Ceramic finger state detection method and system

Abstract

The invention relates to the technical field of ceramic finger detection, in particular to a ceramic finger state detection method and system, and the method comprises the steps: respectively collecting dynamic data and static data based on the movement and pause of a ceramic finger in a single working cycle; collecting dynamic data and static data of the ceramic finger in multiple working cycles in the initial health state, and generating a health state set; obtaining a mechanical vibration response value of an initial section, a stress wave release value of a middle section and a resistance value of an idle section in a working cycle to form a health signature of the ceramic finger; in a subsequent working cycle, detecting a real-time signal of the ceramic finger, comparing the real-time signal with the health signature, and calculating an instant health score; and integrating multiple times of instant health scores, obtaining a score decreasing trend, and generating a detection result. According to the method, the health signature is constructed by fusing the dynamic data and the static data, and the real-time health score and trend are calculated to analyze the health state of the ceramic finger, so that the detection accuracy and reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of ceramic finger detection technology, and in particular to a method and system for detecting the state of ceramic fingers. Background Technology

[0002] Ceramic fingers are special structural components used in high-temperature, highly corrosive, or high-stress environments. They are typically made of high-performance ceramic materials and possess properties such as wear resistance, heat resistance, and electrical insulation. However, due to the brittleness of ceramic materials, ceramic fingers are susceptible to microcracks, structural fatigue, or changes in conductivity during service, leading to shortened service life or even failure. Monitoring and evaluating the operating status of ceramic fingers during long-term equipment operation can identify structural deterioration, crack propagation, or abnormal resistance in advance, thereby preventing sudden failures.

[0003] Existing methods often rely on a single time point or a small number of sample signals for judgment. Vibration, stress wave and resistance signals are analyzed independently. Single signals are greatly affected by noise, and small damage or early crack propagation is difficult to detect reliably, which can easily lead to delayed warnings. Moreover, they lack trend analysis based on multiple test results and cannot accurately reflect the dynamic process of ceramic fingers from health to degradation.

[0004] Therefore, there is a need for a method and system that can detect the health status of ceramic fingers in real time, accurately and trend-based under complex working conditions. Summary of the Invention

[0005] In view of at least one of the above technical problems, the present invention provides a method and system for detecting the state of ceramic fingers, which adopts multi-source signal fusion and health signature construction, calculates real-time scores and trend analysis of the health status of ceramic fingers, so as to realize the quantitative assessment of the health status of ceramic fingers and improve the accuracy and reliability of detection.

[0006] This invention provides a method for detecting the state of a ceramic finger, comprising the following steps: S1: Based on the movement and pauses of the ceramic finger within a single working cycle, dynamic data and static data are collected respectively; the dynamic data includes stress wave signals and vibration signals, and the static data includes surface resistance signals. S2: Collect the dynamic data and static data of the ceramic finger during multiple working cycles in its initial healthy state, and generate a health status set; S3: Based on the health status set, obtain the mechanical vibration response value of the initial segment, the stress wave release value of the middle segment, and the resistance value of the idle segment within the working cycle, and compose the health signature of the ceramic finger according to the vibration response value, stress wave release value, and resistance value. S4: In subsequent work cycles, detect the real-time signal of the ceramic finger, compare the real-time signal with the health signature, and calculate the instant health score; S5: Based on the real-time health scores mentioned multiple times, obtain the decreasing trend of the scores and generate the test results.

[0007] In some embodiments of the present invention, step S2, generating a health status set, includes: The dynamic data and static data are collected in multiple complete work cycles, and the data from different work cycles are classified and stored according to the start segment, middle segment and idle segment to form a staged health status set.

[0008] In some embodiments of the present invention, step S3, which composes the health signature of the ceramic finger, includes: Frequency domain analysis is performed on the vibration signal of the initial segment, and the fundamental frequency amplitude and harmonic energy are extracted as the vibration response value. The energy accumulation of the stress wave signal in the middle section is calculated, and the maximum impact peak value is extracted to obtain the stress wave release value; The surface resistance signal in the idle segment is filtered for stable intervals, and its average value and variance are calculated to obtain the resistance value. The vibration response value, the stress wave release value, and the resistance value are combined in chronological order to form a multidimensional feature vector, which is defined as the health signature of the ceramic finger.

[0009] In some embodiments of the present invention, the health signature combination process includes: The vibration response value, the stress wave release value, and the resistance value are concatenated in the order of working cycle time to form an initial feature vector; By applying dynamic weights to different time periods in the initial feature vector based on environmental variables, a weighted feature vector is obtained; Uncertainty information is introduced for each feature in the weighted feature vector, and an extended feature vector is formed based on the uncertainty information and the weighted feature vector. The extended feature vector is denoted as the health signature.

[0010] In some embodiments of the present invention, step S4, calculating the instant health score includes: In subsequent work cycles, real-time vibration signals, real-time stress wave signals and real-time resistance signals are collected and real-time vibration response values, real-time stress wave release values ​​and real-time resistance values ​​are extracted respectively and combined into a real-time feature vector. The real-time feature vector is consistent with the health signature in the same dimension. The real-time feature vector is mapped to an energy parameter, and a health baseline parameter is obtained by statistical analysis of the health status set. The energy parameter and the health baseline parameter are compared and the difference is calculated to obtain the energy offset value. The energy offset values ​​are recorded as a time series in the order of the working cycle and then smoothed to form an energy trajectory curve; Based on the energy trajectory curve, calculate the offset and offset rate relative to the health benchmark parameter; The instant health score is calculated based on the offset and the offset rate.

[0011] In some embodiments of the present invention, the energy parameter and the health benchmark parameter are used to perform a difference calculation, including: The vibration response value within the health state set is mapped to a structural integrity parameter, the stress wave release value is mapped to a crack evolution parameter, and the resistance value is mapped to a conductivity stability parameter. Statistical analysis was performed on the structural integrity parameters, crack evolution parameters, and electrical stability parameters to obtain their mean, variance, and distribution range, thus forming the corresponding health baseline energy parameters. Based on the aforementioned health baseline energy parameters, mechanical path baseline curves, crack path baseline curves, and resistance path baseline curves are established. In subsequent work cycles, the energy parameters are compared with the corresponding baseline curves, and the difference values ​​are calculated and independently recorded as three types of degradation paths. The energy offset value is obtained by weighted fusion of the difference values ​​corresponding to the three types of degradation paths.

[0012] In some embodiments of the present invention, calculating the instantaneous health score based on the offset and the offset rate includes: The over-offset d is calculated based on the offset, and the over-rate r is calculated based on the offset rate; The adjustment amount is P = α·d + β·r + γ·d·r; Calculate the instant health score S = 100 / (1+P); Where α, β, and γ are weighting coefficients greater than 0.

[0013] In some embodiments of the present invention, the specific calculation process for the over-offset d and the over-rate r is as follows: The over-offset d=max(0,d0-t) d The supervelocity quantity r = max(0, r0 - t) r ); Where d0 is the offset; r0 is the offset rate; t dThe offset threshold is a preset constant; t r The offset rate threshold is a preset constant.

[0014] The present invention also provides a ceramic finger state detection system, comprising: The data acquisition and generation module collects dynamic and static data based on the movement and pauses of the ceramic finger in a single work cycle. It also collects the dynamic and static data of the ceramic finger in multiple work cycles in the initial healthy state and generates a health status set. The component module acquires the mechanical vibration response value of the initial segment, the stress wave release value of the middle segment, and the resistance value of the idle segment within the working cycle based on the health status set, and composes the health signature of the ceramic finger based on the vibration response value, stress wave release value, and resistance value. The detection and calculation module detects the real-time signal of the ceramic finger in subsequent work cycles, compares the real-time signal with the health signature, and calculates an instant health score. The comprehensive results module integrates multiple real-time health scores, obtains the decreasing trend of the scores, and generates test results.

[0015] In some embodiments of the present invention, the detection calculation module includes: The real-time acquisition unit acquires real-time vibration signals, real-time stress wave signals, and real-time resistance signals in subsequent work cycles, and extracts real-time vibration response values, real-time stress wave release values, and real-time resistance values ​​respectively, combining them into a real-time feature vector. The real-time feature vector is consistent with the health signature in the same dimension. The mapping calculation unit maps the real-time feature vector into an energy parameter, obtains a health benchmark parameter based on statistical analysis of the health status set, and calculates the difference between the energy parameter and the health benchmark parameter to obtain an energy offset value. The trajectory recording unit records the energy offset value as a time series according to the working cycle sequence, and performs smoothing processing to form an energy trajectory curve; The scoring calculation unit calculates the offset and offset rate relative to the health benchmark parameter based on the energy trajectory curve, and calculates the instantaneous health score based on the offset and the offset rate.

[0016] The beneficial effects of this invention are as follows: This invention constructs a health signature containing both dynamic and static features by simultaneously acquiring multi-source signals such as vibration, stress waves, and surface resistance within a single working cycle. In subsequent runs, real-time signals are compared and scored to comprehensively characterize the structural integrity, crack evolution, and electrical stability of the ceramic finger. Furthermore, by introducing an instantaneous health score based on offset degree and offset rate, a quantitative evaluation of the health status is achieved. Through trend analysis based on the scoring results of multiple cycles, the accelerated degradation process can be identified in a timely manner, generating reliable detection results. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the steps of the ceramic finger state detection method in an embodiment of the present invention; Figure 2 This is a schematic diagram of the steps in the health signature combination process in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the steps of calculating an instant health score in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the steps of calculating the difference between energy parameters and health baseline parameters in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0022] This invention provides a method such as Figures 1 to 4 The ceramic finger state detection method shown includes the following steps: S1: Based on the movement and pauses of the ceramic finger in a single working cycle, dynamic data and static data are collected respectively; dynamic data includes stress wave signals and vibration signals, and static data includes surface resistance signals. During a single working cycle of a ceramic finger, it typically undergoes a motion phase and a pause phase. To comprehensively reflect the mechanical and electrical characteristics of the ceramic finger, this embodiment acquires dynamic signals during the motion phase and static signals during the pause phase. Dynamic signals include vibration signals and stress wave signals. Vibration signals can be obtained by using an accelerometer, while stress wave signals can be obtained by using an acoustic emission sensor or a piezoelectric sensor. Static signals include surface resistance signals, which are preferably measured using the four-probe method to ensure the accuracy and stability of the measurement results.

[0023] S2: Collect dynamic and static data of the ceramic finger during multiple working cycles in its initial healthy state, and generate a health status set; In the initial stage of ceramic finger operation, it is in a healthy state. The dynamic and static data collected from multiple complete work cycles accurately reflect the various parameters of ceramic finger in a healthy state, which can be used as a benchmark for subsequent work. In order to facilitate subsequent feature extraction and comparative analysis, the data of different work cycles are divided into the initial segment, the middle segment and the idle segment according to time segments, forming a multi-dimensional and segmented health status set.

[0024] S3: Based on the health status set, obtain the mechanical vibration response value of the initial segment, the stress wave release value of the middle segment, and the resistance value of the idle segment within the working cycle, and compose the health signature of the ceramic finger according to the vibration response value, stress wave release value, and resistance value. S4: In subsequent work cycles, detect the real-time signal of the ceramic finger, compare the real-time signal with the health signature, and calculate the instant health score; S5: By combining multiple real-time health scores, the decreasing trend of the scores is obtained and the detection result is generated. In order to avoid misjudgment caused by a single abnormality, the real-time health scores in multiple work cycles are comprehensively analyzed and the real-time scores are used to form a sequence curve. Trend fitting and slope analysis can be performed on it. When a continuous decreasing trend of the score is detected and the decrease exceeds the preset threshold, a degradation detection result is generated and a warning message can be output.

[0025] As a critical component operating under high temperature, high load, and complex conditions, ceramic fingers are susceptible to degradation or even failure due to vibration, impact, crack initiation, and changes in conductivity. This embodiment simultaneously collects vibration, stress wave, and resistance signals within a single work cycle to establish a health signature containing both dynamic and static characteristics. In subsequent operations, real-time signals are compared with this health signature to calculate an instantaneous health score. The detection result is then generated by judging the comprehensive trend of multiple scores. This achieves a quantitative, unified, and intuitive evaluation of the operating status of ceramic fingers and can dynamically capture degradation trends. Compared with traditional detection methods that rely on a single indicator, this method is more accurate and reliable, and can provide early warning of potential failures, thereby improving the safety of equipment operation and the timeliness of maintenance.

[0026] In some embodiments of the present invention, step S2, generating a health status set, includes: Dynamic and static data are collected in multiple complete work cycles, and the data from different work cycles are classified and stored according to the start segment, middle segment and idle segment to form a staged health status set.

[0027] When the ceramic finger is in its initial healthy state, it is preferable to continuously collect signal data for no less than 20 complete working cycles. Each working cycle includes both dynamic and static data. The initial segment usually corresponds to the time interval during which the ceramic finger begins to move and gradually loads itself, which can significantly reflect the performance of the ceramic finger in terms of stiffness and response characteristics. The middle segment corresponds to the process when the ceramic finger is under stable force or in constant motion. The stress wave signal in this stage can best characterize the crack initiation and propagation. The idle segment corresponds to the process when the ceramic finger stops moving or is in a standby state. During this stage, the resistance signal is relatively stable, which is convenient for use as a reference for conductivity performance.

[0028] Preferably, the stage division can be achieved by combining time segmentation with trigger signals; for example, a single work cycle can be preset to 10 seconds, with the first 2 seconds defined as the start segment, the middle 6 seconds as the middle segment, and the last 2 seconds as the idle segment; or the motor control signal or displacement signal collected by the sensor can be used as the trigger condition to achieve precise segmentation.

[0029] In some embodiments of the present invention, step S3, which composes the health signature of the ceramic finger, includes: Frequency domain analysis was performed on the vibration signal of the initial segment, and the fundamental frequency amplitude and harmonic energy were extracted as vibration response values. Since the ceramic finger is just beginning to move, the vibration signal is most sensitive to structural integrity. Therefore, the Fast Fourier Transform (FFT) method was preferred for frequency domain analysis of the vibration signal to extract the fundamental frequency amplitude and several harmonic energies. The fundamental frequency amplitude reflects the overall stiffness change, while the harmonic energy reflects the nonlinear response caused by potential cracks or defects. The extracted results were defined as the vibration response values.

[0030] The stress wave signal in the middle section is subjected to energy accumulation calculation, and the maximum impact peak value is extracted to obtain the stress wave release value. The ceramic finger is under stable stress or constant motion. The stress wave signal can reflect the initiation and propagation of cracks. In this embodiment, the stress wave signal in the middle section is subjected to energy accumulation calculation, that is, the signal is squared and integrated to obtain the total stress release energy. The maximum impact peak value in this interval is further extracted to enhance the sensitivity to local crack propagation. The above two results are combined and defined as the stress wave release value.

[0031] The surface resistance signal in the idle segment is screened for stable intervals, and its average value and variance are calculated to obtain the resistance value. The ceramic finger is no longer subjected to external force, and its resistance signal is in a stable state, which is suitable as a reference for conductivity. The resistance signal is screened for stable intervals. The sliding window method is preferred to determine the signal fluctuation interval. When the variance is less than the set threshold, it is considered as a stable interval. The average value and variance of the resistance signal are calculated within this interval and used as the resistance value.

[0032] The vibration response value, stress wave release value, and resistance value are combined in chronological order to form a multidimensional feature vector, which is defined as the health signature of the ceramic finger. This is used to comprehensively characterize its mechanical and electrical properties in a healthy state, and has higher stability and sensitivity, providing a solid foundation for subsequent real-time health scoring and trend analysis.

[0033] In some embodiments of the present invention, such as Figure 2 As shown, the combined process of health signature includes: The vibration response value, stress wave release value, and resistance value are concatenated in the order of working cycle time to form an initial feature vector; By applying dynamic weights to different time periods in the initial feature vector based on environmental variables, a weighted feature vector is obtained. Uncertainty information is introduced for each feature in the weighted feature vector, and an extended feature vector is formed based on the uncertainty information and the weighted feature vector. The extended feature vector is recorded as the health signature.

[0034] Considering that the operating status of the ceramic finger is affected by environmental factors such as temperature, load, or rotational speed, this embodiment applies corresponding weights to features at different stages based on the initial feature vector. Under higher temperatures, the weight of resistance features in the overall health signature can be reduced; under increased load, the weight of stress wave features can be increased, making the contribution of features under different operating conditions to the overall health signature more reasonable. To reflect the fluctuations present during signal acquisition, uncertainty information is introduced. Uncertainty information is the fluctuation range or confidence interval of the same type of feature under healthy conditions. For example, for vibration response values, the degree of fluctuation can be measured by their standard deviation over multiple health cycles; the same method can be used for stress wave release values ​​and resistance values. The weighted features and uncertainty information are combined to form an expanded feature vector, which is defined as the health signature of the ceramic finger. Compared with the traditional method that only contains a single value, this health signature not only reflects typical feature values ​​but also includes expressions of environmental adaptability and fluctuation range, maintaining higher stability and reliability for state detection under different operating conditions.

[0035] In some embodiments of the present invention, such as Figure 3 As shown, in step S4, calculating the immediate health score includes: In subsequent work cycles, using the same processing methods, real-time vibration signals, real-time stress wave signals, and real-time resistance signals are collected, and real-time vibration response values, real-time stress wave release values, and real-time resistance values ​​are extracted respectively. The three are then combined into a real-time feature vector according to the time sequence of the initial segment, the middle segment, and the idle segment. The real-time feature vector is consistent with the health signature in the same dimension, thereby ensuring the effectiveness of subsequent comparisons. The real-time feature vector is mapped to the energy parameter. Based on the statistical analysis of the health status set, such as calculating the corresponding mean, variance or distribution interval, the health baseline parameter is obtained. The energy parameter and the health baseline parameter are compared and the difference is calculated to obtain the energy offset value. The energy offset values ​​are recorded as a time series in the order of the working cycle. In order to reduce the influence of environmental noise and random fluctuations, it is preferable to use methods such as moving average or low-pass filtering to smooth the time series and form an energy trajectory curve, which intuitively reflects the trend of the health status of the ceramic finger over time. Based on the energy trajectory curve, the deviation and deviation rate relative to the health baseline parameters are calculated; the deviation reflects the size of the gap between the health state and the baseline, while the deviation rate reflects the speed of the degradation process. Real-time health scores are calculated based on offset and offset rate. When the offset is small and the offset rate is low, the score is high, indicating that the ceramic finger is in good operating condition. When the offset continues to increase and the offset rate is fast, the score gradually decreases, suggesting that the ceramic finger may be entering a degenerative stage. This significantly reduces the impact of single-point abnormalities on the judgment and can also sensitively capture early deterioration trends, enabling intuitive quantification and dynamic tracking of the operating status of the ceramic finger. Based on the above embodiments, such as Figure 4 As shown, the difference between energy parameters and health baseline parameters is calculated, including: The vibration response values ​​within the healthy state set are mapped to structural integrity parameters to characterize the overall stiffness and mechanical stability of the ceramic finger; the stress wave release values ​​are mapped to crack evolution parameters to reflect the activity of crack initiation and propagation; and the resistance values ​​are mapped to electrical stability parameters to measure the electrical performance stability of the ceramic finger under unloaded conditions. Statistical analysis was performed on the structural integrity parameters, crack evolution parameters, and electrical stability parameters to obtain their mean, variance, and distribution range, thus forming the corresponding health baseline energy parameters. Based on the health baseline energy parameters, mechanical path baseline curves, crack path baseline curves, and resistance path baseline curves are established. These baseline curves can be constructed using fitting or piecewise statistical methods, so that each type of signal has a clear physical meaning and statistical reference, thereby enabling the quantification of their respective degradation degree through differential calculation in subsequent operation. In subsequent work cycles, the energy parameters are compared with the corresponding baseline curves, the difference values ​​are calculated and independently recorded as three types of degradation paths. The three types of degradation paths independently track the degradation trend of structural integrity, the expansion trend of crack evolution, and the fluctuation of electrical conductivity. By weighted and fused the difference values ​​corresponding to the three types of degradation paths, an energy offset value is obtained. This value not only comprehensively reflects the overall health status of the ceramic finger, but also allows for flexible adjustment of the weights of different features according to the application scenario. It achieves systematic integration of multi-source information, resulting in more comprehensive, accurate, and robust detection results. It can more sensitively capture early degradation trends and provide a reliable basis for safe operation and maintenance decisions of the equipment.

[0036] In some embodiments of the present invention, calculating an instantaneous health score based on offset and offset rate includes: The over-offset d is calculated based on the offset, and the over-rate r is calculated based on the offset rate; The adjustment amount is P = α·d + β·r + γ·d·r; α·d represents the absolute difference between the current state of the ceramic finger and the healthy baseline parameters. When the mechanical properties, electrical properties and other parameters of the ceramic finger have deviated from the normal range, it indicates that the structure is at risk of deterioration. Therefore, this part is added to reflect the degree of decline in the current health level of the ceramic finger.

[0037] β·r represents the rate at which the health of the ceramic finger deteriorates. When crack propagation accelerates, fatigue accumulates rapidly, or resistance fluctuations intensify, even if the absolute deviation is small, it indicates a potential for rapid failure. This part ensures that the instant health score can capture the degradation trend and avoid delayed pre-tightening.

[0038] γ·d·r reflects the superposition effect of offset amplitude and degradation rate. Considering that the ceramic finger has already degraded significantly (large offset) and the degradation rate is relatively fast, the failure risk is often not simply the sum of the two, but rather exhibits an amplification or even acceleration effect. By introducing the product term of the over-offset d and the over-rate r, the instantaneous health score S can impose additional penalties on high-risk working conditions, achieving a more sensitive detection effect.

[0039] Calculate the immediate health score S = 100 / (1 + P); Where α, β, and γ are weighting coefficients greater than 0.

[0040] The offset is further converted into a super-offset d, meaning that the calculation is only made when the offset exceeds a preset threshold, avoiding misjudgments caused by minor fluctuations. The offset rate is converted into a super-rate r, which only contributes when it exceeds a threshold, thereby enhancing sensitivity to abnormal changes. An adjustment quantity P is constructed, jointly determined by the super-offset d and the super-rate r. The contribution corresponding to the offset reflects the absolute decline in health level, while the contribution corresponding to the offset rate reflects the deterioration trend of health level. The interaction term of the two reflects the amplification effect when the magnitude and speed of deterioration are superimposed. By setting weight coefficients, the influence of different factors in the instant health score S can be flexibly adjusted. For example, the weight of offset is increased when emphasizing stability, and the weight of rate is increased when focusing on sudden changes. The instant health score is calculated using an inverse proportional formula. The larger the adjustment quantity, the lower the instant health score S. When the adjustment quantity is close to zero, the instant health score S approaches the full score, ensuring the intuitiveness and consistency of the scoring system, facilitating rapid judgment in practical applications, significantly reducing the risk of misjudgment caused by small fluctuations, and enhancing sensitivity to early deterioration trends.

[0041] Based on the above embodiments, the specific calculation process for the over-offset d and the over-rate r is as follows: Overoffset d=max(0,d0-t) d The overspeed quantity r = max(0, r0 - t) r ); Where d0 is the offset; r0 is the offset rate; t d The offset threshold is a preset constant; t r The offset rate threshold is a preset constant.

[0042] During the condition detection of ceramic fingers, environmental noise, operating condition disturbances, or measurement errors are inevitably mixed into the real-time signal. If the offset d0 and offset rate r0 are directly used as the calculation basis, the real-time health score S is prone to fluctuation due to slight fluctuations, resulting in false alarms or false alarms. Therefore, in this embodiment, when the actual offset d0 exceeds the preset threshold, only the excess part is included in the calculation; otherwise, it is recorded as zero. When the actual offset rate r0 exceeds the preset threshold, the excess part is taken as the valid value, filtering out the small fluctuations within the normal range, so that the scoring system only responds to real abnormal changes. The threshold is set manually, usually based on the allowable fluctuation range in the actual application scenario. For example, the offset threshold can be set according to the upper limit of normal fluctuation of ceramic fingers in a healthy state, and the offset rate threshold can be set according to the acceptable deterioration rate during long-term operation.

[0043] The present invention also provides a ceramic finger state detection system, comprising: The data acquisition and generation module collects dynamic and static data based on the movement and pauses of the ceramic finger in a single work cycle. It also collects dynamic and static data of the ceramic finger in multiple work cycles in its initial healthy state and generates a health status set. The module obtains the mechanical vibration response value of the initial segment, the stress wave release value of the middle segment, and the resistance value of the idle segment within the working cycle based on the health status set. The health signature of the ceramic finger is then composed based on the vibration response value, stress wave release value, and resistance value. The detection and calculation module detects the real-time signal of the ceramic finger in subsequent work cycles, compares the real-time signal with the health signature, and calculates the instant health score. The comprehensive results module combines multiple real-time health scores, obtains the decreasing trend of the scores, and generates test results.

[0044] The detection system described above in this invention can effectively realize a method for detecting the state of ceramic fingers, and the technical effects it can achieve are as described in the above embodiments, which will not be repeated here.

[0045] In some embodiments of the present invention, the detection calculation module includes: The real-time acquisition unit acquires real-time vibration signals, real-time stress wave signals, and real-time resistance signals in subsequent working cycles, and extracts real-time vibration response values, real-time stress wave release values, and real-time resistance values ​​respectively, combining them into a real-time feature vector. The real-time feature vector maintains the same dimension as the health signature. The mapping calculation unit maps real-time feature vectors to energy parameters, obtains health baseline parameters based on statistical analysis of the health status set, and calculates the difference between the energy parameters and the health baseline parameters to obtain the energy offset value. The trajectory recording unit records the energy offset values ​​as a time series according to the working cycle sequence, and performs smoothing processing to form an energy trajectory curve; The scoring calculation unit calculates the offset and offset rate relative to the health baseline parameters based on the energy trajectory curve, and calculates the instantaneous health score based on the offset and offset rate.

[0046] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the methods provided by the present invention, which will not be elaborated here.

[0047] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A ceramic finger state detection method, characterized by, The method comprises the following steps: S1: Collecting dynamic data and static data based on the movement and pause of the ceramic finger in a single working cycle; the dynamic data includes stress wave signals and vibration signals, and the static data includes surface resistance signals; S2: Collecting the dynamic data and the static data of the ceramic finger in multiple working cycles in the initial healthy state to generate a health state set; S3: Obtaining the mechanical vibration response value of the starting segment, the stress wave release value of the middle segment and the resistance value of the idle segment in the working cycle based on the health state set, and composing the health signature of the ceramic finger according to the vibration response value, the stress wave release value and the resistance value; S4: In the subsequent working cycle, detecting the real-time signals of the ceramic finger, comparing the real-time signals with the health signature, and calculating the instant health score; S5: Integrating multiple instant health scores to obtain a decreasing trend of the score and generating a detection result.

2. The ceramic finger state detection method according to claim 1, characterized by, In step S2, the health state set is generated, including: Collecting the dynamic data and the static data in multiple complete working cycles, and storing the data of different working cycles according to the starting segment, the middle segment and the idle segment to form a phased health state set.

3. The ceramic finger state detection method according to claim 2, characterized by, In step S3, the health signature of the ceramic finger is composed, including: Performing frequency domain analysis on the vibration signal of the starting segment, extracting the fundamental frequency amplitude and harmonic energy as the vibration response value; Performing energy accumulation calculation on the stress wave signal of the middle segment, and extracting the maximum impact peak value to obtain the stress wave release value; Performing stable interval screening on the surface resistance signal of the idle segment, and calculating the average value and variance to obtain the resistance value; Combining the vibration response value, the stress wave release value and the resistance value into a multi-dimensional feature vector in time sequence, which is defined as the health signature of the ceramic finger.

4. The ceramic finger state detection method according to claim 3, characterized by, The combination process of the health signature includes: Splicing the vibration response value, the stress wave release value and the resistance value in the working cycle time sequence to form an initial feature vector; Applying dynamic weights to different time segments in the initial feature vector according to environmental variables to obtain a weighted feature vector; Introducing uncertainty information for each feature in the weighted feature vector, and forming an extended feature vector based on the uncertainty information and the weighted feature vector, which is recorded as the health signature.

5. The ceramic finger state detection method according to claim 1, characterized by, In step S4, the instant health score is calculated, including: In the subsequent working cycle, collecting real-time vibration signals, real-time stress wave signals and real-time resistance signals and extracting real-time vibration response values, real-time stress wave release values and real-time resistance values to combine into a real-time feature vector, which is consistent with the health signature in the same dimension; Mapping the real-time feature vector into an energy parameter, obtaining a health benchmark parameter through statistical analysis of the health state set, and calculating the energy offset value by comparing the energy parameter with the health benchmark parameter; Recording the energy offset value in the working cycle sequence as a time series and performing smoothing processing to form an energy trajectory curve; Based on the energy trajectory curve, calculating the offset degree and offset rate relative to the health benchmark parameter; The instant health score is calculated based on the offset degree and the offset rate.

6. The ceramic finger state detection method according to claim 5, characterized by, The energy parameter is differentially calculated corresponding to the health benchmark parameter, including: The vibration response value in the initial health state set is mapped as a structural integrity parameter, the stress wave release value is mapped as a crack evolution parameter, and the resistance value is mapped as a conductive stability parameter; Statistical analysis is respectively performed on the structural integrity parameter, the crack evolution parameter, and the conductive stability parameter to obtain their mean value, variance, and distribution interval, and a corresponding health benchmark energy parameter is formed; Based on the health benchmark energy parameter, a mechanical path benchmark curve, a crack path benchmark curve, and a resistance path benchmark curve are established; In the subsequent working cycle, the energy parameter is compared with the corresponding benchmark curve, and a differential value is respectively calculated and independently recorded as three types of degradation paths; The differential values corresponding to the three types of degradation paths are weighted and fused to obtain the energy offset value.

7. The ceramic finger state detection method according to claim 5, characterized by, The instant health score is calculated based on the offset degree and the offset rate, including: According to the super offset amount d calculated according to the offset degree, and according to the super rate amount r calculated according to the offset rate; An adjustment amount P = α · d + β · r + γ · d · r is constructed; The instant health score S = 100 / (1+P) is calculated; Wherein, α, β, γ are weight coefficients greater than 0.

8. The ceramic finger state detection method according to claim 7, characterized by, The specific calculation process of the super offset amount d and the super rate amount r is: the overdisplacement d = max(0, d0- t d ); the overrate r = max(0, r0- t r ); Wherein, d0 is the offset degree; r0 is the offset rate; t d is the offset degree threshold, which is a constant artificially preset; t r is the offset rate threshold, which is a constant artificially preset.

9. A ceramic finger state detection system, which applies the ceramic finger state detection method according to any one of claims 1 to 8, characterized by, Including: The acquisition generation module acquires dynamic data and static data based on the movement and pause of the ceramic finger in a single working cycle, acquires the dynamic data and the static data of the ceramic finger in multiple working cycles in the initial health state, and generates a health state set; The acquisition module acquires the mechanical vibration response value of the initial segment, the stress wave release value of the middle segment, and the resistance value of the idle segment in the working cycle based on the health state set, and acquires the health signature of the ceramic finger according to the vibration response value, the stress wave release value, and the resistance value; The detection calculation module detects the real-time signal of the ceramic finger in the subsequent working cycle, compares the real-time signal with the health signature, and calculates the instant health score; The comprehensive result module integrates multiple instant health scores to obtain a score decreasing trend and generates a detection result.

10. The ceramic finger status detection system of claim 9, wherein, The detection calculation module includes: The real-time acquisition unit acquires real-time vibration signals, real-time stress wave signals, and real-time resistance signals in the subsequent working cycle, and extracts real-time vibration response values, real-time stress wave release values, and real-time resistance values, respectively, to form a real-time feature vector, and the real-time feature vector and the health signature maintain the same dimension; The mapping calculation unit maps the real-time feature vector to an energy parameter, obtains a health benchmark parameter through statistical analysis based on the health state set, differentially calculates the energy parameter corresponding to the health benchmark parameter to obtain an energy offset value; The trajectory recording unit records the energy offset value as a time sequence in the order of working cycles and performs smoothing processing to form an energy trajectory curve; A score calculation unit calculates a deviation degree and a deviation rate with respect to the health baseline parameter based on the energy trajectory curve, and calculates the instant health score based on the deviation degree and the deviation rate.

Images