CN-122020919-A - Stress calculation method for spring operating mechanism with multiple fault couplings

CN122020919ACN 122020919 ACN122020919 ACN 122020919ACN-122020919-A

Abstract

The invention discloses a stress calculation method of a spring operation mechanism with multiple fault couplings, which relates to the technical field of power equipment mechanical simulation and fault analysis and comprises the steps of constructing an N multiplied by N order fault coupling coefficient matrix of the spring operation mechanism; the stress field data of the component under the single fault working condition is calculated by adopting a stress exclusive simulation method, a submatrix and a coupling coefficient of the corresponding order of the target compound fault are extracted from an N multiplied by N order fault coupling coefficient matrix, the stress field data of the single fault in the target compound fault is subjected to weighted linear superposition to obtain a preliminary coupling stress field, an attention enhancement-multiscale feature fusion neural network model is constructed and trained, network parameters are corrected through a self-adaptive weight attenuation optimization algorithm, and the nonlinear distortion area in the preliminary coupling stress field is accurately corrected. The method solves the problem of large calculation deviation caused by stress nonlinear distortion under composite faults in the prior art.

Inventors

YANG HAO
ZHAO YANGZHEN
TENG XIAOYI
WU QIANHUA
LIU YIXUAN

Assignees

西安工程大学

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (7)

1. The stress calculation method of the spring operating mechanism with multiple fault couplings is characterized by comprising the following steps of: s1, constructing an N multiplied by N order fault coupling coefficient matrix of a spring operating mechanism, and quantifying stress influence weights among multiple faults through the coefficient matrix; s2, calculating stress field data of the component under a single fault condition by adopting a stress exclusive simulation method; S3, extracting a submatrix and a coupling coefficient of the corresponding order of the target composite fault from the NxN order fault coupling coefficient matrix, and performing weighted linear superposition on stress field data of a single fault in the target composite fault to obtain a preliminary coupling stress field; And S4, constructing and training an attention enhancement-multiscale feature fusion neural network model, correcting network parameters through a self-adaptive weight attenuation optimization algorithm, accurately correcting a nonlinear distortion region in the primary coupling stress field, and outputting a composite fault stress distribution result of the spring operating mechanism.
2. The method for calculating stress of a spring operator with multiple fault couplings according to claim 1, wherein constructing a spring operator n×n order fault coupling coefficient matrix in S1 comprises the following sub-steps: s11, respectively constructing N types of single fault experiment platforms and multiple types of composite fault experiment platforms, and collecting stress data of key components under each platform; S12, calculating the absolute value of the sum of the actual measurement stress value of the composite fault and the stress value of the single fault as stress interference intensity aiming at each type of composite fault; S13, establishing a hierarchical structure model, wherein a target layer is used for determining a coupling coefficient, a criterion layer comprises stress interference intensity, fault occurrence probability and mechanism structure relevance, a scheme layer is used for combining various faults, constructing a judgment matrix and carrying out consistency test, and a subjective weight vector is obtained after normalization; S14, carrying out standardization processing on the composite fault sample data to obtain a normalized matrix, calculating information entropy of the fault, and calculating an objective weight vector based on the information entropy; s15, fusing the subjective weight and the objective weight by adopting a variable weight coefficient to obtain a mixed weight; S16, generating an NxN order fault coupling coefficient matrix, wherein the matrix is a symmetric matrix, diagonal elements are 1, and non-diagonal elements The method comprises the following steps: = × × Wherein, the Is the first The mixed weight of the class fault, Is the first The mixed weight of the class fault, Is a fault And (3) with Is used for the stress disturbance intensity of the (c), ≠ ,0.1≤ ≤0.8。
3. The method for calculating stress of a multi-fault-coupled spring actuator of claim 2, characterized in that the information entropy of the fault The method comprises the following steps: = ln( +ε) Wherein, the For the number of composite fault sample data sets, For a positive indicator, ε is a positive value that prevents nonsensical log entries, Is the first Group composite fault sample of the first The original measured stress data of the class fault, In all the composite fault samples, the first Column vectors are formed by the fault-like original measured stress data.
4. The method for calculating stress of a multi-fault-coupled spring actuator of claim 2, characterized in that the weight-changing coefficient The method comprises the following steps: Wherein, the Is an objective weight vector Is used for the coefficient of variation of (a), Is subjective weight vector Coefficient of variation of (2); The mixing weights The method comprises the following steps: = + Wherein, the Is the first The subjective weight of the class fault, Is the first Objective weight of class failure.
5. The method of claim 2, wherein the off-diagonal elements of the nxn order fault coupling coefficient matrix satisfy: the coupling coefficient of cam abrasion to the clamping of the energy storage shaft is 0.5-0.7, the coupling coefficient of cam abrasion to the fatigue of the closing spring is 0.2-0.4, the coupling coefficient of clamping of the energy storage shaft to the fatigue of the closing spring is 0.1-0.3, the coupling coefficient of cam abrasion to the clamping of the output shaft is 0.4-0.6, the coupling coefficient of clamping of the energy storage shaft to the clamping of the output shaft is 0.3-0.5, and the coupling coefficient of clamping of the output shaft to the fatigue of the closing spring is 0.5-0.7.
6. The method for calculating stress of a multi-fault coupled spring actuator according to claim 1, wherein in S3, the stress field data of a single fault in the target composite fault is weighted and linearly superimposed, and the formula is: Wherein, the As the data of the stress field after the superposition, Is the first The class fault stress value is set to be, In order to normalize the mixing weights, Is a fault And (3) with Is used for the coupling coefficient of the (c), Is the kind of fault.
7. The method for calculating stress of a spring actuator with multiple fault couplings according to claim 1, wherein the structure of the attention-enhancing-multiscale feature fusion neural network model in S4 is as follows: The input layer node comprises a stress peak value, a stress distribution interval median value, a stress change slope, a fault 1 grade parameter, a fault 2 grade parameter, a fault combination type code and a multi-fault coupling strength comprehensive factor which are subjected to linear superposition; The hidden layer comprises a first layer, a second layer, a third layer and a characteristic fusion layer, wherein the first layer is a multi-scale characteristic extraction layer, the second layer is a spatial attention module, and the third layer is a characteristic fusion layer; The output layer node comprises a corrected stress peak value, a high stress area occupation ratio, a nonlinear distortion correction coefficient and an area distortion grade label.

Description

Stress calculation method for spring operating mechanism with multiple fault couplings Technical Field The invention relates to the technical field of mechanical simulation and fault analysis of power equipment, in particular to a stress calculation method of a spring operating mechanism with multiple fault couplings. Background The CT26 type spring operating mechanism is a core power executing component of the high-voltage circuit breaker, and the running state of the CT26 type spring operating mechanism directly determines the switching-on and switching-off reliability of the circuit breaker. In a long-term complex operating environment, the mechanism does not only generate single faults, but often generates composite faults such as cam abrasion, energy storage shaft clamping, closing spring fatigue, output shaft clamping and the like, the faults are superposition of a plurality of single faults, mutual interference exists among the faults, so that stress distribution of key components presents strong nonlinear distortion, and stress fields of the key components are not simple superposition of stress fields of the single faults. In the prior art, the stress calculation method aiming at the CT26 mechanism focuses on a single fault scene, and only can obtain stress data of independent faults such as cam abrasion, transmission jam and the like through simulation or experiment. For composite faults, the existing method generally adopts a simplified mode of directly adding single fault stress fields, and does not consider the coupling interference effect among faults, on one hand, the single fault can change the stress transmission path of another fault, such as the stress angle deviation of a transmission mechanism caused by cam abrasion, so that the stress concentration of a clamping part is further increased, on the other hand, the stress distortion under the composite fault has nonlinear characteristics, and the simple linear superposition can cause extremely large deviation between a calculation result and actual stress. In addition, the prior art lacks an effective means for quantifying the coupling relation between faults, and cannot accurately judge the influence degree of a single fault on the stress distribution of another fault, so that the composite fault stress calculation is strong in subjectivity and poor in repeatability. The calculation result with insufficient precision can mislead the judgment of operation and maintenance personnel on the severity degree of the mechanism fault, thereby causing insufficient maintenance or excessive maintenance, and even causing the large-area power failure accident caused by the mechanism failure. Therefore, developing a composite fault stress calculation method for quantifying a fault coupling relation and correcting nonlinear distortion becomes a technical problem to be solved in the field of CT26 type mechanism mechanics analysis at present. Disclosure of Invention The invention provides a stress calculation method for a spring operating mechanism with multiple fault couplings, which solves the technical problems that the stress distribution of a single fault of the spring operating mechanism of a CT26 type circuit breaker can only be calculated in the prior art, the complex fault scenes such as cam abrasion, transmission clamping, closing spring fatigue, energy storage shaft clamping and output shaft clamping in actual operation cannot be accurately dealt with, and the calculation deviation is large due to stress nonlinear distortion under the complex faults. In order to achieve the aim of the invention, the technical scheme adopted by the invention is that the stress calculation method of the spring operating mechanism with multiple fault couplings comprises the following steps: s1, constructing an N multiplied by N order fault coupling coefficient matrix of a spring operating mechanism, and quantifying stress influence weights among multiple faults through the coefficient matrix; s2, calculating stress field data of the component under a single fault condition by adopting a stress exclusive simulation method; S3, extracting a submatrix and a coupling coefficient of the corresponding order of the target composite fault from the NxN order fault coupling coefficient matrix, and performing weighted linear superposition on stress field data of a single fault in the target composite fault to obtain a preliminary coupling stress field; And S4, constructing and training an attention enhancement-multiscale feature fusion neural network model, correcting network parameters through a self-adaptive weight attenuation optimization algorithm, accurately correcting a nonlinear distortion region in the primary coupling stress field, and outputting a composite fault stress distribution result of the spring operating mechanism. Further, the step S1 of constructing a spring operating mechanism NxN order fault coupling coefficient matrix comprises the following sub-steps: