Search

CN-122021138-A - Multi-physical-field reliability assessment method for electronic equipment based on data driving

CN122021138ACN 122021138 ACN122021138 ACN 122021138ACN-122021138-A

Abstract

The invention belongs to the technical field of reliability analysis of electronic equipment, and particularly relates to a multi-physical-field reliability evaluation method of electronic equipment based on data driving. The traditional reliability analysis method has obvious defects when the problem of multi-physical field coupling is solved, and the defects are effectively overcome by organically combining the systematic data analysis method with the multi-physical field coupling simulation technology. Firstly, a complete reliability evaluation system is established, and then a technical path fused with theoretical calculation and numerical simulation is adopted, so that the reliability and accuracy analysis of the electronic equipment under the coupling action of multiple physical fields is realized. The method has the characteristics of good economy, high efficiency and strong practicability, can rapidly and accurately evaluate the reliability level of the electronic equipment, and provides an innovative reliability analysis technical scheme and engineering practice support for the industry.

Inventors

  • QI WENLIANG
  • CHEN FENGJUN
  • LU ZHENKUN
  • YANG RUI

Assignees

  • 温州理工学院

Dates

Publication Date
20260512
Application Date
20260116

Claims (10)

  1. 1. And (3) analyzing the operation conditions: The system collects historical operation data of the electronic equipment in a real working environment, and builds a complete data set containing multidimensional parameters such as power supply voltage fluctuation, environmental temperature change, mechanical vibration frequency spectrum and the like. On the basis, the main working mode of the K-means cluster analysis and identification equipment is utilized, extreme operating conditions are scientifically captured through extreme value statistical methods such as generalized pareto distribution, and finally a typical working condition set with engineering representativeness and safety boundary coverage is formed.
  2. 2. Orthogonal test design: Based on key design parameters such as a typical working condition set, device layout and the like, a high-efficiency simulation matrix is constructed by adopting an orthogonal experimental design principle. By selecting proper orthogonal table, the level and combination of each parameter are reasonably arranged, and the number of the full factor simulation times is reduced by 1-2 orders of magnitude on the premise of guaranteeing the integrity of the verification information.
  3. 3. Multiple physical field coupling simulation calculation: And establishing an accurate thermal-electric-force multi-physical field coupling simulation model by utilizing finite element analysis platforms such as ANSYS, COMSOL and the like. Through sequential coupling or direct coupling strategies, the system executes all simulation working conditions in the orthogonal experimental matrix, and key performance parameters such as junction temperature distribution, power supply integrity, signal integrity, warp deformation and the like under each sample point are completely obtained.
  4. 4. Multiple physical field base database: and carrying out grid convergence analysis and experimental verification on the simulation result, and ensuring the numerical accuracy and engineering credibility of the data. And constructing a structured database architecture, and establishing a mapping relation among working condition parameters, design variables and performance response to realize unified storage, quick retrieval and traceable management of data.
  5. 5. Multidimensional weighting analysis: And extracting a standardized performance parameter matrix from the database, and performing double weighting analysis by adopting CRITIC method and entropy weighting method respectively. The CRITIC method is used for identifying independent information quantity of each physical field index by calculating correlation coefficient evaluation conflict among indexes, and the entropy weight rule is used for evaluating information content of each index from the data fluctuation angle based on the variation degree of index data quantized by the information entropy theory.
  6. 6. Combining weighting models: And establishing a combined weighting model based on game theory or least square optimization, optimally fusing objective weight obtained by CRITIC method and information weight obtained by entropy weight method, and forming a final weight coefficient integrating the characteristics of each physical field in a linear weighting mode.
  7. 7. Multiple physical field coupling calculation: for specific electronic equipment products, coupling field calculation covering physical phenomena such as heat conduction, thermal-electric coupling, thermal-force coupling and the like is carried out.
  8. 8. Dimensionless treatment of each physical field evaluation index: And carrying out dimensionless treatment on the original evaluation indexes of each physical field by adopting a polar difference standardization method. Positive standardization is adopted for benefit type indexes, negative standardization is adopted for cost type indexes, influences of different physical dimensions and magnitude orders on comprehensive evaluation results are eliminated, and comparability of the indexes in a comprehensive evaluation system is ensured.
  9. 9. Reliability characterization model: And organically integrating the weight coefficient of each physical field obtained by combining the weighted models with the standardized evaluation index to construct a reliability comprehensive characterization model.
  10. 10. Reliability analysis and evaluation: And (3) applying the constructed reliability characterization model to comprehensively analyze and evaluate the reliability of the electronic equipment. The key influence parameters are identified through parameter sensitivity analysis, the optimization direction of each design parameter is defined by combining a response surface method, and a quantization basis is provided for design improvement of device layout, heat dissipation structures and the like.

Description

Multi-physical-field reliability assessment method for electronic equipment based on data driving Technical Field The invention belongs to the technical field of reliability analysis of electronic equipment, and particularly relates to a multi-physical-field reliability evaluation method of electronic equipment based on data driving. Background With the rapid development of electronic devices in the high-density, high-frequency and high-power directions, the internal multi-physical field coupling effect of the electronic devices is increasingly remarkable, and the reliability problem caused by the interaction of heat, electricity and force has become a key bottleneck for limiting the service life and performance of products. Traditional reliability assessment methods rely on single physical field analysis or simplified empirical models, and it is difficult to accurately characterize complex failure mechanisms under multiple field coupling. Although the multi-physical field coupling simulation technology has become the main means for analyzing the problems, two prominent limitations still exist in practical application, namely firstly, the selection of working condition samples based on simulation depends on the experience of engineers, a selection method of a scientific system is lacked, so that data samples are difficult to cover typical working states and extreme boundary conditions at the same time, and the evaluation results are deviated or omitted, secondly, when a comprehensive reliability evaluation model is constructed, the weight distribution of different physical field indexes such as heat, electricity, force and the like generally depends on subjective experience or simple equalization treatment, and an objective weighting mechanism based on the data is lacked, so that the accuracy and universality of the model are severely restricted. Therefore, developing a reliability evaluation method capable of automatically selecting typical working conditions and realizing objective comprehensive evaluation based on data driving has become a technical problem to be solved urgently in the industry. Disclosure of Invention In view of the above, the invention provides a data-driven electronic equipment multi-physical field reliability evaluation method, which is used for establishing a reliability evaluation model considering multi-physical field coupling effect by constructing a thermo-electric multi-physical field coupling simulation database. The method combines numerical simulation and theoretical calculation, and realizes accurate prediction of the reliability of the electronic equipment. Compared with the traditional method, the method can more economically, rapidly and effectively complete reliability analysis of the electronic equipment under the action of multi-physical field coupling, and provides reliable technical support for engineering application of domestic electronic equipment. In order to achieve the technical purpose, the invention adopts the following specific technical scheme: the method for evaluating the reliability of the multiple physical fields of the electronic equipment based on data driving is characterized by comprising the following operation steps: 1. and (3) analyzing the operation conditions: The system collects historical operation data of the electronic equipment in a real working environment, and builds a complete data set containing multidimensional parameters such as power supply voltage fluctuation, environmental temperature change, mechanical vibration frequency spectrum and the like. On the basis, the main working mode of the K-means cluster analysis and identification equipment is utilized, extreme operating conditions are scientifically captured through extreme value statistical methods such as generalized pareto distribution and the like, and finally a typical working condition set with engineering representativeness and safety boundary coverage is formed; 2. orthogonal test design: based on key design parameters such as a typical working condition set, device layout and the like, a high-efficiency simulation matrix is constructed by adopting an orthogonal experimental design principle. By selecting a proper orthogonal table, reasonably arranging the level and combination of each parameter, and reducing the number of the full factor simulation times by 1-2 orders of magnitude on the premise of guaranteeing the integrity of verification information; 3. Multiple physical field coupling simulation calculation: And establishing an accurate thermal-electric-force multi-physical field coupling simulation model by utilizing finite element analysis platforms such as ANSYS, COMSOL and the like. Through sequential coupling or direct coupling strategies, the system executes all simulation working conditions in the orthogonal experimental matrix to completely acquire key performance parameters such as junction temperature distribution, power supply integrity, signal integrity,