CN114580282A - Engineering machinery service life determining method, device, equipment and storage medium - Google Patents
Engineering machinery service life determining method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114580282A CN114580282A CN202210209504.9A CN202210209504A CN114580282A CN 114580282 A CN114580282 A CN 114580282A CN 202210209504 A CN202210209504 A CN 202210209504A CN 114580282 A CN114580282 A CN 114580282A
- Authority
- CN
- China
- Prior art keywords
- machine learning
- learning model
- life
- target component
- engineering machinery
- 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
Images
Classifications
-
- 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
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- 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
-
- 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
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the technical field of computers, and provides a method, a device, equipment and a storage medium for determining the service life of engineering machinery, wherein the method comprises the following steps: acquiring real-time load information of the engineering machinery; inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component; based on the fatigue life of the target component, the remaining life of the target component is determined. Therefore, the stress value of each dangerous point can be obtained through the real-time load information and the first machine learning model, the residual life of the target component is rapidly obtained through the second machine learning model, the problem that a sensor is pasted at each dangerous point on the engineering machinery to conduct real-time collection is avoided, the method is rapid and accurate, the implementation is simple, and the cost is reduced.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method, a device, equipment and a storage medium for determining the service life of engineering machinery.
Background
At present, many engineering machines have severe working environment, high working level, complex structure and higher safety requirement. However, as the service life of the engineering machinery continues, the service lives of all parts and the like of the engineering machinery inevitably deteriorate, and accordingly, the service life of the whole engineering machinery is reduced, so that normal operation is affected, and potential safety hazards are caused. Timely repairing and replacing of each part is an effective way for prolonging the service life of the parts and further prolonging the service life of the whole machine. For this reason, the determination of the service life of the construction machine is a very important link, wherein the determination of the service life of the construction machine includes the determination of the service life of a component of the construction machine, and further includes the determination of the service life of the whole machine affected by the service life of the component. However, how to evaluate the condition of each component to monitor the service life of each component to determine the optimal admission period for repairing and replacing each component is a key problem.
In the prior art, sensors are mainly adhered to dangerous points of engineering machinery, real-time stress values and strain values of the dangerous points of the engineering machinery in a running state are collected, then accumulated damage of the vibration dangerous points is calculated according to a damage accumulation method, and the service lives of corresponding parts are monitored. However, the working environment of the engineering machinery is severe, the accuracy of the measurement result of the sensor is low, the measurement operation is complex, and the cost is high.
Disclosure of Invention
The invention provides a method, a device, equipment and a storage medium for determining the service life of an engineering machine, which are used for solving the defects that in the prior art, when the service life of a corresponding part of the engineering machine is monitored, because the working environment of the engineering machine is severe, a sensor stuck at a dangerous point has lower accuracy on the measurement results of a real-time stress value and a strain value, the measurement operation is complex, the cost is high, the residual service life of a target part is quickly and accurately obtained, the realization is simple, and the cost is reduced.
The invention provides a method for determining the service life of engineering machinery, which comprises the following steps:
acquiring real-time load information of the engineering machinery;
inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points;
inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component; the second machine learning model is obtained by training based on stress value samples of all the dangerous points and fatigue life samples of all the target components;
determining a remaining life of the target component based on the fatigue life of the target component.
According to the engineering machinery service life determining method provided by the invention, the first machine learning model is obtained by the following method:
constructing a finite element model of the engineering machinery;
inputting the load information sample into the finite element model to obtain a first stress value of each node of each part in the engineering machinery;
if the node is the dangerous point, taking a first stress value of the node as a stress value sample of the dangerous point;
and training a first initial machine learning model based on the load information samples and the stress value samples of all the dangerous points to obtain the first machine learning model.
According to the method for determining the service life of the engineering machinery, provided by the invention, the finite element model is constructed in the following way:
constructing an initial finite element model of the engineering machinery;
inputting preset load information into the initial finite element model, and outputting a second stress value and a second strain value of the node;
if the second stress value of the node is larger than or equal to a first threshold value and/or the second strain value is larger than or equal to a second threshold value, determining that the node is the dangerous point and the part where the node is located is the target part;
collecting an actual stress value and an actual strain value of the dangerous point;
and optimizing the initial finite element model based on the actual stress value and the actual strain value of the dangerous point to obtain the finite element model.
According to the engineering machinery service life determining method provided by the invention, the second machine learning model is obtained by the following method:
inputting the stress value samples of all the dangerous points into a fatigue model, and calculating to obtain a fatigue life sample of each target component;
and training a second initial machine learning model based on the stress value samples of all the dangerous points and the fatigue life samples of all the target components to obtain the second machine learning model.
According to the engineering machinery service life determining method provided by the invention, the method further comprises the following steps:
and when the residual life of the target component is less than or equal to a third threshold value, sending out prompt information for prompting the replacement of the target component.
According to the method for determining the service life of the engineering machinery, provided by the invention, the first machine learning model and the second machine learning model are the first machine learning model and the second machine learning model corresponding to the current working condition of the engineering machinery; before the real-time load information is input into the first machine learning model and the stress value of each dangerous point of each target component in the engineering machinery is obtained, the method further comprises the following steps:
and determining the current working condition of the engineering machinery according to the real-time working condition information of the engineering machinery.
According to the engineering machinery service life determining method provided by the invention, the method further comprises the following steps:
and taking the minimum value of the residual lives of all the target components as the residual life of the whole engineering machine.
The invention also provides an engineering machinery service life determining device, which comprises:
the load information acquisition module is used for acquiring real-time load information of the engineering machinery;
the stress obtaining module is used for inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points;
the fatigue obtaining module is used for inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component; the second machine learning model is obtained by training based on stress value samples of all the dangerous points and fatigue life samples of all the target components;
a life determination module to determine a remaining life of the target component based on the fatigue life of the target component.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the engineering machinery service life determining method.
The present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of determining a life of a construction machine as described in any one of the above.
According to the method for determining the service life of the engineering machinery, the real-time load information of the engineering machinery is acquired and input into the first machine learning model to obtain the stress value of each dangerous point of each target component in the engineering machinery, then the stress values of all the dangerous points are input into the second machine learning model to obtain the fatigue life of each target component, and further the residual life of the target component is quickly obtained.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed to be 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 it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for determining the life of a construction machine according to the present invention;
FIG. 2 is a second schematic flow chart of a method for determining the life of a construction machine according to the present invention;
FIG. 3 is a third schematic flow chart of a method for determining the life of a construction machine according to the present invention;
FIG. 4 is a schematic structural diagram of a life determination device for construction machinery provided by the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious 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.
The method for determining the life of the construction machine according to the present invention will be described with reference to fig. 1 to 3.
The method for determining the service life of the engineering machine provided by the embodiment of the invention can be executed by the engineering machine or software and/or hardware in the engineering machine, and can also be executed by a server or software and/or hardware in the server. The server may be a cloud server or a physical server. The engineering machinery can be a crane, exemplarily, an automobile crane, a crawler crane, a tower crane or the like, and also can be a pump truck, a rotary drilling rig, an ascending vehicle or the like.
Fig. 1 is a schematic flow chart of a construction machine determination method according to the present invention.
As shown in fig. 1, the method for determining the service life of the engineering machine provided by this embodiment at least includes:
In practical application, a data acquisition system is arranged in the engineering machine, and the data acquisition system is used for acquiring various parameters of the engineering machine, such as load information. If the work machine is a crane, the load information may be information that is indicative of the mass of the weight lifted by the work machine, such as a force signal required to lift the weight. If the method for determining the service life of the engineering machinery is applied to the engineering machinery, the real-time load information of the engineering machinery can be directly acquired through the data acquisition system. If the method for determining the service life of the engineering machine is applied to a server, the engineering machine can upload real-time load information acquired by a data acquisition system to the server, such as a cloud server, namely a cloud platform. Therefore, the server can obtain the real-time load information of the engineering machinery.
102, inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points.
The target component herein refers to a component having a dangerous point among components of the construction machine. The risk point is a node where damage such as a crack is likely to occur in the component. There may be one or more points of danger in the target component.
The number of stress cycles required for a material to fail fatigue is referred to as the fatigue life. Thus, the second machine learning model is able to output the number of stress cycles for each target component.
And 104, determining the residual life of the target component based on the fatigue life of the target component.
Specifically, the number of stress cycles of the target member may be directly used as the remaining life of the target member, or the remaining life of the target member may be determined based on the ratio of the number of stress cycles of the target member to the number of stress cycles per unit time. The unit time may be years, etc. Illustratively, the number of stress cycles of the target component is 10000, and the number of stress cycles per year is 1000, then the remaining life of the target component is 10000/1000-10 years.
The digital twin is a conceptual system of interaction of the physical world and the digital space, and is a digital twin technology which presents real world physical entities in a virtual digital form. In this embodiment, the first machine learning model and the second machine learning model are models in a digital space, and both the first machine learning model and the second machine learning model can obtain a stress value and a fatigue life corresponding to a dangerous point of a target component of a physical entity of the construction machine with respect to a dangerous point of the target component of the physical entity of the construction machine, and the first machine learning model and the second machine learning model can be considered to constitute a digital twin of the construction machine in the digital space. Referring to fig. 2, for example, the cloud platform inputs the acquired real-time load information into the digital twin and outputs the fatigue life of the target component.
In this embodiment, through obtaining the real-time load information of the engineering machine, input to the first machine learning model, obtain the stress value of each dangerous point of each target part in the engineering machine, then, input the stress value of all dangerous points to the second machine learning model, obtain the fatigue life of each target part, and then obtain the remaining life of the target part fast, so, just can obtain the stress value of each dangerous point through real-time load information and first machine learning model, need not to paste the sensor in every dangerous point on the engineering machine and carry out real-time collection, not only fast accurate, and realize simply, the cost is reduced.
As described above, the service life determination of the construction machine may further include, in addition to the service life determination of the target component, the service life determination of the whole construction machine, and based on this, the method for determining the service life of the construction machine according to this embodiment may further include: and taking the minimum value of the residual lives of all the target components as the residual life of the whole engineering machine. In practical application, the danger points of the parts affecting the service life of the whole engineering machine can be mainly focused, correspondingly, the target parts can be the parts affecting the service life of the whole engineering machine, so that the service life of the whole engineering machine is ended when the service life of one target part in all the target parts of the engineering machine is ended, and on the basis, the residual service life of the target part with the minimum residual service life in all the target parts can be used as the residual service life of the whole engineering machine, so that the residual service life of the whole engineering machine can be accurately obtained.
Based on the above embodiment, the first machine learning model and the second machine learning model are the first machine learning model and the second machine learning model corresponding to the current working condition of the engineering machine. In practical application, the first machine learning model and the second machine learning model corresponding to each working condition can be trained respectively according to different working conditions of the engineering machine, such as dynamic compaction and suspension working conditions.
Based on this, before the real-time load information is input to the first machine learning model and the stress value of each dangerous point of each target component in the engineering machine is obtained, the method for determining the service life of the engineering machine according to the embodiment may further include: and determining the current working condition of the engineering machinery according to the real-time working condition information of the engineering machinery. For example, if the engineering machine is a crane, the real-time operating condition information may include real-time load information and the like, and may further include boom combination information, a working radius, a ground gradient and the like. The arm support combination information is information input by a user. In actual practice, input controls may be provided. The user can input working condition information such as arm support combination information in advance through the input control. Based on the method, the real-time load information is input into a first machine learning model corresponding to the current working condition, and the stress values of all the dangerous points are input into a second machine learning model corresponding to the current working condition. Therefore, the first machine learning model and the second machine learning model are matched with specific working conditions, and the finally obtained residual life of the target component is more accurate.
Based on the above embodiment, the first machine learning model is obtained by:
firstly, constructing a finite element model of the engineering machinery.
In practical application, a finite element model corresponding to each working condition of the engineering machine can be constructed. The finite element model here is a three-dimensional simulation model.
The finite element model is used for carrying out transient dynamics analysis on the current working condition of the engineering machinery by a finite element analysis method based on the load information to obtain the stress value and the strain value of each node of each part of the engineering machinery.
The finite element analysis method is an analysis method using a finite element method, also called a finite element method, for analyzing a static or dynamic physical object or physical system, and is a numerical technique for solving an approximate solution of an edge value problem of a partial differential equation. In this method, an object or system is decomposed into a geometric model consisting of a plurality of interconnected simple parts, the number of which is limited and is therefore referred to as a finite element.
Transient dynamics analysis is a method for determining the dynamic response of a structure subjected to an arbitrary time-varying load, and is capable of outputting time-varying stress values, strain values, and the like.
In addition, the finite element model also needs to input the working environment (such as ground gradient) of the current working condition, the weight center of gravity of the whole machine, action planning and the like.
And secondly, inputting the load information sample into the finite element model to obtain a first stress value of each node of each part in the engineering machinery.
In practical application, the rated load information of the construction machine can be used as a load information sample, and certainly, more load information samples can be constructed based on the rated load information. And inputting the load information sample into the finite element model, and outputting a first stress value and a first strain value of each node of each part in the engineering machine. Wherein the stress values of each node form a stress cloud.
And thirdly, if the node is the dangerous point, taking the first stress value of the node as a stress value sample of the dangerous point.
And fourthly, training a first initial machine learning model based on the load information samples and the stress value samples of all the dangerous points to obtain the first machine learning model.
Specifically, the load information samples may be input to a first initial machine learning model to obtain predicted stress values of all dangerous points, a first loss function is determined based on the predicted stress values of the dangerous points and the stress value samples, and when the first loss function is less than or equal to a first preset value, training is stopped to obtain the first machine learning model.
The first initial machine learning model may be a neural network model. The specific value of the first preset value may be set according to actual conditions, and is not specifically limited herein.
In practical application, referring to fig. 3, for example, a curve of a load information sample of a dangerous point over time and a curve of a stress value sample of the dangerous point over time are introduced into an ansys Twinbuilder, and the finite element model is reduced in order to obtain a reduced-order model of the finite element model, i.e., a first machine learning model, which is a one-dimensional simulation model. The ansys twinbuild is finite element analysis software which can automatically learn by a machine under the set parameters and reduce a three-dimensional simulation model into a one-dimensional simulation model.
In this embodiment, the input and the output of the finite element model are utilized, and a machine learning technology is combined to train to obtain a first machine learning model, that is, a finite element model is reduced, and the first machine learning model is a reduced model of the finite element model. Although the finite element model can also accurately calculate the stress value of the dangerous point, as mentioned above, the essence of the finite element analysis is to solve the partial differential equation, the high-order partial differential equation only has an approximate solution, and the process of solving the approximate solution needs continuous iteration, which needs a lot of time and is not suitable for real-time calculation.
Illustratively, the finite element model is constructed by:
firstly, constructing an initial finite element model of the engineering machinery.
And secondly, inputting preset load information into the initial finite element model, and outputting a second stress value and a second strain value of the node.
And thirdly, if the second stress value of the node is larger than or equal to a first threshold value and/or the second strain value is larger than or equal to a second threshold value, determining that the node is the dangerous point and the part where the node is located is the target part.
In this step, the nodes with the second stress value greater than or equal to the first threshold and/or the second strain value greater than or equal to the second threshold are selected as the dangerous points, so that more dangerous points can be covered, that is, the dangerous points are covered more completely.
Specific values of the first threshold and the second threshold may be set according to actual situations, and are not specifically limited herein.
And fourthly, collecting an actual stress value and an actual strain value of the dangerous point.
In practical application, a sensor, such as a strain gauge, may be arranged at each dangerous point of the engineering machine in advance, and an actual stress value and an actual strain value of each dangerous point may be acquired. The dangerous point conditions of the engineering machinery of the same model are similar, so that the actual stress value and the actual strain value of each dangerous point can be acquired once for one engineering machinery in the engineering machinery of the same model, and the acquired acquisition result can be universal in the engineering machinery of the same model.
And fifthly, optimizing the initial finite element model based on the actual stress value and the actual strain value of the dangerous point to obtain the finite element model.
The initial finite element model may be optimized by selecting different unit types, unit sizes, connection modes between components, modeling of weld joints, and other existing modes, and reference may be made to related technologies, which are not described herein again. And the stress value and the strain value of the output dangerous point of the optimized finite element model are closer to the actual stress value and the actual strain value.
In the embodiment, the dangerous points are determined by using the stress values and the strain values of the nodes output by the constructed initial finite element model, more dangerous points can be covered, the method is more comprehensive, and the initial finite element model is optimized by using the collected actual stress values and the collected actual strain values of the dangerous points, so that a more accurate finite element model is obtained.
Based on the above embodiment, the second machine learning model is obtained by:
firstly, inputting the stress value samples of all the dangerous points into a fatigue model, and calculating to obtain a fatigue life sample of each target component.
And the fatigue model also adopts a finite element method, calculates a fatigue life sample of the target component where the danger point is located based on the stress value sample of the danger point, and if the danger point of the target component is multiple, takes the minimum value in the fatigue life samples corresponding to the danger points as the final fatigue life sample of the target component. Based on the stress value samples of all the dangerous points are input into a fatigue model, and fatigue life samples of each target component are output.
In practical application, the fatigue evaluation method can be realized by adopting the existing fatigue analysis software, wherein a fatigue model is constructed in the fatigue analysis software, the stress values of all dangerous points are read into the fatigue analysis software, the SN curve of the material of each target component is input, the stress time history for fatigue evaluation is determined, the fatigue solving parameters are set for fatigue solving, and the fatigue life of the target component can be output.
The fatigue properties of a material are described in terms of the relation between the stress range S applied and the lifetime N to failure, i.e. the S (stress) -N (lifetime) curve.
And secondly, training a second initial machine learning model based on the stress value samples of all the dangerous points and the fatigue life samples of all the target components to obtain the second machine learning model.
Specifically, the stress value samples of all the dangerous points may be input to a second initial machine learning model to obtain predicted fatigue lives of all the target components, a second loss function is determined based on the predicted fatigue lives and the fatigue life samples of the target components, and when the second loss function is less than or equal to a second preset value, the training is stopped to obtain the second machine learning model.
The second initial machine learning model may be a neural network model. The specific value of the second preset value may be set according to actual conditions, and is not specifically limited herein.
In this embodiment, a second machine learning model, that is, a reduced-order model of the fatigue model, is obtained by training using the input and output of the fatigue model and combining a machine learning technique. Although the fatigue model can also accurately calculate the fatigue life of the target component, the calculation process also needs a lot of time and is not suitable for real-time calculation, and the fatigue life of the target component can be quickly obtained by the second machine learning model obtained by reducing the fatigue model, so that the real-time requirement for determining the residual life of the target component is met. In the implementation, when the accuracy is reduced to less than 10%, the calculation speed is reduced from 30 hours to within 3 seconds.
Based on the above embodiment, the method for determining the service life of the engineering machine provided by this embodiment may further include: and when the residual life of the target component is less than or equal to a third threshold value, sending out prompt information for prompting the replacement of the target component. In practical applications, the specific value of the third threshold may be set according to actual needs, and is not specifically limited herein. Specifically, the prompt information can be sent to the engineering machinery so that the operator can conveniently check the prompt information. In the embodiment, when the residual life of the target component is short, the target component is prompted to be replaced, and early warning is realized so as to facilitate timely treatment.
The invention provides a method for determining the service life of an engineering machine based on digital twins, which comprises the steps of analyzing all working conditions of normal work through finite elements to further determine the positions of a target component and a dangerous point, acquiring a stress value and a strain value at the dangerous point based on a distributed data acquisition system of a sensor, optimizing a finite element model, inputting the result of the optimized finite element model into fatigue analysis software, carrying out fatigue analysis, reducing the order of the finite element model and the fatigue model through machine learning to construct a digital twinbody, acquiring a force signal representing load information based on a data acquisition system of a physical entity of the engineering machine to realize real-time output of the fatigue life of the dangerous point, and prompting a user to replace the target component when the fatigue life of the output target component is smaller than or equal to a third threshold value. And taking the minimum value of the residual lives of all target components as the residual life of the whole engineering machine.
The following describes the device for determining the service life of the construction machine according to the present invention, and the device for determining the service life of the construction machine described below and the method for determining the service life of the construction machine described above may be referred to in correspondence with each other.
Fig. 4 is a schematic structural diagram of a service life determination device for construction machinery according to the present invention.
As shown in fig. 4, the present embodiment provides an engineering machine life determining apparatus, including:
a load information obtaining module 401, configured to obtain real-time load information of the engineering machine;
a stress obtaining module 402, configured to input the real-time load information into a first machine learning model, to obtain a stress value of each dangerous point of each target component in the engineering machine; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points;
a fatigue obtaining module 403, configured to input the stress values of all the dangerous points into a second machine learning model, so as to obtain a fatigue life of each target component; the second machine learning model is obtained by training based on stress value samples of all the dangerous points and fatigue life samples of all the target components;
a life determination module 404 for determining a remaining life of the target component based on the fatigue life of the target component.
Based on the above embodiment, the first machine learning model is obtained by:
constructing a finite element model of the engineering machinery;
inputting the load information sample into the finite element model to obtain a first stress value of each node of each part in the engineering machinery;
if the node is the dangerous point, taking a first stress value of the node as a stress value sample of the dangerous point;
and training a first initial machine learning model based on the load information samples and the stress value samples of all the dangerous points to obtain the first machine learning model.
Based on the above embodiment, the finite element model is constructed by:
constructing an initial finite element model of the engineering machinery;
inputting preset load information into the initial finite element model, and outputting a second stress value and a second strain value of the node;
if the second stress value of the node is larger than or equal to a first threshold value and/or the second strain value is larger than or equal to a second threshold value, determining that the node is the dangerous point and the part where the node is located is the target part;
collecting an actual stress value and an actual strain value of the dangerous point;
and optimizing the initial finite element model based on the actual stress value and the actual strain value of the dangerous point to obtain the finite element model.
Based on the above embodiment, the second machine learning model is obtained by:
inputting the stress value samples of all the dangerous points into a fatigue model, and calculating to obtain a fatigue life sample of each target component;
and training a second initial machine learning model based on the stress value samples of all the dangerous points and the fatigue life samples of all the target components to obtain the second machine learning model.
Based on the above embodiment, further include:
and the service life prompting module is used for sending out prompting information for prompting the target component to be replaced when the residual service life of the target component is less than or equal to a third threshold value.
Based on the above embodiment, the first machine learning model and the second machine learning model are the first machine learning model and the second machine learning model corresponding to the current working condition of the engineering machine; further comprising:
and the working condition determining module is used for determining the current working condition of the engineering machinery according to the real-time working condition information of the engineering machinery.
Based on the above embodiment, the lifetime determination module is further configured to:
and taking the minimum value of the residual lives of all the target components as the residual life of the whole engineering machine.
Fig. 5 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 5: a processor (processor)510, a communication Interface (Communications Interface)520, a memory (memory)530 and a communication bus 540, wherein the processor 510, the communication Interface 520 and the memory 530 communicate with each other via the communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a method of engineering machine life determination, the method comprising:
acquiring real-time load information of the engineering machinery;
inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points;
inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component; the second machine learning model is obtained by training based on stress value samples of all the dangerous points and fatigue life samples of all the target components;
determining a remaining life of the target component based on the fatigue life of the target component.
Furthermore, the logic instructions in the memory 530 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, which when executed by a computer, enable the computer to perform the method for determining the life of a work machine provided by the above methods, the method comprising:
acquiring real-time load information of the engineering machinery;
inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points;
inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component; the second machine learning model is obtained by training based on stress value samples of all the dangerous points and fatigue life samples of all the target components;
determining a remaining life of the target component based on the fatigue life of the target component.
In still another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, the computer program being implemented by a processor to perform the method for determining the life of a construction machine provided by the above methods, the method including:
acquiring real-time load information of the engineering machinery;
inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery; the first machine learning model is obtained by training based on load information samples and stress value samples of all the dangerous points;
inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component; the second machine learning model is obtained by training based on stress value samples of all the dangerous points and fatigue life samples of all the target components;
determining a remaining life of the target component based on the fatigue life of the target component.
The above-described embodiments of the apparatus are merely illustrative, and the 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 modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment may be implemented by software plus a necessary general hardware platform, and may also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but 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; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for determining the service life of engineering machinery is characterized by comprising the following steps:
acquiring real-time load information of the engineering machinery;
inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery;
inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component;
determining a remaining life of the target component based on the fatigue life of the target component.
2. The method of claim 1, wherein the first machine learning model is obtained by:
constructing a finite element model of the engineering machinery;
inputting the load information sample into the finite element model to obtain a first stress value of each node of each part in the engineering machinery;
if the node is the dangerous point, taking a first stress value of the node as a stress value sample of the dangerous point;
and training a first initial machine learning model based on the load information samples and the stress value samples of all the dangerous points to obtain the first machine learning model.
3. The method of claim 2, wherein the finite element model is constructed by:
constructing an initial finite element model of the engineering machinery;
inputting preset load information into the initial finite element model, and outputting a second stress value and a second strain value of the node;
if the second stress value of the node is larger than or equal to a first threshold value and/or the second strain value is larger than or equal to a second threshold value, determining that the node is the dangerous point and the part where the node is located is the target part;
collecting an actual stress value and an actual strain value of the dangerous point;
and optimizing the initial finite element model based on the actual stress value and the actual strain value of the dangerous point to obtain the finite element model.
4. The method of claim 1, wherein the second machine learning model is obtained by:
inputting the stress value samples of all the dangerous points into a fatigue model, and calculating to obtain a fatigue life sample of each target component;
and training a second initial machine learning model based on the stress value samples of all the dangerous points and the fatigue life samples of all the target components to obtain the second machine learning model.
5. The method for determining the life of a construction machine according to claim 1, further comprising:
and when the residual life of the target component is less than or equal to a third threshold value, sending out prompt information for prompting the replacement of the target component.
6. The method for determining the service life of the engineering machine according to claim 1, wherein the first machine learning model and the second machine learning model are the first machine learning model and the second machine learning model corresponding to the current working condition of the engineering machine; before the real-time load information is input into the first machine learning model and the stress value of each dangerous point of each target component in the engineering machinery is obtained, the method further comprises the following steps:
and determining the current working condition of the engineering machinery according to the real-time working condition information of the engineering machinery.
7. The method for determining the life of a construction machine according to any one of claims 1 to 6, further comprising:
and taking the minimum value of the residual lives of all the target components as the residual life of the whole engineering machine.
8. An apparatus for determining a life of a construction machine, comprising:
the load information acquisition module is used for acquiring real-time load information of the engineering machinery;
the stress obtaining module is used for inputting the real-time load information into a first machine learning model to obtain a stress value of each dangerous point of each target component in the engineering machinery;
the fatigue obtaining module is used for inputting the stress values of all the dangerous points into a second machine learning model to obtain the fatigue life of each target component;
a life determination module to determine a remaining life of the target component based on the fatigue life of the target component.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for determining the life of a construction machine according to any one of claims 1 to 7 when executing the program.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method for determining the life of a working machine according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210209504.9A CN114580282A (en) | 2022-03-04 | 2022-03-04 | Engineering machinery service life determining method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210209504.9A CN114580282A (en) | 2022-03-04 | 2022-03-04 | Engineering machinery service life determining method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114580282A true CN114580282A (en) | 2022-06-03 |
Family
ID=81773878
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210209504.9A Pending CN114580282A (en) | 2022-03-04 | 2022-03-04 | Engineering machinery service life determining method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114580282A (en) |
-
2022
- 2022-03-04 CN CN202210209504.9A patent/CN114580282A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111737811B (en) | Helicopter movable part service life management method, device and medium based on digital twin | |
| CN111435557B (en) | Fault detection device for detecting machine part problems | |
| US20190102494A1 (en) | System for tracking incremental damage accumulation | |
| CN104217094B (en) | For calculating the fatigue of structure and the method for fatigue rupture | |
| JP5887405B2 (en) | Stress calculation system for work machines | |
| US10436669B2 (en) | Interpolation engine for analysis of time-varying load data signals | |
| CN107092735A (en) | A kind of bridge fatigue state appraisal procedure | |
| KR101653116B1 (en) | Strain estimation system, global and local safety evaluation system, and method for structure through impact hammer test | |
| CN111091310B (en) | Excavation equipment health monitoring system and method | |
| CN114772466A (en) | Crane stability analysis method, crane anti-overturning control method and device | |
| CN112364550B (en) | Fatigue life prediction method and device for thin-wall welding structure | |
| CN111583067A (en) | Urban underground large space construction safety early warning and emergency decision-making method and system | |
| CN116049989A (en) | Bogie digital twin model construction method, system, electronic equipment and medium | |
| CN102252861A (en) | Progressive method for identifying support generalized displacement of damaged cable based on space coordinate monitoring | |
| CN112801199B (en) | Excavator movable arm service life prediction method and device, electronic equipment and storage medium | |
| US10288530B2 (en) | Method for improving severity estimates | |
| CN114580282A (en) | Engineering machinery service life determining method, device, equipment and storage medium | |
| CN110955933B (en) | Mechanical structure fuzzy fatigue reliability calculation method based on response surface method | |
| CN112948925B (en) | Bridge health state evaluation method, system and storage medium | |
| CN113515848A (en) | Method and system for optimizing structural design of mining equipment and readable storage medium | |
| CN113651245A (en) | Crane bearing capacity monitoring system | |
| CN113267217A (en) | Bridge group monitoring system and bridge group monitoring method | |
| CN102305723A (en) | Health monitoring method based on mixed monitoring and identification of damaged cable and generalized displacement of support | |
| CN113282988B (en) | Method and system for evaluating state of superstructure of large-span bridge | |
| Stamboliska et al. | FEM in Proactive Condition Monitoring |
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 |