CN113357138B - Method and device for predicting remaining service life of hydraulic pump and terminal equipment - Google Patents
Method and device for predicting remaining service life of hydraulic pump and terminal equipment Download PDFInfo
- Publication number
- CN113357138B CN113357138B CN202110826044.XA CN202110826044A CN113357138B CN 113357138 B CN113357138 B CN 113357138B CN 202110826044 A CN202110826044 A CN 202110826044A CN 113357138 B CN113357138 B CN 113357138B
- Authority
- CN
- China
- Prior art keywords
- hydraulic pump
- degradation
- health state
- characteristic
- health
- 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
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
Abstract
The invention is suitable for the technical field of hydraulic pumps, and provides a method, a device and terminal equipment for predicting the residual service life of a hydraulic pump, wherein the method comprises the following steps: acquiring vibration data and return oil flow data of a hydraulic pump; calculating the health state index of the hydraulic pump according to the vibration data and the return oil flow data; inputting the health state index into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump; and determining the remaining service life of the hydraulic pump according to the health state prediction curve. The method and the device can accurately predict the residual service life of the hydraulic pump.
Description
Technical Field
The invention belongs to the technical field of hydraulic pumps, and particularly relates to a method and a device for predicting the residual service life of a hydraulic pump and terminal equipment.
Background
The residual service life of the hydraulic pump is accurately and scientifically predicted, maintenance is facilitated according to the state of the hydraulic pump, and the method and the device have important significance for guaranteeing the reliability and safety of the hydraulic pump.
However, since the operation environment of the hydraulic pump is harsh and the degradation mechanism is complex, the sensor information from a single source is not sufficient to accurately reflect the degradation process of the hydraulic pump, which is not beneficial to accurately predicting the remaining service life of the hydraulic pump. There is a need in the art for a method to accurately predict the remaining useful life of a hydraulic pump.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method, an apparatus, and a terminal device for predicting a remaining service life of a hydraulic pump, so as to accurately predict the remaining service life of the hydraulic pump.
The first aspect of the embodiment of the invention provides a method for predicting the remaining service life of a hydraulic pump, which comprises the following steps:
acquiring vibration data and return oil flow data of a hydraulic pump;
calculating the health state index of the hydraulic pump according to the vibration data and the return oil flow data;
inputting the health state index into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump;
and determining the remaining service life of the hydraulic pump according to the health state prediction curve.
Optionally, calculating the health status index of the hydraulic pump according to the vibration data and the return oil flow data includes:
extracting a plurality of first degradation features from the vibration data; wherein the first degradation characteristic is a time domain characteristic and/or a frequency domain characteristic;
screening out a first degradation characteristic with better prediction performance from the plurality of first degradation characteristics to obtain a second degradation characteristic;
extracting a third degradation characteristic from the oil return flow data;
a state of health indicator of the hydraulic pump is calculated based on the second degradation characteristic and the third degradation characteristic.
Optionally, screening a degradation feature with better prediction performance from the plurality of first degradation features to obtain a second degradation feature, including:
calculating monotonicity indexes, robustness indexes and correlation indexes of the first degradation characteristics;
calculating the comprehensive score of each first degradation characteristic according to a TOPSIS comprehensive evaluation method and the monotonicity index, the robustness index and the correlation index of each first degradation characteristic;
and determining a first degradation characteristic with the comprehensive score higher than a first preset threshold value as a second degradation characteristic.
Optionally, calculating the health indicator of the hydraulic pump according to the second degradation characteristic and the third degradation characteristic includes:
combining the second degradation features and the third degradation features to obtain a multi-source feature vector;
acquiring a historical second degradation characteristic value and a historical third degradation characteristic value of the hydraulic pump in a healthy state, combining the historical second degradation characteristic value and the historical third degradation characteristic value to obtain a historical multisource characteristic vector, and estimating an expected multisource characteristic vector corresponding to the multisource characteristic vector according to the historical multisource characteristic vector and a preset autocorrelation kernel regression algorithm;
and determining the health state index of the hydraulic pump according to the distance similarity and the space direction similarity between the multi-source characteristic vector and the expected multi-source characteristic vector.
Optionally, the health indicator of the hydraulic pump is determined according to the following formula:
in the formula, HI (t) is a health index of a hydraulic pump, f obs (t) is a multi-source feature vector,
in order to expect a multi-source feature vector,
is composed of
A diagonal covariance matrix of variance constructions.
Optionally, the process of establishing the hydraulic pump health state prediction model includes:
establishing a degradation model based on a 3-time B-order spline basis function;
taking the random coefficient of the degradation model as the particles of a preset particle filter model, setting the number of the particles, and initializing each particle according to a maximum likelihood estimation method;
iteratively updating the particles according to the particle filter model, and deleting the particles which do not meet the monotonicity constraint in each iteration process until the only particles which meet the monotonicity constraint are obtained;
and setting the value of the particles meeting the monotonicity constraint as a random coefficient of the degradation model to obtain a hydraulic pump health state prediction model.
Optionally, determining the remaining service life of the hydraulic pump according to the health status prediction curve includes:
taking the moment when the health state prediction curve is larger than a second preset threshold value for the first time as the damage moment of the hydraulic pump;
and determining the remaining service life of the hydraulic pump according to the damage moment and the current moment.
A second aspect of an embodiment of the present invention provides a remaining service life prediction apparatus for a hydraulic pump, including:
the acquisition module is used for acquiring vibration data and return oil flow data of the hydraulic pump;
the calculation module is used for calculating the health state index of the hydraulic pump according to the vibration data and the return oil flow data;
the prediction module is used for inputting the health state indexes into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump;
and the determining module is used for determining the remaining service life of the hydraulic pump according to the health state prediction curve.
A third aspect of the embodiments of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for predicting the remaining service life of a hydraulic pump as described above.
A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the method for predicting the remaining service life of a hydraulic pump as described above.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
according to the embodiment of the invention, the health state index of the hydraulic pump is formed by combining the vibration data and the return flow data of the hydraulic pump, so that the health state of the hydraulic pump can be more accurately reflected. Furthermore, a health state prediction curve of the hydraulic pump can be obtained through the health state index and a pre-established hydraulic pump health state prediction model, the health state prediction curve accurately reflects the degradation process of the hydraulic pump, and the remaining service life of the hydraulic pump can be accurately predicted according to the health state prediction curve.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a schematic flow chart illustrating a method for predicting the remaining service life of a hydraulic pump according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a process for establishing a health prediction model of a hydraulic pump according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart illustrating an implementation of a method for predicting the remaining service life of a hydraulic pump according to an embodiment of the present invention;
FIG. 4 is a graphical illustration of the scoring of various degradation features provided by embodiments of the present invention;
FIG. 5 is a schematic diagram of a hydraulic pump health indicator provided by an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of a remaining service life prediction device of a hydraulic pump according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
In order to explain the technical means of the present invention, the following description will be given by way of specific examples.
Referring to fig. 1, an embodiment of the present invention provides a method for predicting a remaining service life of a hydraulic pump, including the following steps:
step S101, acquiring vibration data and return oil flow data of the hydraulic pump.
In the embodiment of the invention, the vibration data can be collected by a vibration sensor preset on the hydraulic pump, and the return oil flow data can be collected by a return oil flowmeter preset on the hydraulic pump.
And step S102, calculating the health state index of the hydraulic pump according to the vibration data and the return oil flow data.
In the embodiment of the invention, the vibration data contains abundant hydraulic pump performance state information, so the health state of the hydraulic pump is mainly reflected and predicted through the vibration data. Meanwhile, in consideration of the limitation of sensor information with a single source, the embodiment of the invention also combines the oil return flow data of the hydraulic pump to more accurately reflect and predict the health state of the hydraulic pump.
And step S103, inputting the health state indexes into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump.
In the embodiment of the invention, the pre-established hydraulic pump health state prediction model can predict the health state prediction curve of the hydraulic pump according to the current health state index of the hydraulic pump, and the health state prediction curve can accurately reflect the degradation process of the hydraulic pump.
And step S104, determining the remaining service life of the hydraulic pump according to the health state prediction curve.
Therefore, the health state index of the hydraulic pump is formed by combining the vibration data and the return flow data of the hydraulic pump, and the health state of the hydraulic pump can be reflected more accurately. Furthermore, a health state prediction curve of the hydraulic pump can be obtained through the health state index and a pre-established hydraulic pump health state prediction model, the health state prediction curve accurately reflects the degradation process of the hydraulic pump, and the remaining service life of the hydraulic pump can be accurately predicted according to the health state prediction curve.
Optionally, in a possible implementation manner, the health status index of the hydraulic pump is calculated according to the vibration data and the oil return flow data, which may be detailed as follows:
extracting a plurality of first degradation features from the vibration data; wherein the first degenerate feature is a time-domain feature and/or a frequency-domain feature;
screening out a first degradation characteristic with better prediction performance from the plurality of first degradation characteristics to obtain a second degradation characteristic;
extracting a third degradation characteristic from the oil return flow data;
a state of health indicator of the hydraulic pump is calculated based on the second degradation characteristic and the third degradation characteristic.
In the embodiment of the present invention, 15 time-domain features and 13 frequency-domain features are extracted from the vibration data as the first degradation features, the time-domain features are shown in table 1, and the frequency-domain features are shown in table 2.
TABLE 1 time domain signature table
TABLE 2 frequency domain characterization table
The predicted performance of the degradation characteristics of the hydraulic pump has a significant impact on the prediction of the remaining useful life of the hydraulic pump. If the degradation characteristics are well characterized in the life-cycle degradation process of the hydraulic pump from a healthy state to a failed state, even a simple predictive model can obtain an accurate prediction result. Conversely, even if the predictive model performs well, inefficient degradation features can lead to poor predictive results. Therefore, after the first degradation characteristics are extracted, the first degradation characteristics with better prediction performance are screened out from the first degradation characteristics to obtain second degradation characteristics, and then the health state index of the hydraulic pump is formed by combining the third degradation characteristics extracted from the oil return flow data.
Optionally, in a possible implementation manner, screening a degradation feature with better prediction performance from the plurality of first degradation features to obtain a second degradation feature, including:
calculating monotonicity indexes, robustness indexes and correlation indexes of the first degradation characteristics;
calculating the comprehensive score of each first degradation characteristic according to a TOPSIS comprehensive evaluation method and the monotonicity index, the robustness index and the correlation index of each first degradation characteristic;
a first degradation characteristic with a composite score above a first preset threshold is determined as a second degradation characteristic.
In embodiments of the present invention, the degradation characteristic that is well predicted should satisfy three conditions: (1) Monotonically increasing or monotonically decreasing to describe a monotonically degrading process of the hydraulic pump from a healthy state to a failed state; (2) robustness to outliers; (3) The degradation process of the hydraulic pump can be tracked along with time with good correlation. Therefore, the monotonicity index Mon (x), the robustness index Rob (x) and the correlation index Corr (x) are introduced to evaluate the prediction performance of each first degradation feature. The calculation formula of each index is shown in table 3. In table 3, x (i), i =1,2.. N is the i-th sample point of the degradation feature x, N is the length of the degradation feature, x (i) t And x (i) r The ith sample value, x (i), of the degradation characteristic trend term and the residual term, respectively t And x (i) r Can be obtained by smoothing the degradation characteristic x and satisfies the equation x (i) = x (i) t +x(i) r 。
TABLE 3 evaluation index Table
Any single evaluation index in table 3 is difficult to make a complete and accurate evaluation on the prediction performance of the degradation feature, so that the degradation feature needs to be evaluated and selected by combining the above three evaluation indexes.
In the embodiment of the invention, the degradation characteristics are evaluated and selected by adopting a TOPSIS comprehensive evaluation method, which comprises the following steps:
(1) Construction evaluation matrix
Assuming that the number of degradation features to be evaluated is n, and the number of evaluation indexes is m (n is 28, m is 3 in the present embodiment), the evaluation matrix may be configured as:
the evaluation matrix was normalized by the following formula:
wherein, a ij (i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to m) is the j-th evaluation index of the i-th degradation characteristic.
(2) Definition of Baseline
Defining the positive baseline of the jth evaluation index as the maximum of the jth evaluation index:
defining the negative baseline of the jth evaluation index as the minimum value of the jth evaluation index:
(3) Distance calculation
The distance of the metric of the i-th degradation feature from the positive baseline is:
the distance between the metric of the ith degradation feature and the negative baseline is:
(4) Calculating composite score
The overall score for the ith degeneration characteristic was:
the higher the composite score, the better the predictive performance of the degradation features. Therefore, a score threshold (typically 0.6) may be set, and the degradation feature having a composite score higher than the score threshold is used as a degradation feature having a better prediction performance, that is, the first degradation feature having a composite score higher than the score threshold is used as the second degradation feature.
Optionally, in a possible implementation, the health indicator of the hydraulic pump is calculated according to the second degradation characteristic and the third degradation characteristic, which may be detailed as:
combining the second degradation characteristic and the third degradation characteristic to obtain a multi-source characteristic vector;
acquiring a historical second degradation characteristic value and a historical third degradation characteristic value of the hydraulic pump in a healthy state, combining the historical second degradation characteristic value and the historical third degradation characteristic value to obtain a historical multisource characteristic vector, and estimating an expected multisource characteristic vector corresponding to the multisource characteristic vector according to the historical multisource characteristic vector and a preset autocorrelation kernel regression algorithm;
and determining the health state index of the hydraulic pump according to the distance similarity and the space direction similarity between the multi-source characteristic vector and the expected multi-source characteristic vector.
Optionally, in a possible implementation, the health indicator of the hydraulic pump is determined according to the following formula:
in the formula, HI (t) is a health index of a hydraulic pump, f obs (t) is a multi-source feature vector,
in order to expect a multi-source feature vector,
is composed of
A diagonal covariance matrix of variance constructions.
In the embodiment of the invention, the health state index of the hydraulic pump is obtained by fusing the second degradation characteristic and the third degradation characteristic through an improved Auto-associated Kernel Regression (AAKR). The traditional AAKR usually adopts euclidean distance as a similarity criterion to measure the difference between a source domain and a target domain, and the traditional AAKR algorithm has poor performance because the euclidean distance has certain limitation when processing multi-source data with complex correlation. Still other scholars propose to use mahalanobis distance instead of euclidean distance to improve the performance of AAKR in processing complex correlated data, however mahalanobis distance is less effective when the data dimension is higher. Therefore, the embodiment of the present invention provides an improved AAKR algorithm, which uses distance similarity and spatial direction similarity as similarity criteria, where the distance similarity is measured by mahalanobis distance, and the spatial direction similarity is measured by cosine distance.
The principle of the improved AAKR algorithm for calculating the health status index of the hydraulic pump is as follows:
k-1 second degradation characteristics f i (1. Ltoreq. I. Ltoreq.K-1) and a third degradation characteristic f K Combining to obtain multi-source feature vector f = [ f ] 1 ,f 2 ...f K ]. If the multi-source feature vector at the moment t is f obs (t)＝[f 1 obs (t),f 2 obs (t)...f K obs (t)]The AAKR algorithm will f obs (t) mapping to expected multi-source feature vectors in healthy states
Wherein, the first and the second end of the pipe are connected with each other,f obs-nc the matrix is L multiplied by K, and shows that K degradation characteristics are observed for L times under the actual health state of the hydraulic pump;
is the mapping of the source space to the target space.
Expecting multi-source feature vectors
Can be prepared from
Calculated to obtain the weight value w l (1. Ltoreq. L. Ltoreq.L) from f obs-nc And f obs Similarity between the two is calculated, and as with the traditional AAKR, the improved AAKR also adopts a Gaussian radial basis function as a mapping kernel, and a weight calculation formula is as follows:
where the parameter h is the width of the mapping kernel, with a typical value of 0.05.
The embodiment of the invention adopts the comprehensive measurement criterion of distance similarity and space direction similarity to calculate the Gaussian kernel function
The formula is:
wherein the covariance matrix
(K is more than or equal to 1 and less than or equal to K) is the K-th degradation characteristic observation vector f under the health state of the hydraulic pump k obs-nc The variance of (c).
Finally, the comprehensive measurement criterion of distance similarity and space direction similarity is adoptedTo calculate the currently observed multi-source feature vector f obs (t) and expected multi-source feature vector
The difference between them:
the greater the value of the state of health indicator HI (t), the greater the degree to which the hydraulic pump deviates from a healthy operating state, and the more severe the degradation of the hydraulic pump.
Optionally, in a possible implementation manner, the process of establishing the hydraulic pump health state prediction model may be detailed as follows:
establishing a degradation model based on a 3-time B-order spline basis function;
taking the random coefficient of the degradation model as the particles of a preset particle filter model, setting the number of the particles, and initializing each particle according to a maximum likelihood estimation method;
iteratively updating the particles according to the particle filter model, and deleting the particles which do not meet the monotonicity constraint in each iteration process until the only particles which meet the monotonicity constraint are obtained;
and setting the value of the particles meeting the monotonicity constraint as a random coefficient of the degradation model to obtain a hydraulic pump health state prediction model.
In an embodiment of the invention, a B-spline basis function with infinite support is more suitable for degradation modeling of truncated signals than a classical B-spline basis function. And the 3-order B-spline basis function has better nonlinear modeling capability and shows better flexibility in the aspect of random coefficient regression prediction than other low-order and high-order B-spline basis functions.
In the interval t ∈ [ gamma ] p ,γ p+1 ]The inner 3-degree B-spline basis function can be expressed as:
b j,3 (t)＝α 0,p,j +α 1,p,j t+α 2,p,j t 2 +α 3,p,j t 3
wherein alpha is r,p,j R =0, …,3 are nodal coefficients, and a 3-degree B-spline basis function based degradation model belongs to [ γ ] in the interval t ∈ p ,γ p+1 ]Inner can be represented as:
wherein
Are the coefficients of the B-spline basis functions in the degradation model.
Let it be assumed that an infinitely supported B-spline is defined at a set of nondecreasing nodes Γ = { γ } 0 ≤γ 1 ≤…≤γ k-1 ,k>3} at, let eta r,p (r =0,1,2,3) is a cubic polynomial Q 3 (t) in the interval t ∈ [ gamma ] p ,γ p+1 ]Random coefficient of (3), monotonicity constraint condition C of 3-degree B spline h Comprises the following steps:
in the embodiment of the invention, the degradation process of the hydraulic pump is simulated by establishing a particle filter model and using a four-order Markov model, t h The state variable at a time can be passed by x h ＝f(x h-1 ,x h-2 ,...,x h-4 ,θ h ,ω h ) And (5) transferring. To track the local details of the hydraulic pump degradation trajectory to obtain sufficient degradation information, the state variables are updated using sliding windows of length 5, each of which is a fourth order markov unit with a monotonicity constraint. The observed value of the h-th cell can be defined as z h ＝g(x h ,x h+1 ,x h+2 ,x h+3 ,x h+4 ,v h ) H is more than or equal to 1. In order to ensure monotonicity of the predicted degradation track, 5 points in time are taken as 5 nodes of a 3-degree B spline basis function, and monotonicity is restricted by C h Applied to a 3-fold B-spline regression model. Randomization of 3-th order B-spline basis functionsCoefficient theta h In the monotonicity constraint C h The process is completed.
Thus, obeying a truncated normal distribution
The posterior probability density function of (a) can be approximately expressed as:
when N particles are used to approximate the posterior parameters, assume that there are N c (N c Less than or equal to N) particles fall into the constraint space C h The posterior parameters such as mean value and covariance are respectively as follows:
the weight of the monotonicity constraint particle is updated according to the following formula:
referring to fig. 2, the particle filter updates the random coefficient θ of the degradation model h The process of (2) is as follows:
(1) Initializing a B spline basis function degradation model for 3 times;
(2) Initializing particles (random coefficients) of the degradation model by a maximum likelihood estimation method;
(3) Sampling N particles from a prior distribution of a degradation model;
(4) Carrying out updating, resampling, normalization and other processing on the particles;
(5) Estimating posterior probability density distribution under an unconstrained condition;
(6) Judging whether each particle meets the monotonicity constraint condition C h And deleting the particles which do not meet the monotonicity constraint condition;
(6) Estimating a posterior probability density distribution p (theta) with monotonicity constraints h |z 1:h ；θ h ∈C h )；
(7) Judging whether a new observation value is available, when the new observation value is available, returning to the step (3), otherwise, stopping updating and keeping the latest random coefficient theta h 。
Optionally, in a possible implementation, the remaining service life of the hydraulic pump is determined according to the health state prediction curve, which may be detailed as:
and taking the moment when the health state prediction curve is larger than a second preset threshold value for the first time as the damage moment of the hydraulic pump, and determining the remaining service life of the hydraulic pump according to the damage moment and the current moment.
In the embodiment of the invention, the health state prediction curve accurately reflects the degradation process of the hydraulic pump, and the larger the value of the health state prediction curve is, the more serious the degradation of the hydraulic pump is.
According to the above, referring to fig. 3, an embodiment of the present invention provides a flow chart of a method for predicting remaining service life of a hydraulic pump:
(1) Acquiring vibration data of a hydraulic pump, and extracting a first degradation characteristic from the vibration data;
(2) Screening the degeneration characteristics with the score larger than 0.6 from the first degeneration characteristics by a TOPSIS comprehensive evaluation method to obtain second degeneration characteristics;
(3) Acquiring oil flow data, and extracting a third degradation characteristic from the oil return flow data;
(4) Fusing the second degradation characteristic and the third degradation characteristic through AAKR to obtain a health state index of the hydraulic pump;
(5) Establishing a degradation model based on 3-time B-spline basis functions, and updating random coefficients of the degradation model through monotonicity-constrained particle filtering to obtain a hydraulic pump health state prediction model;
(6) Inputting the health state index into a hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump;
(7) And taking the moment when the health state prediction curve is larger than the second preset threshold value for the first time as the damage moment of the hydraulic pump, and determining the residual service life of the hydraulic pump according to the damage moment and the current moment.
The feasibility verification of the method provided by the embodiment of the invention is performed as follows.
After obtaining vibration data of the hydraulic pump for a certain period of time, 28 degradation features were extracted from the vibration data, and TOPSIS scores for the respective degradation features were calculated as shown in fig. 4. The number of degeneration characteristics meeting the requirement that TOPSIS score is more than or equal to 0.6 is 6, namely F2, F3, F6, F7, F8 and F17. The health status index of the hydraulic pump in the time period is obtained by acquiring the oil return flow data of the hydraulic pump in the time period and extracting the degradation characteristics and fusing the degradation characteristics through the AAKR as shown in FIG. 5.
The health status indicator is input into a health status prediction model for prediction, and the uncertainty of the prediction is measured using a 95% confidence interval of the Remaining Useful Life (RUL) distribution. Analysis was performed in comparison with other prediction methods in the field (AOPF, PEPE, EPF)
To evaluate the performance of the respective prediction results, wherein
Er i Indicating the error rate of the i-th prediction,
ActRUL i and PreRUL i The actual remaining service life and the predicted remaining service life of the ith prediction are respectively obtained, and the comparison results of each model are shown in table 4.
TABLE 4 comparison of predicted results
As can be seen from Table 4, the prediction accuracy of the method for predicting the remaining service life of the hydraulic pump provided by the embodiment of the invention is higher than that of the other three methods, and the 95% confidence interval for predicting the RUL is in a reasonable level. Also, as the degradation data increases, the prediction error rate of the embodiments of the present invention decreases. The embodiment of the invention can more accurately predict the residual service life of the hydraulic pump.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
An embodiment of the present invention provides a device for predicting remaining service life of a hydraulic pump, and as shown in fig. 6, the device 60 includes:
and the acquisition module 61 is used for acquiring vibration data and return oil flow data of the hydraulic pump.
And the calculating module 62 is used for calculating the health state index of the hydraulic pump according to the vibration data and the return oil flow data.
And the prediction module 63 is configured to input the health state index into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump.
A determination module 64 for determining a remaining useful life of the hydraulic pump based on the state of health prediction curve.
Optionally, in a possible implementation, the computing module 62 is configured to:
extracting a plurality of first degradation features from the vibration data; wherein the first degenerate feature is a time-domain feature and/or a frequency-domain feature;
screening out first degradation characteristics with better prediction performance from the plurality of first degradation characteristics to obtain second degradation characteristics;
extracting a third degradation characteristic from the oil return flow data;
a state of health indicator of the hydraulic pump is calculated based on the second degradation characteristic and the third degradation characteristic.
Optionally, in a possible implementation, the calculation module 62 is configured to:
calculating monotonicity indexes, robustness indexes and correlation indexes of the first degradation characteristics;
calculating the comprehensive score of each first degradation characteristic according to a TOPSIS comprehensive evaluation method and the monotonicity index, the robustness index and the correlation index of each first degradation characteristic;
a first degradation characteristic with a composite score above a first preset threshold is determined as a second degradation characteristic.
Optionally, in a possible implementation, the calculation module 62 is configured to:
combining the second degradation characteristic and the third degradation characteristic to obtain a multi-source characteristic vector;
acquiring a historical second degradation characteristic value and a historical third degradation characteristic value of the hydraulic pump in a healthy state, combining the historical second degradation characteristic value and the historical third degradation characteristic value to obtain a historical multisource characteristic vector, and estimating an expected multisource characteristic vector corresponding to the multisource characteristic vector according to the historical multisource characteristic vector and a preset autocorrelation kernel regression algorithm;
and determining the health state index of the hydraulic pump according to the distance similarity and the space direction similarity between the multi-source characteristic vector and the expected multi-source characteristic vector.
Optionally, in a possible implementation, the calculation module 62 is configured to:
determining a health indicator for the hydraulic pump according to the following equation:
in the formula, HI (t) is a health index of a hydraulic pump, f obs (t) is a multi-source feature vector,
in order to expect a multi-source feature vector,
is made of
A diagonal covariance matrix of variance constructions.
Optionally, in a possible implementation manner, the prediction module 63 is further configured to:
establishing a degradation model based on a 3-time B-time spline basis function;
taking the random coefficient of the degradation model as the particles of a preset particle filter model, setting the number of the particles, and initializing each particle according to a maximum likelihood estimation method;
iteratively updating the particles according to the particle filter model, and deleting the particles which do not meet the monotonicity constraint in each iteration process until the only particles which meet the monotonicity constraint are obtained;
and setting the value of the particles meeting the monotonicity constraint as a random coefficient of the degradation model to obtain a hydraulic pump health state prediction model.
Optionally, in a possible implementation manner, the determining module 64 is configured to:
taking the moment when the health state prediction curve is larger than a second preset threshold value for the first time as the damage moment of the hydraulic pump;
and determining the remaining service life of the hydraulic pump according to the damage moment and the current moment.
Fig. 7 is a schematic diagram of a terminal device 70 according to an embodiment of the present invention. As shown in fig. 7, the terminal device 70 of this embodiment includes: a processor 71, a memory 72, and a computer program 73 stored in the memory 72 and executable on the processor 71. The processor 71, when executing the computer program 73, implements the steps in the above-described embodiments of the remaining useful life prediction method of the respective hydraulic pumps, such as the steps S101 to S104 shown in fig. 1. Alternatively, the processor 71, when executing the computer program 73, implements the functions of the modules in the above-described device embodiments, such as the functions of the modules 61 to 64 shown in fig. 6.
Illustratively, the computer program 73 can be divided into one or more modules/units, which are stored in the memory 72 and executed by the processor 71 to carry out the invention. One or more of the modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution of the computer program 73 in the terminal device 70. For example, the computer program 73 may be divided into the acquisition module 61, the calculation module 62, the prediction module 63, and the determination module 64 (module in the virtual device), and each module has the following specific functions:
and the acquisition module 61 is used for acquiring vibration data and return oil flow data of the hydraulic pump.
And the calculating module 62 is used for calculating the health status index of the hydraulic pump according to the vibration data and the return flow data.
And the prediction module 63 is configured to input the health state index into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump.
And a determination module 64 for determining the remaining service life of the hydraulic pump according to the health state prediction curve.
The terminal device 70 may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. Terminal equipment 70 may include, but is not limited to, a processor 71, a memory 72. Those skilled in the art will appreciate that fig. 7 is merely an example of terminal device 70 and does not constitute a limitation of terminal device 70, and may include more or fewer components than shown, or some components may be combined, or different components, e.g., terminal device 70 may also include input-output devices, network access devices, buses, etc.
The Processor 71 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 72 may be an internal storage unit of the terminal device 70, such as a hard disk or a memory of the terminal device 70. The memory 72 may also be an external storage device of the terminal device 70, such as a plug-in hard disk provided on the terminal device 70, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like. Further, the memory 72 may also include both an internal storage unit of the terminal device 70 and an external storage device. The memory 72 is used to store computer programs and other programs and data required by the terminal device 70. The memory 72 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules, so as to perform all or part of the functions described above. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described apparatus/terminal device embodiments are merely illustrative, and for example, a module or a unit may be divided into only one type of logic function, and another division manner may be provided in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method according to the embodiments of the present invention may also be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of the embodiments of the method. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U.S. disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution media, and the like. It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, in accordance with legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunications signals.
The above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (7)
1. A method for predicting the residual service life of a hydraulic pump is characterized by comprising the following steps:
acquiring vibration data and return oil flow data of a hydraulic pump;
calculating a health status index of the hydraulic pump according to the vibration data and the return oil flow data;
inputting the health state index into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump;
determining the remaining service life of the hydraulic pump according to the health state prediction curve;
calculating a health status indicator of the hydraulic pump according to the vibration data and the return oil flow data, including:
extracting a plurality of first degradation features from the vibration data, and screening out first degradation features with better prediction performance from the plurality of first degradation features to obtain second degradation features; wherein the first degenerate feature is a time-domain feature and/or a frequency-domain feature;
extracting a third degradation feature from the return oil flow data;
combining the second degradation feature and the third degradation feature to obtain a multi-source feature vector;
acquiring a historical second degradation characteristic value and a historical third degradation characteristic value of the hydraulic pump in a healthy state, combining the historical second degradation characteristic value and the historical third degradation characteristic value to obtain a historical multisource characteristic vector, and estimating an expected multisource characteristic vector corresponding to the multisource characteristic vector according to the historical multisource characteristic vector and a preset autocorrelation kernel regression algorithm;
determining a health state index of the hydraulic pump according to the distance similarity and the spatial direction similarity between the multi-source feature vector and the expected multi-source feature vector;
the formula for determining the health indicator of the hydraulic pump is:
in which HI (t) is a health index of the hydraulic pump, f obs (t) is a multi-source feature vector,
in order to expect a multi-source feature vector,
is composed of
A diagonal covariance matrix of variance constructions.
2. The method for predicting the remaining service life of a hydraulic pump according to claim 1, wherein the step of screening out a first degradation characteristic with better predicted performance from a plurality of first degradation characteristics to obtain a second degradation characteristic comprises the steps of:
calculating monotonicity indexes, robustness indexes and correlation indexes of the first degradation characteristics;
calculating the comprehensive score of each first degradation characteristic according to a TOPSIS comprehensive evaluation method and the monotonicity index, the robustness index and the correlation index of each first degradation characteristic;
a first degradation characteristic with a composite score above a first preset threshold is determined as a second degradation characteristic.
3. The method for predicting the remaining service life of a hydraulic pump according to claim 1, wherein the process of establishing the hydraulic pump health prediction model includes:
establishing a degradation model based on a 3-time B spline basis function;
taking the random coefficient of the degradation model as the particles of a preset particle filter model, setting the number of the particles, and initializing each particle according to a maximum likelihood estimation method;
iteratively updating the particles according to the particle filter model, and deleting the particles which do not meet the monotonicity constraint in each iteration process until the only particles which meet the monotonicity constraint are obtained;
and setting the value of the particle meeting the monotonicity constraint as a random coefficient of the degradation model to obtain a hydraulic pump health state prediction model.
4. The method of predicting the remaining useful life of a hydraulic pump as set forth in claim 1, wherein determining the remaining useful life of the hydraulic pump from the state of health prediction curve comprises:
taking the moment when the health state prediction curve is larger than a second preset threshold value for the first time as the damage moment of the hydraulic pump;
and determining the remaining service life of the hydraulic pump according to the damage moment and the current moment.
5. A remaining service life prediction apparatus for a hydraulic pump, comprising:
the acquisition module is used for acquiring vibration data and return oil flow data of the hydraulic pump;
the calculation module is used for calculating the health state index of the hydraulic pump according to the vibration data and the return oil flow data;
the prediction module is used for inputting the health state index into a pre-established hydraulic pump health state prediction model to obtain a health state prediction curve of the hydraulic pump;
the determining module is used for determining the residual service life of the hydraulic pump according to the health state prediction curve;
the calculation module is specifically used for extracting a plurality of first degradation features from the vibration data, and screening out first degradation features with better prediction performance from the plurality of first degradation features to obtain second degradation features; wherein the first degenerate feature is a time-domain feature and/or a frequency-domain feature;
extracting a third degradation feature from the return oil flow data;
combining the second degradation feature and the third degradation feature to obtain a multi-source feature vector;
acquiring a historical second degradation characteristic value and a historical third degradation characteristic value of the hydraulic pump in a healthy state, combining the historical second degradation characteristic value and the historical third degradation characteristic value to obtain a historical multisource characteristic vector, and estimating an expected multisource characteristic vector corresponding to the multisource characteristic vector according to the historical multisource characteristic vector and a preset autocorrelation kernel regression algorithm;
determining a health state index of the hydraulic pump according to the distance similarity and the spatial direction similarity between the multi-source feature vector and the expected multi-source feature vector;
the formula for determining the health indicator of the hydraulic pump is:
in which HI (t) is a health index of the hydraulic pump, f obs (t) is a multi-source feature vector,
in order to expect a multi-source feature vector,
is composed of
A diagonal covariance matrix of variance constructions.
6. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 4 when executing the computer program.
7. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110826044.XA CN113357138B (en) | 2021-07-21 | 2021-07-21 | Method and device for predicting remaining service life of hydraulic pump and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110826044.XA CN113357138B (en) | 2021-07-21 | 2021-07-21 | Method and device for predicting remaining service life of hydraulic pump and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113357138A CN113357138A (en) | 2021-09-07 |
| CN113357138B true CN113357138B (en) | 2022-10-11 |
Family
ID=77540203
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110826044.XA Active CN113357138B (en) | 2021-07-21 | 2021-07-21 | Method and device for predicting remaining service life of hydraulic pump and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113357138B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113864665B (en) * | 2021-10-09 | 2024-02-13 | 重庆邮电大学 | Fluid pipeline leakage positioning method based on adaptive ICA and improved RLS filter |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101539137B (en) * | 2009-04-17 | 2010-06-16 | 北京航空航天大学 | Residual life gray prediction method of aerospace hydraulic pump based on delta filter |
| CN106801670A (en) * | 2015-11-26 | 2017-06-06 | 中国人民解放军军械工程学院 | Life for Hydraulic Pump testing stand |
| CN107358347A (en) * | 2017-07-05 | 2017-11-17 | 西安电子科技大学 | Equipment cluster health state evaluation method based on industrial big data |
| CN111222290B (en) * | 2020-01-13 | 2024-04-09 | 浙江工业大学 | Multi-parameter feature fusion-based method for predicting residual service life of large-scale equipment |
| CN111258297B (en) * | 2020-01-17 | 2021-06-04 | 北京大学 | Equipment health index construction and service life prediction method based on data fusion network |
| CN111680661A (en) * | 2020-06-19 | 2020-09-18 | 哈尔滨工业大学 | Mechanical rotating part performance degradation tracking method based on multi-feature fusion |
| CN111950201A (en) * | 2020-08-11 | 2020-11-17 | 成都一通密封股份有限公司 | Full life cycle monitoring system and method for pump sealing device |
| CN112197890B (en) * | 2020-09-08 | 2021-07-13 | 西安交通大学 | Calorimeter performance degradation evaluation method, storage medium and equipment |
| CN112036051B (en) * | 2020-11-05 | 2021-01-26 | 中国人民解放军国防科技大学 | Method, device, equipment and medium for predicting residual service life of magnetic suspension system |
| CN112685961B (en) * | 2021-01-12 | 2022-06-21 | 武汉大学 | Method and system for predicting remaining service life of analog circuit |
| CN112906157A (en) * | 2021-02-20 | 2021-06-04 | 南京航空航天大学 | Method and device for evaluating health state of main shaft bearing and predicting residual life |
| CN112785092A (en) * | 2021-03-09 | 2021-05-11 | 中铁电气化局集团有限公司 | Turnout residual life prediction method based on self-adaptive deep layer feature extraction |
-
2021
- 2021-07-21 CN CN202110826044.XA patent/CN113357138B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113357138A (en) | 2021-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021089013A1 (en) | Spatial graph convolutional network training method, electronic device and storage medium | |
| JP2015230727A (en) | Method for detecting anomalies in time series data | |
| US20090043715A1 (en) | Method to Continuously Diagnose and Model Changes of Real-Valued Streaming Variables | |
| CN106709588B (en) | Prediction model construction method and device and real-time prediction method and device | |
| CN109426912B (en) | Wind control system optimization method, system and device and electronic equipment | |
| CN111382906A (en) | Power load prediction method, system, equipment and computer readable storage medium | |
| WO2018211721A1 (en) | Abnormal information estimation device, abnormal information estimation method, and program | |
| CN112597610B (en) | Optimization method, device and equipment for lightweight design of mechanical arm structure | |
| CN113357138B (en) | Method and device for predicting remaining service life of hydraulic pump and terminal equipment | |
| CN116307745B (en) | Intelligent risk supervision and early warning method and system for engineering project | |
| CN115686908A (en) | Data processing method and related equipment | |
| CN109065176B (en) | Blood glucose prediction method, device, terminal and storage medium | |
| Hlávka et al. | Change-point methods for multivariate time-series: paired vectorial observations | |
| CN111193627A (en) | Information processing method, device, equipment and storage medium | |
| CN107463486B (en) | System performance analysis method and device and server | |
| CN117407771A (en) | Bearing health state assessment method and device based on digital twin and related equipment | |
| CN111091099A (en) | Scene recognition model construction method, scene recognition method and device | |
| CN113876354B (en) | Fetal heart rate signal processing method and device, electronic equipment and storage medium | |
| CN115730845A (en) | Power system dynamic estimation method, device, equipment and readable storage medium | |
| JP2016520220A (en) | Hidden attribute model estimation device, method and program | |
| Raza et al. | Application of extreme learning machine algorithm for drought forecasting | |
| CN114331349A (en) | Scientific research project management method and system based on Internet of things technology | |
| CN113609445A (en) | Multi-source heterogeneous monitoring data processing method, terminal device and readable storage medium | |
| CN113516275A (en) | Power distribution network ultra-short term load prediction method and device and terminal equipment | |
| CN112346995A (en) | Construction method and device of test risk estimation model based on banking industry |
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 |