CN115326397B - Method and related device for establishing crankshaft bearing wear degree prediction model and prediction method - Google Patents
Method and related device for establishing crankshaft bearing wear degree prediction model and prediction method Download PDFInfo
- Publication number
- CN115326397B CN115326397B CN202210903822.5A CN202210903822A CN115326397B CN 115326397 B CN115326397 B CN 115326397B CN 202210903822 A CN202210903822 A CN 202210903822A CN 115326397 B CN115326397 B CN 115326397B
- Authority
- CN
- China
- Prior art keywords
- vibration signal
- crankshaft bearing
- wear degree
- bearing wear
- memory
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 58
- 238000013528 artificial neural network Methods 0.000 claims abstract description 31
- 238000002922 simulated annealing Methods 0.000 claims abstract description 30
- 230000015654 memory Effects 0.000 claims abstract description 28
- 230000001133 acceleration Effects 0.000 claims abstract description 19
- 230000007787 long-term memory Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 21
- 238000000605 extraction Methods 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 101000872083 Danio rerio Delta-like protein C Proteins 0.000 claims description 2
- 201000010099 disease Diseases 0.000 claims description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 2
- 239000000243 solution Substances 0.000 description 9
- 238000005299 abrasion Methods 0.000 description 8
- 239000011159 matrix material Substances 0.000 description 6
- 230000006403 short-term memory Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000000137 annealing Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The application relates to a method for establishing a crankshaft bearing wear degree prediction model, a prediction method and a related device, wherein the method for establishing the crankshaft bearing wear degree prediction model comprises the following steps: acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor; denoising the obtained vibration signal x (t) by a wavelet threshold value, and processing by a soft threshold value method with a threshold value being an unbiased likelihood estimation threshold value to obtain a processed vibration signal x' (t); optimizing and improving the integrated empirical mode decomposition by using a simulated annealing algorithm to decompose, extracting the characteristics of the processed vibration signal x' (t) by using a singular value decomposition method, and storing the characteristics in a characteristic set; and taking the feature set as input of the long-period memory artificial neural network model, and optimizing the long-period memory artificial neural network model by using a simulated annealing algorithm to generate a crankshaft bearing wear degree prediction model. According to the application, the optimal parameters of the long-term memory artificial neural network are found by using the simulated annealing algorithm, so that the prediction accuracy of the system can be effectively improved.
Description
Technical Field
The application relates to a diesel engine, in particular to a method and a related device for establishing a crankshaft bearing wear degree prediction model.
Background
The diesel engine is a power source of a plurality of vehicles, and the diesel engine is faulty, so that potential safety hazards can be brought to personnel on the vehicles, and due to the fact that the structure is complex, a large amount of manpower and material resources are needed in the traditional mechanical fault diagnosis method, the machinery is required to be disassembled, even parts are damaged, and the fault diagnosis method based on vibration signals can avoid the problems, realize disassembly-free detection and save a large amount of cost. Therefore, fault diagnosis methods based on vibration signals have become a hot spot for many scholars to study in recent years.
Chen Y et al, which uses fast fourier transform (Fast Fourier transform, FFT) to analyze and process the vibration signal, but has the characteristic of easy aggregation, cannot be well identified, and requires human intervention; zhang Lingling, etc., the wavelet packet-AR spectrum is adopted to extract the characteristics of the bearing faults of the transmission, but the method has the defect of difficult selection of wavelet base; cheng J et al, using empirical mode decomposition (Empirical Mode Decomposition, EMD) to decompose vibration signals, but EMD suffers from mode confusion; xia, W, etc., the magnitude of the oil injection advance angle is analyzed by using an adaptive wavelet and integrated empirical mode decomposition (Ensemble Empirical Mode Decomposition, EEMD) to decompose signals, but because the actual operation is limited, the influence of the added white noise cannot be well eliminated, reconstruction errors can be caused, while the influence of the white noise on the signals can be well eliminated by using a complementary integrated empirical mode decomposition method (complete-tary Ensemble Empirical Mode Decompo-position, CEEMD), false components can appear, zheng Jinde, etc., the problems can be alleviated by using an improved integrated empirical mode decomposition MEEMD (CPEMD), but the problems of more parameters and poor stability can occur, the ultra-short-term prediction can be performed on wind power by using an improved integrated empirical mode decomposition and Least Squares Support Vector Machine (LSSVM), but the fitting precision of the algorithm is in proportion to the number of points, and more data are needed.
Disclosure of Invention
In view of the above, the present application provides a method and related apparatus for establishing a model for predicting the wear of a crankshaft bearing.
In order to solve the problems, the technical scheme provided by the application is as follows:
in a first aspect, the present application provides a method for building a predictive model of the wear level of a crankshaft bearing, comprising the steps of:
s11: acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
s12: denoising the obtained vibration signal x (t) by a wavelet threshold value, and processing by a soft threshold value method with a threshold value being an unbiased likelihood estimation threshold value to obtain a processed vibration signal x' (t);
s13: optimizing and improving the integrated empirical mode decomposition by using a simulated annealing algorithm to decompose, extracting the characteristics of the processed vibration signal x' (t) by using a singular value decomposition method, and storing the characteristics in a characteristic set;
s14: and taking the feature set as input of the long-period memory artificial neural network model, and optimizing the long-period memory artificial neural network model by using a simulated annealing algorithm to generate a crankshaft bearing wear degree prediction model.
Further, the step S13 includes the following steps:
s131: an improved aggregate empirical mode decomposition process vibration signal x' (t);
s132: the simulated annealing algorithm optimizes the improved aggregate empirical mode decomposition on the processed vibration signal x' (t);
s133: performing feature extraction on the processed vibration signal x' (t) by using a singular value decomposition method;
s134: and storing the decomposed features into a feature set.
In a second aspect, the present application provides a method for predicting the wear degree of a crankshaft bearing, comprising the steps of:
s21: acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
s22: inputting the diesel engine vibration signal x (t) into a crankshaft bearing wear degree prediction model, and outputting a prediction result of the crankshaft bearing wear degree, wherein the crankshaft bearing wear degree prediction model is built according to the method for building the crankshaft bearing wear degree prediction model according to any one of claims 1-2.
In a third aspect, the present application provides an apparatus for modeling a predicted wear level of a crankshaft bearing, the apparatus comprising:
an acquisition unit for acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
the processing unit is used for carrying out wavelet threshold denoising on the obtained vibration signal x (t), and processing by a soft threshold method of selecting the threshold as an unbiased likelihood estimation threshold to obtain a processed vibration signal x' (t);
the extraction unit is used for optimizing and improving the integrated empirical mode decomposition by using a simulated annealing algorithm to decompose, and the singular value decomposition method is used for extracting the characteristics of the processed vibration signal x' (t) and storing the characteristics in the characteristic set.
And the generating unit is used for taking the vibration signal characteristics in the characteristic set as the input of the long-period memory artificial neural network model, optimizing the long-period memory artificial neural network model by using a simulated annealing algorithm, and generating a crankshaft bearing wear degree prediction model.
Further, the extraction unit includes:
a first decomposition unit for improved aggregate empirical mode decomposition processing of the vibration signal x' (t);
the second decomposition unit is used for simulating the vibration signal x' (t) after the processing of the empirical mode decomposition of the set of optimization and improvement of the annealing algorithm;
an extraction subunit, configured to perform feature extraction on the processed vibration signal x' (t) using a singular value decomposition method;
and the storage unit is used for storing the decomposed characteristics into the characteristic set.
In a fourth aspect, the present application provides a crankshaft bearing wear degree prediction apparatus, the apparatus comprising:
a vibration signal acquisition unit for acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
a predicted result output unit for inputting the diesel engine vibration signal x (t) acquired by the industrial acceleration sensor into a crankshaft bearing wear degree prediction model established according to the method of establishing a crankshaft bearing wear degree prediction model according to any one of claims 1 to 2, and outputting the crankshaft bearing wear degree prediction result.
In a fifth aspect, the present application provides a terminal device, comprising:
a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to invoke instructions in the memory to perform the method of building a predictive model of wear level of a crankshaft bearing of any one of claims 1-2.
In a sixth aspect, the present application provides a terminal device, including:
a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is used for calling instructions in a memory to execute the method for predicting the wear degree of the crankshaft bearing according to claim 3.
From this, the application has the following advantages and beneficial effects:
the application adopts the characteristic of extracting signals by combining singular value decomposition and simulated annealing algorithm optimized and improved ensemble empirical mode decomposition, and uses the simulated annealing algorithm to optimize the long-term and short-term memory artificial neural network to predict faults. Decomposing a crankshaft bearing vibration signal of the diesel engine by using a plurality of decomposition methods, analyzing various indexes, selecting an optimal decomposition method, optimizing and re-decomposing by using a simulated annealing algorithm, and extracting the characteristics of the data by using a singular value decomposition method and taking the characteristics as a characteristic set of a neural network module. The long-term and short-term memory artificial neural network is suitable for multiple classification problems and is suitable for predicting the wear degree of a crankshaft bearing. In addition, the optimal parameters of the long-term memory artificial neural network are found by using a simulated annealing algorithm, so that the prediction accuracy of the system can be effectively improved.
The improved integrated empirical mode decomposition SA-CPEMD optimized by the simulated annealing algorithm can solve the problems of excessive CPEMD parameters and poor stability;
CPEMD can reduce EEMD and CEEMD from adding excessive noise, and the problem of false components occurs;
the mode confusion phenomenon of EMD, CPEMD can better eliminate this phenomenon, make the reconstruction error smaller, the completeness is better.
2) SVD singular value decomposition has good stability, SA-CPEMD-SVD, and can well extract characteristics of vibration signals.
3) The LSTM long-term artificial neural network can process any length of input, is also suitable for multiple output problems, and is suitable for predicting the abrasion degree of the crankshaft bearing.
4) The long-short-period artificial neural network optimized by the SA-LSTM simulated annealing algorithm improves the prediction accuracy of the original long-period artificial neural network model.
Drawings
The application is further described in detail below with reference to the attached drawings and examples:
FIG. 1 is a flow chart of a method for establishing a predictive model of crankshaft bearing wear;
FIG. 2 is a block diagram of LSTM;
FIG. 3 is a flowchart of optimizing a long and short term memory artificial neural network using a simulated annealing algorithm:
FIG. 4 is a flowchart of a method for predicting the wear level of a crankshaft bearing;
FIG. 5 is a block diagram of an apparatus for modeling the wear level of a crankshaft bearing;
fig. 6 is a structural diagram of a crank bearing wear degree prediction device.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Example 1
As shown in fig. 1, a method for establishing a model for predicting the wear degree of a crankshaft bearing according to this embodiment, taking a diesel engine as an example, includes the following steps:
s11: placing an industrial acceleration sensor at a designated test point of a diesel engine of a research object, acquiring an acquired diesel engine vibration signal x (t) acquired by the industrial acceleration sensor, and taking the acquired signal as an original signal of a system; in the embodiment, an industrial acceleration sensor is placed at the left and right positions of a 3-cylinder at the joint of the oil bottom and the cylinder body, and is attracted to the surface of the test part by additionally installing a strong magnetic seat.
S12: and (3) carrying out wavelet threshold denoising on the obtained vibration signal x (t) due to noise interference in the outside, and processing by a soft threshold method of selecting the threshold as an unbiased likelihood estimation threshold to obtain a processed vibration signal x' (t).
S13: optimizing and improving the integrated empirical mode decomposition by using a simulated annealing algorithm to decompose, extracting the characteristics of the processed vibration signal x' (t) by using a singular value decomposition method, and storing the characteristics in a characteristic set;
s14: and taking the feature set as input of a long-term memory artificial neural network model, optimizing the long-term memory artificial neural network model by using a simulated annealing algorithm, and generating a long-term memory artificial neural network (SA-LSTM) model optimized based on the simulated annealing algorithm, wherein the long-term memory artificial neural network (SA-LSTM) model optimized based on the simulated annealing algorithm is a crankshaft bearing wear degree prediction model.
Wherein, the step S13 includes the following steps:
s131: an improved aggregate empirical mode decomposition process vibration signal x' (t);
with a first vibration signal x' 1 (t) for example, first vibration signal x 'is decomposed using a Complementary Ensemble Empirical Mode Decomposition (CEEMD)' 1 (t) decomposing, t being the sequence length, comprising the steps of:
white noise of random equal length, given standard deviation and opposite signAnd added to the vibration signal: obtain->Wherein, the liquid crystal display device comprises a liquid crystal display device,is the u th 1 A signal after white noise is added for the second time; decomposing the signal after the first addition of positive white noise using the EMD algorithm>The following are examples: />i＝1，h 0 ＝r i-1 (t)，j＝1。
Obtaining an ith eigenmode function, which comprises the following steps:
(a) Finding intermediate variable transition h j-1 Local extremum points of (t);
(b) Couple h j-1 Respectively performing cubic spline function interpolation on the maximum value and the minimum value point of (t) to form an upper envelope line and a lower envelope line;
(c) Calculating the average value m of the upper and lower envelope curves j-1 (t)；
(d) Let h j (t)＝h j-1 (t)-m j-1 (t)；
(e) If h j (t) is an eigenmode function, imf i (t)＝h j (t); otherwise, j=j+1, go to step (a).
Let r i (t)＝r i-1 (t)-imf i (t)。
If r i (t) if the number of extreme points is still more than 2, i=i+1, and turning to the step of continuing to find the eigenmode function; otherwise, the decomposition is finished, and the obtained r is finally obtained i And (t) is a residual component. r is (r) n The (t) residual component is generally denoted by r, such as x= imf1+ imf2+ … +imfn+r. The algorithm can finally obtain: and obtaining a preset number of eigenvalue function components and remainder.
And so onJ obtained by decomposition using empirical mode decomposition 1 The components of the intrinsic mode functions are respectively->And->The remainder components are +.>And->Summing the components and the remainder of each eigenmode function and averaging to obtain the following value:
to the final j 1 A plurality of eigenmode functions; and r is the remainder component after decomposition. k is the number of added positive white noise, wherein the number of added negative white noise is equal to the number of added positive white noise.
Sequential detectionIs assigned to the first eigenmode function of the first signal>For example, consider the length N time series +.>And carrying out space reconstruction to obtain the following time sequence:
…
…
wherein: m is the embedding dimension and λ is the time delay.
IMF (i) * ) M vectors of (2): rearranged in ascending order, i.e.)>
If presentThen press j * Is arranged by the magnitude of the value of j, i.e. when j * w1 <j * w2 When the method is used, the following steps are included: />
Therefore, any one of the vectors IMF (i * ) A set of symbol sequences is available: s (g) = [ J 1 ,J 2 ,…,J m ]Shares m-! The different permutations correspondingly share the probability P of the occurrence of k symbol sequences 1 ,P 2 ,…,P k Wherein:wherein N is g In order to generate the number of times of the same arrangement of the g type, k is the total number of types (k is less than or equal to m) after the repeated arrangement is removed,/>correspond to->Time series->The permutation entropy of (a) can be defined in terms of Shannon entropy as:
when P g =1/m-! When Hp (m) reaches a maximum value ln (m|), the permutation entropy Hp (m) is normalized by ln (m|), that is: hp=Hp (m)/ln (m-Is used for the permutation entropy value of (a).
Setting the permutation entropy reference value to be 0.6, and comparing the permutation value of the obtained eigenfunction with the reference value. If the eigenvalue is larger than the reference value, the eigenvalue is singular, otherwise, the eigenvalue is nonsingular. After the first singular component is obtained, the entropy value calculation step is repeatedly arranged on the modal function component of the evidence until the non-singular component is obtained.
The original signal x' 1 (t) subtracting all singular components to obtain a new signal x' with the singular components removed 1 (t) vs. x 1 And (t) performing EMD decomposition to obtain the final required eigenmode function.
S132: the simulated annealing algorithm optimizes the improved aggregate empirical mode decomposition on the processed vibration signal x' (t);
in order to solve the problem of poor multi-stability of improved set empirical mode decomposition (CPEMD) parameters, a simulated annealing algorithm (SA) is used for optimizing the improved set empirical mode decomposition, a set average parameter N is added, a white noise logarithm Ne is added, an entropy order m is replaced, an entropy threshold theta0 is arranged as an independent variable, a mean square error MSE and orthogonality ort are adopted, Y=MSE+8x ort is adopted as an index, and a simulated annealing algorithm is utilized for finding a Y minimum value and obtaining an optimal parameter.
Y is taken as an objective function f () of a simulated annealing algorithm, and then the set average parameter N is set as S1, the added white noise logarithm Ne is set as S2, the substitution entropy order m is set as S3, the permutation entropy threshold theta0 is set as S4, and the algorithm steps of the algorithm are as follows:
initializing: onset temperature T 0 (high enough), initial solution S (iteration start), iteration number L, termination temperature T 1 。
Secondly, k=1, …, and L are the first to the sixth steps.
A new solution S' is created. S '= [ S1' S2 'S3' S4]
The increment Δc=f (S') -f (S) is calculated, f () being the objective function.
If DeltaC is used for the treatment of five diseases<0, receiving S' as a new current solution; otherwise, an in [0,1 ] is generated]And (3) the random number epsilon in the interval is compared with the probability P line, if epsilon is more than P, a new solution is received, and otherwise S is reserved. Wherein the method comprises the steps ofK is a constant.
If the iteration times are not met, turning to a third step;
see if the target temperature is met, if not, T is gradually reduced, and T>T 1 And then turning to the third step.
And obtaining optimal parameters and solutions.
S133: performing feature extraction on the processed vibration signal x' (t) by using a singular value decomposition method;
taking a vibration signal as an example, the first 5 eigenvalue function components after CPEMD decomposition with optimal parameters form an initial vector matrix E e×g The matrix has rank r', singular value decomposition is performed, and e=udw T Where U is an eXe unitary matrix, D is a semi-positive e x g diagonal matrix, and W T I.e., the conjugate transpose of W, is a g×g unitary matrix.，/> Δ r′×r′ ＝diag(σ 1 ,σ 2 ,…,σ r′ )，Is a matrix E T Characteristic value of E. Obtaining E e×g Singular value σ i (i=1, 2, …, r'), the resulting feature component σ i As a feature set of the neural network. Ith of nth signal 2 The characteristic component is->Taking the first feature set as an example. N=1, i 2 The first piece of data for =1, 2, …, i.e. feature set, is: sigma (sigma) 12 、σ 12 、...、σ 16 The nth data is sigma n2 、σ n2 、...、σ n6 。
S134: and storing the decomposed features into a feature set.
In the step S14, a long-term memory artificial neural network LSTM is constructed and training and testing are performed:
as shown in FIG. 2, the obtained feature set T (n) is divided into training sets T for long-term and short-term memory artificial neural network 1 And test set T 2 Wherein the training set T 1 The label of (2) is y 1 The number of the test sets is N, and the test set T 2 The label of (2) is y 2 . The basic steps for constructing the LSTM are as follows: in training set T 1 Label y 1 For input, training is performed, a test set T is input 2 Performing network test and outputting a predictive label y 2 ' let the correct number be N 0 Let N 0 =0, if y' 2 (a)-y 2 (a) =0 (a=1, 2, …, N) then N 0 ＝N 0 +1, cycle from 1 to N, by the formulaWhere s is the accuracy of the classification prediction. The parameters and weights of the long-term and short-term memory artificial neural network are deduced as follows:
let x be the input of the neural network, o be the output, and y' be the label of the training set.
Training a neural network: the signal propagates forward and the error propagates backward as follows:
1) Updating the forget gate output: f (t) =σ (w xf x t +w hf h t-1 +b f ) Wherein σ is a sigmoid function;
2) Updating two-part output of an input gate: i (t) =σ (w xi x t +w hi h t-1 +b i )，g t ＝tanh(w xg x t +w hg h t-1 +b g )；
3) Updating cell state c t ＝c t-1 ⊙f t +g t ⊙i t ；
4) Updating the output gate output: o (t) =σ (w xo x t +w ho h t-1 +b o )；h t ＝o t ⊙tanh(c t ) Maintaining short term memory;
5) Updating the current period prediction output:
6) Error:
forward propagation: the forward propagation formula at time t is:
c t ＝c t-1 ⊙f t +g t ⊙i t
m t ＝tanh(c t )
h t ＝o t ⊙m t
y t ＝w yh h t +by
the back propagation formula: j is the error.
It is known that:
node gradient update:
parameter gradient update:
as shown in fig. 3, the long-term memory artificial neural network model is optimized using a simulated annealing algorithm:
taking the classification prediction accuracy obtained by LSTM as an objective function f (), and setting the batch size as an independent variable x 1 The node number of the hidden layer is set as an independent variable x of an objective function 2 Optimizer type z, s= [ x1 x2 z]The method comprises the following steps:
initializing: onset temperature T 0 (high enough), initial solution S (iteration start), iteration number L, termination temperature T 1 。
Secondly, k=1, …, and L are the first to the sixth steps.
A new solution S' is created. S '= [ x1' x2'z' ].
The increment Δc=f (S') -f (S) is calculated, f () being the objective function.
If delta C is more than 0, receiving S' as a new current solution; otherwise, an in [0,1 ] is generated]Random numbers epsilon in the interval, and comparing with the probability P rows,if epsilon > P, then receive new solution, otherwise keep S. Wherein the method comprises the steps ofK is a constant.
If the iteration times are not met, turning to a third step; see if the target temperature is met, if not, T is gradually reduced, and T>T 1 And then turning to the third step.
And obtaining the optimal classification prediction accuracy S.
In the embodiment, the system with more parameters and poor stability is subjected to optimization algorithm to search and tune parameters, and the optimal parameters are found to obtain the optimal system and the prediction accuracy.
Example 2
As shown in fig. 4, the method for predicting the wear degree of the crankshaft bearing according to the present embodiment includes the following steps:
s21: placing an industrial acceleration sensor at a designated position of a diesel engine to obtain acceleration vibration data of a diesel engine vibration signal x (t); in this embodiment, the 4 th crankshaft bearing of the diesel engine is taken as a research object. A PCB601A01 type ICP industrial acceleration sensor is selected and placed at the position of the left and right of a 3-cylinder at the joint of the oil bottom and the cylinder body, and is attracted to the surface of a test part by additionally installing a strong magnetic seat. The bearing clearance is defined to be 0.05-0.11mm in normal working condition, 0.11-0.15mm in slight abrasion, 0.15-0.20mm in moderate abrasion and more than 0.20mm in heavy abrasion. The vibration signals of the crankshaft bearing under normal working conditions, slight abrasion, moderate abrasion and severe abrasion at the rotating speed of 1800r/min are collected, the sampling frequency of the sensor is 12.8kHz, and 16384 points are sampled each time to serve as one piece of data after the running is stable.
S22: inputting the vibration signal x (t) of the diesel engine into a crankshaft bearing wear degree prediction model, and outputting a prediction result of the crankshaft bearing wear degree.
The accuracy S of the classification prediction of the wear degree of the crankshaft bearing of the diesel engine by adopting the method for predicting the wear degree of the crankshaft bearing reaches 96.875 percent, and the accuracy is improved by more than 9 percent compared with the basic LSTM network prediction. The model can predict the abrasion degree of the crankshaft through the obtained vibration signals.
Example 3
As shown in fig. 5, a device for establishing a crankshaft bearing wear degree prediction model according to the present embodiment includes:
an acquisition unit 301 for acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
because of noise interference, the processing unit 302 is configured to perform wavelet threshold denoising on the obtained vibration signal x (t), and select a soft threshold method with an unbiased likelihood estimation threshold value to perform processing, so as to obtain a processed vibration signal x' (t);
the extracting unit 303 is configured to optimize the improved ensemble empirical mode decomposition by using a simulated annealing algorithm to perform decomposition, perform feature extraction on the processed vibration signal x' (t) by using a singular value decomposition method, and store the feature extraction in a feature set;
and the generating unit 304 is configured to take the vibration signal characteristics in the characteristic set as input of the long-term memory artificial neural network model, optimize the long-term memory artificial neural network model by using a simulated annealing algorithm, and generate a crankshaft bearing wear degree judging model.
The extraction unit 303 includes:
a first decomposition unit 3031 for improved ensemble empirical mode decomposition processing of the vibration signal x' (t);
a second decomposition unit 3032, configured to simulate an annealing algorithm to optimize the improved ensemble empirical mode decomposition on the processed vibration signal x' (t);
an extraction subunit 3033, configured to perform feature extraction on the processed vibration signal x' (t) by using a singular value decomposition method;
a storage unit 3034, configured to store the decomposed features in a feature set.
The device of the embodiment can solve the problems of more improved integrated empirical mode decomposition (CPEMD) parameters and poor stability.
Example 4
As shown in fig. 6, the present embodiment is a crank bearing wear degree prediction apparatus including:
a vibration signal acquisition unit 401 for acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
and a prediction result output unit 402 for inputting the diesel engine vibration signal x (t) acquired by the industrial acceleration sensor into the crank bearing wear degree prediction model and outputting the crank bearing wear degree prediction result.
The present embodiment predicts that the device suppresses CEEMD artifacts for CPEMD. The generation of the pseudo component is that the positive and negative white total noise is added in the signal, and although the decomposition and the average are carried out, the total noise cannot be completely eliminated in the calculation iteration, so that the arrangement entropy Hp is introduced as an index to remove the residual white noise, the influence of the residual white noise is reduced, and the SA-CPEMD improves the stability of the CPEMD.
Embodiments 3 and 4 are presented in terms of functional modularization to provide a device for establishing a model for predicting the wear degree of a crankshaft bearing and a device for predicting the wear degree of a crankshaft bearing. Next, a description will be given of a terminal device for establishing a model for predicting the wear degree of a crank bearing and a terminal device for predicting the wear degree of a crank bearing provided in an embodiment of the present application from the viewpoint of hardware processing.
Example 5
The embodiment provides a terminal device, including: a processor and a memory; a memory for storing program code and transmitting the program code to the processor; and the processor is used for calling instructions in the memory to execute the method for establishing the crankshaft bearing wear degree prediction model provided by the embodiment of the application.
Example 6
The present embodiment provides a terminal device, including: a processor and a memory; a memory for storing program code and transmitting the program code to the processor; and the processor is used for calling instructions in the memory to execute the crankshaft bearing wear degree prediction method provided by the embodiment of the application.
It should be noted that, in the present description, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different manner from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system or device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant points refer to the description of the method section.
It is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.
Claims (6)
1. A method for establishing a predictive model of the wear level of a crankshaft bearing, comprising the steps of:
s11: acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
s12: denoising the obtained vibration signal x (t) by a wavelet threshold value, and processing by a soft threshold value method with a threshold value being an unbiased likelihood estimation threshold value to obtain a processed vibration signal x' (t);
s13: optimizing and improving the integrated empirical mode decomposition by using a simulated annealing algorithm to decompose, extracting the characteristics of the processed vibration signal x' (t) by using a singular value decomposition method, and storing the characteristics in a characteristic set;
s14: the feature set is used as input of a long-period memory artificial neural network model, and a simulated annealing algorithm is used for optimizing the long-period memory artificial neural network model to generate a crankshaft bearing wear degree prediction model;
the basic steps for constructing the LSTM are as follows: in training set T 1 Label y 1 For input, training is performed, a test set T is input 2 Performing network test and outputting a predictive label y 2 ' let the correct number be N 0 Let N 0 =0, if y' 2 (a)-y 2 (a) =0, a=1, 2, …, N, then N 0 ＝N 0 +1, cycle from 1 to N, by the formulaWherein s is the accuracy of classification prediction;
the method for optimizing the long-term memory artificial neural network model by using the simulated annealing algorithm comprises the following steps of:
taking the classification prediction accuracy obtained by LSTM as an objective function f (), and setting the batch size as an independent variable x 1 The node number of the hidden layer is set as an independent variable x of an objective function 2 Optimizer type z, s= [ x1 x2 z]The method comprises the following steps:
initializing: onset temperature T 0 High enough, the initial solution S starts to iterate, the iteration times L and the termination temperature T 1 ；
Secondly, k=1, … and L are subjected to the first steps;
generating a new solution S'; s '= [ x1' x2'z' ];
calculating the increment delta C=f (S') -f (S), wherein f () is an objective function;
if DeltaC is used for the treatment of five diseases>0, receiving S' as a new current solution; otherwise, an in [0,1 ] is generated]Random number epsilon in interval and comparing with probability P, if epsilon>And P, receiving a new solution, otherwise, reserving S. Wherein the method comprises the steps ofK is a constant;
if the iteration times are not met, turning to a third step; see if the target temperature is met, if not, T is gradually reduced, and T>T 1 Then turning to the third step;
and obtaining the optimal classification prediction accuracy S.
2. The method for modeling a predicted wear level of a crank bearing according to claim 1, wherein said step S13 comprises the steps of:
s131: an improved aggregate empirical mode decomposition process vibration signal x' (t);
s132: the simulated annealing algorithm optimizes the improved aggregate empirical mode decomposition on the processed vibration signal x' (t);
s133: performing feature extraction on the processed vibration signal x' (t) by using a singular value decomposition method;
s134: and storing the decomposed features into a feature set.
3. The method for predicting the wear degree of the crankshaft bearing is characterized by comprising the following steps of:
s21: acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
s22: inputting the diesel engine vibration signal x (t) into a crankshaft bearing wear degree prediction model, and outputting a prediction result of the crankshaft bearing wear degree, wherein the crankshaft bearing wear degree prediction model is built according to the method for building the crankshaft bearing wear degree prediction model according to any one of claims 1-2.
4. A crankshaft bearing wear degree prediction apparatus, characterized by comprising:
a vibration signal acquisition unit (401) for acquiring a diesel engine vibration signal x (t) acquired by an industrial acceleration sensor;
a predicted result output unit (402) for inputting the diesel engine vibration signal x (t) acquired by the industrial acceleration sensor into a crankshaft bearing wear degree prediction model established according to the method of establishing a crankshaft bearing wear degree prediction model according to any one of claims 1 to 2, and outputting the predicted result of the crankshaft bearing wear degree.
5. A terminal device, the terminal device comprising:
a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to invoke instructions in the memory to perform the method of building a predictive model of wear level of a crankshaft bearing of any one of claims 1-2.
6. A terminal device, the terminal device comprising:
a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is used for calling instructions in a memory to execute the method for predicting the wear degree of the crankshaft bearing according to claim 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210903822.5A CN115326397B (en) | 2022-07-28 | 2022-07-28 | Method and related device for establishing crankshaft bearing wear degree prediction model and prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210903822.5A CN115326397B (en) | 2022-07-28 | 2022-07-28 | Method and related device for establishing crankshaft bearing wear degree prediction model and prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115326397A CN115326397A (en) | 2022-11-11 |
| CN115326397B true CN115326397B (en) | 2023-10-27 |
Family
ID=83920545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210903822.5A Active CN115326397B (en) | 2022-07-28 | 2022-07-28 | Method and related device for establishing crankshaft bearing wear degree prediction model and prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115326397B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115577643A (en) * | 2022-11-23 | 2023-01-06 | 广东电网有限责任公司中山供电局 | Temperature prediction method and device for cable terminal |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3465127A1 (en) * | 2016-05-24 | 2019-04-10 | Nuovo Pignone Technologie S.r.l. | Method and system to monitor the health status of a rolling bearing of a machinery, and machinery equipped with said system |
| CN110610035A (en) * | 2019-08-28 | 2019-12-24 | 武汉科技大学 | Rolling bearing residual life prediction method based on GRU neural network |
| CN113435322A (en) * | 2021-06-25 | 2021-09-24 | 西安交通大学 | Main shaft bearing fault detection method, system, equipment and readable storage medium |
| CN113465923A (en) * | 2021-06-08 | 2021-10-01 | 华能国际电力股份有限公司上海石洞口第一电厂 | Residual life prediction method for grinding roller bearing of coal mill |
| CN114264478A (en) * | 2021-12-21 | 2022-04-01 | 北京石油化工学院 | Diesel engine crankshaft bearing wear degree prediction method and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111220921A (en) * | 2020-01-08 | 2020-06-02 | 重庆邮电大学 | Lithium battery capacity estimation method based on improved convolution-long-and-short-term memory neural network |
| CN111562358B (en) * | 2020-05-06 | 2021-03-16 | 武汉大学 | Transformer oil gas content prediction method and system based on combined model |
-
2022
- 2022-07-28 CN CN202210903822.5A patent/CN115326397B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3465127A1 (en) * | 2016-05-24 | 2019-04-10 | Nuovo Pignone Technologie S.r.l. | Method and system to monitor the health status of a rolling bearing of a machinery, and machinery equipped with said system |
| CN110610035A (en) * | 2019-08-28 | 2019-12-24 | 武汉科技大学 | Rolling bearing residual life prediction method based on GRU neural network |
| CN113465923A (en) * | 2021-06-08 | 2021-10-01 | 华能国际电力股份有限公司上海石洞口第一电厂 | Residual life prediction method for grinding roller bearing of coal mill |
| CN113435322A (en) * | 2021-06-25 | 2021-09-24 | 西安交通大学 | Main shaft bearing fault detection method, system, equipment and readable storage medium |
| CN114264478A (en) * | 2021-12-21 | 2022-04-01 | 北京石油化工学院 | Diesel engine crankshaft bearing wear degree prediction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115326397A (en) | 2022-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101958674B1 (en) | Actually-measured marine environment data assimilation method based on sequence recursive filtering three-dimensional variation | |
| CN110361778B (en) | Seismic data reconstruction method based on generation countermeasure network | |
| CN110212528B (en) | Power distribution network measurement data missing reconstruction method | |
| CN115326397B (en) | Method and related device for establishing crankshaft bearing wear degree prediction model and prediction method | |
| CN110175541B (en) | Method for extracting sea level change nonlinear trend | |
| CN112364291B (en) | Empirical mode decomposition method and device for pre-filtering extreme point optimization set | |
| CN112926644A (en) | Method and system for predicting residual service life of rolling bearing | |
| CN113239022A (en) | Method and device for complementing missing data in medical diagnosis, electronic device and medium | |
| CN110570487B (en) | Undersampling model generation method, image reconstruction method, device and computer equipment | |
| CN115422687A (en) | Service life prediction method of rolling bearing | |
| CN116384244A (en) | Electromagnetic field prediction method based on physical enhancement neural network | |
| CN115829157A (en) | Chemical water quality index prediction method based on variational modal decomposition and auto former model | |
| CN114264478A (en) | Diesel engine crankshaft bearing wear degree prediction method and system | |
| CN117200208B (en) | User-level short-term load prediction method and system based on multi-scale component feature learning | |
| CN113657244A (en) | Fan gearbox fault diagnosis method and system based on improved EEMD and speech spectrum analysis | |
| CN116465624B (en) | Gearbox fault diagnosis method based on HNGO optimization KELM | |
| CN117058079A (en) | Thyroid imaging image automatic diagnosis method based on improved ResNet model | |
| CN117155806A (en) | Communication base station flow prediction method and device | |
| CN116754213A (en) | Electric gate valve fault diagnosis method, device and equipment based on strong noise background | |
| CN115859048A (en) | Noise processing method and device for partial discharge signal | |
| CN116340724A (en) | ICCEEMDAN-LRTC-based power load completion method and system | |
| CN114720129A (en) | Rolling bearing residual life prediction method and system based on bidirectional GRU | |
| CN113158714B (en) | Fault feature self-adaptive extraction method based on wavelet entropy and EEMD | |
| CN115063499A (en) | Slice self-adaptive determination method for radial sampling trajectory of magnetic resonance imaging | |
| CN111126659A (en) | Power load prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |