caused by long-term continuous sampling.

[0040] Preferably, the synchronous acquisition can be achieved using a unified clock triggering method, preferably a master-slave clock synchronization structure, wherein the master clock is output by the embedded processing unit, and the acoustic emission acquisition unit and the grounding brush current acquisition unit serve as slave clock nodes; or IEEE 1588 PTP combined with hardware timestamp, IRIG-B code synchronization or FPGA phase-locked loop synchronization method can be used to ensure the consistency of subsequent event timestamps.

[0041] Preferably, in order to reduce the impact of abnormal spikes on the calibration results, a coarse pre-filter can be performed during the calibration acquisition stage of this step to remove saturation pulses, packet loss segments, or obvious communication error segments before proceeding to subsequent statistical processing.

[0042] It should be noted that the "health status" refers to the operating state of the wind turbine generator after inspection and confirmation that there are no bearing peeling, abnormal brush contact, or abnormal discharge phenomena. Data collected under this state can reflect the normal noise level of the equipment, thus providing a statistical benchmark for subsequent anomaly detection.

[0043] S12. Calculate the waveform baseline statistics and the event extraction amplitude threshold; The peak amplitude is extracted from the standard diagnostic impulse response waveform and recorded as the reference response amplitude. ; For the acoustic emission background waveform under healthy conditions, the peak sequence is extracted according to a uniform time window and denoted as... ,in, Indicates the first Background peak of acoustic emission, This is the length of the sequence; For the background waveform of the grounding brush current under healthy conditions, the peak sequence is extracted according to a uniform time window and denoted as... ,in, Indicates the first One current background peak This is the length of the sequence.

[0044] Preferably, the peak extraction is achieved using the sliding window maximum value method, that is, the point with the maximum amplitude within the preset sampling window is taken as the peak value; before peak extraction, it is preferable to perform DC removal processing and detrending processing to reduce the impact of drift baseline on threshold calculation.

[0045] Furthermore, statistical analysis is performed on the peak values ​​of the acoustic emission background waveform and the grounding brush current background waveform, and then the following calculations are performed: Calculate the mean of the peak sequence of the acoustic emission background waveform under healthy conditions. and standard deviation The calculation formulas are as follows: Mean peak value of acoustic emission background: ; Standard deviation of background acoustic emission: .

[0046] Calculate the mean value of the peak value sequence of the background waveform of the grounding brush current under healthy conditions. and standard deviation The calculation formula is as follows: Average peak value of background current: ; Standard deviation of current background peak: .

[0047] Preferably, before the statistical peak value is obtained, the acoustic emission background waveform is filtered by a 4th-order Butterworth bandpass filter and then the peak value is extracted. The grounding brush current background waveform is preferably filtered by a low-pass or bandpass filter and then the peak value is extracted, so as to reduce the influence of high-frequency interference and low-frequency drift on the statistical quantity.

[0048] Furthermore, based on The criteria are used to calculate the acoustic emission event extraction amplitude threshold for extracting individual anomalous events. Extracting amplitude thresholds for grounding brush current pulse events The calculation formulas are as follows: ; ; Among them, the amplitude threshold for acoustic emission events is extracted. Used to extract sudden transient events from acoustic emission waveforms, and to extract amplitude thresholds for ground brush current pulse events. Used to extract high-frequency pulse events from current waveforms.

[0049] It should be noted that, under healthy conditions, the vast majority of random noise peaks are distributed around the mean. Using the mean plus three standard deviations as a threshold can effectively filter out background noise, retaining only abnormal transient events with sudden characteristics, thereby improving the accuracy of event extraction.

[0050] S13. Calculate the event count anomaly threshold; The acoustic emission background waveform and the grounding brush current background waveform under healthy conditions acquired in step S11 are segmented according to a preset time window W; preferably, the time window W is preferably set to 10 minutes.

[0051] Within each of the aforementioned preset time windows, the peak value of the acoustic emission background waveform exceeds the acoustic emission event extraction amplitude threshold calculated according to step S12. The number of events is used to generate a sequence of acoustic emission event counts. ;in, This represents the number of acoustic emission events within the k-th time window. Total number of time windows; Similarly, if the peak value of the background waveform of the grounding brush current exceeds the amplitude threshold for extracting the grounding brush current pulse event described in step S12, then... The number of events forms a current pulse event counting sequence. ;in, This represents the number of current pulse events within the k-th time window.

[0052] The event counting preferably adopts a dead zone mechanism: counting starts when the threshold is exceeded, three consecutive sampling points exceeding the threshold are considered the start of an event, and continuous sampling points below the threshold for more than 20 μs are considered the end of an event, so as to avoid the same physical event being counted repeatedly due to ringing or wake waves.

[0053] For a cluster of events that consistently exceeds the threshold, the system preferably counts them as a single event and records them in the cache as additional attributes of the event, including the event peak value, duration, and cumulative energy.

[0054] Furthermore, the calculations are performed using the following formulas: Mean number of acoustic emission events: ; Standard deviation of acoustic emission event count: ; Average count of current pulse events: ; Standard deviation of current pulse event count: ; Based on 3 The criteria are as follows: calculate the abnormal threshold for acoustic emission event counts using the following formulas. and the abnormal threshold for grounding brush current pulse event counting : ; ; in, This is used to determine whether the number of acoustic emission events is abnormally high within a macroscopic time window. This is used to determine whether the number of current pulse events is abnormally high within the macroscopic time window.

[0055] It should be noted that the event extraction amplitude threshold is used to identify individual abnormal events, while the event count abnormality threshold is used to determine whether there has been a significant increase in abnormal events over a macroscopic time range. These two thresholds correspond to single event identification and overall abnormality frequency determination, respectively, and can avoid false alarms caused by occasional noise.

[0056] In another preferred embodiment, the event anomaly counting threshold can also be segmented and corrected in conjunction with the fan speed or nacelle temperature to adapt to changes in background noise under low temperature, low speed or high load conditions.

[0057] S14. Set the coupling state threshold and time overlap threshold; Based on standard diagnostic impulse response waveforms and background statistics, a coupling state threshold is generated. Time overlap threshold ; Wherein, the coupling state threshold Used to define the minimum permissible response ratio of the acoustic emission sensor coupling layer; preferably, a coupling state threshold. The value range is 0.5 to 0.7, indicating that the current response amplitude must not be lower than 50% to 70% of the reference amplitude.

[0058] The time overlap threshold The value was obtained by calibration using synchronous discharge test data of known electro-corrosion samples, and its value was taken as the mean of the time coincidence of the samples minus twice the standard deviation; specifically, the time coincidence sequence was statistically analyzed in the known electro-corrosion samples. Calculate its mean and standard deviation Therefore, the time overlap threshold The calculation formula is: ; Preferably, the time overlap threshold A value of 0.9 is preferred so that an electrical corrosion alarm is only output when a strong synchronization relationship is established.

[0059] Preferably, the calibration samples for the time overlap threshold are preferably from a grounding brush abnormality test bench or an electro-corrosion simulation bench, covering at least different rotational speeds and different load levels, in order to improve the versatility of the threshold.

[0060] It should be noted that shaft current discharge typically generates both high-frequency pulsed current and transient acoustic emission impact simultaneously, thus these two types of events exhibit strong synchronicity on the time axis. By setting a time overlap threshold, electro-corrosion faults and pure mechanical impact faults can be effectively distinguished.

[0061] S15, Solidification calibration parameter library; Output calibration parameter library, the calibration parameter library includes and This serves as a unified basis for subsequent coupling determination and fault classification.

[0062] The calibration parameter library is formed by writing the data into the non-volatile memory area of ​​the embedded processing unit.

[0063] Preferably, the calibration parameter library is stored in Flash, EEPROM or external industrial-grade memory, and a version number and timestamp are set so that the parameters can be recalibrated and traced after changing the fan model or sensor.

[0064] The threshold values ​​mentioned above in this step can be preset to default values ​​at the factory. On-site, only one initialization calibration needs to be performed based on the actual installation conditions, wind field environment, and bearing model to complete threshold coverage for most operating scenarios. For extreme marine salt spray environments, extremely cold high-altitude environments, or high-power variable speed wind fields, step S1 can be performed again for local recalibration.

[0065] It should be noted that the calibration parameter library is used for subsequent online monitoring steps. Different wind farm environments or different bearing models only need to be re-performed with initialization calibration to establish a new health benchmark, which has good engineering adaptability.

[0066] S2. Self-diagnosis of the coupling status of the acoustic emission sensor; like Figure 2 As shown, the specific steps are as follows: S21. Data acquisition and calculation of coupling state coefficients; The standard diagnostic pulse generation unit is controlled to emit standard diagnostic pulses according to a preset cycle.

[0067] The acoustic emission acquisition unit receives the corresponding response signal and extracts the current response peak value. .

[0068] Preferably, the current response peak value is obtained by peak search within the same window as the standard diagnostic pulse, and the response signal can be bandpass filtered and baseline corrected before calculation to reduce the impact of short-term noise.

[0069] Furthermore, the current coupling state coefficients are calculated using the following formula. : ;in, This represents the current peak pulse response, expressed in volts (V). The reference response amplitude obtained from step S12 is expressed in volts (V); therefore, It is a dimensionless quantity.

[0070] It should be noted that the standard diagnostic pulse is a manually controllable verification stimulus, which does not participate in the normal operating load of the bearing and will not substantially interfere with the operating status of the main bearing.

[0071] When the sensor coupling layer becomes loose, ages, or delaminates, the response amplitude will decrease significantly. Therefore, the sensor coupling status can be reflected in real time by the ratio of the current response peak value to the reference response peak value.

[0072] S22, Status determination and data splitting; when When the state of the coupling layer of the acoustic emission sensor is determined to be abnormal, a coupling failure mark is generated and the coupling failure mark is passed to step S5, where step S5 performs the highest priority classification judgment based on the coupling failure mark. when When the acoustic emission sensor is in an effective coupling state, an effective channel marker is generated, and step S3 is initiated.

[0073] Furthermore, if the embedded processing unit makes multiple consecutive determinations... This can further generate a continuous coupling attenuation prompt to remind maintenance personnel to check for problems such as aging of the coupling agent, corrosion of the magnetic base, or loosening of the pressure plate.

[0074] It should be noted that online coupling status diagnosis can avoid false or missed detections caused by sensor installation failure, thereby improving the reliability of long-term online monitoring.

[0075] S3, Event Stream Extraction and Synchronous Acquisition; like Figure 3 As shown, the specific steps are as follows: S31. When the coupling state determination unit outputs a valid channel mark, the acoustic emission acquisition unit enters the continuous sampling mode; the grounding brush current acquisition unit synchronously acquires the high-frequency pulse current signal on the grounding brush line.

[0076] Preferably, the sampling frequency of the acoustic emission acquisition unit is 1MHz~5MHz, and the preferred value in this embodiment is 2MHz, with an A / D conversion accuracy of 16 bits.

[0077] Preferably, the sampling frequency of the grounding brush current acquisition unit is 100kHz.

[0078] Furthermore, to balance continuous sampling and memory usage, the acoustic emission acquisition unit preferably uses a circular buffer to store the raw waveforms of the most recent period. The main loop reads and processes the cached data according to a fixed time window. When the buffer reaches the preset upper limit, the oldest unprocessed data is overwritten first to ensure that the latest abnormal events are not lost during continuous operation.

[0079] S32. The two signals are timestamped using the same time synchronization protocol to form the original waveform sequence of acoustic emission and the original waveform sequence of grounding brush current.

[0080] Preferably, the time synchronization protocol adopts IEEE 1588 PTP combined with hardware timestamps, or adopts IRIG-B code and FPGA / processor hardware lock synchronization to improve timestamp consistency.

[0081] S33. Process the acquired raw acoustic emission waveform sequence and the raw grounding brush current waveform sequence; specifically as follows: Furthermore, the original acoustic emission waveform sequence is subjected to bandpass filtering and envelope extraction to extract the acoustic emission peak value that exceeds the acoustic emission event extraction amplitude threshold described in step S12. At the moment, generate a time series of acoustic emission events. ; Preferably, the lower cutoff frequency of the acoustic emission channel bandpass filter is 50kHz, the upper cutoff frequency is 500kHz, the passband ripple is no greater than 1dB, and the stopband attenuation is no less than 40dB / decade, so as to ensure sufficient distinction between the fault impact signal and the low-frequency mechanical background noise. The bandpass filter is preferably implemented using a 4th-order Butterworth digital bandpass filter, or it can be implemented using a hybrid method of analog bandpass filtering and digital FIR filtering. The analog filter is used to suppress high-frequency aliasing, and the digital filter is used to further improve the passband stability. The envelope extraction is preferably achieved by superimposing a low-pass filter after full-wave rectification, and the cutoff frequency of the low-pass filter is preferably 2kHz; in another optional embodiment, the Hilbert transform can also be used to extract the envelope to obtain the transient amplitude change trend of the event.

[0082] Furthermore, high-frequency pulse extraction is performed on the original waveform sequence of the grounding brush current. The extracted current peak value exceeds the grounding brush current pulse event extraction amplitude threshold in step S12. At the moment of generating the grounding brush current pulse event time sequence ; Meanwhile, within the time window W, the total number of acoustic emission events m and the total number of current pulse events n are counted respectively.

[0083] Preferably, the event counting in this step also adopts a dead zone mechanism: counting starts when the threshold is exceeded, three consecutive sampling points exceeding the threshold are considered the start of an event, and continuous periods below the threshold for more than 20 μs are considered the end of an event, so as to avoid the same physical event being counted repeatedly due to ringing or wake waves.

[0084] When a continuous peak group exceeds the threshold multiple times in a short period of time, the system prefers to count it as a single event and record it in the cache as an additional attribute of the event, including the event peak value, duration, and cumulative energy.

[0085] Furthermore, the event extraction unit can also jointly record the amplitude, duration, and event interval for subsequent fault review or maintenance analysis.

[0086] It should be noted that this step achieves synchronous correlation between the two types of events through a unified time axis, providing a basis for subsequent time overlap analysis; compared with the traditional single-channel monitoring method, this embodiment can simultaneously analyze the correlation between mechanical impact and discharge pulse.

[0087] S4. Calculate time overlap; The time overlap calculation unit receives the acoustic emission event sequence obtained in step S3. and the grounding brush current pulse event time sequence Set the allowable time deviation .

[0088] Preferably, combined with a 2MHz sampling rate, time deviation is allowed. Set to 5 times the sampling period, i.e. ;when When this happens, the corresponding event is determined to be a same-origin matching event; among which, These are the event moments in the time series of acoustic emission events. This refers to the event moment in the current pulse event sequence.

[0089] To avoid many-to-many repeated matching, the same acoustic emission event or the same current pulse event is only allowed to establish a homologous matching relationship with the closest opposite event within a time window; if there are multiple candidate matching events, the one with the smallest time difference is preferred as the homologous matching object.

[0090] Furthermore, count the number of homologous matching events, denoted as . ; Calculate time overlap : ; in, This represents the smaller of the total number of acoustic emission events and the total number of current pulse events.

[0091] When any event sequence is empty, the system records the time overlap as 0 to avoid division by zero and maintain logical continuity.

[0092] It should be noted that when electro-corrosion occurs, the high-frequency current pulse and the acoustic emission impact usually originate from the same discharge physical process, and therefore have a high degree of synchronization on the time axis; through time overlap analysis, the accuracy of electro-corrosion identification can be effectively improved.

[0093] S5. Fault classification and determination; like Figure 4 As shown, the fault classification and determination unit receives: valid channel marker, coupling failure marker, and time overlap. The total number of acoustic emission events m, the total number of current pulse events n, and the values ​​in the calibration parameter library The uniqueness classification is performed according to the following priority: Condition zero (highest priority): If a coupling failure flag is received, it is directly determined as a sensor maintenance alarm; Condition 1: When and and When this occurs, it is determined to be an axial current electrical corrosion alarm.

[0094] It should be noted that, under this condition, the two types of events are not only highly synchronous, but the number of events also significantly exceeds the healthy state benchmark, which is consistent with the typical characteristics of axial current discharge leading to electro-corrosion.

[0095] Condition 2: When condition 1 is not satisfied, and ,and At that time, it was determined to be a mechanical peeling alarm.

[0096] It should be noted that this state indicates a significant increase in acoustic emission impact events, but there is no corresponding abnormal current pulse, which is consistent with the characteristics of a pure mechanical peeling fault.

[0097] Condition 3: When condition 1 is not satisfied, and ,and When this occurs, it is determined to be an abnormal alarm in the grounding brush circuit.

[0098] It should be noted that this state indicates an abnormal pulse current in the grounding circuit, but has not yet caused significant bearing impact damage. Therefore, the fault location is more likely to be in the grounding brush or bypass circuit.

[0099] Condition 4: When condition 1 is not met, but and ,and When this occurs, it is determined to be a compound abnormal alarm.

[0100] It should be noted that this state indicates that both types of abnormalities are frequent, but the timing synchronization is insufficient, and there may be situations where mechanical damage and electrical abnormalities occur simultaneously.

[0101] Condition 5: If none of the above conditions are met, it is considered normal.

[0102] In another preferred embodiment, if multiple consecutive time windows W are detected to be of the same type of alarm, the corresponding alarm can be upgraded to a high-priority alarm to prompt priority handling.

[0103] It should be noted that this embodiment adopts a dual judgment mechanism based on event synchronization relationship and event frequency, which can significantly reduce the false judgment rate and improve the fault differentiation ability compared with the traditional single threshold scheme.

[0104] S6. Alarm output and linkage handling; The alarm output unit receives the fault classification result from step S5 and generates the corresponding maintenance instruction.

[0105] When a shaft current electro-corrosion alarm is detected, the following output will be displayed: Check the grounding brush and shaft current bypass circuit.

[0106] When a mechanical spalling alarm is detected, the output is: Check the main bearing raceway and rolling elements.

[0107] When an alarm is detected as an abnormality in the grounding brush circuit, the output is: Check the grounding brush resistance and contact status.

[0108] When a composite abnormality alarm is detected, the following output will be displayed: Synchronously check the main bearing and grounding brush circuit.

[0109] When step S5 determines that a sensor maintenance alarm has been triggered, the output is: Check the acoustic emission sensor coupling layer.

[0110] Furthermore, the alarm output unit sends maintenance commands to the wind turbine main control system via the industrial Ethernet protocol.

[0111] Preferably, the industrial Ethernet protocol may be: OPC UA protocol or Modbus TCP protocol.

[0112] When the wind farm communication link is interrupted, the alarm unit can cache the alarm result in the local non-volatile storage area and retransmit it to the main control system after the communication is restored, so as to avoid the loss of alarm information.

[0113] The alarm information may include the current time window number, time overlap, total number of events, coupling state coefficient, and threshold comparison results, so that maintenance personnel can quickly verify the fault type.

[0114] The alarm output unit can also simultaneously display the fault type, suggested repair objects, and suggested processing priority on the local display module, so that on-site inspection personnel can quickly identify the key maintenance points.

[0115] It should be noted that this embodiment, through a closed-loop mechanism of fault identification, fault location, and maintenance command output, can directly provide targeted handling suggestions to operation and maintenance personnel, thereby reducing manual troubleshooting time and improving the operation and maintenance efficiency of wind turbines.

[0116] In summary, this invention addresses the failure mechanism of a specific object—the main bearing of a wind turbine generator. It reconstructs the cause and effect of the failure by using standard pulse excitation to ensure the reliability of the source through response diagnosis, and by using isochronous sampling of acoustic and electrical synchronization and time coincidence analysis. Furthermore, it quantifies and binds all judgment thresholds (coupling attenuation ratio, event extraction threshold, count anomaly boundary, and synchronization judgment standard) to the health status and known fault samples in the field, forming a complete technical solution with rigorous physical interpretability, from self-diagnosis of the acquisition link and event synchronization correlation to root cause classification.

[0117] Example 2 Reference Figure 5 This is the second embodiment of the present invention, which differs from the first embodiment in that: the present invention also provides an acoustic emission status monitoring system for the main bearing of a wind turbine generator, which is used to implement the acoustic emission status monitoring method for the main bearing of a wind turbine generator described in Embodiment 1 above.

[0118] The system includes: an embedded processing unit, an acoustic emission acquisition unit, a grounding brush current acquisition unit, a standard diagnostic pulse generation unit, and an alarm output unit; wherein, the embedded processing unit is communicatively connected to the acoustic emission acquisition unit, the grounding brush current acquisition unit, the standard diagnostic pulse generation unit, and the alarm output unit to perform data exchange and control command transmission.

[0119] The embedded processing unit, as the core control and computing component of the system, is used to run all data processing, logical judgment, and flow control tasks in the method described in Embodiment 1; specifically, the embedded processing unit is configured to perform the following operations: Perform the calculation of coupling state coefficients and the determination of coupling failure / effective channels in step S2 of Example 1; The event extraction in step S3 of embodiment 1 is performed, that is, according to the amplitude threshold for the acoustic emission event extraction and the amplitude threshold for the grounding brush current pulse event extraction, the time sequence of the acoustic emission event and the time sequence of the grounding brush current pulse event are generated from the original waveform sequence, and the total number of events is counted. Perform the same-origin matching event count and time overlap calculation in step S4 of Example 1; The fault classification determination in step S5 of Implementation Example 1 is performed, that is, the final alarm type is output according to the priority logic of condition zero to condition five. The calibration parameter library generated in step S15 of Example 1 is stored and invoked. The calibration parameter library is stored in the non-volatile memory area of ​​the embedded processing unit.

[0120] Preferably, the embedded processing unit is a microcontroller, industrial PLC, or industrial edge computing terminal based on the ARM Cortex-M series; the embedded processing unit preferably uses C language to write the control program, wherein the interrupt service routine is responsible for high-speed sampling triggering and timestamp recording, the direct memory access (DMA) transfer is responsible for moving the sampled data from the A / D conversion circuit to the random access memory (RAM) buffer, and the main loop is responsible for threshold comparison, event statistics, and classification judgment to meet the real-time requirements of continuous online operation.

[0121] The acoustic emission acquisition unit is used to acquire acoustic emission waveforms generated by mechanical impact, material deformation or electrical spark discharge during the operation of the main bearing.

[0122] Preferably, the front end of the acoustic emission acquisition unit includes a low-noise preamplifier circuit, a bandpass filter circuit, an analog anti-aliasing circuit, and an A / D conversion circuit connected in series. The lower cutoff frequency of the bandpass filter circuit is preferably 50kHz, and the upper cutoff frequency is preferably 500kHz. The sampling frequency of the A / D conversion circuit is preferably 2MHz, and the conversion accuracy is 16 bits.

[0123] The grounding brush current acquisition unit is used to acquire the high-frequency pulse current signal generated by shaft current discharge in the grounding brush circuit.

[0124] Preferably, the front end of the grounding brush current acquisition unit includes a Rogowski coil or a high-frequency current transformer, a shaping circuit, and an A / D conversion circuit, wherein the sampling frequency of the A / D conversion circuit is preferably 100kHz.

[0125] The standard diagnostic pulse generation unit is used to generate standard diagnostic pulses with fixed amplitude and fixed width according to a preset period to perform online verification of the coupling state of the acoustic emission sensor; this unit corresponds to the function of the standard diagnostic pulse generation unit in step S2 of embodiment 1.

[0126] Preferably, the standard diagnostic pulse has an amplitude of 5V, a pulse width of 100μs, and a repetition period of 24h. The standard diagnostic pulse generation unit can be implemented using a field-programmable gate array (FPGA) or a dedicated signal generator. The diagnostic pulse output by this unit is coupled to the acoustic emission sensor through a dedicated signal line or waveguide rod.

[0127] The alarm output unit is used to receive the fault classification judgment result output in step S5 of Embodiment 1, and output the corresponding differentiated maintenance instructions to the wind turbine main control system according to the judgment result.

[0128] The acoustic emission acquisition unit, the grounding brush current acquisition unit, the standard diagnostic pulse generation unit, and the embedded processing unit exchange data through a digital communication interface. The digital communication interface can be SPI, CAN, RS485, or Ethernet, and opto-isolation or magnetic isolation is preferred to further enhance anti-interference capability.

[0129] To suppress electromagnetic interference and common-mode interference within the wind turbine nacelle, both the acoustic emission acquisition unit and the grounding brush current acquisition unit are powered by isolated power supplies, preferably isolated DC-DC modules. Each signal input terminal is preferably equipped with a shielding layer and adopts a single-point grounding structure to reduce ground loop noise.

[0130] To ensure the accuracy of timestamp alignment in Embodiment 1, the embedded processing unit outputs a unified hardware synchronization trigger signal to the acoustic emission acquisition unit and the grounding brush current acquisition unit; or time synchronization is achieved by combining the IEEE 1588 PTP protocol with hardware timestamps.

[0131] When the wind farm communication link is interrupted, the embedded processing unit caches the alarm results in the local non-volatile storage area and retransmits them to the wind turbine main control system after communication is restored to avoid the loss of alarm information. The alarm information may include the current time window number, time overlap, total number of events (total number of acoustic emission events and total number of current pulse events), coupling state coefficient, and threshold comparison results, so that maintenance personnel can quickly verify the fault type.

[0132] It should be noted that the monitoring system provided in this embodiment coordinates the collaborative work of various units through an embedded processing unit, realizing a complete closed loop from the establishment of calibration parameter library, self-diagnosis of sensor coupling status, synchronous acquisition of acoustic emission and current pulse events, time overlap analysis to fault classification and differential alarm output; the system is deeply integrated with the main bearing hardware and grounding brush circuit, which can effectively distinguish between three fault types: mechanical spalling, electrical corrosion and grounding circuit abnormality, and perform real-time online self-diagnosis of the coupling status of acoustic emission sensors, thereby improving the accuracy, reliability and maintainability of wind turbine main bearing condition monitoring.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for monitoring the acoustic emission status of a wind turbine main bearing, characterized in that: Includes the following steps: S1. Establish a calibration parameter library, which includes at least the acoustic emission event extraction amplitude threshold, the grounding brush current pulse event extraction amplitude threshold, the acoustic emission event count anomaly threshold, the current pulse event count anomaly threshold, the coupling state threshold, and the time overlap threshold. S2. Periodically send standard diagnostic pulses to the acoustic emission sensor, calculate the current coupling state coefficient and compare it with the coupling state threshold. If it is determined to be a coupling failure, generate a coupling failure flag; if it is determined to be a valid coupling state, generate a valid channel flag. S3. When the acoustic emission sensor is in an effective coupling state, the original waveform sequence of acoustic emission and the original waveform sequence of grounding brush current are acquired synchronously. After time stamp alignment, the time sequence of acoustic emission events and the time sequence of grounding brush current pulse events are extracted respectively, and the total number of acoustic emission events and the total number of grounding brush current pulse events are counted. S4. Calculate the time overlap between the acoustic emission event time series and the grounding brush current pulse event time series; S5. Based on the comparison results of the coupling failure mark, time overlap, total number of acoustic emission events and total number of grounding brush current pulse events with the acoustic emission event count abnormal threshold and the grounding brush current pulse event count abnormal threshold, it is determined to be one of the following: sensor maintenance alarm, shaft current electrical corrosion alarm, mechanical peeling alarm, grounding brush circuit abnormal alarm, composite abnormal alarm or normal state. S6. Output the corresponding differentiated maintenance instructions to the wind turbine main control system based on the judgment result.

2. The method for monitoring the acoustic emission status of a wind turbine main bearing as described in claim 1, characterized in that: In step S1, the acoustic emission event extraction amplitude threshold and the grounding brush current pulse event extraction amplitude threshold are used to extract a single abnormal event. The acoustic emission event count abnormal threshold and the grounding brush current pulse event count abnormal threshold are used to determine whether the number of abnormal events within a preset time window exceeds the limit. The coupling state threshold is a preset ratio threshold obtained based on the standard diagnostic pulse response amplitude calibration. The time overlap threshold is a synchronization determination threshold obtained based on the calibration of known electro-corrosion samples.

3. The method for monitoring the acoustic emission status of a wind turbine main bearing as described in claim 2, characterized in that: In step S5, the uniqueness classification determination is performed according to the following priority: If the aforementioned coupling failure marker exists, it is determined to be a sensor maintenance alarm; If the time overlap is greater than or equal to the time overlap threshold, and the total number of acoustic emission events is greater than or equal to the acoustic emission event count abnormal threshold, and the total number of grounding brush current pulse events is greater than or equal to the grounding brush current pulse event count abnormal threshold, then it is determined to be an shaft current electrical corrosion alarm. If the aforementioned conditions are not met, and the total number of acoustic emission events is greater than or equal to the abnormal threshold for acoustic emission event counting, and the total number of grounding brush current pulse events is less than the abnormal threshold for grounding brush current pulse event counting, then it is determined to be a mechanical peeling alarm. If the aforementioned conditions are not met, and the total number of acoustic emission events is less than the abnormal threshold for acoustic emission event counting, and the total number of grounding brush current pulse events is greater than or equal to the abnormal threshold for grounding brush current pulse event counting, then it is determined to be an abnormal alarm of the grounding brush circuit. If the aforementioned conditions are not met, and the total number of acoustic emission events is greater than or equal to the abnormal threshold for acoustic emission event counting, and the total number of grounding brush current pulse events is greater than or equal to the abnormal threshold for grounding brush current pulse event counting, but the time overlap is less than the time overlap threshold, then it is determined to be a composite abnormal alarm. If none of the above conditions are met, it is considered normal.

4. The method for monitoring the acoustic emission status of a wind turbine main bearing as described in claim 3, characterized in that: In step S6, the following differentiated maintenance instructions are output based on different judgment results: When a sensor maintenance alarm is triggered, the output will check the coupling layer of the acoustic emission sensor. When a shaft current electrical corrosion alarm is triggered, the output will check the grounding brush and the shaft current bypass circuit. When a mechanical spalling alarm is triggered, an output will be sent to check the raceway and rolling elements of the main bearing. When a grounding brush circuit malfunction alarm is triggered, the output will check the grounding brush resistance and contact status. When a compound abnormality alarm occurs, the output synchronously checks the main bearing and grounding brush circuit.

5. The method for monitoring the acoustic emission status of a wind turbine main bearing as described in claim 4, characterized in that: The standard diagnostic pulse is generated by a diagnostic pulse generation unit located near the acoustic emission sensor and coupled to the acoustic emission sensor via a signal line or waveguide structure.

6. The method for monitoring the acoustic emission status of a wind turbine main bearing as described in claim 5, characterized in that: The acoustic emission sensor is mounted on the main bearing housing and maintains acoustic coupling with the main bearing housing.

7. A system for monitoring the acoustic emission status of a wind turbine main bearing, used to execute the method for monitoring the acoustic emission status of a wind turbine main bearing as described in any one of claims 1 to 6, characterized in that, The system includes an embedded processing unit, an acoustic emission acquisition unit, a grounding brush current acquisition unit, a standard diagnostic pulse generation unit, and an alarm output unit; The embedded processing unit is communicatively connected to the acoustic emission acquisition unit, the grounding brush current acquisition unit, the standard diagnostic pulse generation unit, and the alarm output unit, and stores a calibration parameter library. The embedded processing unit is used to perform calibration parameter generation, coupling state determination, event extraction, time overlap calculation, fault classification determination, and alarm output control. The acoustic emission acquisition unit is used to acquire the acoustic emission waveform during the operation of the main bearing; The grounding brush current acquisition unit is used to acquire high-frequency pulse current signals in the grounding brush circuit. The standard diagnostic pulse generation unit is used to generate standard diagnostic pulses to perform online verification of the coupling state of the acoustic emission sensor; The alarm output unit is used to output corresponding maintenance commands to the main control system.

8. The acoustic emission status monitoring system for the main bearing of a wind turbine as described in claim 7, characterized in that: The acoustic emission acquisition unit includes an acoustic emission sensor and a preamplifier circuit, a filter circuit, and an A / D conversion circuit connected in sequence; the grounding brush current acquisition unit includes a current sensor, a shaping circuit, and an A / D conversion circuit.

9. The acoustic emission status monitoring system for the main bearing of a wind turbine generator as described in claim 8, characterized in that: The analog front-end of the acoustic emission acquisition unit and the grounding brush current acquisition unit is powered by an isolated power supply; the communication interface between the embedded processing unit and each acquisition unit is connected in an isolated manner.

10. The acoustic emission status monitoring system for the main bearing of a wind turbine generator as described in claim 9, characterized in that: When the communication link is interrupted, the embedded processing unit caches the fault classification and judgment results and related data in the local storage area, and retransmits them to the wind turbine main control system after the communication is restored; the related data includes the current time window number, time overlap, total number of acoustic emission events, total number of current pulse events and coupling state coefficient.

Images