Damping sealing material state evaluation method and system

Abstract

The invention discloses a damping sealing material state evaluation method and system, and relates to the technical field of damping sealing material state evaluation, and the method comprises the steps: obtaining multi-type response data generated by a damping sealing material under a preset excitation condition and corresponding environment working condition data; performing time sequence alignment, noise suppression and compensation processing on the multi-type response data based on the environmental condition data to obtain standardized characterization data; performing multi-dimensional feature extraction and fusion based on the standardized representation data, and constructing a state fingerprint vector; inputting the state fingerprint vector into a physical constraint state estimation model, and resolving to obtain a health index, a degradation level and a degradation parameter; and performing evaluation and life prediction based on the health index, the degradation grade, the degradation parameter and the environment working condition data to obtain a failure risk probability and a residual service life evaluation result. According to the method, the comparability of state data is improved, the recognition accuracy and interpretability are improved, and the comprehensive evaluation of the current state, the degradation mechanism and the future failure risk is realized.

Description

Technical Field

[0001] This invention relates to the field of damping sealing material condition assessment technology, specifically to a method and system for assessing the condition of damping sealing materials. Background Technology

[0002] Damping sealing materials are widely used in aero-engines, compressors, pump and valve equipment, rail transit equipment, and high-end rotating machinery. They serve multiple functions, including media isolation and sealing, vibration dissipation, impact buffering, and interface stability maintenance. As equipment evolves towards higher speeds, higher loads, greater temperature differences, and more complex media environments, damping sealing materials are subjected to the combined effects of multiple factors during long-term service, such as cyclic mechanical loads, thermal stress fluctuations, media penetration and erosion, and environmental humidity and heat changes. Their performance evolution exhibits significant nonlinear, time-varying, and multi-field coupling characteristics. Regarding condition monitoring and life assessment of these materials, related technologies have gradually evolved from traditional single static parameter detection to comprehensive assessment techniques combining dynamic response testing, thermal observation, electrical analysis, and signal processing methods. The assessment scope has also expanded from single material parameters to include material service status, degradation evolution trends, and failure risk assessment.

[0003] However, existing technologies for assessing the condition of damping sealing materials still have significant shortcomings. First, current methods primarily rely on hardness, compression set, local stress, single vibration response, or single thermal response for detection. This results in limited data dimensions, making it difficult to simultaneously reflect internal material degradation, interfacial behavior changes, and the impact of environmental disturbances. Consequently, the ability to identify early and complex degradation patterns is insufficient. Second, existing technologies typically use acquired signals directly for threshold judgment, lacking a unified mechanism for time-series alignment, noise suppression, and compensation processing in conjunction with environmental conditions. This makes signal drift caused by temperature, humidity, load fluctuations, and medium contact states easily misinterpreted as actual material degradation, leading to a lack of comparability in assessment results under different conditions. Third, existing methods generally lack state characterization models that can integrate multi-source features and reflect the true evolution of materials. In particular, they lack a technical path to extract a unified state fingerprint from multi-type response data and further combine it with physical constraints for state calculation. Therefore, it is difficult to simultaneously output health indices, degradation levels, and internal degradation state parameters that can be used for lifetime prediction. Fourth, existing technologies in life assessment mostly rely on empirical life estimation or single-point condition judgment, making it difficult to jointly predict the probability of failure risk and the remaining service life based on historical degradation trajectories and changes in environmental conditions. Therefore, it is impossible to achieve a coordinated assessment of the current state, degradation trend and future failure risk of damping sealing materials. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing methods for assessing the condition of damping sealing materials have problems such as a single dimension of condition representation, difficulty in distinguishing between environmental disturbances and actual material degradation, lack of unified fusion and physical constraint solution mechanism for multi-source response data, and how to achieve the collaborative output of current condition assessment, degradation trend identification and remaining service life prediction of damping sealing materials.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a method for evaluating the condition of damping sealing materials, comprising the following steps: S1: Acquire multi-type response data and corresponding environmental condition data generated by the damping sealing material under preset excitation conditions; S2: Based on the environmental condition data, the multi-type response data are subjected to time alignment, noise suppression and compensation processing to obtain standardized state characterization data; S3: Based on the standardized state characterization data, perform multi-dimensional feature extraction and fusion to construct the state fingerprint vector corresponding to the damping sealing material; S4: Input the state fingerprint vector into the preset physical constraint state estimation model to calculate the health index, degradation level and degradation state parameters corresponding to the damping sealing material; S5: Generate a status assessment result of the damping sealing material based on the health index and the degradation level, and predict the remaining service life based on the degradation status parameters and the environmental operating condition data to obtain the failure risk probability and remaining service life assessment result.

[0007] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S1 specifically includes: Based on the installation location, service load type, medium type, and evaluation target of the damping sealing material to be evaluated, determine the preset excitation conditions; Under the preset excitation conditions, the vibration response data, thermal response data, electrical response data, and acoustic response data generated by the damping sealing material are acquired. Simultaneously acquire environmental condition data corresponding to the vibration response data, thermal response data, electrical response data, and acoustic response data. The environmental condition data includes temperature, humidity, load level, contact pressure, medium type, medium contact state, and equipment operating cycle time. The vibration response data, thermal response data, electrical response data, acoustic response data, and environmental condition data are supplemented with time identifiers, measurement point identifiers, and condition identifiers to form an original data set.

[0008] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S2 specifically includes: Read the original data set and establish a data index for each channel based on the time identifier and the measurement point identifier; The time markers are extracted from the vibration response data, thermal response data, electrical response data, acoustic response data, and environmental condition data. A reference time axis is determined based on a unified time reference, and the vibration response data, thermal response data, electrical response data, acoustic response data, and environmental condition data are mapped to the reference time axis to complete the time alignment. The vibration response data and acoustic response data after time alignment are subjected to noise suppression processing, which includes outlier identification, frequency band filtering and smoothing. The time-aligned thermal response data is subjected to noise suppression processing, which includes moving average and local baseline correction. The time-aligned electrical response data is subjected to noise suppression processing, which includes stable segment extraction, outlier removal, and low-frequency drift subtraction.

[0009] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S2 further includes: Based on the environmental operating condition data, temperature compensation, humidity compensation, load compensation, medium influence compensation, and operating cycle compensation are performed on the vibration response data, thermal response data, electrical response data, and acoustic response data after noise suppression. Using data under standard reference conditions as a benchmark, the correspondence between environmental conditions and response offset is established through calibration tests. Based on the synchronously acquired environmental condition data, the corresponding correction rules are called to perform condition normalization processing on each response data after compensation. The vibration response data, thermal response data, electrical response data, and acoustic response data after normalization of operating conditions are appended with unified time identifiers, measurement point identifiers, and operating condition identifiers. Data of different dimensions are converted into relative changes, normalized amplitudes, segmented trend quantities, and window statistics to obtain standardized state characterization data.

[0010] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S3 specifically includes: Multidimensional feature extraction is performed on the vibration response data, thermal response data, electrical response data, and acoustic response data in the standardized state characterization data, respectively; The vibration response data is used to extract dynamic response amplitude variation characteristics, vibration attenuation characteristics, frequency band energy distribution characteristics, hysteresis variation trend characteristics, and local wave intensity characteristics. The thermal response data is used to extract features such as temperature rise rate, steady-state temperature difference, thermal diffusion uniformity, local thermal hysteresis, and temperature fluctuation stability. The amplitude drift feature, frequency band response change feature, polarization change trend feature, conductivity change trend feature, and stable section offset feature are extracted from the electrical response data. The acoustic response data is used to extract burst event density features, local energy enhancement features, transient fluctuation intensity features, and abnormal event distribution features to form an initial feature set.

[0011] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S3 further includes: The initial feature set is subjected to correlation test, stability test and sensitivity test. Feature items that are not sensitive to state changes, fluctuate too much under different working conditions and are highly repetitive with other features are removed to obtain a feature subset. The feature subset is subjected to uniform scaling to convert each feature into a normalized feature value; The normalized features are combined in a preset order to construct a fixed-dimensional state fingerprint vector.

[0012] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S4 specifically includes: The state fingerprint vector is input into a preset physical constraint state estimation model; The physical constraint rules in the physical constraint state estimation model include state evolution continuity constraints, degradation change direction constraints, parameter value boundary constraints, environmental condition consistency constraints, and multi-feature response coordination constraints. Based on the state mapping rules and physical constraint rules in the physical constraint state estimation model, the state fingerprint vector is subjected to state space projection, constraint correction and parameter calculation to obtain the initial state estimation result; The consistency of the initial state estimation results is verified. When the initial state estimation result satisfies the physical constraint rules, the corresponding degradation state parameters are output.

[0013] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S4 further includes: The degradation state parameters are obtained by solving the physical constraint state estimation model. The degradation state parameters include the damping performance retention rate, the change in interface contact stiffness, the degree of thermal response hysteresis, and the cumulative activity of abnormal events. Based on the degradation state parameters, a health index is generated according to a preset mapping rule; A degradation level is generated based on the health index, the degree of change of the degradation state parameters, and the state evolution trend; When the level corresponding to the health index is inconsistent with the level corresponding to the degradation state parameter, the higher degradation level shall be used for judgment. The health index, the degradation level, and the degradation state parameters are used as the output of step S4.

[0014] As a preferred embodiment of the damping sealing material condition assessment method of the present invention, step S5 specifically includes: Based on the health index and the degradation level, a state assessment result of the damping sealing material is generated according to a preset assessment rule library; Based on the degradation state parameters and the environmental condition data, a preset failure criterion is invoked, and the future degradation trajectory is estimated based on the historical change trend of the degradation state parameters. The future degradation trajectory is compared with the failure criterion to determine the predicted failure time, and the remaining service life assessment result is generated based on the time length between the current time and the predicted failure time. Based on the probability of reaching the preset failure criterion within a predetermined future time window, a failure risk probability is generated. Output the status assessment results, the failure risk probability, and the remaining useful life assessment results.

[0015] Secondly, embodiments of the present invention provide a damping sealing material condition assessment system, comprising: Data acquisition module: Acquires various types of response data and corresponding environmental condition data generated by the damping sealing material under preset excitation conditions; Data processing module: Based on the environmental condition data, performs time-series alignment, noise suppression, and compensation processing on the multi-type response data to obtain standardized state characterization data; Fingerprint construction module: Based on the standardized state characterization data, multi-dimensional feature extraction and fusion are performed to construct the state fingerprint vector corresponding to the damping sealing material; State estimation module: Input the state fingerprint vector into the preset physical constraint state estimation model to calculate the health index, degradation level and degradation state parameters corresponding to the damping sealing material; Assessment and prediction module: Based on the health index and the degradation level, it generates a status assessment result of the damping sealing material, and predicts the remaining service life based on the degradation status parameters and the environmental operating condition data, thereby obtaining the failure risk probability and the remaining service life assessment result.

[0016] The beneficial effects of this invention are as follows: By acquiring multi-type response data and environmental condition data generated by damping sealing materials under preset excitation conditions, and performing time-series alignment, noise suppression, and compensation processing on the data, the comparability and reliability of state data under different operating conditions can be improved. By constructing a state fingerprint vector and introducing a physical constraint state estimation model, the joint calculation of health index, degradation level, and degradation state parameters can be achieved, thereby improving the accuracy and interpretability of state identification. Furthermore, by combining environmental condition data to predict the remaining service life, the failure risk probability and life assessment results can be output, thereby achieving a comprehensive assessment of the current state, degradation mechanism, and future failure risk of the damping sealing material. Attached Figure Description

[0017] 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 A flowchart illustrating the overall process of a method for assessing the condition of a damping sealing material, as provided in the first embodiment of the present invention; Figure 2 The module connection diagram is provided for a damping sealing material condition assessment system according to the third embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0019] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for evaluating the condition of damping sealing materials is provided.

[0020] S1: Acquire various types of response data and corresponding environmental condition data generated by the damping sealing material under preset excitation conditions.

[0021] In this embodiment, step S1 is used to obtain raw basic data that can characterize the current state of the damping sealing material, providing a unified, correlated and traceable data source for subsequent standardization processing, state fingerprint construction, state estimation and remaining service life prediction.

[0022] It should be noted that the "preset excitation conditions" in this invention refer to external conditions pre-set to stimulate differences in the material's state, without compromising the normal structural integrity and service relationship of the damping sealing material. The purpose is to enable the damping sealing material to produce observable responses reflecting changes in its mechanical properties, thermal conductivity, and electrical response under controllable conditions, while avoiding excessive excitation that could cause additional damage. Furthermore, the preset excitation conditions can be set according to the installation location of the damping sealing material, the type of service load, the type of medium used, and the evaluation target. For example, frequency sweep excitation, random excitation, impact excitation, periodic mechanical loading, pulsed thermal excitation, electrical scanning excitation, or a composite excitation method matched to actual operating conditions can be used. Frequency sweep excitation is suitable for exposing the dynamic characteristics of the material at different frequency bands, random excitation is suitable for simulating broadband disturbances in the actual service environment, and impact excitation is suitable for obtaining the transient response characteristics of the material.

[0023] In one embodiment, the mechanical excitation can be set as a periodic disturbance corresponding to the equipment's operating cycle, the thermal excitation can be set as a short-duration thermal pulse, and the electrical excitation can be set as a staged frequency scan to ensure that the damping sealing material under different conditions can output comparable response information. It should also be noted that the excitation intensity setting should meet two conditions: first, it should ensure that the response signal is higher than the background noise level; second, it should not exceed the allowable deformation range of the damping sealing material under the current operating conditions, thereby ensuring that the acquired data is both valuable for identification and does not introduce new, unexpected damage.

[0024] It should be noted that the "multi-type response data" in this invention refers to a set of data generated by the damping sealing material under the preset excitation conditions, which can reflect its state changes from different physical perspectives. This set of data is not a repeated collection of the same response signal, but a joint characterization of the current state of the damping sealing material from different dimensions. Furthermore, the multi-type response data may include vibration response data reflecting the material's dynamic deformation, vibration attenuation, hysteresis changes, or contact behavior changes; thermal response data reflecting changes in the material's surface temperature rise, thermal diffusion differences, local thermal hysteresis, or heat dissipation capacity; electrical response data reflecting the material's internal polarization characteristics, conductivity characteristics, dielectric properties, or interfacial electrical change trends; and acoustic response data reflecting the initiation of microcracks, interfacial microslippage, or local damage events within the material. In one embodiment, the multi-type response data may specifically include two or more of the following: vibration response data, thermal response data, electrical response data, and acoustic response data. Vibration response data can be used to characterize the dynamic stiffness and damping energy dissipation characteristics of the material under stimulated conditions; thermal response data can be used to characterize the material's heat conduction and local heat accumulation characteristics; electrical response data can be used to characterize the material's electrical stability and dielectric change characteristics; and acoustic response data can be used to characterize local damage events within the material. Damping sealing materials are typically affected by multiple factors during actual service, including mechanical loads, temperature fluctuations, media contact, and long-term aging. Their degradation often manifests as a coupled change of multiple phenomena, such as decreased viscoelastic energy dissipation capacity, abnormal local heat dissipation, electrical response drift, and enhanced local damage signals. Therefore, by acquiring multi-type response data, a joint observation basis for the same material state can be established from different physical dimensions, providing necessary support for subsequent extraction of multi-dimensional features and construction of state fingerprint vectors.

[0025] It should also be noted that, to ensure the acquired response data has sufficient representativeness and comparability, when implementing step S1, the excitation area and response acquisition area should be determined based on the structural morphology, installation boundary, force direction, and expected failure area of ​​the damping sealing material. Generally, the excitation area should be preferably located in a position that can effectively transfer energy to the material body and interface area, while the response acquisition area should cover the material body area, the adjacent interface area, and areas sensitive to changes in stress, heat, or electrical response. Furthermore, in specific implementations, multiple acquisition locations can be set for the same damping sealing material, and different acquisition locations can be assigned corresponding measurement point identifiers and time identifiers. This allows the acquired data to reflect not only the overall state of the material but also the state differences in local sensitive areas. The advantage of this setup is that it avoids misjudgments caused by relying solely on local single-point data, enabling the subsequent state estimation process to simultaneously consider both the overall degradation trend and local abnormal evolution. Furthermore, when the damping sealing material is in an offline detection state, the preset excitation conditions can be applied using a controlled excitation device. When the damping sealing material is already installed in the actual equipment and is inconvenient to disassemble, the periodic loads, fluid pulsations, structural vibrations, or thermal disturbances inherent in the equipment during operation can also be used as excitation sources to acquire the various types of response data without interrupting equipment operation. By simultaneously supporting offline excitation and online operating condition excitation, this invention can adapt to the state assessment needs of different application scenarios.

[0026] It should be noted that the "environmental operating condition data" in this invention refers to the data corresponding to the external environment and operating conditions of the damping sealing material when acquiring the multi-type response data. Its function is not simply to record the test background, but rather to serve as an important basis for subsequent time-series alignment, noise suppression, compensation processing, and prediction of remaining service life. Furthermore, the environmental operating condition data may include one or more of the following at the time of acquisition: temperature, humidity, load level, contact pressure, vibration background, medium type, medium contact state, running time, start-stop frequency, pressure fluctuation, and equipment operating cycle time. In one embodiment, ambient temperature can be used to characterize the temperature sensitivity changes of the material's viscoelastic properties; humidity and medium contact state can be used to characterize the influence of internal adsorption, penetration, or interface changes on the response signal; load level and operating cycle time can be used to characterize the influence of external mechanical disturbances on the response amplitude and baseline drift; and medium type can be used to distinguish the differences in the effects of different working media on the material's aging rate and electrical response drift. The reason for simultaneously acquiring environmental operating condition data is that the response data of damping sealing materials is usually significantly affected by environmental factors. For example, temperature changes can cause overall drift in dynamic and thermal responses, media intrusion can cause changes in electrical responses, and load fluctuations can lead to abnormal transient signal amplitudes. If environmental operating condition data is not recorded simultaneously, it will be difficult to effectively distinguish whether the response changes are due to actual material degradation or due to changes in operating conditions. Therefore, in step S1 of this invention, the corresponding acquisition of environmental operating condition data and multiple types of response data is emphasized. Essentially, this provides background constraints and auxiliary criteria for subsequent state normalization, degradation mechanism identification, and life trend prediction.

[0027] Furthermore, the specific implementation of step S1 can be as follows: First, an excitation scheme is preset according to the type of damping sealing material to be evaluated, the service scenario, and the key failure concerns. Then, during the excitation process, multiple types of response data are synchronously collected or continuously collected according to a predetermined time sequence according to a unified acquisition strategy. Environmental operating condition data at corresponding times are recorded synchronously during the acquisition process, ultimately forming a raw data set with time stamps, measurement point stamps, and operating condition stamps. In one embodiment, the sampling rhythm of different types of response data can be set according to their rate of change. For example, a higher sampling frequency is used for rapidly changing vibration or acoustic data, a longer sampling window is used for slowly changing thermal response data, and electrical response data is gradually collected according to a preset frequency band, thus taking into account the effective acquisition needs of different types of data. It should also be noted that, for the convenience of subsequent processing and quality traceability, the acquired multi-type response data and environmental operating condition data can be uniformly packaged into a raw data frame with a timestamp, data source stamp, measurement point stamp, and data quality stamp after acquisition. The data quality stamp is used to mark whether there are over-range, abnormal jumps, sampling interruptions, or data loss. In this way, the output of step S1 is not a collection of scattered measurement values, but a data base unit that can be directly entered into subsequent steps, thus providing a unified interface for timing alignment, noise suppression and compensation processing in step S2.

[0028] S2: Based on the environmental condition data, perform time-series alignment, noise suppression, and compensation processing on the multi-type response data to obtain standardized state characterization data.

[0029] In this embodiment, step S2 is used to uniformly process the multi-type response data obtained in step S1 to obtain standardized state characterization data that can be directly used for subsequent feature extraction. It should be noted that the input to step S2 is the multi-type response data output from step S1 and the corresponding environmental condition data, and the output is the standardized state characterization data. Since different types of response data usually come from different data channels and are affected by sampling frequency, acquisition delay, environmental fluctuations, and operating cycle time, the original data often suffer from inconsistent time bases, noise superposition, and operating condition offsets. Therefore, it is necessary to combine the environmental condition data with sequential time alignment, noise suppression, and compensation processing.

[0030] It should be noted that the aforementioned time alignment refers to adjusting various types of response data and environmental condition data to corresponding time sequences according to a unified time reference. Further, in specific implementations, the time identifiers in various types of response data and environmental condition data can be extracted first, and a reference time axis can be determined based on a unified sampling clock, unified timestamp, or unified trigger event. Then, various types of data can be mapped onto the reference time axis. For vibration or acoustic data with high sampling rates, segmentation and recombination can be performed according to a preset time window; for thermal response data and environmental condition data with low sampling rates, mapping to the same time interval can be achieved through time interpolation, nearest neighbor matching, or moving average. In one embodiment, the sampling time point of the vibration response data can be used as the reference time axis. Linear interpolation or spline interpolation can be used to obtain the corresponding thermal response value for the thermal response data. For acoustic response data with a sampling rate significantly higher than that of the vibration response data, downsampling, peak hold, or time window feature convergence methods can be used to extract representative values. When the damping sealing material is in the online operating state of the equipment, time offset correction can also be performed on various types of response data by combining the equipment operating cycle, excitation start time, response peak time, or feature trigger point.

[0031] Furthermore, the noise suppression refers to weakening or eliminating background interference components, random fluctuation components, sensor background noise, and sampling anomalies superimposed on various response data. For vibration response data and acoustic response data, non-target components can be weakened by combining anomaly identification, frequency band filtering, smoothing, segmented denoising, or wavelet threshold denoising. For thermal response data, moving average, local baseline correction, or temperature rise trend filtering can be used. For electrical response data, stable segment extraction, anomaly removal, low-frequency drift subtraction, or band-limited filtering can be used. In one embodiment, for vibration response data, a wavelet basis function adapted to the current signal frequency band characteristics can be selected for multi-level decomposition, and high-frequency detail components can be thresholded and contracted. For thermal response data, smoothing windows of different widths can be set according to the signal change rate. For acoustic response data, a bandpass filtering interval can be set according to the frequency band range corresponding to the target damage event.

[0032] It should be noted that the compensation processing is used to correct the intriguing offsets caused by changes in external operating conditions in multi-type response data using environmental condition data. Furthermore, the compensation processing may include one or more of the following: temperature compensation, humidity compensation, load compensation, media influence compensation, operating cycle time compensation, and background disturbance compensation. Specifically, when the ambient temperature changes, baseline correction or amplitude correction can be performed on the corresponding channel data based on the correspondence between ambient temperature and historical response offset; when humidity and media contact state change, conditional normalization can be performed on the corresponding data based on media type, media contact duration, and humidity level; when load level, contact pressure, and operating cycle time change, vibration response and acoustic response can be corrected according to load range and operating state range. Furthermore, the compensation processing can be implemented based on pre-established operating condition correction rules or based on calibration test data of the same type of material under different environmental conditions. In one embodiment, when the damping sealing material is in its initial healthy state, reference data can be collected under different temperature, pressure, humidity and medium conditions. The data under the standard reference condition is used as a benchmark to establish the correspondence between the environmental conditions and the response offset. During the real-time evaluation process, the corresponding correction rules are called according to the environmental condition data synchronously obtained in step S1 to perform condition normalization processing on the current response data.

[0033] Furthermore, after completing time alignment, noise suppression, and compensation processing, the various types of response data should be standardized and integrated to obtain the standardized state characterization data. It should be noted that the standardized state characterization data refers to a data set that has been uniformly processed in terms of time reference, data quality, environmental impact, and dimensional expression. In specific implementation, unified time identifiers, measurement point identifiers, and operating condition identifiers can be added to the response data processed from different channels first, and then the data with different dimensions can be converted into a unified characterization form, such as relative change, normalized amplitude, piecewise trend, window statistics, or dimensionless values ​​within a unified range. In one embodiment, the standardized state characterization data can be organized in chronological order into a multi-channel data matrix or data frame sequence, where each row corresponds to one time sampling unit, and each column corresponds to one type of response channel; environmental operating condition data is stored together as constraint data associated with each time sampling unit. For data segments with packet loss, over-range, abrupt distortion, or compensation failure, anomaly marking can be performed, and these segments can be removed, replaced, or downweighted in subsequent processing.

[0034] Furthermore, the specific implementation process of step S2 can be as follows: First, read the original data frame output in step S1 and establish data indexes for each channel based on the time identifier and measurement point identifier; then, align various types of response data with environmental condition data according to a unified time reference; based on the aligned data, perform noise suppression according to data type; then, perform compensation processing on the corresponding channels in conjunction with environmental condition data; finally, integrate the processed multi-type response data in a unified format and output standardized state characterization data. Through the above processing, step S2 completes the transformation from the original measurable signal to standardized state characterization data, providing a data foundation for multi-dimensional feature extraction and fusion in step S3.

[0035] S3: Based on the standardized state characterization data, perform multi-dimensional feature extraction and fusion to construct the state fingerprint vector corresponding to the damping sealing material.

[0036] In this embodiment, step S3 is used to extract multi-dimensional features that reflect the differences in the current state of the damping sealing material based on the standardized state characterization data obtained in step S2, and to uniformly fuse the multi-dimensional features to construct a state fingerprint vector corresponding to the damping sealing material. It should be noted that the input to step S3 is the standardized state characterization data output in step S2, and the output is the state fingerprint vector. The state fingerprint vector is not a simple concatenation of the original data, but rather a structured characterization result that summarizes information related to material degradation state, damage evolution degree, and operating condition response characteristics from different types of response data. Its function is to convert multi-source, heterogeneous, and dimensionally different state information into a unified state description unit that can be directly input into the subsequent state estimation model.

[0037] It should be noted that the multidimensional feature extraction refers to extracting feature information that can represent the state change laws of different physical dimensions from the standardized state characterization data. Furthermore, in specific implementations, feature extraction can be performed on one or more of the mechanical response, thermal response, electrical response, and acoustic response, based on the signal type corresponding to the standardized state characterization data. For vibration response data, features reflecting changes in dynamic response amplitude, vibration attenuation capability, frequency band energy distribution, hysteresis change trend, and local fluctuation intensity can be extracted. In one embodiment, features such as root mean square value, peak value, kurtosis, waveform factor, impulse factor, damping ratio, natural frequency, and frequency response amplitude can be further extracted to characterize the material's dynamic stiffness change and damping energy dissipation capability. For thermal response data, features reflecting the temperature rise rate, steady-state temperature difference, heat diffusion uniformity, local thermal hysteresis degree, and temperature fluctuation stability can be extracted. In one embodiment, thermal response time constant, temperature peak time, and temperature distribution uniformity index can be further extracted to characterize changes in the material's heat generation and heat conduction capabilities. For electrical response data, features reflecting amplitude drift, frequency band response changes, polarization trends, conductivity trends, or shifts in stable regions can be extracted. For acoustic response data, features reflecting the density of sudden events, local energy enhancement, transient fluctuation intensity, and distribution patterns of anomalous events can be extracted. In one embodiment, features such as event counts, cumulative energy, peak frequency, centroid frequency, and frequency band energy distribution can be further extracted to characterize the activity level of microcrack initiation, interfacial microslip, or local damage events within the material.

[0038] Furthermore, the multidimensional feature extraction includes not only instantaneous feature extraction at a single moment, but also statistical feature extraction and trend feature extraction for the state change process within a certain time window. It should be noted that the degradation of damping sealing materials is usually not fully characterized by a single-point anomaly, but rather by the slow drift, local accumulation, and staged abrupt changes of certain features over a continuous period. Therefore, in implementing step S3, a predetermined analysis window can be set for the standardized state characterization data, and peak features, mean features, fluctuation features, stability features, segmented change features, and trend change features can be extracted within the analysis window, so that the obtained features can reflect both the state performance at a certain moment and the evolution over a period of time. In one embodiment, a shorter analysis window can be used for vibration and acoustic data to retain local anomaly change information; a longer analysis window can be used for thermal response data and data with strong environmental coupling to characterize their slow change trend. It should also be noted that the above features can be selectively extracted according to the specific type of damping sealing material, service conditions, and main failure modes; not all features are necessary. For example, in scenarios where the main focus is on changes in damping energy dissipation capacity, the focus can be on extracting damping features from the vibration response and heat conduction features from the thermal response; in scenarios where the main focus is on sealing integrity and local damage identification, the focus can be on extracting abnormal event features from the acoustic response and stiffness change features from the mechanical response.

[0039] It should be noted that, to avoid excessive redundancy, scale inconsistencies, or imbalanced correlations among different types of features, the extracted features should be screened and organized after multidimensional feature extraction. Furthermore, in practical implementation, correlation, stability, and sensitivity tests can be performed on the initial feature set to eliminate features that are insensitive to state changes, fluctuate excessively under different operating conditions, or are highly repetitive with other features, retaining only a subset of features that can stably characterize the state differences of damping sealing materials. Furthermore, the retained features can be standardized in scale to ensure comparability of features with different dimensions and value ranges in subsequent fusion; in one implementation, each feature can be converted into a normalized feature value within a unified range before participating in subsequent fusion processing. Through these steps, the features ultimately participating in the fusion can be both representative and maintain good discriminative power and robustness in subsequent state estimation.

[0040] Furthermore, feature fusion refers to combining feature information from different data channels, different physical dimensions, and different time scales according to unified rules to form a state fingerprint vector that can comprehensively characterize the state of the damping sealing material. It should be noted that feature fusion can be performed by grouping features by category and then combining them sequentially, by weighting features based on their importance, or by first compressing dimensions and then uniformly encoding them. In one embodiment, the extracted features can be first divided into mechanical, thermal, electrical, and acoustic feature groups, then normalized and integrated within each feature group, and finally concatenated into a unified state vector in a preset order. In another embodiment, the contribution of each type of feature to state differentiation can be determined based on historical sample data or calibration sample data, and then different weights can be assigned to different features according to their contribution before fusion. For scenarios with a large number of features and strong correlations between features, dimensionality reduction can be performed on the features before fusion to eliminate redundant information and reduce the computational complexity of subsequent models. By using the above methods, we can avoid the excessive weight of single physical dimension features in the fusion process, while improving the ability of state fingerprint vectors to comprehensively express multiple degradation modes.

[0041] It should be noted that the "state fingerprint vector" in this invention refers to a feature vector obtained by fusing multi-dimensional features to characterize the current state of the damping sealing material. The state fingerprint vector can be understood as a comprehensive state code of the damping sealing material under the current operating conditions, current degradation stage, and current response characteristics, which can serve as input to the subsequent physical constraint state estimation model. Furthermore, the data in each dimension of the state fingerprint vector are not required to maintain the original physical quantity form, but can be a sequence of feature values ​​after standardization, screening, intra-group integration, and unified encoding, as long as they can stably reflect the state differences of the damping sealing material. In one embodiment, the state fingerprint vector can be constructed with fixed dimensions, so that the state fingerprint vectors output by the same type of damping sealing material at different sampling times and under different service conditions have a unified structure, thereby facilitating subsequent batch processing, horizontal comparison, and vertical tracking by the model. For example, when selecting several damping features in the vibration response, heat conduction features in the thermal response, and abnormal event features in the acoustic response as core features, the normalized feature values ​​can be arranged sequentially according to a preset order to form a fixed-dimensional state fingerprint vector.

[0042] Furthermore, the specific implementation process of step S3 can be as follows: First, read the standardized state characterization data output from step S2; then, extract multi-dimensional features according to the data type and analysis window to form an initial feature set; then, filter, classify, and unify the scale of the initial feature set; after completing the feature processing, combine various features according to preset fusion rules to obtain the state fingerprint vector corresponding to the damping sealing material. Through the above processing, step S3 completes the conversion from standardized state characterization data to a unified state description vector, enabling the subsequent step S4 to perform state estimation based on input data with a unified structure and stable expression. It should also be noted that the state fingerprint vector output by step S3 not only retains the multi-dimensional characterization ability of multiple types of response data for the current state of the damping sealing material, but also improves the efficiency and consistency of subsequent model processing through a unified encoding method.

[0043] S4: Input the state fingerprint vector into the preset physical constraint state estimation model to calculate the health index, degradation level and degradation state parameters corresponding to the damping sealing material.

[0044] In this embodiment, step S4 is used to input the state fingerprint vector obtained in step S3 into a preset physical constraint state estimation model to calculate the health index, degradation level, and degradation state parameters corresponding to the damping sealing material. It should be noted that the input of step S4 is the state fingerprint vector, and the output is the health index, degradation level, and degradation state parameters. The physical constraint state estimation model is not simply an empirical classification model that fits sample data, but rather a state calculation model that introduces the service law, degradation evolution law, and environmental sensitivity law of the damping sealing material during the state identification process. Its function is to convert the multidimensional feature information represented by the state fingerprint vector into a state evaluation result with clear engineering significance.

[0045] It should be noted that the "physical constraints" in the preset physical constraint state estimation model refer to the prior constraints imposed on the state evolution of the damping sealing material during the state solution process. Further, the physical constraints may include one or more of the following: state evolution continuity constraints, degradation change direction constraints, parameter value boundary constraints, environmental condition consistency constraints, and multi-feature response coordination constraints. Specifically, the state evolution continuity constraint is used to limit drastic jumps in the state estimation results between adjacent time points that are inconsistent with the actual degradation law. For example, the change in the health index within two adjacent sampling periods should not exceed 15, or the change in the same degradation state parameter within two consecutive sampling periods should not exceed 20% of its initial reference value. The degradation change direction constraint is used to limit the changes of some key state quantities of the material within permissible ranges during long-term service, thereby preventing the model from misjudging transient disturbances as degradation reversals. For example, the increase in damping performance retention rate relative to the previous time point under maintenance-free conditions should not exceed 5%, and the decrease in thermal response hysteresis relative to the previous time point should not exceed 5%. The decrease should not exceed 5%; parameter value boundary constraints are used to ensure that the calculated state variables remain within the range actually achievable by the material, for example, the health index is limited to 0 to 100, and the damping performance retention rate is limited to 0 to 1; environmental condition consistency constraints are used to ensure that the state estimation results under the same environmental conditions are consistent with the input conditions; multi-feature response coordination constraints are used to ensure that there are no continuous contradictions between the state changes corresponding to mechanical, thermal, electrical, and acoustic characteristics, for example, when the vibration energy dissipation characteristic decreases by more than 10% for three consecutive sampling periods, the thermal stability characteristic and the local damage activity characteristic should not simultaneously remain within the initial reference range for a long period of time. By introducing the above physical constraints, the state estimation results can not only match the input data, but also be consistent with the actual service mechanism of the damping sealing material.

[0046] Furthermore, in specific implementation, a model structure for the physical constraint state estimation model can be pre-established. This model structure can include a state input unit, a state association unit, a constraint correction unit, and a state output unit. Specifically, the state input unit receives the state fingerprint vector; the state association unit establishes the correspondence between the state fingerprint vector and the degenerate state parameters; the constraint correction unit performs consistency correction on the initial state estimation results based on preset physical constraints; and the state output unit outputs the degenerate state parameters and generates a health index and a degradation level based on these parameters. In one embodiment, the state fingerprint vector can be input into a pre-trained or pre-calibrated state estimation model. Through the model's internal state mapping rules, the multi-dimensional feature vector is mapped to a preset state space, and the mapping result is then corrected based on physical constraints to obtain the final state estimation result. It should also be noted that the model can employ rule-based state estimation, sample-based state estimation, or a hybrid state estimation method combining rule constraints and sample learning, as long as it can achieve stable calculation of the health index, degradation level, and degenerate state parameters.

[0047] Furthermore, the state estimation model can be pre-set with parameters by incorporating historical sample data or calibration sample data during implementation. Specifically, state fingerprint vectors and corresponding state labels of the damping sealing material under different degradation stages and environmental conditions can be pre-collected to establish a mapping relationship between the state fingerprint vectors and degradation state parameters. This mapping relationship can be adjusted based on newly collected data during model operation. It should also be noted that when the model calculates the current state, it can simultaneously refer to the current state fingerprint vector and historical state results from the previous 1 or 3 to 10 time points to ensure the continuity and stability of the state estimation results over time.

[0048] It should be noted that the degradation state parameters are a set of parameters used to characterize the degree of evolution of the internal state of the damping sealing material, and are an important input for subsequent prediction of remaining service life. Further, the degradation state parameters may include one or more of the following: parameters reflecting the degree of damping performance degradation, parameters reflecting the degree of interface stiffness degradation, parameters reflecting the degree of thermal stability change, parameters reflecting the degree of local damage activity, and parameters reflecting the overall degree of degradation evolution. In one embodiment, degradation state parameters may be exemplified by damping performance retention rate, interface contact stiffness change, thermal response hysteresis, and cumulative activity of abnormal events. For example, damping performance retention rate can be expressed as the ratio of the current damping characteristic to the initial reference damping characteristic; interface contact stiffness change can be expressed as the offset of the current interface stiffness characteristic relative to the initial reference value; thermal response hysteresis can be expressed as the delay of the thermal response peak value relative to the excitation peak value; and cumulative activity of abnormal events can be expressed as the change in the number of abnormal events or the accumulated energy per unit time relative to the initial stage reference value. It should be noted that the degradation state parameters are not required to correspond one-to-one with a single physical quantity, but can be internal state characterization quantities obtained by the state estimation model based on the state fingerprint vector, as long as they can stably reflect the degradation evolution process of the damping sealing material.

[0049] It should be noted that the health index is used to characterize the overall health level of the damping sealing material, which can be understood as a comprehensive quantitative result of the degree of quality of the current state. Furthermore, in specific implementations, the health index can be set as a state evaluation value with continuously changing characteristics. In one embodiment, the health index can be set as a continuous value between 0 and 100, where a higher value indicates that the current state of the damping sealing material is closer to its initial healthy state, and a lower value indicates that the current state of the material is closer to the failure boundary. Furthermore, the health index is preferably generated based on degradation state parameters according to a preset mapping rule. For example, when the damping performance retention rate is greater than 0.9, the change in interface contact stiffness is less than 10% of the initial reference value, and the change in thermal response hysteresis is less than 10%, the health index can be between 80 and 100. When the damping performance retention rate is between 0.6 and 0.9, the change in interface contact stiffness reaches 10% to 30% of the initial reference value, and the change in thermal response hysteresis reaches 10% to 25%, the health index can be between 40 and 80. When the damping performance retention rate is less than 0.6, the change in interface contact stiffness exceeds 30% of the initial reference value, and the change in thermal response hysteresis exceeds 25%, the health index can be between 0 and 40. By setting the health index, the state results under different sampling times and different service conditions can be compared.

[0050] Furthermore, the degradation level is used to characterize the current degradation stage of the damping sealing material. It should be noted that the degradation level is a staged evaluation result obtained by further discretizing the degradation state parameters and health index. Its purpose is to convert continuously changing state estimation results into a level result that facilitates engineering judgment and maintenance decisions. In one embodiment, the degradation level can be divided into normal state, mild degradation state, moderate degradation state, and severe degradation state; it can also be divided into preset level ranges such as level 1, level 2, level 3, and level 4. In specific implementation, the degradation level can be determined jointly based on the range of the health index, the degree of change of the degradation state parameters, and the state evolution trend; when the level corresponding to the health index is inconsistent with the level corresponding to the degradation state parameters, the higher degradation level is used for judgment. For example, when the health index is greater than 80 and the change in each degradation parameter relative to the initial reference value is less than 10%, it can be judged as a normal state or Level 1; when the health index is between 60 and 80, or at least one degradation parameter changes by 10% to 20%, it can be judged as a mild degradation state or Level 2; when the health index is between 40 and 60, or at least one degradation parameter changes by 20% to 30%, it can be judged as a moderate degradation state or Level 3; when the health index is less than 40, or at least one degradation parameter changes by more than 30%, or shows a continuous deterioration trend for more than three consecutive sampling periods, it can be judged as a severe degradation state or Level 4. "Continuous deterioration trend" can refer to a continuous decline in the health index within three or more consecutive sampling periods, with a cumulative decrease exceeding 10, or at least one degradation parameter showing a unidirectional deterioration for three or more consecutive sampling periods.

[0051] Furthermore, in implementing step S4, the state fingerprint vector output in step S3 can be first input into a preset physical constraint state estimation model. The model then performs state space projection, constraint correction, and parameter calculation on the state fingerprint vector according to preset constraint rules and state mapping relationships to obtain an initial state estimation result. Next, a consistency check is performed on the initial state estimation result to confirm whether it meets the physical constraint rules. When the preset constraint conditions are met, the corresponding degenerate state parameters are output, and a health index and degradation level are further generated. When the preset constraint conditions are not met, the state estimation result is readjusted or the constraints are substituted back until a final state result that meets the constraint requirements is output. In one embodiment, joint calculation can be performed based on the state fingerprint vectors from multiple consecutive sampling times, ensuring continuous change in the state estimation results at adjacent times to improve the stability of the estimation results. In another embodiment, independent calculation can be performed for a single sampling time, and then combined with historical state results for posterior correction.

[0052] Furthermore, the specific implementation process of step S4 can be as follows: First, read the state fingerprint vector output in step S3; then, call the preset physical constraint state estimation model and input the state fingerprint vector into the model; next, perform state calculation on the state fingerprint vector according to the state mapping rules and physical constraint rules within the model; then, generate the corresponding degradation state parameters, and generate a health index and degradation level based on the degradation state parameters according to the preset mapping rules; finally, pass the health index, degradation level, and degradation state parameters as the output results of step S4 to the subsequent step S5. Through the above processing, step S4 completes the conversion from the state fingerprint vector to the engineering state evaluation result, enabling subsequent steps to generate state evaluation results and predict remaining service life based on the calculated state parameters.

[0053] S5: Generate a status assessment result of the damping sealing material based on the health index and the degradation level, and predict the remaining service life based on the degradation status parameters and the environmental operating condition data to obtain the failure risk probability and remaining service life assessment result.

[0054] In this embodiment, step S5 is used to generate a condition assessment result for the damping sealing material based on the health index and degradation level output in step S4, and to predict the remaining service life based on the degradation state parameters and the environmental condition data, thereby obtaining the failure risk probability and the remaining service life assessment result. It should be noted that the inputs to step S5 include the health index, degradation level, degradation state parameters, and environmental condition data, and the outputs include the condition assessment result, the failure risk probability, and the remaining service life assessment result. The condition assessment result characterizes the overall condition level and degradation stage of the damping sealing material at the current moment, while the failure risk probability and the remaining service life assessment result characterize the likelihood of failure and the duration of continued use of the damping sealing material during subsequent service, thus forming a joint assessment result for current condition determination and future service life prediction.

[0055] It should be noted that the status assessment result is a comprehensive assessment conclusion generated based on the health index and degradation level. Its form may include one or more of the following: numerical assessment result, graded assessment result, and textual assessment conclusion. Furthermore, in specific implementations, the health index and degradation level can be mapped to corresponding status assessment results according to a preset assessment rule base. In one embodiment, when the health index is greater than 80 and the degradation level is level 1, the status assessment result can be determined as a normal usable state; when the health index is between 60 and 80 and the degradation level is level 2, the status assessment result can be determined as a slightly degraded state; when the health index is between 40 and 60 and the degradation level is level 3, the status assessment result can be determined as a moderately degraded state; and when the health index is less than 40 and the degradation level is level 4, the status assessment result can be determined as a severely degraded state or a near-failure state. It should also be noted that the health index and degradation level are preferably the results of the comprehensive judgment in step S4; when the state description corresponding to the health index is inconsistent with the state description corresponding to the degradation level, the final state evaluation result can be generated according to the result corresponding to the more severe state, so as to avoid making an overly optimistic judgment on the current state of the material.

[0056] Furthermore, the condition assessment results, in addition to the condition category determination, may also include descriptive results regarding the current service status of the damping sealing material. In one embodiment, the condition assessment results may include the current condition level, the current health index range, and the current main degradation direction. The main degradation direction can be determined based on the parameter type with the largest change among the degradation state parameters. For example, when the damping performance retention rate decreases the most, the main degradation direction can be determined as damping performance degradation; when the change in interface contact stiffness is the largest, the main degradation direction can be determined as interface contact degradation; when the change in thermal response hysteresis is the most significant, the main degradation direction can be determined as thermal stability degradation; when the cumulative activity of abnormal events continues to increase, the main degradation direction can be determined as local damage activation. In this way, the condition assessment results can not only provide a level determination of the current material condition but also further reflect the main manifestations of the current degradation.

[0057] It should be noted that the remaining service life prediction is based on degradation state parameters and environmental condition data to extrapolate the future degradation evolution trend of the damping sealing material, in order to estimate its remaining usable time before reaching the preset failure condition. Furthermore, the remaining service life prediction is not based on a direct extrapolation of the health index at a single moment, but rather on the degree of internal state evolution reflected by the degradation state parameters, combined with the environmental conditions that may persist in the present and future, to dynamically extrapolate the degradation trend. Since the degradation rate of damping sealing materials is usually closely related to environmental conditions such as temperature, load, pressure fluctuations, media contact state, start-stop frequency, and operating cycle time, environmental condition data needs to be used as an important input for service life prediction in step S5 to improve the consistency between the prediction results and actual service conditions.

[0058] Furthermore, in specific implementation, failure criteria for the damping sealing material can be pre-set, and these failure criteria can be used as the endpoint of the remaining service life during the service life prediction process. It should be noted that the failure criteria can be determined based on one or more of the following: health index, degradation level, or degradation state parameters. In one embodiment, any one of the following conditions can be used as the failure criteria: health index below 20, degradation level reaching 4, damping performance retention rate below 0.5, interface contact stiffness change exceeding 40% of the initial reference value, thermal response hysteresis change exceeding 30%, or cumulative abnormal event activity exceeding twice the initial reference level. When the prediction result indicates that the damping sealing material will reach any of the above failure criteria in the future, the corresponding time can be determined as the predicted failure time, and the time length between the current time and the predicted failure time can be used as the remaining service life assessment result.

[0059] Furthermore, the remaining useful life prediction can be based on the historical variation trends of key degradation state parameters to extrapolate the degradation trajectory. Specifically, one or more parameters with a strong correlation to material failure can be selected from the degradation state parameters as prediction indicators, such as damping performance retention rate, interface contact stiffness change, or thermal response hysteresis. Then, a degradation trajectory is established based on the variation trends of the prediction indicators within the historical sampling period, and this trajectory is extrapolated to future time periods until it intersects with a preset failure criterion. In one embodiment, the degradation trajectory can be described using one of the following: an exponential variation trajectory, a power-law variation trajectory, a polynomial variation trajectory, or a state-space trajectory. It should also be noted that when multiple degradation state parameters jointly affect failure, the future variation trajectories of multiple degradation state parameters can be jointly extrapolated, and the predicted failure time can be determined based on the time corresponding to the parameter that first reaches the failure criterion.

[0060] It should be noted that the failure risk probability is used to characterize the likelihood of the damping sealing material failing within a predetermined future time interval. Further, in specific implementations, the probability of reaching the failure criterion within a predetermined future time window can be calculated based on the current value and rate of change of the degradation state parameters, as well as environmental operating condition data. In one embodiment, a future operating cycle, 10 operating cycles, 24 hours, 100 hours, or other predetermined prediction duration can be set as the risk assessment window, and the probability value of the damping sealing material reaching the failure criterion within the window can be calculated. For example, if the prediction result indicates that the probability of the damping sealing material reaching the failure criterion within the next 100 hours is greater than 0.8, it can be classified as a high-risk state; when the probability is between 0.5 and 0.8, it can be classified as a medium-risk state; and when the probability is less than 0.5, it can be classified as a low-risk state. Further, in one embodiment, the failure risk probability can be obtained by generating multiple possible future degradation trajectories and statistically analyzing the proportion of trajectories that reach the failure criterion within the predetermined time window, thereby reflecting the uncertainty of the life prediction result under future operating condition fluctuations.

[0061] Furthermore, environmental operating condition data not only serves as a basis for correcting the current state in the remaining service life prediction, but can also be used as a condition for predicting future operating conditions to participate in the degradation trajectory calculation. In one embodiment, the service life prediction can be combined with the expected future operating condition spectrum, which may include the temperature variation range, pressure variation range, media contact duration, and operating cycle information within the future service cycle. For example, when the expected future ambient temperature rises, pressure fluctuations increase, or media contact duration increases, the corresponding degradation rate can be increased accordingly, and the remaining service life assessment result can be shortened accordingly; when the expected future operating conditions remain stable, the corresponding degradation rate can remain at a low level, and the remaining service life assessment result can be extended accordingly. By incorporating the expected future operating conditions into the service life prediction process, the prediction results can be made closer to the actual service scenario.

[0062] Furthermore, the remaining service life assessment result can be output in one or more forms, such as time length, number of operating cycles, or equivalent working time period. In one embodiment, when the damping sealing material is applied to continuously operating equipment, the remaining operating hours can be used as the output form, for example, "remaining service life is 120 hours"; when the damping sealing material is applied to periodically start-stop equipment, the remaining operating cycles can be used as the output form, for example, "remaining service life is 350 working cycles"; when the damping sealing material is applied to intermittent load scenarios, the equivalent working time period can also be used as the output form. Furthermore, the remaining service life assessment result can also include lifespan range information, for example, "remaining service life is 100 to 140 hours," to reflect the fluctuation range of the predicted result. It should also be noted that the remaining service life assessment result can be output in conjunction with the condition assessment result, so that the user can simultaneously obtain information on both the current state and future lifespan of the material.

[0063] Furthermore, in implementing step S5, the health index, degradation level, and degradation state parameters output in step S4 can be read first, and combined with the environmental operating condition data synchronously acquired in step S1 or retained after processing in step S2, to generate the current state assessment result. Then, based on the current value and evolution trend of the degradation state parameters, combined with the load level, temperature conditions, medium conditions, and operating cycle corresponding to the environmental operating condition data, the future degradation trajectory is calculated. Next, the calculated future degradation trajectory is compared with the preset failure criteria to determine the predicted failure time, and the remaining service life between the current time and the predicted failure time is calculated. Simultaneously, based on the probability of reaching the failure criteria within a predetermined future time window, a failure risk probability is generated. Finally, the state assessment result, failure risk probability, and remaining service life assessment result are output. Through the above processing, step S5 completes the conversion from state calculation results to engineering maintenance decision information.

[0064] It should also be noted that the condition assessment results and life prediction results generated in step S5 can be further used for maintenance prompts, overhaul arrangements, or replacement warnings. In one embodiment, when the condition assessment result is a normal and usable state and the failure risk probability is less than 0.3, a prompt to continue operation can be output; when the condition assessment result is a slightly degraded state, or the failure risk probability is between 0.3 and 0.6, a prompt to strengthen monitoring can be output; when the condition assessment result is a moderately degraded state, or the failure risk probability is between 0.6 and 0.8, a prompt to plan maintenance can be output; when the condition assessment result is a severely degraded state, or the failure risk probability is greater than 0.8, a prompt to replace as soon as possible can be output. In this way, the condition assessment method of the present invention can not only realize condition identification and life prediction, but also directly serve the operation and maintenance management of damping sealing materials.

[0065] Example 2 is the second embodiment of the present invention, which differs from the previous embodiment in that: If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art or the current technical solution, can be embodied in the form of a software product. This current computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0066] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0067] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0068] Example 3, referring to Figure 2 As an embodiment of the present invention, a damping sealing material condition assessment system is provided, which includes a data acquisition module, a data processing module, a fingerprint construction module, a condition estimation module, and an assessment and prediction module; Data acquisition module: Acquires various types of response data and corresponding environmental condition data generated by the damping sealing material under preset excitation conditions; Data processing module: Based on the environmental condition data, performs time-series alignment, noise suppression, and compensation processing on the multi-type response data to obtain standardized state characterization data; Fingerprint construction module: Based on the standardized state characterization data, multi-dimensional feature extraction and fusion are performed to construct the state fingerprint vector corresponding to the damping sealing material; State estimation module: Input the state fingerprint vector into the preset physical constraint state estimation model to calculate the health index, degradation level and degradation state parameters corresponding to the damping sealing material; Assessment and prediction module: Based on the health index and the degradation level, it generates a status assessment result of the damping sealing material, and predicts the remaining service life based on the degradation status parameters and the environmental operating condition data, thereby obtaining the failure risk probability and the remaining service life assessment result.

[0069] Example 4 is an embodiment of the present invention, which provides a method for evaluating the condition of damping sealing materials. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation / comparative experiments.

[0070] This embodiment uses an interstage damping sealing ring of a certain type of aero-engine compressor as the object. The damping sealing material condition assessment method described in this invention is used to continuously track its condition evolution during service, and the results are compared with traditional single-index assessment methods to verify the effectiveness of this invention in multi-source condition identification, degradation trend capture, and remaining service life prediction. The test objects are six fluorosilicone rubber-based damping sealing ring samples manufactured in the same batch, numbered D01, D02, D03, D04, D05, and D06. All samples underwent initial condition calibration before installation, and the vibration response, thermal response, electrical response, and acoustic response baselines under standard reference conditions were recorded. The standard reference conditions were set as follows: ambient temperature 25℃, ambient pressure 101.3 kPa, relative humidity 45%, and no medium contact. The test platform consists of an electromagnetic excitation system, a programmable environmental chamber, a pressure regulating device, a dielectric scanning unit, an acoustic emission acquisition unit, and a multi-channel data synchronous acquisition system. The preset excitation conditions employ a combination of broadband random excitation and staged thermal and electrical combined excitation. The mechanical excitation frequency range is set to 50Hz to 5000Hz, and the root mean square acceleration value is set to 0.5g. Thermal excitation uses short-time pulsed thermal loading, and electrical excitation uses segmented scanning from 1kHz to 1MHz. The programmable environment chamber simulates the actual service temperature and pressure spectra, with temperature cycling between 25℃ and 150℃ and pressure cycling between 0.1MPa and 1.2MPa, completing one operating cycle every 24 hours.

[0071] During the experiment, test nodes were set at equivalent running times of 0h, 500h, 1000h, 1500h, 2000h and 2500h to correspond to different degradation stages.

[0072] At each test node, step S1 is first executed, applying preset excitation conditions under standard reference conditions, and simultaneously acquiring vibration response data, thermal response data, electrical response data, and acoustic response data, while simultaneously recording environmental condition data such as ambient temperature, pressure, load level, and medium contact state. Vibration response is acquired by a triaxial accelerometer at a sampling frequency of 25.6 kHz; thermal response is acquired by a patch thermocouple at a sampling frequency of 10 Hz; electrical response is acquired by a dielectric scanning unit; and acoustic response is acquired by a broadband acoustic emission sensor at a sampling frequency of 2 MHz.

[0073] Then, step S2 is executed, using the vibration response sampling time point as a unified reference time axis. Spline interpolation is used to map the thermal response data, and peak hold and time window feature convergence are used to downsample the acoustic response data. Timing alignment is then performed on the data from each channel. In the noise suppression stage, db4 wavelet thresholding is used for denoising the vibration response data, a 5-point moving average is used for the thermal response data, a 100kHz to 400kHz bandpass filter is used for the acoustic response data, and stable segment extraction and low-frequency drift subtraction are performed on the electrical response data. Afterwards, based on the temperature-amplitude correction curve and pressure-damping correction curve obtained from the calibration test, operating condition compensation is performed on each response data to obtain a standardized state characterization data matrix.

[0074] Next, step S3 is executed to extract vibration, thermal, electrical, and acoustic features from the standardized state characterization data. Vibration features include root mean square (RMS) value, damping ratio, and frequency response amplitude; thermal features include thermal response time constant and temperature rise rate; electrical features include dielectric constant change rate and loss response offset; and acoustic features include event count, cumulative energy, and peak frequency. After feature extraction, the initial feature set undergoes correlation and stability tests, redundant terms are removed, and the retained features are uniformly converted into normalized feature values ​​to construct a fixed-dimensional state fingerprint vector.

[0075] Then, step S4 is executed, where the state fingerprint vector is input into the preset physical constraint state estimation model. The model embeds degradation continuity constraints, parameter boundary constraints, and multi-feature coordination constraints to solve the state of each test node and output degradation state parameters such as health index, degradation level, damping performance retention rate, interface contact stiffness change, thermal response hysteresis degree, and abnormal event cumulative activity.

[0076] Finally, step S5 is executed to generate the current state assessment result based on the health index and degradation level. Furthermore, based on the historical trends of degradation state parameters and the current environmental conditions, the future degradation trajectory is calculated to obtain the remaining service life assessment result and the probability of failure within the next 1000 hours. As a control, a traditional single-index method is simultaneously used, i.e., the sample state is graded according to the compression set rate, to compare the differences between the two schemes in early degradation identification and service life prediction. Key data recorded are shown in Table 1: Table 1: Experimental Data Recording Table

[0077] The experimental data above demonstrates that the advantages of this invention in assessing the condition of damping sealing materials are mainly reflected in the following aspects. First, the invention's ability to identify early degradation is significantly superior to traditional single-parameter assessment methods. Taking specimen D03 as an example, after 1000 hours of equivalent operation, its compressive permanent deformation rate was only 8.7%, which would still be considered "normal" according to traditional methods. However, the health index of this invention has decreased to 81.4, the degradation level is determined to be level 2, the damping performance retention rate has decreased to 0.918, the thermal response hysteresis has increased to 3.12 ms, and the cumulative activity of abnormal events has increased by 1.49 times. This indicates that before the compressive permanent deformation has significantly deteriorated, the material's internal vibration energy dissipation capacity, thermal response characteristics, and local damage activity have already changed. This invention, through joint characterization of multiple types of response data, enables the early identification of this type of latent degradation, while traditional methods, relying only on a single deformation index, are unable to reflect the evolution of internal material damage and interface behavior in a timely manner.

[0078] Secondly, this invention can distinguish different degrees of degradation and provide more engineering-valuable quantitative results for mid-to-late-stage degradation trends. For example, at 1500 hours, traditional methods only indicated a "caution" status for specimen D04, while this invention identified its health index as 67.9, classifying it as Level 2 degradation, and predicting a remaining service life of 1432 hours. At 2000 hours, traditional methods gave a "warning" status for specimen D05, while this invention not only classified it as Level 3 degradation but also provided a remaining service life of 968 hours and a failure risk probability of 0.56 within the next 1000 hours. Therefore, this invention can not only determine "whether degradation has occurred currently" but also quantify "the degree of degradation" and "how long it may take to reach the failure threshold." This is particularly important for preventative maintenance of high-end equipment, as actual maintenance decisions rely not only on the current level but also on the quantitative prediction of future failure risks.

[0079] Furthermore, this invention also demonstrates strong risk characterization capabilities during severe degradation. Specimen D06, after 2500 hours, exhibited a damping performance retention rate of only 0.556, an interface contact stiffness change of 41.6%, a thermal response hysteresis increase to 6.18 ms, and an abnormal event cumulative activity level of 5.21 times. This corresponds to a health index of 28.4, a degradation level of 4, a predicted remaining service life of only 438 hours, and a failure probability of 0.88 within the next 1000 hours. Compared to traditional methods that only provide a "critical" state, this invention further quantifies the degree of risk and the remaining usable time, thereby elevating maintenance planning from a static judgment of "whether to replace" to a dynamic decision of "when to replace and how great the risk is."

[0080] Finally, from the overall data trend, the health index output by this invention does not mechanically correspond to a single parameter, but is jointly correlated with the damping performance retention rate, the change in interface contact stiffness, the degree of thermal response hysteresis, and the cumulative activity of abnormal events, demonstrating the role of multi-source feature fusion and physical constraint state estimation model. The health index gradually decreased from 98.2 to 28.4, the degradation level gradually increased from level 1 to level 4, the remaining service life prediction shortened from 2364h to 438h, and the failure risk probability increased from 0.02 to 0.88. This evolution trajectory is consistent with the actual degradation law of materials in long-term service. This indicates that this invention does not simply superimpose multiple signals, but achieves a unified characterization and reliable prediction of complex degradation behavior through environmental condition compensation, state fingerprint construction, and physical constraint solution. Compared with the prior art, this invention can significantly improve the sensitivity of state identification, the ability to distinguish degradation mechanisms, and the reliability of life prediction under complex service environments, and has significant inventiveness and engineering application value.

[0081] 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 assessing the condition of damping sealing materials, characterized in that, include: S1: Acquire multi-type response data and corresponding environmental condition data generated by the damping sealing material under preset excitation conditions; S2: Based on the environmental condition data, the multi-type response data are subjected to time alignment, noise suppression and compensation processing to obtain standardized state characterization data; S3: Based on the standardized state characterization data, perform multi-dimensional feature extraction and fusion to construct the state fingerprint vector corresponding to the damping sealing material; S4: Input the state fingerprint vector into the preset physical constraint state estimation model to calculate the health index, degradation level and degradation state parameters corresponding to the damping sealing material; S5: Generate a status assessment result of the damping sealing material based on the health index and the degradation level, and predict the remaining service life based on the degradation status parameters and the environmental operating condition data to obtain the failure risk probability and remaining service life assessment result.

2. The method for evaluating the condition of damping sealing materials as described in claim 1, characterized in that, Step S1 specifically includes: Based on the installation location, service load type, medium type, and evaluation target of the damping sealing material to be evaluated, determine the preset excitation conditions; Under the preset excitation conditions, the vibration response data, thermal response data, electrical response data, and acoustic response data generated by the damping sealing material are acquired. Simultaneously acquire environmental condition data corresponding to the vibration response data, thermal response data, electrical response data, and acoustic response data. The environmental condition data includes temperature, humidity, load level, contact pressure, medium type, medium contact state, and equipment operating cycle time. The vibration response data, thermal response data, electrical response data, acoustic response data, and environmental condition data are supplemented with time identifiers, measurement point identifiers, and condition identifiers to form an original data set.

3. The method for evaluating the condition of damping sealing materials as described in claim 2, characterized in that, Step S2 specifically includes: Read the original data set and establish a data index for each channel based on the time identifier and the measurement point identifier; The time markers are extracted from the vibration response data, thermal response data, electrical response data, acoustic response data, and environmental condition data. A reference time axis is determined based on a unified time reference, and the vibration response data, thermal response data, electrical response data, acoustic response data, and environmental condition data are mapped to the reference time axis to complete the time alignment. The vibration response data and acoustic response data after time alignment are subjected to noise suppression processing, which includes outlier identification, frequency band filtering and smoothing. The time-aligned thermal response data is subjected to noise suppression processing, which includes moving average and local baseline correction. The time-aligned electrical response data is subjected to noise suppression processing, which includes stable segment extraction, outlier removal, and low-frequency drift subtraction.

4. The method for evaluating the condition of damping sealing materials as described in claim 3, characterized in that, Step S2 also includes: Based on the environmental operating condition data, temperature compensation, humidity compensation, load compensation, medium influence compensation, and operating cycle compensation are performed on the vibration response data, thermal response data, electrical response data, and acoustic response data after noise suppression. Using data under standard reference conditions as a benchmark, the correspondence between environmental conditions and response offset is established through calibration tests. Based on the synchronously acquired environmental condition data, the corresponding correction rules are called to perform condition normalization processing on each response data after compensation. The vibration response data, thermal response data, electrical response data, and acoustic response data after normalization of operating conditions are appended with unified time identifiers, measurement point identifiers, and operating condition identifiers. Data of different dimensions are converted into relative changes, normalized amplitudes, segmented trend quantities, and window statistics to obtain standardized state characterization data.

5. The method for evaluating the condition of damping sealing materials as described in claim 4, characterized in that, Step S3 specifically includes: Multidimensional feature extraction is performed on the vibration response data, thermal response data, electrical response data, and acoustic response data in the standardized state characterization data, respectively; The vibration response data is used to extract dynamic response amplitude variation characteristics, vibration attenuation characteristics, frequency band energy distribution characteristics, hysteresis variation trend characteristics, and local wave intensity characteristics. The thermal response data is used to extract features such as temperature rise rate, steady-state temperature difference, thermal diffusion uniformity, local thermal hysteresis, and temperature fluctuation stability. The amplitude drift feature, frequency band response change feature, polarization change trend feature, conductivity change trend feature, and stable section offset feature are extracted from the electrical response data. The acoustic response data is used to extract burst event density features, local energy enhancement features, transient fluctuation intensity features, and abnormal event distribution features to form an initial feature set.

6. The method for evaluating the condition of damping sealing materials as described in claim 5, characterized in that, Step S3 also includes: The initial feature set is subjected to correlation test, stability test and sensitivity test. Feature items that are not sensitive to state changes, fluctuate too much under different working conditions and are highly repetitive with other features are removed to obtain a feature subset. The feature subset is subjected to uniform scaling to convert each feature into a normalized feature value; The normalized features are combined in a preset order to construct a fixed-dimensional state fingerprint vector.

7. The method for evaluating the condition of damping sealing materials as described in claim 6, characterized in that, Step S4 specifically includes: The state fingerprint vector is input into a preset physical constraint state estimation model; The physical constraint rules in the physical constraint state estimation model include state evolution continuity constraints, degradation change direction constraints, parameter value boundary constraints, environmental condition consistency constraints, and multi-feature response coordination constraints. Based on the state mapping rules and physical constraint rules in the physical constraint state estimation model, the state fingerprint vector is subjected to state space projection, constraint correction and parameter calculation to obtain the initial state estimation result; The consistency of the initial state estimation results is verified. When the initial state estimation result satisfies the physical constraint rules, the corresponding degradation state parameters are output.

8. The method for evaluating the condition of damping sealing materials as described in claim 7, characterized in that, Step S4 also includes: The degradation state parameters are obtained by solving the physical constraint state estimation model. The degradation state parameters include the damping performance retention rate, the change in interface contact stiffness, the degree of thermal response hysteresis, and the cumulative activity of abnormal events. Based on the degradation state parameters, a health index is generated according to a preset mapping rule; A degradation level is generated based on the health index, the degree of change of the degradation state parameters, and the state evolution trend; When the level corresponding to the health index is inconsistent with the level corresponding to the degradation state parameter, the higher degradation level shall be used for judgment. The health index, the degradation level, and the degradation state parameters are used as the output of step S4.

9. The method for assessing the condition of damping sealing materials as described in claim 8, characterized in that, Step S5 specifically includes: Based on the health index and the degradation level, a state assessment result of the damping sealing material is generated according to a preset assessment rule library; Based on the degradation state parameters and the environmental condition data, a preset failure criterion is invoked, and the future degradation trajectory is estimated based on the historical change trend of the degradation state parameters. The future degradation trajectory is compared with the failure criterion to determine the predicted failure time, and the remaining service life assessment result is generated based on the time length between the current time and the predicted failure time. Based on the probability of reaching the preset failure criterion within a predetermined future time window, a failure risk probability is generated. Output the status assessment results, the failure risk probability, and the remaining useful life assessment results.

10. A damping sealing material condition assessment system, used to implement the damping sealing material condition assessment method as described in any one of claims 1 to 9, characterized in that, include: Data acquisition module: Acquires various types of response data and corresponding environmental condition data generated by the damping sealing material under preset excitation conditions; Data processing module: Based on the environmental condition data, performs time-series alignment, noise suppression, and compensation processing on the multi-type response data to obtain standardized state characterization data; Fingerprint construction module: Based on the standardized state characterization data, multi-dimensional feature extraction and fusion are performed to construct the state fingerprint vector corresponding to the damping sealing material; State estimation module: Input the state fingerprint vector into the preset physical constraint state estimation model to calculate the health index, degradation level and degradation state parameters corresponding to the damping sealing material; Assessment and prediction module: Based on the health index and the degradation level, it generates a status assessment result of the damping sealing material, and predicts the remaining service life based on the degradation status parameters and the environmental operating condition data, thereby obtaining the failure risk probability and the remaining service life assessment result.

Images