CN114154269A - Gear service life prediction method and device - Google Patents
Gear service life prediction method and device Download PDFInfo
- Publication number
- CN114154269A CN114154269A CN202111490708.6A CN202111490708A CN114154269A CN 114154269 A CN114154269 A CN 114154269A CN 202111490708 A CN202111490708 A CN 202111490708A CN 114154269 A CN114154269 A CN 114154269A
- Authority
- CN
- China
- Prior art keywords
- signal
- gear
- current
- characteristic signal
- model
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000012549 training Methods 0.000 claims abstract description 9
- 238000005070 sampling Methods 0.000 claims description 35
- 230000015654 memory Effects 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 11
- 238000012417 linear regression Methods 0.000 claims description 8
- 238000012423 maintenance Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 230000036541 health Effects 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Abstract
The invention provides a method and a device for predicting the service life of a gear, wherein the method comprises the following steps: acquiring a historical vibration acquisition signal and the current moment of a target gear; calculating a characteristic signal of the target gear based on the historical vibration acquisition signal; inputting the characteristic signal into a preset time and characteristic signal relation model, carrying out model training, and determining a model parameter of the preset time and characteristic signal relation model; and generating a gear service life prediction result of the target gear based on the corresponding failure characteristic signal when the preset gear fails, the trained relation model of the preset time and the characteristic signal and the current moment. Therefore, the service life condition of the gear is predicted in real time by utilizing the vibration acquisition signal and the failure characteristic signal of the target gear, so that the automatic and accurate prediction of the service life of the gear is realized without depending on manual experience, the time is saved, the efficiency of the gear service life prediction is improved, an accurate data base is provided for the maintenance and the replacement of the gear, and the safe operation of mechanical equipment provided with the gear is further ensured.
Description
Technical Field
The invention relates to the technical field of equipment fault diagnosis, in particular to a method and a device for predicting the service life of a gear,
background
With the increase of complexity and maintenance cost of mechanical equipment, attention is increasingly paid to fault diagnosis and condition monitoring technologies for equipment components, a gear is one of the most common and easily damaged parts in the mechanical equipment, and if the service life of the gear is expired, faults can easily occur, the normal working condition of the mechanical equipment can be affected, and the mechanical equipment can be damaged.
Therefore, how to accurately predict the service life of the gear has important significance on the safe operation of mechanical equipment, the guarantee of the working efficiency of the mechanical equipment and the like. However, in the prior art, the service life of the gear is mainly predicted by manually observing the meshing condition and the gear form of the gear, the predicted result is greatly influenced by manual experience, and the accuracy of the predicted result is difficult to guarantee.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for predicting a gear life to solve a problem in the prior art that it is difficult to ensure accuracy of a prediction result by using a manual method for predicting a gear life.
According to a first aspect, an embodiment of the present invention provides a method for predicting a life of a gear, including:
acquiring a historical vibration acquisition signal and the current moment of a target gear;
calculating a characteristic signal of the target gear based on the historical vibration acquisition signal, wherein the characteristic signal comprises the meshing frequency at different sampling moments and a side-band energy signal of harmonic waves of the meshing frequency;
inputting the characteristic signal into a preset time and characteristic signal relation model, carrying out model training, and determining a model parameter of the preset time and characteristic signal relation model;
and generating a gear service life prediction result of the target gear based on a corresponding failure characteristic signal when the preset gear fails, a trained relation model of the preset time and the characteristic signal and the current moment.
Optionally, the calculating a characteristic signal of the target gear based on the historical vibration acquisition signal includes:
sequentially extracting a current vibration signal and a current sampling moment corresponding to the current vibration signal according to sampling time from the historical vibration acquisition signal;
carrying out Fourier transform on the current vibration signal to obtain a corresponding frequency domain signal;
calculating a current characteristic signal corresponding to the current vibration signal based on the frequency domain signal;
and adding a mark of the current sampling moment to the current characteristic signal to obtain the current characteristic signal with the current sampling moment.
Optionally, the calculating a current characteristic signal corresponding to the current vibration signal based on the frequency domain signal includes:
acquiring the meshing frequency and the sideband of the target gear;
and calculating the meshing frequency corresponding to the current vibration signal and the sideband energy signal of the harmonic wave of the current vibration signal based on the frequency domain signal, the meshing frequency of the target gear and the sideband.
Optionally, the preset time and characteristic signal relationship model is a linear regression model, and the preset time and characteristic signal relationship model is represented by the following formula:
y^＝wx+b，
wherein y ^ represents the predicted meshing frequency and the sideband energy signal of the harmonic wave thereof, x represents the current moment, and w and b represent two model parameters of the preset time and characteristic signal relation model.
Optionally, the method further comprises:
acquiring a side band energy signal of a current meshing frequency and harmonic thereof of a real-time vibration acquisition signal of a target gear at a current sampling moment;
and inputting the current meshing frequency at the current sampling moment and the side band energy signal of the harmonic wave of the current meshing frequency to a trained preset time and characteristic signal relation model for optimizing model parameters.
Optionally, the mesh frequency corresponding to the current vibration signal and the sideband energy signal of the harmonic thereof are calculated by the following formula:
the Energy _ side represents the meshing frequency corresponding to the current vibration signal and the sideband Energy signal of the harmonic thereof, f represents the meshing frequency of the target gear, d is the sideband of the target gear, and Acc _ f (n) represents the frequency domain signal corresponding to the current vibration signal.
According to a second aspect, an embodiment of the present invention provides a gear life prediction apparatus, including:
the acquisition module is used for acquiring a historical vibration acquisition signal and the current moment of the target gear;
the first processing module is used for calculating a characteristic signal of the target gear based on the historical vibration acquisition signal, wherein the characteristic signal comprises the meshing frequency at different sampling moments and a side band energy signal of harmonic waves of the meshing frequency;
the second processing module is used for inputting the characteristic signal to a preset time and characteristic signal relation model, carrying out model training and determining a model parameter of the preset time and characteristic signal relation model;
and the third processing module is used for generating a gear service life prediction result of the target gear based on a corresponding failure characteristic signal when the preset gear fails, a trained relation model of preset time and the characteristic signal and the current moment.
According to a third aspect, embodiments of the present invention provide a non-transitory computer readable storage medium storing computer instructions which, when executed by a processor, implement the method of the first aspect of the present invention and any one of its alternatives.
According to a fourth aspect, an embodiment of the present invention provides an electronic device, including: a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory having stored therein computer instructions, the processor being configured to execute the computer instructions to perform the method of the first aspect of the present invention and any one of the alternatives thereof.
The technical scheme of the invention has the following advantages:
the embodiment of the invention provides a method and a device for predicting the service life of a gear, which are characterized in that historical vibration acquisition signals and the current moment of a target gear are acquired; calculating characteristic signals of the target gear based on historical vibration acquisition signals, wherein the characteristic signals comprise meshing frequencies at different sampling moments and side-band energy signals of harmonic waves of the meshing frequencies; inputting the characteristic signal into a preset time and characteristic signal relation model, carrying out model training, and determining a model parameter of the preset time and characteristic signal relation model; and generating a gear service life prediction result of the target gear based on the corresponding failure characteristic signal when the preset gear fails, the trained relation model of the preset time and the characteristic signal and the current moment. Therefore, the vibration of the target gear in the using process is utilized to collect signals, the characteristic signals capable of reflecting the health state of the gear are extracted, the relation between the characteristic signals and the sampling time is utilized to train a preset time and characteristic signal relation model, the failure characteristic signals corresponding to the failure of the gear and the service life condition of the gear at the current time are utilized to predict in real time, manual experience is not needed, the automatic and accurate prediction of the service life of the gear is realized, the time is saved, the efficiency of the gear service life prediction is improved, an accurate data base is provided for the overhaul and the replacement of the gear, and the safe operation of mechanical equipment provided with the gear is further ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method of predicting gear life in an embodiment of the present invention;
FIG. 2 is a graph showing the meshing spectrum of a normal cylindrical gear according to an embodiment of the present invention;
FIG. 3 is a graph of the mesh spectrum of a cylindrical gear with pitting failure in an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a gear life prediction device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device in an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
In recent years, rotary machines are becoming larger, more precise, and more intelligent, and health state scoring and remaining life prediction of rotary machines are receiving more and more attention. The gear is used as an important component of the rotary machine, and when the gear works at a high rotating speed or under a heavy load, fatigue damage, degradation and failure are easy to occur, so that the rotary machine is seriously failed or functionally failed, and therefore production stagnation and enterprise benefit loss are brought.
Therefore, how to accurately predict the service life of the gear has important significance on the safe operation of mechanical equipment, the guarantee of the working efficiency of the mechanical equipment and the like. However, in the prior art, the service life of the gear is mainly predicted by manually observing the meshing condition and the gear form of the gear, the predicted result is greatly influenced by manual experience, and the accuracy of the predicted result is difficult to guarantee.
Based on the above problem, an embodiment of the present invention provides a method for predicting a gear life, as shown in fig. 1, the method for predicting a gear life specifically includes the following steps:
step S101: and acquiring a historical vibration acquisition signal and the current moment of the target gear.
The historical vibration acquisition signal can acquire the vibration signal of the target gear by using a vibration sensor, and the sampling time of each vibration signal is extracted.
Step S102: and calculating a characteristic signal of the target gear based on the historical vibration acquisition signal.
The characteristic signal comprises meshing frequencies at different sampling moments and side-band energy signals of harmonics of the meshing frequencies.
Step S103: and inputting the characteristic signal into a preset time and characteristic signal relation model, carrying out model training, and determining a model parameter of the preset time and characteristic signal relation model.
Specifically, the preset time and characteristic signal relation model is a linear regression model, and is represented by the following formula (1):
y^＝wx+b (1)
wherein y ^ represents the predicted mesh frequency and the sideband energy signal of the harmonic wave thereof, x represents the current moment, and w and b represent two model parameters of a preset time and characteristic signal relation model.
Step S104: and generating a gear service life prediction result of the target gear based on the corresponding failure characteristic signal when the preset gear fails, the trained relation model of the preset time and the characteristic signal and the current moment.
The failure characteristic signal can be set according to an empirical value of failure of the target gear, and the gear life prediction result is a predicted residual life value of the target gear at the current moment, such as: the gear life prediction was 1 month, indicating that the gear is expected to fail after 1 month.
By executing the steps, the gear life prediction method provided by the embodiment of the invention extracts the characteristic signal capable of reflecting the health state of the gear by using the vibration acquisition signal of the target gear in the using process, trains the preset time and characteristic signal relation model by using the relation between the characteristic signal and the sampling moment, and predicts the life condition of the gear in real time by using the corresponding failure characteristic signal when the gear fails and the current moment, so that the automatic and accurate prediction of the gear life is realized without depending on manual experience, the time is saved, the gear life prediction efficiency is improved, an accurate data base is provided for the overhaul and the replacement of the gear, and the safe operation of mechanical equipment provided with the gear is further ensured.
Specifically, in an embodiment, the step S102 specifically includes the following steps:
step S201: and sequentially extracting the current vibration signal and the corresponding current sampling moment from the historical vibration acquisition signal according to the sampling time.
Step S202: and carrying out Fourier transform on the current vibration signal to obtain a corresponding frequency domain signal.
Step S203: and calculating a current characteristic signal corresponding to the current vibration signal based on the frequency domain signal.
Specifically, the step S203 specifically includes acquiring the meshing frequency and the sideband of the target gear; and calculating the meshing frequency corresponding to the current vibration signal and the sideband energy signal of the harmonic wave of the current vibration signal based on the frequency domain signal, the meshing frequency of the target gear and the sideband.
The meshing frequency corresponding to the current vibration signal and the sideband energy signal of the harmonic wave thereof are calculated by the following formula (2):
wherein, Energy _ side represents the meshing frequency corresponding to the current vibration signal and the sideband Energy signal of the harmonic thereof, f represents the meshing frequency of the target gear, d is the sideband of the target gear, and AcC _ F (n) represents the frequency domain signal corresponding to the current vibration signal.
Step S204: and adding a label of the current sampling moment to the current characteristic signal to obtain the current characteristic signal with the current sampling moment.
And then inputting the characteristic signal with the time mark into the formula (1) to obtain a predicted time, comparing the difference between the predicted time and the marked real sampling time through a loss function, and adjusting the model parameters until the function value of the loss function reaches the preset requirement and does not change any more, wherein the corresponding model parameter is the optimal model parameter.
Specifically, the relation model between the preset time and the characteristic signal adopted in the step S103 is a linear regression model, and in practical application, other neural network models may also be adopted to implement the relation model, which is not limited in the present invention.
Regression analysis is a predictive modeling technique that studies the relationship between dependent variables (targets) and independent variables (predictors). This technique is commonly used for predictive analysis, time series modeling, and discovering causal relationships between variables. Data points are typically fitted using curves/lines, with the goal of minimizing the difference in curve-to-data point distances. Linear regression is one of the regression problems, linear regression assuming a linear correlation between target values and features, i.e., satisfying a multivariate linear equation. The parameters w and b when the loss function is minimum are solved by constructing the loss function.
Wherein the content of the first and second substances,
for the prediction value, the independent variable x is the timestamp, and the dependent variable y is the health index, which are known. What is desired to be achieved here is to predict what x is newly added, which corresponds to y; or adding a new y and predicting the corresponding x. Therefore, to establish this model relationship, the goal is to solve both w and b parameters in the above model by knowing the data points.
Solving the optimal parameters, the embodiment of the invention adopts a sampling least square method to solve w and b of the loss function and solve the partial derivatives, the solution only when the partial derivatives are 0 is the optimal solution, and the calculation mode is shown as formulas (3) and (4):
specifically, in an embodiment, the step S104 specifically includes the following steps:
step S401: and inputting the failure characteristic signal into a trained preset time and characteristic signal relation model to obtain a corresponding predicted failure moment.
Specifically, by inputting the failure characteristic signal into the above formula (1), the corresponding predicted failure time can be obtained, such as: xx years xx month xx days, etc.
Step S402: and determining a gear life prediction result of the target gear based on the time difference between the predicted failure time and the current time.
For example, assuming that the predicted failure time is 2025 years, 1 month and 1 day, and the current time is 2024 years, 1 month and 1 day, the corresponding gear life prediction result is that the predicted remaining life of the gear is 1 year.
Specifically, in an embodiment, the method for predicting the life of the gear specifically further includes the following steps:
step S105: and acquiring a side band energy signal of the current meshing frequency and harmonic thereof of the real-time vibration acquisition signal of the target gear corresponding to the current sampling moment.
Step S106: and inputting the current meshing frequency at the current sampling moment and the side band energy signal of the harmonic wave of the current meshing frequency to a trained preset time and characteristic signal relation model for optimizing the model parameters.
Specifically, since the residual life of the bearing and the gear is mainly determined by the factors such as load, rotating speed, temperature and lubrication, the characteristic signals of the rolling bearing and the gear are different from the change rule of the service time even if the rolling bearing and the gear are the same in model. The characteristic signal calculation comprises a time stamp of data reported by each data collector. In practical application, in order to update linear regression and reduce memory space and calculation amount in real time, the maximum number of data sets calculated each time is 1000 sample data, and in addition, vibration data reported each time can participate in calculation of linear regression to update model parameters. Therefore, the accuracy of the gear service life prediction result is further improved by optimizing the model parameters in real time.
The method for predicting the life of the gear provided by the embodiment of the invention will be described in detail with reference to specific application examples.
Illustratively, the gearbox vibrates at high frequency due to the meshing function between gears, and the more teeth engaged at any one time, the smoother the performance of the gearbox. Therefore, regardless of the presence of a fault or failure of the gear, the acquired vibration signal will have an amplitude in the frequency domain at the mesh frequency and its harmonic components. Fig. 2 and 3 are a meshing frequency spectrum diagram of a normal cylindrical gear and a meshing frequency spectrum diagram of a cylindrical gear with pitting failure, respectively. It can be known from the comparison between fig. 2 and fig. 3 that the gear has pitting failure, the amplitude at the meshing frequency is not changed greatly, and it is obvious that the energy of the edge band of the meshing frequency and its harmonic is enhanced, therefore, in the embodiment of the present invention, the energy of the edge band of the meshing frequency and its harmonic is used as the life prediction index of the gear, and the specific calculation process is as follows:
1, an acceleration sensor acquires an equal time interval time domain signal Acc (n) (namely a vibration signal of a gear), wherein the interval time is the reciprocal of sampling time, and the sampling time duration is 5 s;
2 Fourier transform (FFT) is carried out on the time domain signal to obtain frequency domain information Acc _ F (n)
3, solving the Energy of the side band Energy _ side of the Acc _ f (n) in the step 2, and recording the time stamp of the sampling signal reported by the data acquisition unit each time, wherein a calculation formula (5) is as follows:
wherein, Energy _ side represents the meshing frequency corresponding to the current vibration signal and the sideband Energy signal of the harmonic thereof, f represents the meshing frequency of the target gear, d is the sideband of the target gear, and Acc _ f (n) represents the frequency domain signal corresponding to the current vibration signal.
Inputting the calculated edge band energy of the meshing frequency and the harmonic thereof into a preset time and characteristic signal relation model for model training, and determining model parameters of the preset time and characteristic signal relation model, wherein the specific process parameters are described in the above embodiment and are not repeated herein. Finally, optimal parameter values w and b of the two model parameters of the linear model are determined.
Then, according to the current value and the empirical value of the normal operation of the gear, the corresponding meshing frequency and the sideband energy Threshold of the harmonic wave thereof under the condition of the gear failure are set, and then the residual life of the gear is calculated by using the following formula (6):
wherein RUL represents the current remaining life of the gear, and t represents0Representing the current time instant, w and b represent two model parameters of the linear model. It should be noted that, if the gear is maintained, repaired or replaced, the collected vibration signal needs to be cleared and the above process needs to be executed again to ensure the consistency of the life prediction result of the gear with the real application scenario.
The vibration signal of the meshing-type failure (wear, crack, damage, etc.) of the gear has a sideband energy of the meshing frequency. The embodiment of the invention takes the sideband energy as the health index of the gear according to the health state of the gear. Because the load, the rotating speed and other working conditions of each pair of gears are different, the meshing frequency side bands of each pair of gears need to be linearly regressed in real time. By setting the failure threshold of the sideband energy, the residual life can be calculated through the linear relation between the time and the index. Therefore, the residual life of the whole life cycle of the gear can be predicted in real time, and the predicted result is more fit for the real use condition.
By executing the steps, the gear life prediction method provided by the embodiment of the invention extracts the characteristic signal capable of reflecting the health state of the gear by using the vibration acquisition signal of the target gear in the using process, trains the preset time and characteristic signal relation model by using the relation between the characteristic signal and the sampling moment, and predicts the life condition of the gear in real time by using the corresponding failure characteristic signal when the gear fails and the current moment, so that the automatic and accurate prediction of the gear life is realized without depending on manual experience, the time is saved, the gear life prediction efficiency is improved, an accurate data base is provided for the overhaul and the replacement of the gear, and the safe operation of mechanical equipment provided with the gear is further ensured.
An embodiment of the present invention further provides a device for predicting a life of a gear, as shown in fig. 4, the device for predicting a life of a gear specifically includes:
the obtaining module 101 is configured to obtain a historical vibration collecting signal of the target gear and a current time. For details, refer to the related description of step S101 in the above method embodiment, and no further description is provided here.
The first processing module 102 is configured to calculate a characteristic signal of the target gear based on the historical vibration acquisition signal, where the characteristic signal includes a sideband energy signal of the meshing frequency and its harmonic at different sampling times. For details, refer to the related description of step S102 in the above method embodiment, and no further description is provided here.
The second processing module 103 is configured to input the feature signal to the preset time and feature signal relationship model, perform model training, and determine a model parameter of the preset time and feature signal relationship model. For details, refer to the related description of step S103 in the above method embodiment, and no further description is provided here.
And the third processing module 104 is configured to generate a gear life prediction result of the target gear based on a failure characteristic signal corresponding to a preset gear failure, a trained preset time and characteristic signal relation model, and a current time. For details, refer to the related description of step S104 in the above method embodiment, and no further description is provided here.
Further functional descriptions of the modules are the same as those of the corresponding method embodiments, and are not repeated herein.
Through the cooperative cooperation of the components, the gear life prediction device provided by the embodiment of the invention extracts the characteristic signal capable of reflecting the health state of the gear by using the vibration acquisition signal of the target gear in the use process, trains the preset time and characteristic signal relation model by using the relation between the characteristic signal and the sampling time, and predicts the life condition of the gear in real time by using the corresponding failure characteristic signal when the gear fails and the current time, so that the automatic and accurate prediction of the gear life is realized without depending on manual experience, the time is saved, the gear life prediction efficiency is improved, an accurate data base is provided for the overhaul and the replacement of the gear, and the safe operation of mechanical equipment provided with the gear is further ensured.
An embodiment of the present invention further provides an electronic device, as shown in fig. 5, the electronic device may include a processor 901 and a memory 902, where the processor 901 and the memory 902 may be connected by a bus or in another manner, and fig. 5 takes the connection by the bus as an example.
The memory 902, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the methods in the embodiments of the present invention. The processor 901 executes various functional applications and data processing of the processor, i.e., implements the above-described method, by executing non-transitory software programs, instructions, and modules stored in the memory 902.
The memory 902 may include a storage program area and a storage data area, wherein the storage program area may store an application program required for operating the device, at least one function; the storage data area may store data created by the processor 901, and the like. Further, the memory 902 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 902 may optionally include memory located remotely from the processor 901, which may be connected to the processor 901 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
One or more modules are stored in the memory 902, which when executed by the processor 901 performs the methods described above.
The specific details of the electronic device may be understood by referring to the corresponding related descriptions and effects in the above method embodiments, and are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, and the implemented program can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above.
The above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A method of predicting gear life, comprising:
acquiring a historical vibration acquisition signal and the current moment of a target gear;
calculating a characteristic signal of the target gear based on the historical vibration acquisition signal, wherein the characteristic signal comprises the meshing frequency at different sampling moments and a side-band energy signal of harmonic waves of the meshing frequency;
inputting the characteristic signal into a preset time and characteristic signal relation model, carrying out model training, and determining a model parameter of the preset time and characteristic signal relation model;
and generating a gear service life prediction result of the target gear based on a corresponding failure characteristic signal when the preset gear fails, a trained relation model of the preset time and the characteristic signal and the current moment.
2. The method of claim 1, wherein said calculating a signature signal of the target gear based on the historical vibration acquisition signal comprises:
sequentially extracting a current vibration signal and a current sampling moment corresponding to the current vibration signal according to sampling time from the historical vibration acquisition signal;
carrying out Fourier transform on the current vibration signal to obtain a corresponding frequency domain signal;
calculating a current characteristic signal corresponding to the current vibration signal based on the frequency domain signal;
and adding a mark of the current sampling moment to the current characteristic signal to obtain the current characteristic signal with the current sampling moment.
3. The method of claim 2, wherein the calculating a current feature signal corresponding to the current vibration signal based on the frequency domain signal comprises:
acquiring the meshing frequency and the sideband of the target gear;
and calculating the meshing frequency corresponding to the current vibration signal and the sideband energy signal of the harmonic wave of the current vibration signal based on the frequency domain signal, the meshing frequency of the target gear and the sideband.
4. The method of claim 1, wherein the predetermined time-to-signature relationship model is a linear regression model, and the predetermined time-to-signature relationship model is represented by the following formula:
y^＝wx+b，
wherein y ^ represents the predicted meshing frequency and the sideband energy signal of the harmonic wave thereof, x represents the current moment, and w and b represent two model parameters of the preset time and characteristic signal relation model.
5. The method according to claim 1, wherein the generating of the gear life prediction result of the target gear based on the failure characteristic signal corresponding to the preset gear failure, the trained relation model between the preset time and the characteristic signal and the current time comprises:
inputting the failure characteristic signal into a trained preset time and characteristic signal relation model to obtain a corresponding predicted failure moment;
and determining a gear life prediction result of the target gear based on the time difference between the predicted failure time and the current time.
6. The method of claim 1, further comprising:
acquiring a side band energy signal of a current meshing frequency and harmonic thereof of a real-time vibration acquisition signal of a target gear at a current sampling moment;
and inputting the current meshing frequency at the current sampling moment and the side band energy signal of the harmonic wave of the current meshing frequency to a trained preset time and characteristic signal relation model for optimizing model parameters.
7. The method of claim 3, wherein the sideband energy signal of the mesh frequency and its harmonics corresponding to the current vibration signal is calculated by the following formula:
the Energy _ side represents the meshing frequency corresponding to the current vibration signal and the sideband Energy signal of the harmonic thereof, f represents the meshing frequency of the target gear, d is the sideband of the target gear, and Acc _ f (n) represents the frequency domain signal corresponding to the current vibration signal.
8. A gear life prediction device, comprising:
the acquisition module is used for acquiring a historical vibration acquisition signal and the current moment of the target gear;
the first processing module is used for calculating a characteristic signal of the target gear based on the historical vibration acquisition signal, wherein the characteristic signal comprises the meshing frequency at different sampling moments and a side band energy signal of harmonic waves of the meshing frequency;
the second processing module is used for inputting the characteristic signal to a preset time and characteristic signal relation model, carrying out model training and determining a model parameter of the preset time and characteristic signal relation model;
and the third processing module is used for generating a gear service life prediction result of the target gear based on a corresponding failure characteristic signal when the preset gear fails, a trained relation model of preset time and the characteristic signal and the current moment.
9. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method of any one of claims 1-7.
10. An electronic device, comprising:
a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory having stored therein computer instructions, the processor being configured to execute the computer instructions to perform the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111490708.6A CN114154269A (en) | 2021-12-08 | 2021-12-08 | Gear service life prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111490708.6A CN114154269A (en) | 2021-12-08 | 2021-12-08 | Gear service life prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114154269A true CN114154269A (en) | 2022-03-08 |
Family
ID=80453717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111490708.6A Pending CN114154269A (en) | 2021-12-08 | 2021-12-08 | Gear service life prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114154269A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5210704A (en) * | 1990-10-02 | 1993-05-11 | Technology International Incorporated | System for prognosis and diagnostics of failure and wearout monitoring and for prediction of life expectancy of helicopter gearboxes and other rotating equipment |
| CN110175369A (en) * | 2019-04-30 | 2019-08-27 | 南京邮电大学 | A kind of gear method for predicting residual useful life based on two-dimensional convolution neural network |
| CN110987167A (en) * | 2019-12-17 | 2020-04-10 | 北京昊鹏智能技术有限公司 | Fault detection method, device, equipment and storage medium for rotary mechanical equipment |
-
2021
- 2021-12-08 CN CN202111490708.6A patent/CN114154269A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5210704A (en) * | 1990-10-02 | 1993-05-11 | Technology International Incorporated | System for prognosis and diagnostics of failure and wearout monitoring and for prediction of life expectancy of helicopter gearboxes and other rotating equipment |
| CN110175369A (en) * | 2019-04-30 | 2019-08-27 | 南京邮电大学 | A kind of gear method for predicting residual useful life based on two-dimensional convolution neural network |
| CN110987167A (en) * | 2019-12-17 | 2020-04-10 | 北京昊鹏智能技术有限公司 | Fault detection method, device, equipment and storage medium for rotary mechanical equipment |
Non-Patent Citations (3)
| Title |
|---|
| 孟宪源: "重载齿轮故障诊断技术的应用研究", 《重型机械》 * |
| 石慧等: "考虑突变状态检测的齿轮实时剩余寿命预测", 《振动与冲击》 * |
| 苗强等: "旋转设备在线健康监控指数研究", 《电子科技大学学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Igba et al. | Analysing RMS and peak values of vibration signals for condition monitoring of wind turbine gearboxes | |
| Tavner et al. | Study of weather and location effects on wind turbine failure rates | |
| CN101532911B (en) | Large steam turbine-generator set rotor crack fault real-time diagnosis method | |
| CN104048825B (en) | A kind of gearbox of wind turbine Fault Locating Method of Multi-sensor Fusion | |
| CN113221280A (en) | Rolling bearing modeling and model updating method and system based on digital twinning | |
| CN110688617B (en) | Fan vibration abnormity detection method and device | |
| CN111191191B (en) | Construction method of combined model for accurately predicting deformation effect of concrete dam | |
| Bangalore et al. | An approach for self evolving neural network based algorithm for fault prognosis in wind turbine | |
| CN105205569A (en) | Draught fan gear box state on-line evaluation model building method and on-line evaluation method | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| WO2023065580A1 (en) | Fault diagnosis method and apparatus for gearbox of wind turbine generator set | |
| Jaramillo et al. | A Bayesian approach for fatigue damage diagnosis and prognosis of wind turbine blades | |
| Kim et al. | Design of wind turbine fault detection system based on performance curve | |
| CN115617606A (en) | Equipment monitoring method and system, electronic equipment and storage medium | |
| CN116629136A (en) | Method, device, equipment and storage medium for updating digital twin model | |
| CN116415126A (en) | Method, device and computing equipment for anomaly detection of doctor blades of paper machine | |
| CN115186701A (en) | Bearing life prediction method, device, electronic device and storage medium | |
| Fan et al. | A particle-filtering approach for remaining useful life estimation of wind turbine gearbox | |
| CN114154269A (en) | Gear service life prediction method and device | |
| Tchakoua et al. | New trends and future challenges for wind turbines condition monitoring | |
| Cao et al. | Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator | |
| CN112486136B (en) | Fault early warning system and method | |
| Martin et al. | Automated machine health monitoring at an expert level | |
| Ni et al. | An adaptive state-space model for predicting remaining useful life of planetary gearbox | |
| Zhang | Comparison of data-driven and model-based methodologies of wind turbine fault detection with SCADA data |
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 | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20231110 |
|
| AD01 | Patent right deemed abandoned |