CN113239495A - Complex structure reliability design method based on vector hybrid agent model - Google Patents
Complex structure reliability design method based on vector hybrid agent model Download PDFInfo
- Publication number
- CN113239495A CN113239495A CN202110603383.1A CN202110603383A CN113239495A CN 113239495 A CN113239495 A CN 113239495A CN 202110603383 A CN202110603383 A CN 202110603383A CN 113239495 A CN113239495 A CN 113239495A
- Authority
- CN
- China
- Prior art keywords
- model
- reliability design
- parameters
- complex structure
- target
- 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]
- G06F30/10—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
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- 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
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a complex structure reliability design method based on a vector hybrid agent model, which comprises the following steps: step 1, obtaining a reliability design target of a complex structure; step 2, acquiring input parameters and response parameters of each reliability design target and using the input parameters and the response parameters as input parameters and response parameters of a complex structure; step 3, performing series-parallel sampling on the input parameters and the response parameters of the complex structure to obtain input parameter samples and response parameter samples; step 4, selecting an agent model for each reliability design target; step 5, training an agent model by using an input parameter sample and a response parameter sample, and obtaining an optimal model parameter by using a self-adaptive modeling method; step 6, establishing an optimal parameter matrix according to the optimal model parameters and establishing a vector hybrid agent model; step 7, evaluating the precision of the vector hybrid agent model and correcting the precision; and 8, performing multi-target reliability design on the complex structure according to the vector hybrid agent model.
Description
Technical Field
The invention relates to a complex structure reliability design method, in particular to a complex structure reliability design method based on a vector hybrid agent model.
Background
Mechanical systems often consist of multiple components, such as a gas turbine engine rotor structural system consisting of a main shaft, a wheel disc and blades, and work in complex multi-physical fields, subject to loads such as vibration, structural, pneumatic, thermal and the like. Even though a single component tends to involve multiple failure modes under multiple loads, turbine blade failure, for example, often involves analysis of multiple failure modes such as stress, strain, deformation, fatigue, and the like. Therefore, the optimization and reliability design of the structure generally relates to a multi-objective and multi-disciplinary complex analysis problem, the essence of which is a multi-objective design problem, and the method is widely applied to a plurality of structural optimization and design in the engineering fields of aviation, aerospace, navigation, industry, civil engineering and the like.
The structure multi-objective optimization and the reliability design also relate to the problems of multivariable parameters (hyper-parameters), high nonlinearity (high nonlinearity) between an analysis target and input parameters, large-scale optimization parameters, a large amount of loop iteration and the like, and directly influence the precision problem of the engineering structure optimization and the reliability design. Research shows that it is a common practice to adopt an agent model to realize multi-objective optimization and reliability design. However, the effectiveness of structural design analysis is directly related to the design analysis of each design target and is determined by the result of the single target analysis. Due to obvious defects in the process of the problems of coupling among design parameters, correlation among analysis targets, cooperativity of analysis results and the like, the precision and the efficiency of the overall 'large' model design analysis are seriously influenced by the multi-objective optimization and the reliability design based on the agent model. The concrete points are as follows: (1) compared with the analysis of an integral model, the single design analysis of a single target generally only relates to partial design parameters of the integral system, the relation between the single target parameters and the integral parameters is not effectively considered during the analysis, and the coupling and the correlation between the design parameters are weakened, so that the problem of the single target design precision is caused, and the design precision of the integral system is influenced; (2) the analysis of a single target cannot effectively consider the correlation between design targets and the correlation between established model parameters, the cooperativity (synchronism) of the design process of each target is difficult to achieve, and the design precision of the whole system is greatly influenced; (3) according to the method, each design target needs to be sampled, modeled and simulated respectively, and then the simulation result is subjected to collaborative analysis to process the whole structure design, the whole design process is carried out independently, and the whole structure design is relatively complicated, long in consumed time and low in design efficiency. (4) Due to different characteristics of each single target analysis, the aspects of parameter scale, nonlinear degree and the like of the analysis design of the single target analysis are different. If the same agent model is used to realize the design analysis of each target, the design analysis is sometimes unreasonable. Therefore, the above 4 aspects seriously affect the effectiveness of structural system design analysis and need to be solved urgently.
Disclosure of Invention
The present invention is made to solve the above problems, and an object of the present invention is to provide a method for designing reliability of a complex structure based on a vector hybrid agent model.
The invention provides a complex structure reliability design method based on a vector hybrid agent model, which is used for carrying out multi-target reliability design on a complex structure and has the characteristics that the method comprises the following steps: step 1, acquiring a reliability design target of a complex structure according to analysis characteristics of the complex structure, wherein the reliability design target is more than or equal to two;
step 2, respectively acquiring input parameters and response parameters of each reliability design target, and taking the input parameters and the response parameters of all the reliability design targets as the input parameters and the response parameters of a complex structure;
step 3, performing series-parallel sampling on the input parameters and the response parameters of the complex structure to obtain input parameter samples and response parameter samples and establish a sample database;
step 4, establishing a proxy model library based on a plurality of proxy models, and selecting a suitable type of proxy model for each reliability design target from the proxy model library;
step 5, training the agent models of the reliability design targets respectively by using the input parameter samples and the response parameter samples corresponding to the reliability design targets, and fitting and optimizing model parameters of the agent models of the reliability design targets by using a self-adaptive modeling method to obtain optimal model parameters of the agent models of each reliability design target;
step 6, establishing an optimal parameter matrix according to the optimal model parameters of each agent model, and then combining the optimal parameter matrix with input parameter samples and response parameter samples corresponding to a plurality of reliability design targets to establish and obtain a vector hybrid agent model;
step 7, evaluating the precision of the established vector hybrid agent model based on a supervised learning method of the test sample, and if the precision is not met, correcting the vector hybrid agent model by adopting test data until the precision is met;
and 8, performing multi-target reliability design on the complex structure by adopting a numerical simulation method according to the vector hybrid agent model.
In the complex structure reliability design method based on the vector hybrid agent model provided by the invention, the method can also have the following characteristics: in the step 3, the hybrid sampling adopts a Latin hypercube sampling method, and all input parameters and response parameters of the complex structure are sampled simultaneously for many times based on the probability statistical characteristics of the input parameters.
In the complex structure reliability design method based on the vector hybrid agent model provided by the invention, the method can also have the following characteristics: when the agent model is selected in the step 4, selecting a type suitable for the agent model for each reliability design target by taking the minimum loss function as the target according to the analysis characteristics of each reliability design target, wherein the types of the agent models in the agent model library comprise a polynomial model, a kriging model, an artificial neural network model, a support vector machine model and a deep learning model.
In the complex structure reliability design method based on the vector hybrid agent model provided by the invention, the method can also have the following characteristics: wherein, step 5 comprises the following substeps:
step 5-1, respectively taking the input parameter samples and the corresponding response parameter samples of each group of multiple random variables as input sample vectors and output sample vectors, and inputting the input sample vectors and the corresponding response parameter samples into an agent model corresponding to a reliability design target for training after data normalization;
step 5-2, in the training process, the model parameters of each agent model are subjected to iterative optimization by adopting a self-adaptive modeling method to obtain optimal model parameters,
the self-adaptive modeling method comprises a least square method, a weighted least square method and a moving least square method.
In the complex structure reliability design method based on the vector hybrid agent model provided by the invention, the method can also have the following characteristics: wherein, step 7 comprises the following substeps:
7-1, evaluating the precision of the established vector hybrid agent model based on a supervised learning method of a test sample;
7-2, if the precision is not met, correcting the vector hybrid agent model by adopting the test data until the precision is met;
and 7-3, if the precision is met, using the obtained vector hybrid agent model for subsequent complex structure multi-target reliability design.
In the complex structure reliability design method based on the vector hybrid agent model provided by the invention, the method can also have the following characteristics: wherein, step 8 comprises the following substeps:
8-1, replacing a real structural model with the established vector hybrid agent model as a research object;
8-2, establishing a multi-target extreme state equation based on the maximum or minimum allowable value of each reliability design target;
8-3, sampling random input parameters in a large scale based on a Latin hypercube series-parallel sampling method, and substituting the random input parameters into a vector hybrid agent model to calculate response parameters corresponding to each reliability design target;
and 8-4, comprehensively judging whether the response parameters of all the reliability design targets fall in the safety domain, when the response parameters do not fall in the safety domain, optimizing the parameters of the corresponding reliability design targets to ensure that the response parameters fall in the safety domain, and after optimizing all the reliability design targets, completing multi-target synchronous reliability design of the complex structure and calculating to obtain the reliability of the complex structure.
Action and Effect of the invention
According to the complex structure reliability design method based on the vector hybrid agent model, due to the fact that the Latin hypercube parallel-serial sampling method is used for conducting parallel-serial sampling, the Latin hypercube parallel-serial sampling method integrates the Latin hypercube sampling technology and the multi-parameter linkage sampling strategy, all input parameters and response parameters of multiple design targets can be subjected to simultaneous combined sampling, the coupling and the correlation between random parameters and analysis targets are effectively considered, the availability of extracted samples and the synchronism of sample extraction can be guaranteed, and the reasonable correlation of the multiple design targets, the parameters and the agent model can be further guaranteed; and meanwhile, the invention searches a proxy model suitable for the invention for a plurality of reliability design targets, establishes an integral vector hybrid proxy model for the plurality of reliability design targets based on different types of proxy models, and can synchronously carry out multi-objective optimization and reliability design through the vector hybrid proxy model, thereby realizing high-efficiency and high-precision multi-objective optimization and reliability design analysis of the complex structure.
Drawings
Fig. 1 is a method for designing reliability of a complex structure based on a vector hybrid agent model in an embodiment of the present invention.
Detailed Description
In order to make the technical means and functions of the present invention easy to understand, the present invention is specifically described below with reference to the embodiments and the accompanying drawings.
< example >
Fig. 1 is a method for designing reliability of a complex structure based on a vector hybrid agent model in an embodiment of the present invention.
As shown in fig. 1, a method for designing reliability of a complex structure based on a vector hybrid agent model according to an embodiment of the present invention is used for performing multi-objective reliability design on a complex structure, and includes the following steps:
step 1, obtaining a reliability design target of the complex structure according to the analysis characteristics of the complex structure, wherein the reliability design target is more than or equal to two.
In this embodiment, the complex structure is a mechanical structure composed of a plurality of components, for example, a rotor structure system of a gas turbine engine is composed of a main shaft, a wheel disc and a blade, and the plurality of reliability design targets are design parameters such as stress, strain, deformation, fatigue, creep and the like of each corresponding component in the complex structure.
And 2, respectively acquiring the input parameters and the response parameters of each reliability design target, and taking the input parameters and the response parameters of all the reliability design targets as the input parameters and the response parameters of the complex structure.
And 3, performing series-parallel sampling on the input parameters and the response parameters of the complex structure to obtain input parameter samples and response parameter samples and establish a sample database.
In the step 3, the hybrid sampling adopts a Latin hypercube sampling method, and all input parameters and response parameters of the complex structure are sampled simultaneously for many times based on the probability statistical characteristics of the input parameters.
And 4, establishing a proxy model library based on the plurality of proxy models, and selecting a suitable type of proxy model for each reliability design target from the proxy model library.
When selecting the agent model in the step 4, selecting the type of the agent model suitable for the agent model for each reliability design target by taking the minimum loss function as the target according to the analysis characteristics of each reliability design target, such as parameter scale, non-linearity degree, whether the reliability design target is transient or not,
the types of the agent models in the agent model library include a polynomial model, a kriging model, an artificial neural network model, a support vector machine model and a deep learning model.
Step 5, training the agent models of the reliability design targets respectively by using the input parameter samples and the response parameter samples corresponding to the reliability design targets, and fitting and optimizing model parameters of the agent models of the reliability design targets by using a self-adaptive modeling method to obtain optimal model parameters of the agent models of each reliability design target;
step 5 comprises the following substeps:
step 5-1, respectively taking the input parameter samples and the corresponding response parameter samples of each group of multiple random variables as input sample vectors and output sample vectors, and inputting the input sample vectors and the corresponding response parameter samples into an agent model corresponding to a reliability design target for training after data normalization;
step 5-2, in the training process, the model parameters of each agent model are subjected to iterative optimization by adopting a self-adaptive modeling method to obtain optimal model parameters,
the adaptive modeling method includes a least square method, a weighted least square method, and a moving least square method.
In this embodiment, the adaptive modeling method is a method developed based on an intelligent optimization algorithm (genetic algorithm, particle swarm algorithm).
And 6, establishing an optimal parameter matrix according to the optimal model parameters of each agent model, and combining the optimal parameter matrix with the input parameter samples and the response parameter samples corresponding to the plurality of reliability design targets to establish and obtain the vector hybrid agent model.
In this embodiment, the vector hybrid agent model is composed of an input variable, an output variable, and a model parameter matrix, where the input parameter sample and response parameter samples corresponding to a plurality of reliability design targets are known, and the obtained optimal parameter matrix is used as the model parameter matrix, so that the vector hybrid agent model can be determined.
And 7, evaluating the precision of the established vector hybrid agent model based on the supervised learning method of the test sample, and if the precision is not met, correcting the vector hybrid agent model by adopting the test data until the precision is met.
Step 7 comprises the following substeps:
7-1, evaluating the precision of the established vector hybrid agent model based on a supervised learning method of a test sample;
7-2, if the precision is not met, correcting the vector hybrid agent model by adopting the test data until the precision is met;
and 7-3, if the precision is met, using the obtained vector hybrid agent model for subsequent complex structure multi-target reliability design.
And 8, performing multi-target reliability design on the complex structure by adopting a numerical simulation method according to the vector hybrid agent model.
In this embodiment, the numerical simulation method includes an MC simulation method, a FORM method, and an sor method.
The step 8 comprises the following substeps:
8-1, replacing a real structural model with the established vector hybrid agent model as a research object;
8-2, establishing a multi-target extreme state equation based on the maximum or minimum allowable value of each reliability design target;
8-3, sampling random input parameters in a large scale based on a Latin hypercube series-parallel sampling method, and substituting the random input parameters into a vector hybrid agent model to calculate response parameters corresponding to each reliability design target;
and 8-4, comprehensively judging whether the response parameters of all the reliability design targets fall in the safety domain, when the response parameters do not fall in the safety domain, optimizing the parameters of the corresponding reliability design targets to ensure that the response parameters fall in the safety domain, and after optimizing all the reliability design targets, completing multi-target synchronous reliability design of the complex structure and calculating to obtain the reliability of the complex structure.
Effects and effects of the embodiments
According to the complex structure reliability design method based on the vector hybrid agent model, the Latin hypercube series-parallel sampling method is used for carrying out series-parallel sampling, the Latin hypercube series-parallel sampling method integrates the Latin hypercube sampling technology and the multi-parameter linkage sampling strategy, all input parameters and response parameters of multiple design targets can be simultaneously and jointly sampled, the coupling and the correlation between random parameters and analysis targets are effectively considered, the availability of sample extraction and the synchronization of sample extraction can be ensured, and the reasonable correlation of the multiple design targets, the parameters and the agent model is further ensured; in addition, the embodiment simultaneously searches for an agent model suitable for the embodiment for a plurality of reliability design targets, establishes an integral vector hybrid agent model for the plurality of reliability design targets based on different types of agent models, and can synchronously perform multi-objective optimization and reliability design through the vector hybrid agent model, thereby realizing high-efficiency and high-precision multi-objective optimization and reliability design analysis of the complex structure.
The above embodiments are preferred examples of the present invention, and are not intended to limit the scope of the present invention.
Claims (6)
1. A complex structure reliability design method based on a vector hybrid agent model is used for carrying out multi-target reliability design on a complex structure, and is characterized by comprising the following steps:
step 1, acquiring a reliability design target of the complex structure according to the analysis characteristics of the complex structure, wherein the reliability design target is more than or equal to two;
step 2, respectively obtaining the input parameters and the response parameters of each reliability design target, and taking the input parameters and the response parameters of all the reliability design targets as the input parameters and the response parameters of the complex structure;
step 3, performing series-parallel sampling on the input parameters and the response parameters of the complex structure to obtain input parameter samples and response parameter samples and establish a sample database;
step 4, establishing a proxy model library based on a plurality of proxy models, and selecting a suitable type of proxy model for each reliability design target from the proxy model library;
step 5, the input parameter samples and the response parameter samples corresponding to the plurality of reliability design targets are used for respectively training the agent models of the reliability design targets, and fitting and model parameter optimization are carried out on the agent models of the plurality of reliability design targets through a self-adaptive modeling method to obtain optimal model parameters of the agent models of the reliability design targets;
step 6, establishing an optimal parameter matrix according to the optimal model parameters of each agent model, and then combining the optimal parameter matrix with the input parameter samples and the response parameter samples corresponding to the plurality of reliability design targets to establish and obtain a vector hybrid agent model;
step 7, evaluating the precision of the established vector hybrid agent model based on a supervised learning method of a test sample, and if the precision is not met, correcting the vector hybrid agent model by adopting test data until the precision is met;
and 8, performing multi-target reliability design on the complex structure by adopting a numerical simulation method according to the vector hybrid agent model.
2. The method of claim 1, wherein the method comprises:
and 3, performing the parallel-serial sampling in the step 3 by adopting a Latin hypercube sampling method, and simultaneously sampling all input parameters and response parameters of the complex structure for multiple times based on the probability statistical characteristics of the input parameters.
3. The method of claim 1, wherein the method comprises:
wherein, when selecting the agent model in the step 4, selecting a type of the agent model suitable for itself for each reliability design target by taking a minimum loss function as a target according to the analysis characteristics of each reliability design target,
the types of the agent models in the agent model library comprise a polynomial model, a kriging model, an artificial neural network model, a support vector machine model and a deep learning model.
4. The method of claim 1, wherein the method comprises:
wherein the step 5 comprises the following substeps:
step 5-1, the input parameter samples and the corresponding response parameter samples of each group of multiple random variables are respectively used as input sample vectors and output sample vectors, and the input sample vectors and the corresponding response parameter samples are substituted into the agent model corresponding to the reliability design target for training after data normalization;
step 5-2, in the training process, the model parameters of each agent model are subjected to iterative optimization by adopting the self-adaptive modeling method to obtain the optimal model parameters,
the self-adaptive modeling method comprises a least square method, a weighted least square method and a moving least square method.
5. The method of claim 1, wherein the method comprises:
wherein the step 7 comprises the following substeps:
7-1, evaluating the precision of the established vector hybrid agent model based on a supervised learning method of a test sample;
7-2, if the precision is not met, correcting the vector hybrid agent model by adopting test data until the precision is met;
and 7-3, if the precision is met, using the obtained vector hybrid agent model for subsequent complex structure multi-target reliability design.
6. The method of claim 1, wherein the method comprises:
wherein, the step 8 comprises the following substeps:
step 8-1, replacing a real structural model with the established vector hybrid agent model as a research object;
8-2, establishing a multi-target extreme state equation based on the maximum or minimum allowable value of each reliability design target;
8-3, sampling random input parameters in a large scale based on a Latin hypercube series-parallel sampling method, and substituting the random input parameters into the vector hybrid agent model to calculate response parameters corresponding to the reliability design targets;
and 8-4, comprehensively judging whether the response parameters of the reliability design targets fall in a safety domain, when the response parameters do not fall in the safety domain, optimizing the parameters of the corresponding reliability design targets to ensure that the response parameters fall in the safety domain, and after the reliability design targets are optimized, completing the multi-target synchronous reliability design of the complex structure and calculating to obtain the reliability of the complex structure.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110603383.1A CN113239495A (en) | 2021-05-31 | 2021-05-31 | Complex structure reliability design method based on vector hybrid agent model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110603383.1A CN113239495A (en) | 2021-05-31 | 2021-05-31 | Complex structure reliability design method based on vector hybrid agent model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113239495A true CN113239495A (en) | 2021-08-10 |
Family
ID=77135952
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110603383.1A Pending CN113239495A (en) | 2021-05-31 | 2021-05-31 | Complex structure reliability design method based on vector hybrid agent model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113239495A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114429091A (en) * | 2022-01-30 | 2022-05-03 | 内蒙古工业大学 | Reliability analysis method for hammer type rotor of straw rubbing crusher based on multiple failure modes |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102867083A (en) * | 2012-08-30 | 2013-01-09 | 浙江大学 | High-rigidity and light-weight design method considering uncertainty of slide block mechanism of press machine |
CN105930562A (en) * | 2016-04-13 | 2016-09-07 | 浙江大学 | Structural performance optimum design method under non-probability conditions |
CN107944547A (en) * | 2017-11-22 | 2018-04-20 | 电子科技大学 | A kind of time-dependent structural reliability appraisal procedure based on RNN |
WO2020197533A1 (en) * | 2019-03-22 | 2020-10-01 | General Electric Company | Surrogate of a simulation engine for power system model calibration |
CN111797535A (en) * | 2020-07-11 | 2020-10-20 | 吉林大学 | Structure reliability analysis self-adaptive point adding method for multiple agent models |
CN112100759A (en) * | 2020-08-16 | 2020-12-18 | 复旦大学 | Distributed cooperative agent model method for approximation analysis of complex engineering structure system |
CN112149253A (en) * | 2020-09-24 | 2020-12-29 | 复旦大学 | Engineering structure reliability evaluation method based on distributed hybrid cooperative agent model |
CN112287484A (en) * | 2020-10-29 | 2021-01-29 | 复旦大学 | Complex engineering system reliability design method based on vector agent model |
CN112529282A (en) * | 2020-12-03 | 2021-03-19 | 清华大学 | Wind power plant cluster short-term power prediction method based on space-time graph convolutional neural network |
CN112784347A (en) * | 2021-02-25 | 2021-05-11 | 中信建筑设计研究总院有限公司 | Cable-stayed bridge cable force reliability assessment method based on bridge tower deformation and considering partial cable failure |
-
2021
- 2021-05-31 CN CN202110603383.1A patent/CN113239495A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102867083A (en) * | 2012-08-30 | 2013-01-09 | 浙江大学 | High-rigidity and light-weight design method considering uncertainty of slide block mechanism of press machine |
CN105930562A (en) * | 2016-04-13 | 2016-09-07 | 浙江大学 | Structural performance optimum design method under non-probability conditions |
CN107944547A (en) * | 2017-11-22 | 2018-04-20 | 电子科技大学 | A kind of time-dependent structural reliability appraisal procedure based on RNN |
WO2020197533A1 (en) * | 2019-03-22 | 2020-10-01 | General Electric Company | Surrogate of a simulation engine for power system model calibration |
CN111797535A (en) * | 2020-07-11 | 2020-10-20 | 吉林大学 | Structure reliability analysis self-adaptive point adding method for multiple agent models |
CN112100759A (en) * | 2020-08-16 | 2020-12-18 | 复旦大学 | Distributed cooperative agent model method for approximation analysis of complex engineering structure system |
CN112149253A (en) * | 2020-09-24 | 2020-12-29 | 复旦大学 | Engineering structure reliability evaluation method based on distributed hybrid cooperative agent model |
CN112287484A (en) * | 2020-10-29 | 2021-01-29 | 复旦大学 | Complex engineering system reliability design method based on vector agent model |
CN112529282A (en) * | 2020-12-03 | 2021-03-19 | 清华大学 | Wind power plant cluster short-term power prediction method based on space-time graph convolutional neural network |
CN112784347A (en) * | 2021-02-25 | 2021-05-11 | 中信建筑设计研究总院有限公司 | Cable-stayed bridge cable force reliability assessment method based on bridge tower deformation and considering partial cable failure |
Non-Patent Citations (3)
Title |
---|
刘品: "《可靠性工程基础》", 10 October 1995, 中国计量出版社 * |
陈潇凯: "《车辆多学科设计优化方法》", 30 June 2018, 北京理工大学出版社 * |
黄贤振 等: "《机械可靠性设计理论与应用》", 30 November 2019, 东北大学出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114429091A (en) * | 2022-01-30 | 2022-05-03 | 内蒙古工业大学 | Reliability analysis method for hammer type rotor of straw rubbing crusher based on multiple failure modes |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110530638B (en) | Digital twin-based method for detecting and diagnosing damage of main bearing of aero-engine | |
CN110210044A (en) | Load prediction method and device of wind generating set | |
CN114528666A (en) | Complex structure system reliability method based on multi-level distributed cooperative agent model | |
CN116305522A (en) | Digital twin-based aircraft structure reliability simulation test model calibration method, device and medium | |
CN112287484B (en) | Complex engineering system reliability design method based on vector agent model | |
CN112580228A (en) | Fan blade structural design optimization method for mixed structure of turbofan engine | |
CN113239495A (en) | Complex structure reliability design method based on vector hybrid agent model | |
Tucker et al. | Cyber-physical systems: A new paradigm for energy technology development | |
CN116910513A (en) | Centrifugal pump rotor fault diagnosis method based on CWGAN-GP and double-flow CNN model | |
Morison et al. | Identification of voltage control areas and reactive power reserve; an advancement in on-line voltage security assessment | |
CN112100759A (en) | Distributed cooperative agent model method for approximation analysis of complex engineering structure system | |
CN116628854A (en) | Wing section aerodynamic characteristic prediction method, system, electronic equipment and storage medium | |
Roehl et al. | A collaborative optimization environment for turbine engine development | |
CN116108749A (en) | Wind turbine flow field prediction method based on deep learning | |
CN111474905B (en) | Parameter drift fault diagnosis method in manufacturing process of electromechanical product | |
CN112849429B (en) | Tracing method for civil aircraft system measurement parameters | |
Rootliep et al. | Evolutionary algorithm for enhanced gas path analysis in turbofan engines | |
CN114329805A (en) | Connecting piece multidisciplinary collaborative design optimization method based on self-adaptive agent model | |
CN113486553A (en) | Complex equipment reliability analysis method based on Thiessen polygon area division | |
CN112750505A (en) | Method for simplifying large-scale detailed chemical reaction model of high-carbon fuel | |
Zhao et al. | Fault detection and diagnosis for sensor in an aero-engine system | |
CN111506998B (en) | Method for constructing parameter drift fault feature sample library in manufacturing process of electromechanical product | |
CN113326182B (en) | Software defect prediction method based on sampling and ensemble learning | |
CN113722818B (en) | Digital test system for complex system of ship | |
CN117454721B (en) | Wind power plant wake superposition effect evaluation method and medium based on digital simulation experiment |
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 |