Large heavy-load transmission gear meshing surface service performance sensing system and method

Abstract

The invention relates to the technical field of gear monitoring, and discloses a large-scale heavy-load transmission gear meshing surface service performance sensing system and method.The method comprises the steps that historical data of a target gear under different working conditions are obtained, and a gear meshing surface service performance evaluation model is established; collecting real-time data of the target gear, and obtaining the current service performance grade of the target gear according to the real-time data and the evaluation model; a lubricating oil pollution state threshold value is obtained, and the threshold value is dynamically calibrated according to the current service performance grade of the target gear; whether the real-time data exceed the calibrated threshold value or not is judged, and if yes, online filtering is conducted on the lubricating oil; the lubricating oil pollution state threshold value can be dynamically adjusted according to the real-time service state of the gear, self-adaptive detection and treatment of lubricating oil pollution are achieved, the service life of the gear is prolonged, and the system reliability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gear monitoring, and more specifically, to a system and method for sensing the service performance of the meshing surface of a large and heavy-loaded transmission gear. Background Art

[0002] With the rapid development of industry, gear transmission devices have been widely used in various mechanical equipment. Gear transmission has the advantages of compact structure, large transmission ratio and high efficiency, and plays a key role in large and heavy-load equipment such as wind turbines, mining machinery, metallurgical equipment and other fields. However, gears are in service for a long time under harsh working conditions, and their meshing surfaces are prone to wear, pitting and other failures, which lead to increased equipment vibration and reduced transmission efficiency. In severe cases, it will cause sudden accidents and cause significant economic losses. Therefore, real-time perception of the service performance of gears and accurate assessment of their health status are crucial to ensure safe and efficient operation of equipment.

[0003] For example, the Chinese patent with publication number CN118690577A discloses a bearing wear warning method, which obtains the actual value of the copper-based abrasive content in the oil, searches for the corresponding relationship between the bearing wear state and the abrasive content under different working conditions recorded in the target mapping database, and judges the degree of bearing wear risk and performs fault warning. This method establishes the association between oil abrasive and bearing wear, providing a basis for state assessment, but oil sampling and detection takes a long time, making it difficult to achieve online real-time monitoring, and the materials and working conditions of bearings and gears are very different, so the characterization ability of oil indicators is limited. Another example is the Chinese patent with authorization announcement number CN118110641B, which discloses an intelligent lubrication system for wind power gearboxes based on environmental perception. It collects gearbox environment and operation data through sensors, builds a performance attenuation model, monitors the impurity content of lubricating oil, performs filtering treatment when pollution exceeds the standard, and automatically supplies new oil when lubrication is insufficient. The system realizes online monitoring of lubricating oil contamination and liquid level, but lacks perception of the dynamic response of the meshing surface, and the pertinence and accuracy of service performance evaluation need to be improved. Another example is a Chinese patent with publication number CN114755007A, which proposes a gear fault diagnosis method that performs period segmentation, interval comparison, and time domain averaging on the vibration signal to obtain synchronous average data to determine whether the gear is faulty. This method overcomes the influence of speed fluctuations on diagnosis, but does not consider the differences in vibration characteristics caused by changes in lubrication conditions, and is insufficient in early diagnosis of microscopic failures of the meshing surface.

[0004] In summary, the existing technology lacks real-time online sensing means for the service status of the gear meshing surface, making it difficult to comprehensively and accurately evaluate the health level of the gear. The timeliness of lubricant contamination monitoring is not high, which affects the lubrication performance of the grease and aggravates tooth surface wear. The relationship between changes in gear service performance and lubrication requirements is ignored, and the adaptability of lubrication status regulation needs to be improved. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a service performance sensing system and method for the meshing surface of large and heavy-loaded transmission gears, which conducts coordinated monitoring of multi-physical field parameters on the meshing tooth surface, characterizes the service status of the gear in real time, and adaptively calibrates the lubricating oil contamination threshold based on this, dynamically optimizes the lubrication status, thereby extending the service life of the gear and improving the system operation reliability.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for sensing the service performance of a meshing surface of a large and heavy-loaded transmission gear, comprising:

[0008] Acquire first data of the target gear, and establish a gear meshing surface service performance evaluation model based on the first data; the first data includes a historical meshing surface conductivity time series, a historical vibration spectrum time series, and a historical temperature gradient time series of the target gear under different working conditions;

[0009] The real-time meshing surface conductivity time series, real-time vibration spectrum time series and real-time temperature gradient time series of the target gear are collected to generate second data; the current service performance level of the target gear is obtained according to the second data and the gear meshing surface service performance evaluation model; the lubricating oil contamination state threshold is obtained according to the first data; the lubricating oil contamination state threshold is dynamically calibrated according to the current service performance level of the target gear to obtain the calibrated lubricating oil contamination state threshold; it is determined whether the second data exceeds the calibrated lubricating oil contamination state threshold, if not, real-time monitoring is continued; if exceeded, the lubricating oil is filtered online to remove pollutants.

[0010] Furthermore, the establishment of a gear meshing surface service performance evaluation model includes:

[0011] The statistical characteristics and morphological characteristics of the historical meshing surface conductivity time series, the historical vibration spectrum time series and the historical temperature gradient time series are extracted to form a historical feature set. With the historical feature set as input and the gear service performance grade as output, a support vector machine (SVM) regression model is trained to obtain a gear meshing surface service performance evaluation model. The gear service performance grade includes four grades: excellent, good, average and poor.

[0012] Furthermore, obtaining the current service performance level of the target gear includes: extracting statistical features and morphological features of the second data to generate a real-time feature set; and obtaining the current service performance level of the target gear based on the real-time feature set and a gear meshing surface service performance evaluation model.

[0013] Furthermore, the lubricating oil contamination state threshold includes three sub-thresholds, and the three sub-thresholds are conductivity sub-thresholds. , vibration spectrum abnormality sub-threshold and temperature gradient anomaly sub-threshold ;

[0014] The step of obtaining the lubricating oil contamination state threshold comprises:

[0015] Calculate the conductivity sub-threshold according to the historical meshing surface conductivity time series ;

[0016] Calculate the vibration spectrum anomaly of each vibration sample point in the historical vibration spectrum time series, obtain the vibration spectrum anomaly sequence, and take the 90% quantile of the vibration spectrum anomaly sequence as the vibration spectrum anomaly sub-threshold ;

[0017] According to the historical temperature gradient time series, the temperature gradient anomaly sub-threshold is obtained .

[0018] Furthermore, obtaining the temperature gradient anomaly degree sub-threshold according to the historical temperature gradient time series includes:

[0019] Calculate the statistical characteristics of the historical temperature gradient time series to form a temperature gradient feature set;

[0020] The principal component analysis method is used to reduce the dimension of the temperature gradient feature set, extract the principal component scores, and obtain the low-dimensional features of the temperature gradient;

[0021] The support vector data description detection model is trained with the low-dimensional features of temperature gradient as input; the temperature gradient anomaly is defined as the distance from the sample point in the historical temperature gradient time series to the hypersphere of the support vector data description detection model, and the temperature gradient anomaly sequence is obtained;

[0022] The temperature gradient anomaly series was statistically analyzed, and the 90% quantile was taken as the temperature gradient anomaly sub-threshold.

[0023] Furthermore, the calibrated lubricating oil contamination state threshold comprises a calibrated conductivity sub-threshold , calibrated vibration spectrum abnormality sub-threshold and the calibrated temperature gradient anomaly sub-threshold ;

[0024] The dynamic calibration of the lubricating oil contamination state threshold according to the current service performance level of the target gear includes:

[0025] If the current service performance level of the target gear is excellent, the three sub-thresholds are increased according to the proportional coefficient a to obtain , and ;

[0026] If the current service performance level of the target gear is good, the three sub-thresholds are increased according to the proportional coefficient b to obtain , and ;

[0027] If the current service performance level of the target gear is average, the three sub-thresholds remain unchanged;

[0028] If the current service performance level of the target gear is poor, the three sub-thresholds are reduced by the proportional coefficient c to obtain , and ; Among them, a>b>1.0>c.

[0029] Further, the determining whether the second data exceeds the calibrated lubricating oil contamination state threshold comprises:

[0030] Traverse the real-time meshing surface conductivity time series to determine whether each conductivity sample point in the real-time meshing surface conductivity time series exceeds the calibrated conductivity sub-threshold , if so, the calibrated conductivity sub-threshold will be exceeded The conductivity sample points are marked as conductivity exceeding limit sample points. If the conductivity exceeding limit sample points appear continuously and the number exceeds n1, then mark M1=1, otherwise, mark M1=0; n1 is the preset conductivity exceeding limit sample number threshold, and M1 is the conductivity abnormality indicator variable;

[0031] Calculate the vibration spectrum anomaly of each vibration sample point in the real-time vibration spectrum time series to obtain a real-time vibration spectrum anomaly sequence; traverse the real-time vibration spectrum anomaly sequence to determine whether each vibration sample point in the real-time vibration spectrum anomaly sequence exceeds the calibrated vibration spectrum anomaly sub-threshold If so, the calibrated vibration spectrum abnormality threshold will be exceeded. The vibration sample points are marked as vibration abnormal sample points. If the vibration abnormal sample points appear continuously and the number exceeds n2, then mark M2=1, if not, then mark M2=0; n2 is the preset vibration abnormal sample number threshold, and M2 is the vibration abnormality indicator variable.

[0032] Further, the determining whether the second data exceeds the calibrated lubricating oil contamination state threshold value further includes:

[0033] Obtain the real-time temperature gradient anomaly sequence of the real-time temperature gradient time series, traverse the real-time temperature gradient anomaly sequence, and determine whether each temperature sample point in the real-time temperature gradient anomaly sequence exceeds the calibrated temperature gradient anomaly sub-threshold If so, the calibrated temperature gradient anomaly threshold will be exceeded. The temperature sample points are marked as temperature anomaly sample points. If the temperature anomaly sample points appear continuously and the number exceeds n3, then mark M3=1, otherwise, mark M3=0; n3 is the preset temperature anomaly sample number threshold, and M3 is the temperature gradient anomaly indicator variable;

[0034] If M1=1 or M2=1 or M3=1, it is determined that the second data exceeds the calibrated lubricant contamination state threshold; if M1=0 and M2=0 and M3=0, it is determined that the second data does not exceed the calibrated lubricant contamination state threshold.

[0035] Furthermore, the real-time temperature gradient anomaly degree sequence of obtaining the real-time temperature gradient time series includes:

[0036] Calculate the statistical characteristics of the real-time temperature gradient time series to form a real-time temperature gradient feature set;

[0037] The principal component analysis method is used to reduce the dimension of the real-time temperature gradient feature set to obtain the low-dimensional features of the real-time temperature gradient;

[0038] The detection model is described based on the low-dimensional features of the real-time temperature gradient and the support vector data, and the real-time temperature gradient anomaly degree sequence is obtained.

[0039] A large-scale heavy-duty transmission gear meshing surface service performance sensing system, which is used to implement the above-mentioned large-scale heavy-duty transmission gear meshing surface service performance sensing method, the system comprises:

[0040] Evaluation model building module: used to obtain first data of the target gear, and establish a gear meshing surface service performance evaluation model based on the first data; the first data includes the historical meshing surface conductivity time series, historical vibration spectrum time series and historical temperature gradient time series of the target gear under different working conditions;

[0041] Service performance evaluation module: used to collect the real-time meshing surface conductivity time series, real-time vibration spectrum time series and real-time temperature gradient time series of the target gear to generate second data; according to the second data and the gear meshing surface service performance evaluation model, the current service performance level of the target gear is obtained;

[0042] Lubricating oil contamination status determination module: used to obtain the lubricating oil contamination status threshold value based on the first data, dynamically calibrate the lubricating oil contamination status threshold value according to the current service performance level of the target gear, and obtain the calibrated lubricating oil contamination status threshold value; determine whether the second data exceeds the calibrated lubricating oil contamination status threshold value, if not, continue real-time monitoring; if exceeded, filter the lubricating oil online to remove pollutants.

[0043] An electronic device includes a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, the above-mentioned method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear is implemented.

[0044] A computer-readable storage medium stores a computer program, which, when executed, implements the above-mentioned method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention establishes an evaluation model by acquiring historical data of gears under different working conditions, and combined with the real-time service data of gears, can accurately evaluate the current service performance status of gears, and provide a basis for subsequent lubricant contamination detection. Multi-parameter thresholds of lubricant contamination status are introduced, including conductivity, vibration spectrum anomaly, and temperature gradient anomaly, which can comprehensively reflect the degree of lubricant contamination. The lubricant contamination threshold is adaptively calibrated according to the real-time service performance level of the gear. When the service performance of the gear decreases, the lubricant contamination threshold is correspondingly reduced to improve the detection rate of lubricant contamination; when the service performance of the gear is good, the contamination threshold is appropriately increased to avoid too frequent lubricant treatment and save costs. The lubricant contamination status is judged in real time. Once the lubricant contamination exceeds the standard, the online filtering process is started in time to realize the early detection and early treatment of lubricant contamination, delay gear wear, and extend the service life of the gear. Through algorithms such as temperature gradient anomaly analysis, abnormal working conditions such as sudden temperature changes can be effectively identified, providing more information for gear service status diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 It is a principle flow chart of a method for sensing the service performance of the meshing surface of a large heavy-load transmission gear in the present invention;

[0049] Figure 2 A flowchart of a method for obtaining first data of a target gear in a service performance sensing method for a meshing surface of a large and heavy-loaded transmission gear according to the present invention;

[0050] Figure 3A flow chart of a method for obtaining a lubricating oil contamination state threshold value according to first data in a large-scale heavy-load transmission gear meshing surface service performance sensing method of the present invention;

[0051] Figure 4 A flow chart of a method for obtaining a temperature gradient abnormality sub-threshold in a service performance sensing method for a meshing surface of a large heavy-load transmission gear of the present invention;

[0052] Figure 5 A flow chart of a method for dynamically calibrating a lubricating oil contamination state threshold value in a large-scale heavy-load transmission gear meshing surface service performance sensing method of the present invention;

[0053] Figure 6 A flow chart of a method for obtaining a real-time temperature gradient anomaly sequence of a real-time temperature gradient time sequence in a service performance sensing method for a meshing surface of a large heavy-load transmission gear of the present invention;

[0054] Figure 7 It is a functional module diagram of a large-scale heavy-load transmission gear meshing surface service performance sensing system in the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Example 1

[0057] See also Figure 1 As shown, this embodiment provides a method for sensing the service performance of a meshing surface of a large and heavy-duty transmission gear, comprising:

[0058] Step S1000, obtaining first data of a target gear, and establishing a gear meshing surface service performance evaluation model based on the first data; the first data includes a historical meshing surface conductivity time series, a historical vibration spectrum time series, and a historical temperature gradient time series of the target gear under different working conditions;

[0059] Furthermore, step S1000 includes:

[0060] Step S1100, obtaining first data of a target gear, wherein the first data includes a historical meshing surface conductivity time series, a historical vibration spectrum time series, and a historical temperature gradient time series of the target gear under different working conditions;

[0061] Furthermore, if Figure 2 As shown, step S1100 includes:

[0062] Step S1110, arranging a conductivity sensor array on the meshing surface of the target gear, collecting conductivity data of the meshing surface, and forming a historical meshing surface conductivity time series;

[0063] Step S1120, arranging an acceleration sensor on the bearing seat of the target gear to collect vibration signals during the operation of the gear, and forming a historical vibration spectrum time series through spectrum analysis;

[0064] Step S1130, by arranging an infrared temperature sensor array at different positions of the meshing surface of the target gear, collecting temperature data of the meshing surface, and calculating the temperature gradient to form a historical temperature gradient time series.

[0065] Specifically, in step S1100, multi-source heterogeneous data of the gear service status are obtained by multi-sensor data fusion. The conductivity of the gear meshing surface reflects the conductive properties of the surface oil film and is closely related to the degree of surface oil contamination. When the oil contamination of the tooth surface intensifies, the impurity particles will reduce the insulation of the oil film, resulting in an increase in conductivity. By arranging an array of conductivity sensors, the time evolution law of the conductivity of the meshing surface can be captured. The vibration spectrum of the gear contains rich dynamic characteristic information. When the wear of the tooth surface intensifies, characteristic frequency components related to the meshing frequency will be stimulated. By performing spectral analysis on the vibration signal and extracting the time trend of these characteristic frequency components, the wear state of the tooth surface can be indirectly evaluated. The temperature distribution of the gear meshing surface is closely related to friction heat. When lubrication fails, the local temperature will rise sharply and the temperature gradient will increase. By collecting and calculating the gradient of the temperature at different positions, the change of the lubrication state of the meshing surface can be sensitively perceived. By comprehensively using multi-source information such as conductivity, vibration and temperature, the internal service state of the gear can be viewed from the three perspectives of oil contamination, wear and lubrication, and comprehensive perception can be achieved.

[0066] Step S1100 obtains the key performance parameters of the gear during service by fusion of multi-sensor data. Conductivity reflects the oil contamination state of the meshing surface and is a direct representation of oil degradation and contamination particles; the vibration spectrum reveals the dynamic characteristics of the gear and is closely related to tooth surface wear; the temperature gradient describes the lubrication state of the friction pair from a thermodynamic perspective and is a sensitive indicator of lubrication failure. The temporal variation patterns of these three types of data contain the inherent mechanism of the evolution of gear service performance and are important inputs for subsequent evaluation models. The collaborative perception of multi-source heterogeneous data expands the limitations of the perception of a single physical quantity, realizes a three-dimensional description of the gear service state, and lays a data foundation for the comprehensive and accurate evaluation of gear performance. At the same time, the use of non-contact detection methods such as conductivity, vibration and temperature overcomes the limitations of traditional manual inspections, realizes online monitoring of gear status, and can promptly detect signs of performance degradation, providing a basis for status maintenance decisions.

[0067] Step S1200: establishing a gear meshing surface service performance evaluation model based on the first data.

[0068] The establishment of the gear meshing surface service performance evaluation model comprises:

[0069] The statistical characteristics and morphological characteristics of the historical meshing surface conductivity time series, the historical vibration spectrum time series and the historical temperature gradient time series are extracted to form a historical feature set. With the historical feature set as input and the gear service performance grade as output, a support vector machine (SVM) regression model is trained to obtain a gear meshing surface service performance evaluation model. The gear service performance grade includes four grades: excellent, good, average and poor.

[0070] Specifically, in step S1200, a machine learning method is used to mine the correlation between multi-source time series data and gear service performance. Data preprocessing is performed on the conductivity time series, vibration spectrum time series and temperature gradient time series, including outlier removal, data normalization and other operations; outliers are usually caused by sensor failures, signal mutations, etc., which will interfere with subsequent feature extraction and model training, and need to be identified and eliminated by methods such as the 3σ criterion. Data normalization unifies time series data of different dimensions to the same scale, facilitating subsequent feature fusion. From the perspectives of time domain and frequency domain, the statistical features and morphological features of multi-source time series are extracted. Statistical features include mean, variance, peak-to-peak value, etc., which characterize the central tendency and discreteness of time series data; morphological features include skewness, kurtosis, etc., which reflect the distribution form of time series data. By combining these features, the evolution law of time series data can be fully characterized. Finally, the extracted features are input into the support vector machine for regression training. Support vector machine is a pattern recognition method based on statistical learning theory. It realizes nonlinear mapping of feature space by seeking the optimal classification hyperplane. Taking the service performance level scored by experts on typical damage patterns as the training target, the support vector machine model parameters are optimized, and a nonlinear mapping relationship between features and performance levels is established to form a service performance evaluation model. The model can quickly estimate the service performance status of the gear based on the input features, providing a quantitative basis for subsequent maintenance decisions. Compared with empirical evaluation, this model makes full use of the degradation laws contained in the data, overcomes the limitations of manual experience, and realizes intelligent perception of service performance.

[0071] Step S1200 uses a machine learning method to construct a gear meshing surface service performance evaluation model. This model is driven by big data as its core concept. By extracting sensitive features from multi-source heterogeneous data, it reveals the degradation mechanism of gears in the service process and breaks through the limitations of traditional physical models. The machine learning model has a strong nonlinear fitting ability and can establish a performance degradation model under complex working conditions. By learning from typical damage cases, the model has mastered the evaluation knowledge of service performance, can independently judge the health status of gears, and reduces dependence on manual experience. This evaluation model is a data-driven state perception tool. Through the continuous accumulation and learning of state data, the generalization ability of the model is continuously enhanced, and the evaluation results are more accurate and reliable. Embedding the evaluation model into the online monitoring system can realize the real-time evaluation of the service performance of gears and assist maintenance personnel in making scientific decisions. At the same time, the evaluation model is also an effective means of knowledge accumulation. By continuously learning the accumulated data, it can continuously optimize the internal parameters, expand the knowledge base of performance degradation, and provide an intelligent tool for the service performance management of gears throughout their life cycle.

[0072] Step S2000, collect the real-time meshing surface conductivity time series, real-time vibration spectrum time series and real-time temperature gradient time series of the target gear to generate the second data; obtain the current service performance level of the target gear according to the second data and the gear meshing surface service performance evaluation model; obtain the lubricating oil contamination state threshold according to the first data; dynamically calibrate the lubricating oil contamination state threshold according to the current service performance level of the target gear to obtain the calibrated lubricating oil contamination state threshold; determine whether the second data exceeds the calibrated lubricating oil contamination state threshold, if not, continue real-time monitoring; if exceeded, execute step S3000.

[0073] Furthermore, step S2000 includes:

[0074] Step S2100, collecting the real-time meshing surface conductivity time series, the real-time vibration spectrum time series and the real-time temperature gradient time series of the target gear to generate second data; extracting the statistical features and morphological features of the second data to generate a real-time feature set;

[0075] Exemplarily, the conductivity data of the gear meshing surface in the last hour is collected by a conductivity sensor array, and the data is sampled once per minute to form a real-time meshing surface conductivity time series containing 60 data points;

[0076] The vibration data of the gear in the last hour is collected through the acceleration sensor on the bearing seat, with a sampling frequency of 10kHz. The collected vibration data is subjected to fast Fourier transform (FFT), and the spectrum data in the range of 0-1000Hz is extracted with a frequency resolution of 1Hz, forming a two-dimensional spectrum matrix containing 1000 frequency points and 60 moments as a real-time vibration spectrum time series;

[0077] The infrared temperature sensor array collects the temperature data of different positions on the gear meshing surface in the last hour, sampling once per minute. The collected data is differentially calculated to obtain the temperature gradient time series between different positions, forming a real-time temperature gradient time series.

[0078] Extract statistical and morphological features of the second data:

[0079] For the real-time meshing surface conductivity time series, statistical features such as mean, variance, skewness and kurtosis are extracted, and morphological features such as trend term, periodic term and random term are extracted.

[0080] For the real-time vibration spectrum time series, statistical features such as frequency band energy, spectrum center of gravity, and spectrum variance are extracted, and morphological features such as the energy proportion of sidebands, center bands, and harmonic bands are extracted.

[0081] For the real-time temperature gradient time series, statistical features such as the temperature gradient mean and temperature gradient variance are extracted, and morphological features such as trend mutation points and periodic peak points are extracted.

[0082] The extracted features are aggregated to form a 25-dimensional real-time feature set.

[0083] Step S2200, obtaining the current service performance level of the target gear according to the real-time feature set and the gear meshing surface service performance evaluation model;

[0084] Specifically, in step S2200, the second data obtained by monitoring is input into the pre-trained gear meshing surface service performance evaluation model, which can quickly determine the current service state of the gear. First, the Z-score normalization method is used to perform dimensionless processing on each time series data to improve the fusion ability of different physical quantity data. Normalization processing can unify data of different orders of magnitude to the same level, which is convenient for subsequent feature extraction and pattern recognition. By extracting the statistical features of the time series data, the distribution law of the data is fully characterized. The mean and root mean square reflect the central trend of the data, the variance reflects the degree of dispersion of the data, and the skewness and kurtosis reflect the skewness and peak characteristics of the data. By extracting morphological features, the evolution law of the time series data is grasped from a macro perspective. The time series decomposition method is used to decompose the time series data into trend terms, periodic terms and random terms. The trend term reflects the long-term change trend of the time series data, the periodic term reflects the periodic fluctuation law in the time series data, and the random term reflects the random disturbance in the time series data. The extracted statistical features and morphological features are input into the support vector machine regression model for state recognition, which can automatically determine the service performance level of the gear. The support vector machine maps the features of different damage modes into high-dimensional space for classification by seeking the maximum margin hyperplane, thus realizing intelligent evaluation of service performance. Through online evaluation of service performance, the health status of gears can be timely understood, and the mode transition from passive maintenance to active prevention can be realized, thus improving the reliability and maintenance economy of the gear system.

[0085] Step S2300: Obtain a lubricating oil contamination state threshold according to the first data, wherein the lubricating oil contamination state threshold includes three sub-thresholds, wherein the three sub-thresholds are conductivity sub-thresholds. , vibration spectrum abnormality sub-threshold and temperature gradient anomaly sub-threshold ;

[0086] Furthermore, if Figure 3 As shown, step S2300 includes:

[0087] Step S2310, calculating the conductivity sub-threshold according to the historical meshing surface conductivity time series;

[0088]

[0089] in:

[0090] : Conductivity sub-threshold, which indicates the critical value of the conductivity of the gear meshing surface under specific working conditions. This threshold is used to monitor the conductivity change in real time and determine whether a possible fault state has been reached. The higher the threshold, the greater the fluctuation of the surface conductivity and the higher the potential failure risk.

[0091] : Weight coefficient of conductivity fluctuation, used to control the influence of variance term in the formula on the final threshold. The larger the value of α, the higher the sensitivity to conductivity fluctuation. If the conductivity of the gear meshing surface fluctuates violently (variance increases), this term will significantly increase the conductivity sub-threshold , which helps to capture anomalies in time when the environment is unstable.

[0092] : The weight coefficient of the weighted average value of conductivity, which is used to control the influence of the weighted average term in the formula on the final threshold. The weighted average term reflects the overall average trend of the conductivity data. β controls the baseline value of conductivity under the current working conditions. In some cases, the overall average value or trend of conductivity is more important. When β is larger, the threshold More dependent on the baseline level of conductivity.

[0093] : The weight coefficient of the instantaneous rate of change, which is used to control the influence of the maximum difference term in the formula on the final threshold. The maximum difference term is used to capture the sudden change of conductivity and reflect the rapid change of conductivity on the gear surface. γ adjusts the influence of this sudden change on the threshold. If γ is large, It will improve quickly, mainly playing a role in monitoring sudden failures, helping to improve real-time response capabilities.

[0094] The value of is obtained by fitting historical data and can usually be set to satisfy , to keep the overall weight balanced. These coefficients are estimated by machine learning models or optimization algorithms, and their specific values ​​depend on the equipment's operating conditions and performance requirements.

[0095] :The first in the historical meshing surface conductivity time series The conductivity data.

[0096] :The first in the historical meshing surface conductivity time series The conductivity data.

[0097] : The average value of the conductivity in the historical meshing surface conductivity time series, indicating the overall level of conductivity during the sampling period.

[0098] : The number of sampling points, that is, the total number of conductivity sample points in the historical meshing surface conductivity time series.

[0099] : No. The environmental weight factor of each conductivity data reflects the impact of the environment on the conductivity, such as temperature, humidity, load, etc. This parameter can be calculated through environmental sensor data or combined with existing empirical models to define the impact weight of conductivity under different environmental factors.

[0100] With the conductivity fluctuation increase, will increase, indicating that the threshold increases when the fluctuation is large; when the weight factor Increase, indicating that the current environment has a stronger impact on conductivity. Increases accordingly; if the instantaneous conductivity changes suddenly ( Larger), It will also be improved to deal with the risk of sudden failure.

[0101] This formula is used to calculate the conductivity sub-threshold under different working conditions, and can realize efficient monitoring of the conductivity state of the gear meshing surface. This formula comprehensively considers the volatility of conductivity, environmental impact and sudden changes (maximum difference), and can keenly capture abnormal conditions, which is suitable for real-time monitoring in complex environments. The variance term in the formula reflects the volatility of conductivity, the weighted average term provides an overall trend judgment, and the maximum difference term is used to identify sudden failures, which improves the accuracy of monitoring from multiple angles. and The formula can be adjusted dynamically under different working conditions to adapt to the gear service status evaluation under various working conditions. Through this formula, the fluctuation of gear conductivity under different working conditions can be effectively predicted and responded to, reducing false alarms and improving the ability to capture abnormalities.

[0102] Step S2320, calculating the vibration spectrum abnormality of each vibration sample point in the historical vibration spectrum time series, obtaining a vibration spectrum abnormality sequence, and taking the 90% quantile of the vibration spectrum abnormality sequence as the vibration spectrum abnormality sub-threshold;

[0103] Specifically, first, the statistical features of the historical vibration spectrum time series are extracted, such as mean, root mean square value, peak factor, waveform factor, etc. These features reflect the energy distribution and waveform characteristics of the vibration signal from the perspectives of time domain and frequency domain. Then, the abnormality of each vibration sample is calculated using anomaly detection algorithms such as Mahalanobis distance. Mahalanobis distance takes into account the correlation between features and can effectively distinguish normal samples from abnormal samples. The abnormality is positively correlated with the Mahalanobis distance. The greater the abnormality, the greater the degree of deviation of the sample from normal working conditions. By setting a suitable quantile such as 90%, the abnormality sub-threshold can be adaptively determined, that is, 90% of historical samples less than the threshold are considered normal. This data-driven threshold determination method makes full use of the statistical laws of historical big data and has better objectivity and robustness.

[0104] Step S2330, obtaining a temperature gradient abnormality sub-threshold according to the historical temperature gradient time series.

[0105] Furthermore, if Figure 4 As shown, step S2330 includes:

[0106] Step S2331, calculating the statistical characteristics of the historical temperature gradient time series to form a temperature gradient feature set;

[0107] Step S2332, using the principal component analysis method to reduce the dimension of the temperature gradient feature set, extracting the principal component scores, and obtaining the low-dimensional features of the temperature gradient;

[0108] Step S2333, using the temperature gradient low-dimensional features as input, training the support vector data description detection model; defining the temperature gradient anomaly as the distance from the temperature sample point in the historical temperature gradient time series to the support vector data description detection model hypersphere, and obtaining the temperature gradient anomaly sequence;

[0109] Step S2334, statistically analyze the temperature gradient anomaly degree sequence, and take the 90% quantile as the temperature gradient anomaly degree sub-threshold.

[0110] Specifically, in step S2330, the temperature gradient anomaly sub-threshold is obtained in a data-driven manner. First, feature extraction is performed on the historical temperature gradient time series data to obtain statistics reflecting the temperature gradient distribution characteristics, such as the mean reflects the average level of the temperature gradient, the variance reflects the fluctuation amplitude of the temperature gradient, and the skewness and kurtosis reflect the skewness and sharpness of the temperature gradient distribution. Then, a dimensionality reduction method such as principal component analysis is used to remove noise and redundancy while retaining the main feature information of the temperature gradient, and obtain a compact low-dimensional feature representation to facilitate the training of subsequent anomaly detection models.

[0111] The training of the anomaly detection model adopts a single-class anomaly detection method based on support vector data description (SVDD). SVDD describes the training samples by finding a hypersphere with the minimum volume, so that the normal samples are surrounded by the hypersphere, while the abnormal samples are outside the hypersphere. When training the SVDD model, only normal state data is required. By minimizing the radius of the hypersphere, the optimal center and radius parameters are obtained. After mapping the temperature sample point to the feature space, the distance from the center of the sphere is calculated. The larger the distance, the greater the degree of deviation of the temperature sample point from the normal data, that is, the higher the degree of anomaly. Therefore, the temperature gradient anomaly reflects the degree of deviation of the temperature gradient from the normal working condition, and is a quantitative indicator for judging whether the temperature gradient is abnormal.

[0112] All temperature gradient anomalies are combined into a temperature gradient anomaly sequence. By statistically analyzing the historical temperature gradient anomaly sequence, a reasonable temperature gradient anomaly sub-threshold can be obtained. Using the 90% quantile as the sub-threshold means that 90% of the historical data has an anomaly below this threshold. This threshold can cover most normal operating conditions and has good sensitivity to abnormal conditions. During online monitoring, the anomaly of the current temperature gradient is calculated and compared with the sub-threshold. If it exceeds the threshold, it can be determined as an abnormal state and an alarm is required.

[0113] The determination of the abnormality threshold can effectively balance the risk of missed detection and false detection. Reasonable threshold selection requires a balance between the sensitivity and reliability of detection. Through a large amount of historical data analysis and statistical modeling, the abnormality thresholds of different working conditions and different equipment can be adaptively determined to improve the accuracy and real-time performance of abnormality detection. Temperature gradient anomaly detection provides a quantitative basis for temperature anomaly diagnosis and fault warning of gearboxes, and is an important means to achieve condition monitoring and fault diagnosis. Combined with the anomaly detection of other state parameters such as vibration and oil quality, a multi-parameter anomaly monitoring system for gearboxes can be constructed to achieve comprehensive perception and evaluation of the operating status of the equipment, providing a reliable basis for condition maintenance and fault prevention. This method makes full use of the degradation laws and statistical information contained in the historical temperature gradient data, and obtains adaptive anomaly quantification indicators and threshold criteria through data-driven modeling. Compared with the traditional empirical threshold, it can better adapt to changes in equipment working conditions and improve the adaptability and robustness of temperature anomaly detection.

[0114] Step S2300 obtains three key threshold indicators of lubricant contamination status through statistical analysis of historical data. The conductivity sub-threshold reflects the critical level of the conductivity of the gear surface oil film and is an important representation of oil degradation and increased contamination. The determination of this threshold requires comprehensive consideration of the statistical characteristics of the conductivity data, including factors such as data volatility, weighted trends and instantaneous mutations, and a reasonable threshold is obtained through weighted calculation. The vibration spectrum anomaly sub-threshold characterizes the degree to which the gear vibration signal deviates from the normal working condition. By extracting the statistical features of the vibration spectrum, such as mean, variance, peak factor, etc., and performing anomaly detection, the threshold boundary of the vibration spectrum anomaly can be determined. The greater the anomaly, the greater the deviation of the vibration signal from the normal working condition, reflecting the nonlinear dynamic response caused by the increased wear of the tooth surface. The temperature gradient anomaly sub-threshold is used to characterize the risk of local temperature rise in the meshing caused by lubrication failure. By performing feature extraction and dimensionality reduction processing on the temperature gradient data, and using the support vector data description method to learn the compact hypersphere boundary under normal working conditions, the temperature gradient anomaly can be sensitively captured. The anomaly quantifies how far the temperature gradient deviates from the normal boundary, reflecting the risk level of lubrication failure.

[0115] The three sub-thresholds are combined to form a complete representation of the lubricant contamination state threshold. This threshold characterizes the critical level of the lubricant contamination state from the three perspectives of electrical performance, kinetics, and thermodynamics. When the actual detected conductivity, vibration spectrum abnormality, or temperature gradient abnormality exceeds the corresponding sub-threshold, it can be determined that the lubricant contamination state exceeds the standard and the corresponding maintenance strategy needs to be adopted. This threshold characterization method that integrates multi-source information overcomes the limitations of a single indicator and improves the reliability of lubricant contamination monitoring and early warning.

[0116] Step S2300 uses historical data to obtain three sub-thresholds. Historical data provides rich background information that can reflect the normal and abnormal conditions of the equipment under different working conditions. Accurate thresholds can be established by analyzing these data. Historical data can capture common patterns and abnormal conditions of gears in long-term operation. The thresholds calculated using these data can better reflect actual working conditions. Through the analysis of historical data, thresholds can be set more accurately, reducing the possibility of false alarms as abnormalities under normal conditions. Historical data helps identify subtle but important changes, ensures a rapid response to abnormal conditions, and improves the sensitivity of the monitoring system. As historical data accumulates, thresholds can be continuously updated and optimized to adapt to changes in equipment status and environment, and improve the robustness of the system.

[0117] Compared with the traditional single indicator threshold, this lubricating oil contamination status threshold has the following advantages:

[0118] It integrates multi-source heterogeneous data to provide comprehensive characterization of electrical performance, kinetics and thermodynamics, making the evaluation more comprehensive and reliable.

[0119] The determination of sub-thresholds is based on the mining and analysis of historical big data, and the adaptive extraction of thresholds is achieved through technologies such as mathematical statistics and machine learning, thus overcoming the subjectivity of empirical thresholds.

[0120] Step S2400: dynamically calibrate the lubricating oil contamination state threshold according to the current service performance level of the target gear to obtain the calibrated conductivity sub-threshold , calibrated vibration spectrum abnormality sub-threshold and the calibrated temperature gradient anomaly sub-threshold , , and forming a calibrated lubricant contamination state threshold;

[0121] Furthermore, if Figure 5 As shown, step S2400 includes:

[0122] Step S2410: If the current service performance level of the target gear is excellent, increase the three sub-thresholds according to the proportional coefficient a to obtain , and ; 1.2 < a ≤ 1.5;

[0123] Step S2420, if the current service performance level of the target gear is good, increase the three sub-thresholds by the proportionality coefficient b to obtain 、 and ; 1.0 < b ≤ 1.2;

[0124] Step S2430, if the current service performance level of the target gear is average, keep the three sub-thresholds unchanged;

[0125] Step S2440, if the current service performance level of the target gear is poor, decrease the three sub-thresholds by the proportionality coefficient c to obtain 、 and ; 0.6 ≤ c < 1.0.

[0126] Specifically, in step S2400, according to the real-time service performance level of the target gear, the previously determined lubricant contamination state threshold is adaptively calibrated, and the threshold level is dynamically adjusted to match the health state of the gear. When the gear service performance is evaluated as "excellent", it indicates that all performance indicators of the gear are within the ideal range, the tooth surface and oil condition are good, and the tolerance of the oil contamination threshold can be appropriately increased. By magnifying the three sub-thresholds by the proportionality coefficient a (1.2 < a ≤ 1.5), the defined range of the threshold is expanded, while ensuring sufficient lubrication margin, unnecessary maintenance intervention is reduced, and the equipment utilization rate and operation and maintenance efficiency are improved. When the gear service performance is evaluated as "good", it indicates that the overall state of the gear is normal, but individual indicators have slightly decreased, and monitoring needs to be moderately strengthened. By slightly increasing the three sub-thresholds by the proportionality coefficient b (1.0 < b ≤ 1.2), a certain observation window period is reserved for state changes on the premise of ensuring safety, avoiding overly sensitive false alarms. When the gear service performance is evaluated as "average", it indicates that a certain degree of performance degradation has occurred in the gear, and the lubricant contamination state needs to be strictly controlled according to the standard requirements, and preventive maintenance should be implemented in a timely manner. At this time, it is advisable to maintain the original oil contamination threshold level to ensure the margin of state monitoring. When the gear service performance is evaluated as "poor", it indicates that the gear is in a sub-healthy or pre-fault state, and problems such as lubrication and wear are more prominent. It is necessary to closely monitor its degradation trend. By reducing the three sub-thresholds by the proportionality coefficient c (0.6 ≤ c < 1.0), the sensitivity to oil contamination is increased, the early warning time point is advanced, and more maintenance window periods are obtained to avoid the chain reaction caused by lubrication failure.

[0127] The proportional coefficients a, b, and c of the threshold calibration can be empirically set according to factors such as the fault history, lubrication characteristics, and reliability requirements of the gear equipment. a>b>1.0>c usually follows the principle of "loose entry and strict exit", that is, the threshold is appropriately relaxed for the condition of good health, and the threshold is timely tightened for the condition of sub-health or high-risk. The setting of the proportional coefficient needs to balance safety and economy on the basis of sufficient experimental verification to ensure the rationality of the threshold adjustment. For example, for a wind turbine gearbox, through a large number of failure case analyses, it is found that when the degree of lubricating oil contamination is relatively light, the conductivity sub-threshold can be relaxed by 20%, and when the service performance of the gearbox is evaluated as "good" and the load is low, the conductivity sub-threshold can be further relaxed to 40%. However, when the service performance drops to "poor", the conductivity sub-threshold needs to be reduced by 30% to keenly capture potential faults caused by lubrication deterioration. This adaptive dynamic threshold adjustment strategy can significantly improve the sensitivity and real-time performance of lubricating oil contamination monitoring, maximize the matching of the health status of the equipment, reduce the false alarm rate while ensuring sufficient safety margin, and provide a more accurate and reliable decision-making basis for condition maintenance.

[0128] Compared with fixed thresholds, the threshold calibration method driven by service performance level has the following advantages:

[0129] Strong adaptability, capable of dynamically adjusting the threshold according to the actual health status of the equipment, improving the sensitivity and accuracy of condition monitoring;

[0130] The degradation law of equipment service performance is fully considered, and the performance level is used as the calibration basis, so the threshold adjustment is more targeted and effective;

[0131] In the stage of mild degradation, the threshold is appropriately relaxed to reduce blind maintenance, improve equipment availability and operation and maintenance efficiency; in the stage of severe degradation, the threshold is tightened in time, the warning time point is moved forward, and more time windows are created for condition-based maintenance decisions.

[0132] The calibration method is simple and easy. It only needs to set three proportional coefficients according to the reliability requirements to achieve flexible adjustment of the threshold. The rules are clear and easy to apply in engineering.

[0133] Step S2500, determine in real time whether the second data exceeds the calibrated lubricant contamination state threshold value, if not, return to step S2100 and continue real-time monitoring; if exceeded, execute step S3000.

[0134] Further, step S2500 includes:

[0135] Step S2510, traverse the real-time meshing surface conductivity time series, and determine whether each conductivity sample point in the real-time meshing surface conductivity time series exceeds the calibrated conductivity sub-threshold value. , if so, the calibrated conductivity sub-threshold will be exceeded The conductivity sample points are marked as conductivity exceeding limit sample points. If the conductivity exceeding limit sample points appear continuously and the number exceeds n1, then mark M1=1, otherwise, mark M1=0; n1 is the preset conductivity exceeding limit sample number threshold, and M1 is the conductivity abnormality indicator variable;

[0136] Step S2520, calculate the vibration spectrum abnormality of each vibration sample point in the real-time vibration spectrum time series to obtain a real-time vibration spectrum abnormality sequence; traverse the real-time vibration spectrum abnormality sequence to determine whether each vibration sample point in the real-time vibration spectrum abnormality sequence exceeds the calibrated vibration spectrum abnormality sub-threshold If so, the calibrated vibration spectrum abnormality threshold will be exceeded. The vibration sample points are marked as vibration abnormal sample points. If the vibration abnormal sample points appear continuously and the number exceeds n2, then mark M2=1, otherwise, mark M2=0; n2 is the preset vibration abnormal sample number threshold, and M2 is the vibration abnormality indicator variable;

[0137] Step S2530, obtaining a real-time temperature gradient anomaly sequence of the real-time temperature gradient time series, traversing the real-time temperature gradient anomaly sequence, and determining whether each temperature sample point in the real-time temperature gradient anomaly sequence exceeds the calibrated temperature gradient anomaly sub-threshold If so, the calibrated temperature gradient anomaly threshold will be exceeded. The temperature sample points are marked as temperature anomaly sample points. If the temperature anomaly sample points appear continuously and the number exceeds n3, then mark M3=1, otherwise, mark M3=0; n3 is the preset temperature anomaly sample number threshold, and M3 is the temperature gradient anomaly indicator variable;

[0138] Furthermore, if Figure 6 As shown, step S2530 includes:

[0139] Step S2531, calculating the statistical characteristics of the real-time temperature gradient time series to form a real-time temperature gradient feature set;

[0140] Step S2532, using a principal component analysis method to reduce the dimension of the real-time temperature gradient feature set to obtain a low-dimensional feature of the real-time temperature gradient;

[0141] Step S2533, describing the detection model according to the real-time temperature gradient low-dimensional features and support vector data, and obtaining the real-time temperature gradient anomaly degree sequence.

[0142] Specifically, in step S2531, statistical features are extracted from the real-time collected temperature gradient time series. The temperature gradient reflects the uneven heat distribution on the gear meshing surface and is closely related to the lubrication state. Common statistical features include mean, variance, peak-to-peak value, root mean square, etc. The mean reflects the overall level of the temperature gradient, the variance reflects the degree of fluctuation of the temperature gradient, the peak-to-peak value reflects the maximum change amplitude of the temperature gradient, and the root mean square comprehensively reflects the energy size of the temperature gradient. These features characterize the distribution law of the temperature gradient time series from different aspects to form a real-time temperature gradient feature set. However, the dimension of the feature set is high, containing a large amount of redundant and noise information, which is not conducive to subsequent abnormality detection. In step S2532, the principal component analysis (PCA) method is used to reduce the dimension of the real-time temperature gradient feature set. PCA is a commonly used linear dimensionality reduction algorithm that maps the original high-dimensional features to a low-dimensional space through orthogonal transformation to extract the main structural information of the data. PCA selects several orthogonal directions with the largest variance as the main components according to the variance of the feature, and discards the components with smaller variance to achieve data compression and noise reduction. PCA dimensionality reduction not only reduces the computational overhead of the feature set, but also highlights the key information contained in the temperature gradient, laying the foundation for subsequent anomaly detection. The data after dimensionality reduction is called the real-time temperature gradient low-dimensional feature.

[0143] In step S2533, the abnormality of the temperature gradient is calculated using the low-dimensional features of the real-time temperature gradient and the pre-trained support vector data description (SVDD) model. SVDD is a single-class anomaly detection algorithm that learns the compact hypersphere of the data under the normal state and judges the data points that deviate from the hypersphere as abnormal. The abnormality can be measured by the distance from the data point to the hypersphere. The larger the distance, the greater the degree to which the data point deviates from the normal state. In order to adapt to the dynamic changes of the temperature gradient, the sliding time window method is used for anomaly detection. The temperature gradient abnormality at the current moment and within a certain time range before and after is calculated each time to form a real-time temperature gradient abnormality sequence. This sequence dynamically reflects the changing trend of the temperature gradient and provides a quantitative indicator for real-time monitoring of the lubrication state.

[0144] Through steps S2531-S2533, the original temperature gradient time series is converted into a real-time temperature gradient anomaly sequence through a series of processing such as feature extraction, principal component analysis and anomaly detection. The anomaly sequence eliminates redundant information and random noise in the temperature gradient, extracts the core features closely related to the lubrication state, and quantifies the degree to which the temperature gradient deviates from the normal state by comparing with the SVDD model. The real-time generation of the anomaly sequence provides a new idea for online evaluation of the lubrication state. By setting a suitable anomaly threshold, the potential risk of lubrication failure can be discovered in time, providing a reliable basis for condition maintenance decisions. Compared with the monitoring of a single physical quantity, the temperature gradient anomaly makes full use of the information of the temperature spatial distribution and has higher sensitivity and diagnostic accuracy to the lubrication state. Integrating this anomaly index into the gear health monitoring system can realize real-time evaluation and early warning of the lubrication state, reduce the probability of accidents, and improve the reliability and safety of the gear transmission system.

[0145] Step S2540, if M1=1 or M2=1 or M3=1, it is determined that the second data exceeds the calibrated lubricant contamination state threshold; if M1=0 and M2=0 and M3=0, it is determined that the second data does not exceed the calibrated lubricant contamination state threshold.

[0146] Specifically, step S2500 comprehensively uses multi-source heterogeneous data such as conductivity, vibration spectrum and temperature gradient, and compares with the corresponding calibration threshold to determine whether the gear meshing surface is in a lubricating oil contamination state. In step S2510, the real-time meshing surface conductivity time series is traversed, and each conductivity sample point is judged one by one whether it exceeds the calibrated conductivity sub-threshold. Conductivity is an important indicator reflecting the degree of lubricating oil contamination. When the number of impurity particles in the lubricating oil increases, the conductivity of the oil film will increase and the conductivity will increase. Therefore, conductivity exceeding the threshold is an important sign of increased lubricating oil contamination. In order to improve the reliability of detection and avoid interference from a single abnormal point, this step sets the criterion for continuous abnormal points, that is, only when the over-limit sample points appear continuously and the number exceeds n1, it is considered that the conductivity anomaly has occurred, and M1 is marked as 1. Here n1 represents the threshold value of the number of conductivity over-limit sample points that appear continuously in the real-time meshing surface conductivity time series. That is, when the number of conductivity out-of-limit sample points that appear continuously exceeds n1, the conductivity is considered abnormal. n1 is an empirical value, usually taken as 10%-20% of the total length of the time series. For example, for a time series with a length of 1000, n1 can be between 100-200. The setting of n1 needs to balance sensitivity and reliability. If it is too small, it will lead to more false alarms, and if it is too large, it may miss the detection. M1 is the conductivity abnormality indicator variable. When the conductivity out-of-limit sample points appear continuously and the number exceeds n1, M1=1, indicating that the conductivity is abnormal; otherwise, M1=0, indicating that the conductivity is normal.

[0147] The idea of ​​step S2520 is similar to that of step S2510, except that the criterion object is replaced by the vibration spectrum abnormality. The vibration spectrum reflects the dynamic characteristics of the gear. When the tooth surface wear intensifies, characteristic frequency components will appear near the meshing frequency, which is manifested as local abnormalities in the spectrum. By extracting the abnormality characteristics of the real-time spectrum and comparing it with the calibration threshold, the wear state of the tooth surface can be judged. Similar to conductivity, vibration anomalies also need to appear for a certain number (>n2) in succession to be confirmed in order to improve reliability. n2 represents the threshold value of the number of consecutive appearances of vibration abnormality sample points in the real-time vibration spectrum abnormality sequence. That is, when the number of consecutive appearances of vibration abnormality sample points exceeds n2, the vibration is considered abnormal. M2 is the vibration abnormality indicator variable. When the vibration abnormality sample points appear continuously and the number exceeds n2, M2=1, indicating that the vibration is abnormal; otherwise, M2=0, indicating that the vibration is normal.

[0148] Step S2530 is for the judgment of temperature gradient abnormality. The temperature gradient reflects the uniformity of heat distribution on the tooth surface and is an important representation of the lubrication state. When lubrication fails, local friction heating will occur and the temperature gradient will increase sharply. Therefore, if the temperature gradient abnormality exceeds the threshold, it means that the lubricant may have failed. Similarly, the temperature anomaly needs to appear continuously enough times (>n3) to be confirmed. n3 represents the threshold value of the number of consecutive appearances of temperature abnormality sample points in the real-time temperature gradient anomaly sequence. That is, when the number of consecutive appearances of temperature abnormality sample points exceeds n3, the temperature gradient is considered abnormal. M3 is the temperature gradient anomaly indicator variable. When temperature abnormality sample points appear continuously and the number exceeds n3, M3=1, indicating that the temperature gradient is abnormal; otherwise M3=0, indicating that the temperature gradient is normal.

[0149] Step S2540 is to combine the above three judgment results to obtain the final lubricant oil contamination status. Since conductivity, vibration and temperature reflect the health level of lubricant oil from different aspects, as long as any one of them is judged to be abnormal, it is considered that the lubricant oil is contaminated and maintenance measures need to be taken. Only when the three criteria are normal can the lubricant oil status be confirmed to be healthy. This fusion criterion helps to improve the reliability of detection and reduce missed reports and false alarms.

[0150] In summary, step S2500 makes full use of multi-source heterogeneous data to comprehensively reflect the lubricant contamination status and avoid the limitations of single information; it adopts relative threshold criteria to overcome the influence of operating condition fluctuations and can adapt to different service conditions; it integrates the detection results of multiple subsystems to improve the accuracy and reliability of diagnosis; it can perform online real-time monitoring to promptly detect signs of lubricant contamination and facilitate status maintenance.

[0151] Exemplarily, step S2510 traverses the real-time meshing surface conductivity time series {0.08, 0.09, 0.12, 0.15, 0.11, 0.13, 0.12, 0.14, 0.18, 0.19, 0.21, 0.14, 0.16, 0.15, 0.16}, assuming that the calibrated conductivity sub-threshold is 0.16. Determine whether each conductivity sample point in the real-time meshing surface conductivity time series exceeds the calibrated conductivity sub-threshold. For the 9th, 10th, and 11th points that exceed the threshold of 0.16, mark them as conductivity exceeding limit sample points. If 3 conductivity exceeding limit sample points appear continuously and exceed the given number n1 (assuming n1=2), mark M1=1.

[0152] Step S2520, calculate the vibration spectrum abnormality of each vibration sample point in the real-time vibration spectrum time series, and obtain the real-time vibration spectrum abnormality sequence {0.12, 0.15, 0.22, 0.18, 0.14, 0.08, 0.13, 0.09, 0.17,0.19, 0.20, 0.26, 0.29, 0.27, 0.28}, assuming that the calibrated vibration spectrum abnormality sub-threshold is 0.25. Traverse the real-time vibration spectrum abnormality sequence to determine whether each vibration sample point exceeds the calibrated vibration spectrum abnormality sub-threshold. For the 12th, 13th, 14th, and 15th points that exceed the threshold of 0.25, mark them as vibration abnormal sample points. If 4 vibration abnormal sample points appear continuously and exceed the given number n2 (assuming n2=3), mark M2=1.

[0153] Step S2530, obtain the real-time temperature gradient anomaly sequence {0.02, 0.05, 0.03, 0.06, 0.09, 0.13, 0.12, 0.15, 0.18, 0.04, 0.10, 0.09, 0.07, 0.06, 0.04} of the real-time temperature gradient time series, assuming that the calibrated temperature gradient anomaly sub-threshold is 0.10. Traverse the real-time temperature gradient anomaly sequence to determine whether each temperature sample point exceeds the calibrated temperature gradient anomaly sub-threshold. For the 6th, 7th, 8th, and 9th points that exceed the threshold of 0.10, mark them as temperature anomaly sample points. If 4 temperature anomaly sample points appear continuously but do not exceed the given number n3 (assuming n3=5), mark M3=0.

[0154] In step S2540, since M1=1, M2=1, and M3=0, "M1=1 or M2=1 or M3=1" is satisfied, it is determined that the second data exceeds the calibrated lubricating oil contamination state threshold, and step S3000 needs to be executed.

[0155] This example simulates the possible situations of conductivity, vibration spectrum abnormality and temperature gradient abnormality under actual working conditions. By judging the number of abnormal points in each time series data, marking M1, M2, and M3, and finally determining whether the calibrated lubricant contamination state threshold is exceeded based on the marked value. This multi-parameter comprehensive judgment method can improve the reliability of oil state monitoring.

[0156] In general, step S2000 integrates multi-source sensor information such as conductivity, vibration and temperature to build a multi-dimensional monitoring system for the service performance of the gear meshing surface. By collecting real-time data of the target gear, calculating and integrating the abnormality index corresponding to each physical quantity, the dynamic changes of the gear service status can be accurately grasped, and early signs of performance degradation can be discovered in time. This data-driven online monitoring mode breaks through the traditional threshold-based single fault diagnosis method and realizes the active perception of the gear service status. The monitoring results are output in the form of service performance level, which intuitively reflects the health level of the gear, making it easier for engineers to quickly understand and make decisions. At the same time, by optimizing the evaluation model parameters, the sensitivity and accuracy of monitoring can be further improved, and the risk of false alarms and missed alarms can be reduced. Applying this solution to the gear transmission system of wind power, rail transit and other equipment can significantly improve the safety and reliability of the system, maximize the service life of the equipment, and reduce the operation and maintenance costs. In addition, this solution also lays a data foundation for gear fault warning and remaining life prediction, which is of great value for realizing the reliability design and intelligent operation and maintenance of gears.

[0157] Step S3000, filtering the lubricating oil online to remove pollutants.

[0158] Specifically, in step S3000, when the monitoring data exceeds the calibrated lubricant contamination state threshold, the lubricant needs to be filtered online to remove pollutants and restore the cleanliness and lubrication performance of the lubricant. Online filtration refers to a method of directly purifying the lubricant without stopping the equipment during operation. Its main purpose is to promptly remove pollutants, moisture, oxidation products and other harmful substances in the lubricant, maintain the performance indicators of the lubricant, extend the service life of the lubricant, and ensure the normal operation of the equipment.

[0159] Online filtration systems usually consist of multi-stage precision filters, heaters, oil pumps, controllers, etc. The precision filter adopts a multi-layer structure, and the filtration accuracy gradually increases from the inside to the outside, which can effectively intercept pollutants of different sizes. For example, the commonly used β (5)1000The filter element with high precision means that the interception rate of the polluted particles with a size of 5μm or more can reach more than 99.9%. The material selection of the filter is also very important. The glass fiber filter material has a good filtering effect on solid particles, while the coalescing filter material can effectively remove moisture. The heater can heat the lubricating oil to a suitable temperature, reduce the viscosity of the oil, accelerate the flow of the oil and the separation of impurities. The oil pump provides stable circulation power to ensure the filtering efficiency. The controller can automatically adjust the working state of the filtration system according to parameters such as oil contamination, temperature, and flow.

[0160] The working principle of online filtration is to introduce a part of the lubricating oil into the filtration system through bypass circulation, and then return it to the main oil circuit after multi-stage precision filtration to achieve continuous purification of the oil. A contamination sensor is installed at the inlet of the filter to monitor the contamination status of the lubricating oil in real time. When the contamination exceeds the standard, the system automatically switches to the filtration mode. At the same time, the blockage status of the filter is also monitored in real time. When the pressure difference exceeds the set value, the controller issues an alarm to remind you to replace the filter element in time. The entire filtration process is fully automatic and does not affect the normal operation of the main equipment, which can significantly improve the operation efficiency and reliability of the equipment.

[0161] Online filtration technology has significant technical advantages and economic benefits. Compared with offline filtration, online filtration does not interrupt equipment operation, can continuously maintain the cleanliness of lubricating oil, and reduce production losses caused by downtime. At the same time, the online filtration system is equipped with a precision filter with high filtration accuracy and high purification efficiency, which can filter a large amount of lubricating oil to the ideal cleanliness within 1 hour. Compared with replacing new oil, online filtration greatly extends the service life of lubricating oil, saves oil storage and disposal costs, and has significant economic benefits. In addition, online filtration control is precise and can be adjusted in real time according to the contamination status of the lubricating oil to avoid waste of filter materials and energy consumption caused by excessive filtration.

[0162] In summary, the online filtration technology in step S3000 is an efficient, economical and environmentally friendly lubricant oil purification method. Through multi-stage precision filtration and real-time monitoring, the quality of lubricant oil can be continuously improved, and the service performance and reliability of gear transmission devices can be improved. The reasonable design and optimized control of online filtration devices are the key links to realize intelligent status monitoring and maintenance, and are of great significance to improving equipment level and economic benefits.

[0163] The present invention can be applied to large-scale heavy-duty transmission gear systems such as wind power gearboxes and industrial transmission devices, and has broad application prospects. Compared with the traditional method of regular inspection and replacement of lubricating oil, the present invention can significantly improve the timeliness and pertinence of inspection, reduce inspection and maintenance costs, and improve equipment reliability, which is of great significance to promoting the safe and economical operation of large-scale transmission devices.

[0164] Example 2

[0165] This embodiment provides a large-scale heavy-duty transmission gear meshing surface service performance sensing system based on the first embodiment. Figure 7 As shown, including:

[0166] Evaluation model building module: used to obtain first data of the target gear, and establish a gear meshing surface service performance evaluation model based on the first data; the first data includes the historical meshing surface conductivity time series, historical vibration spectrum time series and historical temperature gradient time series of the target gear under different working conditions;

[0167] Service performance evaluation module: used to collect the real-time meshing surface conductivity time series, real-time vibration spectrum time series and real-time temperature gradient time series of the target gear to generate second data; according to the second data and the gear meshing surface service performance evaluation model, the current service performance level of the target gear is obtained;

[0168] Lubricating oil contamination status determination module: used to obtain the lubricating oil contamination status threshold value based on the first data, dynamically calibrate the lubricating oil contamination status threshold value according to the current service performance level of the target gear, and obtain the calibrated lubricating oil contamination status threshold value; determine whether the second data exceeds the calibrated lubricating oil contamination status threshold value, if not, continue real-time monitoring; if exceeded, filter the lubricating oil online to remove pollutants.

[0169] In the evaluation model building module, the step of obtaining the first data of the target gear includes:

[0170] Step S1110, arranging a conductivity sensor array on the meshing surface of the target gear, collecting conductivity data of the meshing surface, and forming a historical meshing surface conductivity time series;

[0171] Step S1120, arranging an acceleration sensor on the bearing seat of the target gear to collect vibration signals during the operation of the gear, and forming a historical vibration spectrum time series through spectrum analysis;

[0172] Step S1130, by arranging an infrared temperature sensor array at different positions of the meshing surface of the target gear, collecting temperature data of the meshing surface, and calculating the temperature gradient to form a historical temperature gradient time series.

[0173] The establishment of the gear meshing surface service performance evaluation model comprises:

[0174] The statistical characteristics and morphological characteristics of the historical meshing surface conductivity time series, the historical vibration spectrum time series and the historical temperature gradient time series are extracted to form a historical feature set. With the historical feature set as input and the gear service performance grade as output, a support vector machine (SVM) regression model is trained to obtain a gear meshing surface service performance evaluation model. The gear service performance grade includes four grades: excellent, good, average and poor.

[0175] In the lubricating oil contamination state determination module, obtaining the lubricating oil contamination state threshold according to the first data includes:

[0176] Step S2310, calculating the conductivity sub-threshold according to the historical meshing surface conductivity time series;

[0177] Step S2320, calculating the vibration spectrum abnormality of each vibration sample point in the historical vibration spectrum time series, obtaining a vibration spectrum abnormality sequence, and taking the 90% quantile of the vibration spectrum abnormality sequence as the vibration spectrum abnormality sub-threshold;

[0178] Step S2330, obtaining a temperature gradient abnormality sub-threshold according to the historical temperature gradient time series.

[0179] The step S2330 includes:

[0180] Step S2331, calculating the statistical characteristics of the historical temperature gradient time series to form a temperature gradient feature set;

[0181] Step S2332, using the principal component analysis method to reduce the dimension of the temperature gradient feature set, extracting the principal component scores, and obtaining the low-dimensional features of the temperature gradient;

[0182] Step S2333, using the temperature gradient low-dimensional features as input, training the support vector data description detection model; defining the temperature gradient anomaly as the distance from the temperature sample point in the historical temperature gradient time series to the support vector data description detection model hypersphere, and obtaining the temperature gradient anomaly sequence;

[0183] Step S2334, statistically analyze the temperature gradient anomaly degree sequence, and take the 90% quantile as the temperature gradient anomaly degree sub-threshold.

[0184] In the lubricating oil contamination state determination module, the calibrated lubricating oil contamination state threshold includes a calibrated conductivity sub-threshold , calibrated vibration spectrum abnormality sub-threshold and the calibrated temperature gradient anomaly sub-threshold ;

[0185] The dynamic calibration of the lubricating oil contamination state threshold comprises:

[0186] Step S2410: If the current service performance level of the target gear is excellent, increase the three sub-thresholds according to the proportional coefficient a to obtain , and ; 1.2<a≤1.5;

[0187] Step S2420: If the current service performance level of the target gear is good, increase the three sub-thresholds according to the proportional coefficient b to obtain , and ; 1.0<b≤1.2;

[0188] Step S2430, if the current service performance level of the target gear is average, the three sub-thresholds are kept unchanged;

[0189] Step S2440: If the current service performance level of the target gear is poor, then reduce the three sub-thresholds according to the proportional coefficient c to obtain , and ;0.6≤c<1.0.

[0190] In the lubricating oil contamination state determination module, the real-time determination of whether the second data exceeds the calibrated lubricating oil contamination state threshold comprises:

[0191] Step S2510, traverse the real-time meshing surface conductivity time series, and determine whether each conductivity sample point in the real-time meshing surface conductivity time series exceeds the calibrated conductivity sub-threshold value. , if so, the calibrated conductivity sub-threshold will be exceeded The conductivity sample points are marked as conductivity exceeding limit sample points. If the conductivity exceeding limit sample points appear continuously and the number exceeds n1, then mark M1=1, otherwise, mark M1=0; n1 is the preset conductivity exceeding limit sample number threshold, and M1 is the conductivity abnormality indicator variable;

[0192] Step S2520, calculate the vibration spectrum abnormality of each vibration sample point in the real-time vibration spectrum time series to obtain a real-time vibration spectrum abnormality sequence; traverse the real-time vibration spectrum abnormality sequence to determine whether each vibration sample point in the real-time vibration spectrum abnormality sequence exceeds the calibrated vibration spectrum abnormality sub-threshold If so, the calibrated vibration spectrum abnormality threshold will be exceeded. The vibration sample points are marked as vibration abnormal sample points. If the vibration abnormal sample points appear continuously and the number exceeds n2, then mark M2=1, otherwise, mark M2=0; n2 is the preset vibration abnormal sample number threshold, and M2 is the vibration abnormality indicator variable;

[0193] Step S2530, obtaining a real-time temperature gradient anomaly sequence of the real-time temperature gradient time series, traversing the real-time temperature gradient anomaly sequence, and determining whether each temperature sample point in the real-time temperature gradient anomaly sequence exceeds the calibrated temperature gradient anomaly sub-threshold If so, the calibrated temperature gradient anomaly threshold will be exceeded. The temperature sample points are marked as temperature anomaly sample points. If the temperature anomaly sample points appear continuously and the number exceeds n3, then mark M3=1, otherwise, mark M3=0; n3 is the preset temperature anomaly sample number threshold, and M3 is the temperature gradient anomaly indicator variable;

[0194] Step S2540, if M1=1 or M2=1 or M3=1, it is determined that the second data exceeds the calibrated lubricant contamination state threshold; if M1=0 and M2=0 and M3=0, it is determined that the second data does not exceed the calibrated lubricant contamination state threshold.

[0195] The step S2530 includes:

[0196] Step S2531, calculating the statistical characteristics of the real-time temperature gradient time series to form a real-time temperature gradient feature set;

[0197] Step S2532, using a principal component analysis method to reduce the dimension of the real-time temperature gradient feature set to obtain a low-dimensional feature of the real-time temperature gradient;

[0198] Step S2533, describing the detection model according to the real-time temperature gradient low-dimensional features and support vector data, and obtaining the real-time temperature gradient anomaly degree sequence.

[0199] Example 3

[0200] This embodiment discloses an electronic device, which may include one or more processors and one or more memories. The memories store computer readable codes, which, when executed by the one or more processors, may execute the above-mentioned method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear.

[0201] The method or system according to the embodiment of the present application can also be implemented with the help of the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, can store a large-scale heavy-duty transmission gear meshing surface service performance sensing method provided in the present application. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear may, for example, include: obtaining first data of a target gear, and establishing a gear meshing surface service performance evaluation model based on the first data; the first data includes historical meshing surface conductivity time series, historical vibration spectrum time series, and historical temperature gradient time series of the target gear under different working conditions; collecting real-time meshing surface conductivity time series, real-time vibration spectrum time series, and real-time temperature gradient time series of the target gear to generate second data; obtaining the current service performance level of the target gear based on the second data and the gear meshing surface service performance evaluation model; obtaining a lubricating oil contamination state threshold based on the first data; dynamically calibrating the lubricating oil contamination state threshold based on the current service performance level of the target gear to obtain a calibrated lubricating oil contamination state threshold; judging whether the second data exceeds the calibrated lubricating oil contamination state threshold, and if not, continuing real-time monitoring; if exceeded, filtering the lubricating oil online to remove pollutants.

[0202] Furthermore, the electronic device may also include a user interface. Of course, the architecture disclosed in the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0203] Example 4

[0204] This embodiment discloses a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, a large-scale heavy-duty transmission gear meshing surface service performance sensing method of the embodiment of the present application can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0205] In addition, according to the implementation of the present application, the process described in the above reference flow chart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: obtaining first data of a target gear, and establishing a gear meshing surface service performance evaluation model according to the first data; the first data includes a historical meshing surface conductivity time series, a historical vibration spectrum time series, and a historical temperature gradient time series of the target gear under different working conditions; collecting the real-time meshing surface conductivity time series, the real-time vibration spectrum time series, and the real-time temperature gradient time series of the target gear to generate second data; obtaining the current service performance level of the target gear according to the second data and the gear meshing surface service performance evaluation model; obtaining a lubricant contamination state threshold according to the first data; dynamically calibrating the lubricant contamination state threshold according to the current service performance level of the target gear to obtain a calibrated lubricant contamination state threshold; judging whether the second data exceeds the calibrated lubricant contamination state threshold, if not, continuing real-time monitoring; if exceeded, filtering the lubricant online to remove pollutants. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are performed.

[0206] The methods, systems, and devices of the present application may be implemented in many ways. For example, the methods, systems, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0207] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0208] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear, characterized in that: The method comprises: Acquire first data of the target gear, and establish a gear meshing surface service performance evaluation model based on the first data; the first data includes a historical meshing surface conductivity time series, a historical vibration spectrum time series, and a historical temperature gradient time series of the target gear under different working conditions; The real-time meshing surface conductivity time series, real-time vibration spectrum time series and real-time temperature gradient time series of the target gear are collected to generate second data; the current service performance level of the target gear is obtained according to the second data and the gear meshing surface service performance evaluation model; the lubricating oil contamination state threshold is obtained according to the first data; the lubricating oil contamination state threshold is dynamically calibrated according to the current service performance level of the target gear to obtain the calibrated lubricating oil contamination state threshold; it is determined whether the second data exceeds the calibrated lubricating oil contamination state threshold, if not, real-time monitoring is continued; if exceeded, the lubricating oil is filtered online to remove pollutants.

2. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 1, characterized in that: The establishment of the gear meshing surface service performance evaluation model comprises: The statistical characteristics and morphological characteristics of the historical meshing surface conductivity time series, the historical vibration spectrum time series and the historical temperature gradient time series are extracted to form a historical feature set. With the historical feature set as input and the gear service performance grade as output, a support vector machine (SVM) regression model is trained to obtain a gear meshing surface service performance evaluation model. The gear service performance grade includes four grades: excellent, good, average and poor.

3. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 1, characterized in that: The method of obtaining the current service performance level of the target gear includes: extracting statistical features and morphological features of the second data to generate a real-time feature set; and obtaining the current service performance level of the target gear based on the real-time feature set and a gear meshing surface service performance evaluation model.

4. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 2, characterized in that: The lubricating oil contamination state threshold includes three sub-thresholds, and the three sub-thresholds are conductivity sub-thresholds. , vibration spectrum abnormality sub-threshold and temperature gradient anomaly sub-threshold ; The step of obtaining the lubricating oil contamination state threshold comprises: Calculate the conductivity sub-threshold according to the historical meshing surface conductivity time series ; Calculate the vibration spectrum anomaly of each vibration sample point in the historical vibration spectrum time series, obtain the vibration spectrum anomaly sequence, and take the 90% quantile of the vibration spectrum anomaly sequence as the vibration spectrum anomaly sub-threshold ; According to the historical temperature gradient time series, the temperature gradient anomaly sub-threshold is obtained .

5. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 4, characterized in that: The step of obtaining the temperature gradient anomaly degree sub-threshold according to the historical temperature gradient time series includes: Calculate the statistical characteristics of the historical temperature gradient time series to form a temperature gradient feature set; The principal component analysis method is used to reduce the dimension of the temperature gradient feature set, extract the principal component scores, and obtain the low-dimensional features of the temperature gradient; The support vector data description detection model is trained with the low-dimensional features of temperature gradient as input; the temperature gradient anomaly is defined as the distance from the sample point in the historical temperature gradient time series to the hypersphere of the support vector data description detection model, and the temperature gradient anomaly sequence is obtained; The temperature gradient anomaly series was statistically analyzed, and the 90% quantile was taken as the temperature gradient anomaly sub-threshold.

6. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 5, characterized in that: The calibrated lubricant contamination state threshold includes a calibrated conductivity sub-threshold , calibrated vibration spectrum abnormality sub-threshold and the calibrated temperature gradient anomaly sub-threshold ; The dynamic calibration of the lubricating oil contamination state threshold according to the current service performance level of the target gear includes: If the current service performance level of the target gear is excellent, the three sub-thresholds are increased according to the proportional coefficient a to obtain , and ; If the current service performance level of the target gear is good, the three sub-thresholds are increased according to the proportional coefficient b to obtain , and ; If the current service performance level of the target gear is average, the three sub-thresholds remain unchanged; If the current service performance level of the target gear is poor, the three sub-thresholds are reduced by the proportional coefficient c to obtain , and ; Among them, a>b>1.0>c.

7. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 6, characterized in that: The determining whether the second data exceeds the calibrated lubricating oil contamination state threshold comprises: Traverse the real-time meshing surface conductivity time series to determine whether each conductivity sample point in the real-time meshing surface conductivity time series exceeds the calibrated conductivity sub-threshold , if so, the calibrated conductivity sub-threshold will be exceeded The conductivity sample points are marked as conductivity exceeding limit sample points. If the conductivity exceeding limit sample points appear continuously and the number exceeds n1, then mark M1=1, otherwise, mark M1=0; n1 is the preset conductivity exceeding limit sample number threshold, and M1 is the conductivity abnormality indicator variable; Calculate the vibration spectrum anomaly of each vibration sample point in the real-time vibration spectrum time series to obtain a real-time vibration spectrum anomaly sequence; traverse the real-time vibration spectrum anomaly sequence to determine whether each vibration sample point in the real-time vibration spectrum anomaly sequence exceeds the calibrated vibration spectrum anomaly sub-threshold If so, the calibrated vibration spectrum abnormality threshold will be exceeded. The vibration sample points are marked as vibration abnormal sample points. If the vibration abnormal sample points appear continuously and the number exceeds n2, then mark M2=1, if not, then mark M2=0; n2 is the preset vibration abnormal sample number threshold, and M2 is the vibration abnormality indicator variable.

8. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 7, characterized in that: The determining whether the second data exceeds the calibrated lubricating oil contamination state threshold further includes: Obtain the real-time temperature gradient anomaly sequence of the real-time temperature gradient time series, traverse the real-time temperature gradient anomaly sequence, and determine whether each temperature sample point in the real-time temperature gradient anomaly sequence exceeds the calibrated temperature gradient anomaly sub-threshold If so, the calibrated temperature gradient anomaly threshold will be exceeded. The temperature sample points are marked as temperature anomaly sample points. If the temperature anomaly sample points appear continuously and the number exceeds n3, then mark M3=1, otherwise, mark M3=0; n3 is the preset temperature anomaly sample number threshold, and M3 is the temperature gradient anomaly indicator variable; If M1=1 or M2=1 or M3=1, it is determined that the second data exceeds the calibrated lubricant contamination state threshold; if M1=0 and M2=0 and M3=0, it is determined that the second data does not exceed the calibrated lubricant contamination state threshold.

9. A method for sensing the service performance of the meshing surface of a large and heavy-duty transmission gear according to claim 8, characterized in that: The real-time temperature gradient anomaly degree sequence for obtaining the real-time temperature gradient time series includes: Calculate the statistical characteristics of the real-time temperature gradient time series to form a real-time temperature gradient feature set; The principal component analysis method is used to reduce the dimension of the real-time temperature gradient feature set to obtain the low-dimensional features of the real-time temperature gradient; The detection model is described based on the low-dimensional features of the real-time temperature gradient and the support vector data, and the real-time temperature gradient anomaly degree sequence is obtained.

10. A large-scale heavy-duty transmission gear meshing surface service performance sensing system, which is used to implement a large-scale heavy-duty transmission gear meshing surface service performance sensing method according to any one of claims 1 to 9, characterized in that: The system comprises: Evaluation model building module: used to obtain first data of the target gear, and establish a gear meshing surface service performance evaluation model based on the first data; the first data includes the historical meshing surface conductivity time series, historical vibration spectrum time series and historical temperature gradient time series of the target gear under different working conditions; Service performance evaluation module: used to collect the real-time meshing surface conductivity time series, real-time vibration spectrum time series and real-time temperature gradient time series of the target gear to generate second data; according to the second data and the gear meshing surface service performance evaluation model, the current service performance level of the target gear is obtained; Lubricating oil contamination status determination module: used to obtain the lubricating oil contamination status threshold value based on the first data, dynamically calibrate the lubricating oil contamination status threshold value according to the current service performance level of the target gear, and obtain the calibrated lubricating oil contamination status threshold value; determine whether the second data exceeds the calibrated lubricating oil contamination status threshold value, if not, continue real-time monitoring; if exceeded, filter the lubricating oil online to remove pollutants.

Images