System-oriented bearing fault prediction method applying noise reduction auto-encoder

Abstract

The invention relates to a system-oriented bearing fault prediction method applying a noise reduction auto-encoder, belongs to the technical field of bearing fault prediction, and solves the problem that the influence of complex system noise on bearing fault prediction is not considered in the prior art. Determining a dangerous bearing as a target bearing; performing dynamic simulation on the to-be-tested transmission chain, and obtaining simulation result data as simulation noise; carrying out a part-level degradation test, and obtaining a vibration signal in a degradation process as test data; performing signal mixing on the test data and the simulation result data to obtain a noisy signal; constructing and training a noise reduction auto-encoder model, and outputting denoised bearing vibration data; and constructing a fault prediction model, dividing the degradation process into a healthy stage, a slow degradation stage and a rapid degradation stage, and respectively inputting the denoised bearing vibration data into the three independent models to obtain a predicted value of the highest-precision model in the three independent models.

Description

Technical Field

[0001] The present invention relates to the technical field of bearing fault prediction, and in particular to a system-oriented bearing fault prediction method using a noise reduction autoencoder. Background Art

[0002] Bearings are important components in all kinds of rotating machinery. They play important roles such as supporting, positioning and reducing friction. They bear a variety of cyclic alternating loads. Their health status plays a decisive role in the stable operation of the system. In actual application scenarios, bearings are also a common vulnerable part that has received much attention. If their faults are not identified in a timely and effective manner, they may bring huge economic and personnel safety risks. Therefore, in order to ensure the normal operation and safety of equipment and reduce the cost of use and maintenance, the fault prediction of rolling bearings has theoretical and practical significance.

[0003] During the operation of the bearing, the corresponding vibration signal will be generated due to different fault states. Its frequency domain characteristics and energy level can be used as the basis for fault identification and prediction. Therefore, obtaining its vibration signal in the test and actual work, combined with the fault prediction model, can effectively control the degree of bearing damage and risk.

[0004] With the development of computers and artificial intelligence, fault prediction of industrial machinery can be carried out through industrial information and data processing, and fault prediction analysis can be provided. The health status monitoring and prediction of bearings are gradually shifting from theoretical models to data-driven, and the corresponding preventive maintenance is also gradually turning into predictive maintenance. The development of intelligent operation and maintenance can effectively improve the stability and reliability of equipment operation.

[0005] In the prior art, the entire life cycle of a bearing is divided into stages to distinguish its fault level. Usually, two-stage or multi-stage division is performed. Both methods require the determination of the health stage, and the basis for multi-stage division is the degree of bearing failure, which is reflected in the statistical characteristics of vibration signals or health indicators. The corresponding threshold and probability density are determined according to the actual requirements of the equipment. When the confidence level exceeds the threshold, it is considered that a fault has occurred or a more serious fault is about to occur, which is a general fault prediction method.

[0006] The Chinese patent application, publication number CN112632466A, published on April 9, 2021, is titled "A bearing fault prediction method based on principal component analysis and deep bidirectional long short-term memory network", which discloses a method based on a deep bidirectional long short-term memory network, through which the input data that has been processed and standardized by this network model is obtained and measured. However, this method only considers the vibration signal characteristics of the bearing itself, and the situation of complex system noise in practical applications is not taken into consideration.

[0007] Therefore, an improved technical solution is needed to achieve the task of bearing fault prediction in complex systems taking into account the complex characteristics of noise. Summary of the invention

[0008] In order to make up for the fact that the existing fault prediction technology only focuses on the bearing itself but does not consider the system application and has insufficient coverage of unhealthy conditions, an embodiment of the present invention takes into account the complex characteristics of the system's transmission chain noise and provides a system-oriented bearing fault prediction method using a denoising autoencoder. The method is based on a denoising autoencoder with a gated recurrent unit and an attention mechanism, and extracts the target bearing health indicator from the noisy signal health indicator to input the proposed fault prediction model, and then realizes the degradation stage discrimination and degradation degree quantification covering the entire life cycle, so as to realize system-oriented fault prediction based on component-level degradation test data.

[0009] To achieve the above object, according to an embodiment of the present invention, a system-oriented bearing fault prediction method using a noise reduction autoencoder is provided, comprising the following steps:

[0010] Step S1, obtaining parameters and working condition information of a transmission chain to be tested, and performing bearing life verification on bearings in the transmission chain to be tested, and determining dangerous bearings as target bearings, wherein the transmission chain to be tested is a system including bearings;

[0011] Step S2, performing dynamic simulation on the transmission chain to be tested, and obtaining simulation result data as simulated noise;

[0012] Step S3, performing a component-level degradation test on the target bearing, and acquiring a vibration signal during the degradation process as test data;

[0013] Step S4, performing signal mixing on the obtained test data and simulation result data to obtain a noisy signal, and respectively establishing a bearing degradation process health index of the target bearing based on the test data and a bearing degradation process health index based on the noisy signal;

[0014] Step S5, constructing a denoising autoencoder model, using the bearing degradation process health index based on the test data and the bearing degradation process health index based on the noisy signal to train the denoising autoencoder model, making full use of the test data to accurately extract the target prediction amount in the health index of the noisy signal, obtaining the target bearing health index extracted by denoising, and outputting it as denoised bearing vibration data;

[0015] Step S6, construct a fault prediction model, use the envelope difference method to divide the health status of the bearing, divide the degradation process into a healthy stage, a slow degradation stage and a rapid degradation stage, train three independent models for data in different stages, and the three independent models constitute a fault prediction model. The denoised bearing vibration data are input into the three independent models to generate three sets of prediction values ​​and evaluation indicators. The difference in evaluation indicators is used to determine the current degradation stage and output the prediction value of the highest accuracy model among the three independent models to achieve fault prediction.

[0016] Optionally, the step S1 specifically includes:

[0017] Step S1.1, obtaining gear parameters, bearing characteristic parameters and design operating condition information of the transmission chain to be tested;

[0018] Step S1.2, perform gear stress calibration, calculate the meshing force according to the design working condition information and the gear parameters of the transmission chain, and then obtain the bearing radial load as the equivalent dynamic load, and use this as a basis to perform bearing life calibration, and obtain the bearing life of each bearing as the calibration result, and determine the temperature coefficient g by looking up the table T , Hardness coefficient g H , reliability life correction factor a1, life correction factor a2 and operating condition factor a3, the bearing life is

[0019]

[0020] Where n is the operating speed, P is the equivalent dynamic load, C is the basic rated dynamic load, and ε is the life index;

[0021] Step S1.3, determine the dangerous bearings based on comprehensive comparison of the verification results.

[0022] Optionally, the step S2 specifically includes:

[0023] Step S2.1, using a lumped parameter method to integrate the shaft mass of the transmission shaft in the transmission chain to be tested into the gear, and obtaining a gear force element including the shaft mass;

[0024] Step S2.2, the system dynamics equation of the transmission chain to be tested is:

[0025]

[0026] Where M is the total mass of the system, C is the damping of the system, K is the stiffness matrix of the system, δ is the displacement of the system, and f is the external force vector of the system. The external force vector of the system is the input calculated according to the working condition, the stiffness matrix of the system is calculated according to the setting, and the displacement of the system is the result calculated according to the setting.

[0027] The damping of the system is time-varying mesh damping, expressed as:

[0028]

[0029] Among them, ε g is the gear meshing damping ratio of the system, k(t) represents the time-varying meshing stiffness of the gears of the system, t is the time, I represents the moment of inertia of the gears of the system and the subscripts 1 and 2 represent different gears respectively, r b represents the gear pitch radius of the system;

[0030] Step S2.3, establish a system dynamics model in the simulation software, consider the time-varying meshing damping and stiffness in S2.2, complete the system dynamics calculation and extract the vibration response of the target bearing position as the simulation result data, that is, as the simulated noise.

[0031] Optionally, the step S3 specifically includes:

[0032] Step S3.1, selecting a representative working condition of the transmission chain to be tested from the acquired design working condition information, and converting it into a bearing test load;

[0033] Step S3.2, calculating the single-point fault characteristic frequency of the target bearing in the transmission chain to be tested as a basis for selecting the vibration sensor;

[0034] Step S3.3, select a healthy bearing with the same model as the target bearing in the transmission chain to be tested as a test bearing, and select a constant test condition from the design condition information obtained in step S1 for the test bearing to operate until failure; in this process, the vibration signal of the test bearing is collected by the sensor according to a fixed single sampling time and sampling interval, and the data of the vibration signal is merged to obtain a vibration signal data set for the entire life cycle as test data.

[0035] Optionally, the step S4 specifically includes:

[0036] Step S4.1, using cubic spline interpolation method to match the sampling rate of simulation result data and test data to obtain corrected noise, and copying the corrected noise signal to the same length as the test data to obtain an equal-length noise signal;

[0037] Step S4.2, mixing the equal-length noise signal obtained in the above steps and the test vibration signal in the time domain to obtain a noisy signal;

[0038] Step S4.3, extracting health indicators from the test data and the noisy signal respectively, and evaluating the monotonicity and robustness of the extracted health indicators to obtain the health indicators with the best monotonicity and robustness;

[0039] Step S4.4, using the obtained health index with the best monotonicity and robustness as an indirect prediction quantity, the health index is calculated for the noisy signal and the test data respectively, to obtain the health index of the bearing degradation process based on the noisy signal and the health index of the bearing degradation process based on the test data.

[0040] Optionally, the step S4.3 specifically includes:

[0041] The health index extraction includes selecting the root mean square value, kurtosis, peak value and skewness in the test data and the noisy signal as candidate health indicators;

[0042] The monotonicity and robustness of candidate health indicators are evaluated. For a dynamic sliding window, the input sequence X = (x1, x2, ..., x i ,…,x N ), the RMS value is:

[0043]

[0044] Kurtosis is expressed as:

[0045]

[0046] The peak value is expressed as:

[0047] Peak Value = max(X)

[0048] Skewness is expressed as:

[0049]

[0050] Where N is the sample size, μ is the mean of the distribution, σ is the standard deviation of the distribution, E is the expected value of the distribution, and x i is the input sample, i is the sample number;

[0051] The monotonicity calculation method uses the Spearman correlation coefficient, which is expressed as:

[0052]

[0053] Where ρ represents the Spearman correlation coefficient, which is used as a monotonicity indicator;

[0054] The robustness is calculated as the average difference between the upper and lower envelopes, expressed as:

[0055]

[0056] Among them, u env represents the upper envelope calculation, l env represents the lower envelope calculation, x(t) represents the sample point corresponding to the local maximum or minimum value, t represents the index of the local maximum or minimum value, and rob represents the robustness index;

[0057] The monotonicity and robustness of the candidate health indicators are evaluated by the above monotonicity and robustness expressions to obtain the health indicator with the best monotonicity and robustness.

[0058] Optionally, the step S5 specifically includes:

[0059] Step S5.1, obtaining a complete set of health indicators of the bearing degradation process based on test data and health indicators of the bearing degradation process based on noisy signals, performing training, verification and test set division, and performing tensor quantization to obtain an input sequence;

[0060] Step S5.2, construct the gated recurrent unit of the main structure of the denoising autoencoder model. For the input sequence X = (x1, x2, ..., x N ), the gated recurrent unit is expressed as:

[0061] r i =σ(W r ·[h i-1 ,x i ]+b z )

[0062] z i =σ(W z ·[h i-1 ,x i ]+b r )

[0063]

[0064]

[0065] Among them, r i is the reset gate output, z i is the update gate output, is the candidate hidden state, h i is the final hidden state, x i is the input sample, i represents the current input sequence number, σ is the sigmoid activation function, ⊙ represents element-wise multiplication, W r , W z and W are the weight matrices of the reset gate, update gate and candidate hidden state, respectively. r ,b z and b are the reset gate, update gate and bias of candidate hidden state respectively;

[0066] Step S5.3, based on the gated recurrent unit, build a multi-head attention mechanism of the main structure of the denoising autoencoder model to obtain the main structure of the denoising autoencoder model. For a given query vector Q, a given key vector Key and a given value vector V, and the corresponding weight matrix The multi-head attention mechanism can be expressed as:

[0067]

[0068] MultiHed=Concat(head1,head2,…,head j …,head h )W O

[0069]

[0070] Among them, d k is the dimension of the key vector Key, W O is the weight matrix used to update the input, softmax represents the normalization calculation function, Concat represents the vector merging calculation function, h represents the total number of attention heads, and head J represents the jth attention head, j represents the sequence number of the attention head, Attention represents the self-attention mechanism, and MultiHead represents multi-head attention;

[0071] Step S5.4, adding a neuron parameter release process and a linear layer to the main structure of the denoising autoencoder model built in step S5.3, and outputting the model prediction value;

[0072] Step S5.5, based on the obtained model prediction value, the denoising autoencoder model constructed in steps S5.2 to S5.4 is trained by using the MSE loss function and the adam optimization algorithm through the gradient descent method, so that the model parameters converge, and the changes of the MSE loss function are monitored in the process and the optimal model is output;

[0073] Step S5.6, divide the obtained bearing degradation process health index based on the noisy signal according to the given model sequence length, and input it into the denoising autoencoder model to obtain the target bearing health index extracted by denoising, and output it as denoised bearing vibration data.

[0074] Optionally, the step S6 specifically includes:

[0075] Step S6.1, according to the health index extracted from the test data of the test bearing, the envelope difference method is used to analyze its trend to obtain the envelope difference result;

[0076] Step S6.2, dividing the degradation process of the test bearing obtained in step S3 into stages, performing boundary identification based on the obtained envelope difference result, and dividing the degradation process into a healthy stage, a slow degradation stage, and a fast degradation stage;

[0077] Step S6.3, splitting the full life cycle test data of the test bearing obtained in step S3 into three data sets according to the above three stages, training three independent models respectively, so that each independent model is fully overfitted on the corresponding data set, and the three trained independent models are used as fault prediction models;

[0078] Step S6.4, input the denoised bearing vibration data of the target bearing in the transmission chain to be tested into the three trained independent models, output the prediction values ​​and evaluation indicators respectively, compare the evaluation indicators output by the three independent models, determine the model with the highest prediction accuracy, obtain the prediction value of the model with the highest accuracy, and then identify the health stage of the prediction value.

[0079] Optionally, the overall application method of system-oriented bearing fault prediction is: obtaining bearing component-level degradation test data and simulated system noise data, converting them into health indicators and dividing them into degradation stages, and then training denoising autoencoders and fault prediction models. For a set of real inputs in the application process, the denoising autoencoder first extracts the target bearing health indicators as the input of the fault prediction model, and then the fault prediction model uses this data to achieve qualitative identification of the degradation stage and quantitative identification of the degradation degree, thereby achieving system-oriented transmission chain bearing fault prediction based on component degradation tests.

[0080] Compared with the prior art, an embodiment of the present invention provides a system-oriented bearing fault prediction method using a noise reduction autoencoder, which has at least the following advantages:

[0081] 1. Dynamic simulation is used to simulate the system noise in the actual application of the transmission chain, effectively avoiding the high cost of system-level degradation tests, long test cycles, complex and demanding test equipment, and the randomness of complex systems and the unpredictability caused by the resulting joint degradation of multiple components, effectively controlling the difficulty of signal processing and improving the feasibility of target component processing. At the same time, the application of component-level degradation test data ensures that the target signal variables are controllable and the signal-to-noise ratio is high to the greatest extent, fully compressing the test cost, difficulty, cycle and danger, and thus being able to propose noise suppression algorithms in a targeted and well-founded manner.

[0082] 2. The present invention achieves the suppression of complex noise in noisy signals through a denoising autoencoder including a gated recurrent unit and an attention mechanism. The processing effect of such noise with a frequency band highly overlapping with the target signal, high energy density and non-stationary is significantly better than that of traditional filtering and wavelet threshold denoising methods, and can extract the health indicators of the target bearing from noisy signals with extremely low signal-to-noise ratio.

[0083] 3. The fault prediction model in the present invention includes the ability to process vibration signals of the entire life cycle of the bearing. Compared with the fault prediction method in the existing invention that uses statistical methods to identify whether the bearing deviates from a healthy state, it can further determine the current degradation stage and degree of degradation of the bearing, achieving more complete coverage of the degradation process, and therefore has a wider range of application scenarios.

[0084] The present invention uses a noise reduction autoencoder to implement a system-oriented fault prediction method, which has more practical significance than the existing invention that only considers the fault prediction of the bearing itself; in specific application scenarios, it is obviously not easy to equip each bearing with a high-precision sensor installed close to it, and the noise of the actual transmission chain is still not negligible in this case. At the same time, the simulated noise generated by dynamic simulation can better represent the actual situation in the system application than the white noise used in the existing invention, and has more realistic noise characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] 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 will be briefly introduced below. The features and advantages of the present invention can be more clearly understood by referring to the drawings. The drawings are schematic and should not be understood as limiting the present invention in any way. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0086] Figure 1 A flowchart of a system-oriented bearing fault prediction method using a denoising autoencoder is provided according to an embodiment of the present invention.

[0087] Figure 2 A schematic diagram of an implementation of a system-oriented bearing fault prediction method using a denoising autoencoder provided according to an implementation of the present invention.

[0088] Figure 3 A schematic diagram of a process of obtaining component-level test data in an example of a system-oriented bearing fault prediction method using a denoising autoencoder provided in an embodiment of the present invention.

[0089] Figure 4 A schematic diagram of a process of obtaining simulated noise in an example of a system-oriented bearing fault prediction method using a denoising autoencoder provided in an embodiment of the present invention.

[0090] Figure 5 A schematic diagram of constructing a denoising autoencoder model in an example of a system-oriented bearing fault prediction method using a denoising autoencoder provided in an embodiment of the present invention.

[0091] Figure 6A schematic diagram of constructing a fault diagnosis model in an example of a system-oriented bearing fault prediction method using a denoising autoencoder provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0092] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0093] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0094] A system-oriented bearing fault prediction method using a noise reduction autoencoder provided according to an embodiment of the present invention is described in detail below with reference to the accompanying drawings.

[0095] Explanation of symbols:

[0096] g T Temperature coefficient of bearings;

[0097] g H The hardness coefficient of the bearing;

[0098] a1 Reliability life correction factor of bearing;

[0099] a2 bearing life correction factor;

[0100] a3 bearing operating condition coefficient;

[0101] L bearing life;

[0102] nThe operating speed of the bearing;

[0103] P equivalent dynamic load of the bearing;

[0104] C basic dynamic load rating of bearings;

[0105] ε is the bearing life index;

[0106] M Overall system quality;

[0107] C system damping;

[0108] K is the stiffness matrix of the system;

[0109] δ displacement of the system;

[0110] f The external force vector of the system;

[0111] εg The gear mesh damping ratio of the system;

[0112] k(t) is the time-varying mesh stiffness of the gears in the system;

[0113] Time t;

[0114] I The moment of inertia of the gears in the system;

[0115] r b The gear pitch radius of the system;

[0116] N sample size;

[0117] μ distribution mean;

[0118] σ is the standard deviation of the distribution;

[0119] E distribution expectation;

[0120] x i Input sample;

[0121] i Sample serial number;

[0122] X gives the input sequence;

[0123] ρSpearman correlation coefficient;

[0124] Rob robustness index;

[0125] u env Upper envelope calculation;

[0126] l env Lower envelope calculation;

[0127] x(t) represents the sample point corresponding to the local maximum or minimum value;

[0128] tThe local maximum or minimum index;

[0129] r i Reset gate output;

[0130] z i Update gate output;

[0131] Candidate hidden states;

[0132] h i Final hidden state;

[0133] σsigmoid activation function (S-shaped growth curve activation function);

[0134] W r are the weight matrices of the reset gates respectively;

[0135] W zUpdate the gate weight matrix;

[0136] W is the weight matrix of candidate hidden states;

[0137] b r Reset the gate bias;

[0138] b z Update gate bias;

[0139] b Bias of candidate hidden states;

[0140] Q is a given query vector;

[0141] Key gives the key vector;

[0142] V given value vector;

[0143] The weight matrix of the query vector;

[0144] Weight matrix for a given key vector;

[0145] weight matrix of value vector;

[0146] d k Indicates the dimension of the key vector Key;

[0147] W O The weight matrix used to update the input;

[0148] Softmax normalization calculation function;

[0149] Concat vector merging calculation function;

[0150] head h The hth attention head;

[0151] h the total number of attention heads;

[0152] head j The jth attention head;

[0153] j Attention head number;

[0154] Attention self-attention mechanism;

[0155] MultiHead multi-head attention;

[0156] y i Actual data for predicting step positions;

[0157] Model output value for the predicted step location.

[0158] refer to Figure 1-Figure 6 According to an embodiment of the present invention, a system-oriented bearing fault prediction method using a noise reduction autoencoder is provided, which takes into account the complex characteristics of the noise of the bearing transmission chain and the system and the impact on the fault, and includes the following steps.

[0159] Step S1. Obtain parameters and working condition information of the transmission chain to be tested, perform bearing life verification on the bearings in the transmission chain to be tested, and determine dangerous bearings as target bearings, wherein the transmission chain to be tested is a system including bearings. Dangerous bearings refer to bearings that are most prone to failure in the entire transmission chain system to be tested.

[0160] Comparison Figure 1 In this embodiment, the bearing life verification method in the transmission chain to be tested includes the following steps.

[0161] Step S1.1, obtaining the gear parameters, bearing characteristic parameters and design working condition information of the transmission chain to be tested. Among them, the gear parameters of the transmission chain may include module, number of teeth, tooth width, etc.; the bearing characteristic parameters may include inner and outer diameters, number of sticks and pitch circle radius, etc.; the design working condition information may include torque and speed. The above parameters can be obtained by presetting. The gear parameters and bearing characteristic parameters are shape and position parameters, which can be obtained from the three-dimensional model of the transmission chain design, and the design working condition information can be obtained from the transmission chain design.

[0162] Step S1.2, on this basis, the gear stress is checked, the meshing force is calculated according to the design working condition information and the gear parameters of the transmission chain, and the radial load of the bearing is obtained as the equivalent dynamic load, and the bearing life is checked based on this, and the bearing life of each bearing is obtained as the check result. The temperature coefficient g of the bearing can be determined by looking up the table T , hardness coefficient g H , reliability life correction factor a1, life correction factor a2 and operating condition factor a3, the bearing life can be expressed as

[0163]

[0164] Among them, n is the operating speed of the bearing, P is the equivalent dynamic load of the bearing, C is the basic rated dynamic load of the bearing, and ε is the life index of the bearing. The expected test time can be controlled by adjusting the equivalent dynamic load as a check on the bearing life.

[0165] Step S1.3, determine the dangerous bearing as the target bearing based on the comprehensive comparison of the verification results. In addition, determine the sensor installation position close to the dangerous bearing based on the preset three-dimensional digital model geometric features of the transmission chain casing. The sensor installation position is used as the spatial position basis for system simulation noise acquisition and the basis for achieving system-level fault prediction.

[0166] Step S2: Perform dynamic simulation on the transmission chain to be tested, and obtain simulation result data as simulated noise.

[0167] Comparison Figure 4 , the transmission chain dynamics simulation method in this embodiment includes the following steps.

[0168] Step S2.1, the lumped parameter method is used to integrate the shaft mass of the transmission shaft in the transmission chain to be tested into the gear, and the gear force element including the shaft mass is obtained. According to the requirements of multi-body dynamics simulation, the following three assumptions are set: first, the flexibility of the gear is ignored; second, only the rotation and translational degrees of freedom along the contact path are considered; third, the gear meshing is the main source of noise, so the support bearing is regarded as a linear spring damping unit.

[0169] Step S2.2, the system dynamics equation of the transmission chain to be tested can be expressed as:

[0170]

[0171] Where M is the total mass of the system, C is the damping of the system, K is the stiffness matrix of the system, δ is the displacement of the system, and f is the external force vector of the system. The external force vector of the system is the input calculated according to the working condition, the stiffness matrix of the system is calculated according to the setting, and the displacement of the system is the result calculated according to the setting.

[0172] The damping of the system can be time-varying meshing damping, which can be expressed as:

[0173]

[0174] Among them, ε g is the gear meshing damping ratio of the system, ranging from 0.03 to 0.17, k(t) represents the time-varying meshing stiffness of the gears of the system, t is the time, I represents the moment of inertia of the gears of the system, and the subscripts 1 and 2 represent different gears, respectively, r b Represents the gear pitch circle radius of the system.

[0175] Step S2.3, establish a system dynamics model in the simulation software, consider the time-varying meshing damping and stiffness in S2.2, complete the system dynamics calculation and extract the vibration response of the target bearing position (ie, simulation result data) as simulated noise.

[0176] Step S3: Perform a component-level degradation test on the target bearing determined in step S1, and obtain a vibration signal of the test bearing during the degradation process as test data. The test data is component-level test data.

[0177] Comparison Figure 3 , the degradation test signal acquisition method in this embodiment includes the following steps.

[0178] Step S3.1, selecting a representative working condition of the transmission chain to be tested from the acquired design working condition information, and converting it into a bearing test load.

[0179] Step S3.2, calculate the single-point fault characteristic frequency of the target bearing in the transmission chain to be tested as the basis for selecting the vibration sensor. According to experience, the faults that occur in degradation tests are usually multiple faults, and there are usually characteristic peaks at 2-4 times the frequency of the single-point fault frequency. Therefore, according to the Nyquist sampling theorem, the sensor sampling frequency should be at least 8 times the frequency of the single-point fault. The basis for selecting the vibration sensor can also include the operating temperature and range.

[0180] Step S3.3, according to the test conditions, the tester is set based on the requirements of the above steps, and a healthy bearing of the same model as the target bearing in the transmission chain to be tested is selected as the test bearing, and the test bearing is operated until failure by selecting a constant test condition from the design condition information obtained in step S1; in this process, the vibration signal of the test bearing is collected by the sensor according to a fixed single sampling time and sampling interval, and the data of the vibration signal is merged to obtain a vibration signal data set in the degradation process (i.e., the whole life cycle) as the test data. Among them, the tester may include a drive motor, a loading system and a sensor.

[0181] Step S4, signal mixing is performed on the obtained test data and simulation result data to obtain a noisy signal, and a bearing degradation process health index based on the test data and a bearing degradation process health index based on the noisy signal are respectively established for the target bearing. The established bearing degradation process health index based on the test data and the noisy signal is used as the input of the noise reduction autoencoder model in the subsequent step S5.

[0182] Comparison Figure 2 ,The signal mixing and health index establishment method in this embodiment includes the following.

[0183] Step S4.1, use cubic spline interpolation to match the sampling rate of simulation result data and test data to obtain corrected noise, copy the corrected noise signal to the same length as the test data, and obtain an equal-length noise signal, based on which noise coverage can be achieved throughout the life cycle of the bearing. Due to the limitation of computing performance, simulated noise often differs from the measured signal of the test data in terms of sampling rate and signal length. The above processing method can solve the impact of this difference and is conducive to obtaining more accurate results.

[0184] Step S4.2, mixing the equal-length noise signal obtained in the above steps and the test vibration signal in the time domain to obtain a noisy signal.

[0185] Step S4.3, extract health indicators from the test data and the noisy signal respectively, and evaluate the monotonicity and robustness of the extracted health indicators. This step can solve the problems of excessive signal fluctuation and unclear degradation characteristics in direct fault prediction.

[0186] The health index extraction includes selecting the root mean square value, kurtosis, peak value and skewness in the test data and the noisy signal as candidate health indicators.

[0187] Next, the candidate health indicators are evaluated for monotonicity and robustness. For a dynamic sliding window, given input sequence X = (x1, x2, ..., x i , … ,x N ), the RMS value can be expressed as:

[0188]

[0189] The kurtosis can be expressed as:

[0190]

[0191] The peak value can be expressed as:

[0192] Peak Value = max(X)

[0193] Skewness can be expressed as:

[0194]

[0195] Where N is the sample size, μ is the mean of the distribution, σ is the standard deviation of the distribution, E is the expected value of the distribution, and x i is the input sample, and i is the sample number.

[0196] The monotonicity calculation method uses the Spearman correlation coefficient, which can be expressed as:

[0197]

[0198] Here, ρ represents the Spearman correlation coefficient, which is used as a monotonicity indicator.

[0199] The robustness is calculated as the average difference between the upper and lower envelopes, which can be expressed as:

[0200]

[0201] Among them, u env represents the upper envelope calculation, l env represents the lower envelope calculation, x(t) represents the sample point corresponding to the local maximum or minimum value, t represents the local maximum or minimum index, and rob represents the robustness index.

[0202] The monotonicity and robustness of the candidate health indicators are evaluated by the above monotonicity and robustness expressions to obtain the health indicator with the best monotonicity and robustness.

[0203] Step S4.4, using the obtained health index with the best monotonicity and robustness as an indirect prediction quantity, the health index is calculated for the noisy signal and the test data respectively, to obtain the health index of the bearing degradation process based on the noisy signal and the health index of the bearing degradation process based on the test data.

[0204] Step S5. Construct a denoising autoencoder model, use the bearing degradation process health index based on the test data and the bearing degradation process health index based on the noisy signal to train the denoising autoencoder model, make full use of the test data to accurately extract the target prediction amount from the health index of the noisy signal, and obtain the target bearing health index output after denoising extraction. The output of the denoising autoencoder model is the input of the fault prediction model, that is, the health index of the noisy signal is restored after denoising to obtain the target bearing health index.

[0205] Comparison Figure 5 , the method for constructing a denoising autoencoder model in this embodiment includes the following steps.

[0206] Step S5.1, obtain a complete set of bearing degradation process health indicators based on experimental data and bearing degradation process health indicators based on noisy signals, and divide them into training, verification and test sets, and perform tensor quantization to obtain the input sequence for the adaptation model structure and graphics processing unit (GPU) accelerated training.

[0207] Step S5.2, construct the gated recurrent unit of the main structure of the denoising autoencoder model. … ,x N ), the gated recurrent unit can be expressed as:

[0208] r i =σ(W r ·[h i-1 ,x i ]+b z )

[0209] z i =σ(W z ·[h i-1 ,x i ]+b r )

[0210]

[0211] Among them, r i is the reset gate output, z i is the update gate output, is the candidate hidden state, h i is the final hidden state, x i is the input sample, i represents the current input sequence number, and N is the total number of samples. σ is the sigmoid activation function (S-type growth curve activation function), ⊙ represents element multiplication, and W r , W z and W are the weight matrices of the reset gate, update gate and candidate hidden state, respectively. r ,b z and b are the reset gate, update gate and bias of the candidate hidden state respectively.

[0212] Step S5.3, based on the gated recurrent unit, build a multi-head attention mechanism of the main structure of the denoising autoencoder model to obtain the main structure of the denoising autoencoder model. For a given query vector Q, a given key vector Key and a given value vector V, and the corresponding weight matrix The multi-head attention mechanism can be expressed as:

[0213]

[0214] MultiHead=Concat(head1,head2,…,head j …,head h )W O

[0215]

[0216] Among them, d k is the dimension of the key vector Key, W O is the weight matrix used to update the input, ssoftmax represents the normalization calculation function, Concat represents the vector merging calculation function, h represents the total number of attention heads, and head j It represents the j-th attention head, j represents the sequence number of the attention head, Attention represents the self-attention mechanism, and MultiHead represents multi-head attention.

[0217] Step S5.4, adding a neuron parameter release (dropout) process and a linear layer to the main structure of the denoising autoencoder model built in step S5.3, and outputting the model prediction value. This step can improve the robustness of the denoising autoencoder model.

[0218] Step S5.5, based on the obtained model prediction value, the denoising autoencoder model built in steps S5.2 to S5.4 is trained by using the MSE loss function (mean square error loss function) and the adam optimization algorithm (stochastic optimization method of adaptive momentum) through the gradient descent method, so that the model parameters converge, and the changes of the MSE loss function are monitored in the process and the optimal model is output. The MSE loss function can be expressed as:

[0219]

[0220] Among them, y i is the true value, that is, the actual data of the predicted step position, which comes from the data set, and is the model prediction value, that is, the model output value at the prediction step position, i represents the i-th sample, and N represents the total number of samples.

[0221] Step S5.6, the obtained bearing degradation process health index based on the noisy signal is divided according to the given model sequence length, and input into the denoising autoencoder model, so as to obtain the target bearing health index output after denoising as denoised bearing vibration data. In subsequent steps, the denoised bearing vibration data can be used as the input of the fault prediction model.

[0222] Step S6, construct a fault prediction model, use the envelope difference method to divide the health status of the bearing, divide the degradation process of the test bearing into a healthy stage, a slow degradation stage and a rapid degradation stage, train three independent models for data in different stages, and the three independent models constitute a fault prediction model. Input the denoised bearing vibration data into the three trained independent models respectively to generate three sets of prediction values ​​and evaluation indicators. Use the difference in evaluation indicators to determine the current degradation stage, and output the prediction value of the highest accuracy model among the three independent models to achieve fault prediction.

[0223] Comparison Figure 6 , the fault prediction method of this embodiment includes the following steps.

[0224] Step S6.1, based on the health index extracted from the test data of the test bearing, the envelope difference method is used to analyze its trend to obtain the envelope difference result.

[0225] Step S6.2, divide the degradation process of the test bearing obtained in step S3 into stages, and identify the boundaries based on the obtained envelope difference results, and divide it into healthy stage, slow degradation stage and rapid degradation stage. The bearing has a running-in process in the initial stage of operation, which is reflected in the fluctuations on the health index curve. The first non-zero value of the envelope difference after entering the normal working stage is identified as the fault initiation point, that is, the first prediction time (FPT); at the same time, the maximum value of the envelope difference in the identification process is used as the basis for reflecting the sudden change and further deterioration of the bearing health state, and then the identified fault initiation point and the maximum value of the envelope difference are used as two key points to divide the bearing health state into three stages: healthy, slowly degraded and rapidly degraded.

[0226] In step S6.3, the full life cycle test data of the test bearing obtained from step S3 is split into three data sets according to the above three stages, and three independent models are trained respectively, so that each independent model is fully overfitted on the corresponding data set, and the three trained independent models are used as fault prediction models.

[0227] Step S6.4, input the denoised vibration data of the target bearing in the transmission chain to be tested into the three trained independent models, output the predicted values ​​and evaluation indicators respectively, and compare them to obtain the predicted value of the highest accuracy model among the three independent models. For a given input, the three independent models of the fault prediction model output predicted values ​​and evaluation indicators respectively. Among them, the given input can be, for example, the denoised bearing vibration data obtained in step S5. The evaluation index is the absolute percentage error, which can be expressed as: since the model is fully overfitted on the data of each health stage, only a single model shows a higher prediction accuracy for a single input. Therefore, the evaluation indicators output by the three independent models can be compared to determine the model with the highest prediction accuracy, and then identify the health stage of the corresponding data, so as to achieve qualitative judgment of the degradation state; and output the predicted value of the highest accuracy model, that is, to achieve quantitative analysis of the current health indicators.

[0228] Specifically, in order to ensure the differences in evaluation indicators between different models, the fault prediction model described in step S6.3 and step S6.4 should not have too strong generalization ability so as to make it difficult to distinguish the degradation stage. Therefore, a gated recurrent unit model is used and the neuron parameter release operation is not performed, and the prediction value is directly output through the linear layer.

[0229] According to an embodiment of the present invention, a system-oriented bearing fault prediction method using a denoising autoencoder is generally as follows: bearing component-level degradation test data and simulated system noise data are obtained, converted into health indicators and degradation stage divisions, and then the denoising autoencoder and fault prediction model are trained. For a set of real inputs in the application process, the denoising autoencoder first extracts the test bearing health indicators as the input of the fault prediction model, and then the fault prediction model uses this data to achieve qualitative identification of the degradation stage and quantitative identification of the degradation degree, thereby achieving system-oriented transmission chain bearing fault prediction based on component degradation tests.

[0230] An exemplary embodiment of a system-oriented bearing fault prediction method using a denoising autoencoder according to an embodiment of the present invention is described below.

[0231] The data length used in this exemplary embodiment is between 170,000 and 300,000, depending on the specific degradation time of the bearing; at the same time, the data is divided into 60% training data, 20% verification data, and 20% test data. The input sliding window size of the denoising autoencoder is 1024, the number of better hidden layer units is 512, and the number of better hidden layer units of the fault prediction model is 128. It has been verified that the accuracy rates of the two are 95% and 97% respectively, and the comprehensive accuracy rate of the method is 92%. Therefore, the method provided according to the embodiment of the present invention can achieve effective fault prediction while considering the noise of the transmission chain system.

[0232] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0233] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0234] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A system-oriented bearing fault prediction method using a noise reduction autoencoder, characterized in that: include: Step S1, obtaining parameters and working condition information of a transmission chain to be tested, and performing bearing life verification on bearings in the transmission chain to be tested, and determining dangerous bearings as target bearings, wherein the transmission chain to be tested is a system including bearings; Step S2, performing dynamic simulation on the transmission chain to be tested, and obtaining simulation result data as simulated noise; Step S3, performing a component-level degradation test on the target bearing, and acquiring a vibration signal during the degradation process as test data; Step S4, performing signal mixing on the obtained test data and simulation result data to obtain a noisy signal, and respectively establishing a bearing degradation process health index of the target bearing based on the test data and a bearing degradation process health index based on the noisy signal; Step S5, constructing a denoising autoencoder model, using the bearing degradation process health index based on the test data and the bearing degradation process health index based on the noisy signal to train the denoising autoencoder model, making full use of the test data to accurately extract the target prediction amount in the health index of the noisy signal, obtaining the target bearing health index extracted by denoising, and outputting it as denoised bearing vibration data; Step S6, construct a fault prediction model, use the envelope difference method to divide the health status of the bearing, divide the degradation process into a healthy stage, a slow degradation stage and a rapid degradation stage, train three independent models for data in different stages, and the three independent models constitute a fault prediction model. The denoised bearing vibration data are input into the three independent models to generate three sets of prediction values ​​and evaluation indicators. The difference in evaluation indicators is used to determine the current degradation stage and output the prediction value of the highest accuracy model among the three independent models to achieve fault prediction.

2. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 1 is characterized in that: The step S1 specifically includes: Step S1.1, obtaining gear parameters, bearing characteristic parameters and design operating condition information of the transmission chain to be tested; Step S1.2, perform gear stress calibration, calculate the meshing force according to the design working condition information and the gear parameters of the transmission chain, and then obtain the bearing radial load as the equivalent dynamic load, and use this as a basis to perform bearing life calibration, and obtain the bearing life of each bearing as the calibration result, and determine the temperature coefficient g by looking up the table T , hardness coefficient g H , reliability life correction factor a1, life correction factor a2 and operating condition factor a3, the bearing life is Where n is the operating speed, P is the equivalent dynamic load, C is the basic rated dynamic load, and ε is the life index; Step S1.3, determine the dangerous bearings based on comprehensive comparison of the verification results.

3. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 2 is characterized in that: The step S2 specifically includes: Step S2.1, using a lumped parameter method to integrate the shaft mass of the transmission shaft in the transmission chain to be tested into the gear, and obtaining a gear force element including the shaft mass; Step S2.2, the system dynamics equation of the transmission chain to be tested is: Where M is the total mass of the system, C is the damping of the system, K is the stiffness matrix of the system, δ is the displacement of the system, and f is the external force vector of the system. The external force vector of the system is the input calculated according to the working condition, the stiffness matrix of the system is calculated according to the setting, and the displacement of the system is the result calculated according to the setting. The damping of the system is time-varying mesh damping, expressed as: Among them, ε g is the gear meshing damping ratio of the system, k(t) represents the time-varying meshing stiffness of the gears of the system, t is the time, I represents the moment of inertia of the gears of the system, r b It represents the gear pitch radius of the system, and the subscripts 1 and 2 represent different gears respectively; Step S2.3, establish a system dynamics model in the simulation software, consider the time-varying meshing damping and stiffness in S2.2, complete the system dynamics calculation and extract the vibration response of the target bearing position as the simulation result data, that is, as the simulated noise.

4. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 3 is characterized in that: The step S3 specifically includes: Step S3.1, selecting a representative working condition of the transmission chain to be tested from the acquired design working condition information, and converting it into a bearing test load; Step S3.2, calculating the single-point fault characteristic frequency of the target bearing in the transmission chain to be tested as a basis for selecting the vibration sensor; Step S3.3, select a healthy bearing with the same model as the target bearing in the transmission chain to be tested as a test bearing, and select a constant test condition from the design condition information obtained in step S1 for the test bearing to operate until failure; in this process, the vibration signal of the test bearing is collected by the sensor according to a fixed single sampling time and sampling interval, and the data of the vibration signal is merged to obtain a vibration signal data set for the entire life cycle as test data.

5. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 4 is characterized in that: The step S4 specifically includes: Step S4.1, using cubic spline interpolation method to match the sampling rate of simulation result data and test data to obtain corrected noise, and copying the corrected noise signal to the same length as the test data to obtain an equal-length noise signal; Step S4.2, mixing the equal-length noise signal obtained in the above steps and the test vibration signal in the time domain to obtain a noisy signal; Step S4.3, extracting health indicators from the test data and the noisy signal respectively, and evaluating the monotonicity and robustness of the extracted health indicators to obtain the health indicators with the best monotonicity and robustness; Step S4.4, using the obtained health index with the best monotonicity and robustness as an indirect prediction quantity, the health index is calculated for the noisy signal and the test data respectively, to obtain the health index of the bearing degradation process based on the noisy signal and the health index of the bearing degradation process based on the test data.

6. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 5, characterized in that: The step S4.3 specifically includes: The health index extraction includes selecting the root mean square value, kurtosis, peak value and skewness in the test data and the noisy signal as candidate health indicators; The monotonicity and robustness of candidate health indicators are evaluated. For a dynamic sliding window, the input sequence X = (x1, x2, ..., x i ,…,x N ), the RMS value is: Kurtosis is expressed as: The peak value is expressed as: Peak Value = max(X) Skewness is expressed as: Where N is the sample size, μ is the mean of the distribution, σ is the standard deviation of the distribution, E is the expected value of the distribution, and x i is the input sample, i is the sample number; The monotonicity calculation method uses the Spearman correlation coefficient, which is expressed as: Where ρ represents the Spearman correlation coefficient, which is used as a monotonicity indicator; The robustness is calculated as the average difference between the upper and lower envelopes, expressed as: Among them, u env represents the upper envelope calculation, l env represents the lower envelope calculation, x(t) represents the sample point corresponding to the local maximum or minimum value, t represents the index of the local maximum or minimum value, and rob represents the robustness index; The monotonicity and robustness of the candidate health indicators are evaluated by the above monotonicity and robustness expressions to obtain the health indicator with the best monotonicity and robustness.

7. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 5, characterized in that: The step S5 specifically includes: Step S5.1, obtaining a complete set of health indicators of the bearing degradation process based on test data and health indicators of the bearing degradation process based on noisy signals, performing training, verification and test set division, and performing tensor quantization to obtain an input sequence; Step S5.2, construct the gated recurrent unit of the main structure of the denoising autoencoder model. For the input sequence X = (x1, x2, ..., x N ), the gated recurrent unit is expressed as: r i =σ(W r ·[h i-1 ,x i ]+b z ) z i =σ(W z ·[h i-1 ,x i ]+b r ) Among them, r i is the reset gate output, z i is the update gate output, is the candidate hidden state, h i is the final hidden state, x i is the input sample, i represents the current input sequence number, σ is the sigmoid activation function, ⊙ represents element-wise multiplication, W r , W z and W are the weight matrices of the reset gate, update gate and candidate hidden state, respectively. r ,b z and b are the reset gate, update gate and bias of candidate hidden state respectively; Step S5.3, based on the gated recurrent unit, build a multi-head attention mechanism of the main structure of the denoising autoencoder model to obtain the main structure of the denoising autoencoder model. For a given query vector Q, a given key vector Key and a given value vector V, and the corresponding weight matrix The multi-head attention mechanism can be expressed as: MultiHead=Concat(head1,head2,...,head j ...,head h )W O Among them, d k is the dimension of the key vector Key, W O is the weight matrix used to update the input, softmax represents the normalization calculation function, Concat represents the vector merging calculation function, h represents the total number of attention heads, and head j represents the jth attention head, j represents the sequence number of the attention head, Attention represents the self-attention mechanism, and MultiHead represents multi-head attention; Step S5.4, adding a neuron parameter release process and a linear layer to the main structure of the denoising autoencoder model built in step S5.3, and outputting the model prediction value; Step S5.5, based on the obtained model prediction value, the denoising autoencoder model constructed in steps S5.2 to S5.4 is trained by using the MSE loss function and the adam optimization algorithm through the gradient descent method, so that the model parameters converge, and the changes of the MSE loss function are monitored in the process and the optimal model is output; Step S5.6, divide the obtained bearing degradation process health index based on the noisy signal according to the given model sequence length, and input it into the denoising autoencoder model to obtain the target bearing health index extracted by denoising, and output it as denoised bearing vibration data.

8. The system-oriented bearing fault prediction method using a noise reduction autoencoder according to claim 7, characterized in that: The step S6 specifically includes: Step S6.1, according to the health index extracted from the test data of the test bearing, the envelope difference method is used to analyze its trend to obtain the envelope difference result; Step S6.2, dividing the degradation process of the test bearing obtained in step S3 into stages, performing boundary identification based on the obtained envelope difference result, and dividing the degradation process into a healthy stage, a slow degradation stage, and a fast degradation stage; Step S6.3, splitting the full life cycle test data of the test bearing obtained in step S3 into three data sets according to the above three stages, training three independent models respectively, so that each independent model is fully overfitted on the corresponding data set, and the three trained independent models are used as fault prediction models; Step S6.4, input the denoised bearing vibration data of the target bearing in the transmission chain to be tested into the three trained independent models, output the prediction values ​​and evaluation indicators respectively, compare the evaluation indicators output by the three independent models, determine the model with the highest prediction accuracy, obtain the prediction value of the model with the highest accuracy, and then identify the health stage of the prediction value.

Images