CN-121997603-A - Reliability modeling and risk analysis method for numerical control machining center

CN121997603ACN 121997603 ACN121997603 ACN 121997603ACN-121997603-A

Abstract

The invention discloses a reliability modeling and risk analysis method of a numerical control machining center, which comprises the steps of obtaining on-site operation and maintenance data of the numerical control machining center, dividing the whole machine into a plurality of functional subsystems, constructing a mixed Weibull distribution model aiming at each subsystem, determining an optimal mixed score by using AIC and BIC criteria, carrying out parameter estimation by adopting an EM algorithm to represent life characteristics of multi-failure mechanism mixing, calculating accumulated failure probability CFP of each subsystem and first failure contribution degree FFC of failure of the whole machine based on a competition risk theory, constructing a risk quadrant graph by taking failure frequency and first failure contribution degree as dimensions, evaluating risk area categories of each subsystem and generating a differentiated maintenance strategy. The invention solves the problem that a single distribution model is difficult to adapt to heterogeneous fault data, and can quantitatively evaluate the actual influence of a subsystem on the shutdown of the whole machine, thereby guiding the establishment of a scientific preventive maintenance plan.

Inventors

MOU LIMING
CAO WEI
YU GUANGWEI
WANG HONGXI

Assignees

西安工业大学

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (9)

1. The reliability modeling and risk analysis method for the numerical control machining center is characterized by comprising the following steps of: S1, acquiring on-site operation and maintenance data of a numerical control machining center, calculating fault interval time, and dividing the whole numerical control machining center into a plurality of functional subsystems; Step S2, constructing a mixed Weibull distribution model aiming at each functional subsystem, wherein the mixed Weibull distribution model comprises a plurality of Weibull components and is used for representing life characteristics of different failure mechanisms in the functional subsystem; S3, determining the optimal blending score of each functional subsystem by utilizing a red pool information criterion AIC and a Bayesian information criterion BIC, and estimating model parameters under the optimal blending score by adopting an expectation maximization EM algorithm to obtain shape parameters, scale parameters and blending weights of the functional subsystems; step S4, based on a competition risk theory, regarding each functional subsystem as a competition risk source causing the failure of the whole machine, calculating an accumulated fault probability function of each functional subsystem, and calculating the first fault contribution degree of each functional subsystem to the failure of the whole machine based on the function; And S5, constructing a fault risk quadrant graph, generating a maintenance strategy, constructing the fault risk quadrant graph by taking the fault frequency as an abscissa and the first fault contribution degree as an ordinate, dividing risk categories and outputting differentiated maintenance decision suggestions.
2. The method for modeling and analyzing the reliability of a numerical control machining center according to claim 1, wherein: Probability Density function of hybrid Weibull distribution model described in step S2 Is composed of The individual weibull components are: .................................(1) in the formula, As a function of the time variable, Representing the mixed fraction of the fault data of the system j, wherein the mixed fraction reflects the complexity of the life composition of the system; represented is the weight of the component h, ; Is the scale parameter of the weibull model, Is a shape parameter; Then there is a probability density function And reliability function The method comprises the following steps of: .................................(2) ..........................................(3)。
3. The method for modeling and analyzing the reliability of a numerical control machining center according to claim 2, wherein: The process of determining the best blending score for each functional subsystem using the red pool information criterion AIC and the bayesian information criterion BIC in step S3 includes: For each candidate component, parameter estimation is carried out, corresponding AIC value and BIC value are calculated, the component with the smallest AIC value and BIC value is selected as the optimal blending score of the functional subsystem, if the optimal model corresponds to The functional subsystem is adapted to a single Weibull distribution, if It indicates that there is a significant heterogeneous lifetime feature for the functional subsystem.
4. The method for modeling and analyzing the reliability of a numerical control machining center according to claim 2, wherein: In step S3, the model parameters under the best blending score are estimated by using the expectation maximization EM algorithm, which includes the steps of: e, calculating that the observed data belongs to the first step based on the current parameter estimation value Posterior probability of individual blend components; m, updating the estimated values of the mixed weight, the shape parameter and the scale parameter by maximizing the expectation of the log-likelihood function; and E, repeating the step M until the log-likelihood function converges, and outputting final model parameters.
5. The method for modeling and analyzing the reliability of a numerical control machining center according to claim 1, wherein: The method for calculating the contribution degree of the first fault in the step S4 comprises the following steps: First, a cumulative fault probability function of a functional subsystem is calculated : .....................(8) In the formula, Is a functional subsystem Is a fault density function of (1); For removing functional subsystems Reliability functions of other subsystems than the one; then calculate the functional subsystem First fault contribution degree of (2): .................................(9) in the formula, In order to set the task time of the whole machine, Is the total number of subsystems.
6. The method for modeling and analyzing the reliability of a numerical control machining center according to claim 1, wherein: the step S5 of constructing the fault risk quadrant graph and generating the maintenance strategy comprises the following steps: establishing a coordinate system by taking the fault frequency of each functional subsystem as a horizontal axis and the first fault contribution degree as a vertical axis, calculating the average fault frequency of all the functional subsystems And average contribution degree And by Dividing the coordinate plane into four quadrants for the origin, dividing each subsystem into four quadrants according to the coordinates The four quadrants are respectively defined as a first quadrant high-frequency high-contribution, a second quadrant low-frequency high-contribution, a third quadrant high-frequency low-contribution and a fourth quadrant low-frequency low-contribution four-class risk area, a first quadrant function subsystem performs preventive replacement and key monitoring, a second quadrant system performs state monitoring and design improvement, a third quadrant function subsystem performs maintenance flow optimization and spare part inventory management, and a fourth quadrant system performs routine inspection or post-maintenance.
7. The system for evaluating the reliability and analyzing the risk of the numerical control machining center is characterized by comprising the following components: the data acquisition module is used for acquiring fault data of the numerical control machining center and dividing the functional subsystem; The model construction module is used for constructing a mixed Weibull distribution model considering data heterogeneity; the calculation processing module is used for obtaining the shape parameters, the scale parameters and the mixing weights of the functional subsystems; And the decision output module is used for generating a fault risk quadrant graph and a maintenance strategy instruction.
8. 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 of any one of claims 1 to 6 when the program is executed by the processor.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any one of claims 1 to 6.

Description

Reliability modeling and risk analysis method for numerical control machining center Technical Field The invention belongs to the technical field of reliability engineering of numerical control machine tools, relates to a reliability modeling and risk analysis method of a numerical control machining center, and particularly relates to a reliability modeling, competitive risk assessment and maintenance decision generation method of a subsystem of the numerical control machining center considering fault data heterogeneity. Background The numerical control machining center is core equipment manufactured by modern intelligence and is widely applied to the fields of aerospace, automobile manufacturing, precision machinery and the like. The numerical control machining center is formed by integrating a plurality of functional subsystems, has a complex structure and is composed of a plurality of subsystems such as hydraulic subsystem, pneumatic subsystem, electric subsystem, main shaft subsystem and the like. In the actual service process, the failure modes of all subsystems are influenced by alternating load and complex working conditions, and the failure modes have the characteristic of multi-mechanism mixing, such as random failure caused by early installation defects and aging failure caused by later abrasion. Existing machining center reliability modeling methods generally assume that the system or component is subject to a single life distribution. However, the field fault data often has multimodal or long tail characteristics, and a single distribution model is difficult to accurately describe the complex life rule caused by the mixing of multiple failure mechanisms, so that the model fitting precision is low, and the reliability characteristics of the equipment in different life cycle stages cannot be accurately distinguished. Although research attempts have been made to introduce a hybrid distribution model, in engineering practice, a systematic, data-driven method is lacking to determine the component quantities of the hybrid model, relying on manual experience setting, resulting in a modeling process that is highly subjective and poorly generalizable. In addition, in the aspect of overall risk assessment, the existing analysis method generally regards each subsystem as an independent risk source, and mainly ranks the subsystems according to static indexes such as failure frequency or Mean Time Between Failures (MTBF). The method ignores the competition failure relation of all subsystems in the whole system. In a serial system, the influence of the failure of different subsystems on the shutdown of the whole machine is competitive, and the probability of the failure of the whole machine caused by the subsystem cannot be accurately reflected only by the frequency of the failure. Although the failure frequency of some subsystems is low, once the failure frequency is often the primary cause of the shutdown of the whole machine, the traditional assessment method easily covers the key risk, and the allocation of maintenance resources is not matched with the actual risk requirement. Therefore, a reliability analysis method capable of considering the heterogeneity of fault data, automatically determining the order of a hybrid model, and quantitatively evaluating the contribution of each subsystem to the whole machine fault by combining with the competition failure theory is needed to support the accurate maintenance decision of the numerical control machining center. Disclosure of Invention The invention aims to provide a numerical control machining center reliability modeling and risk analysis method based on mixed Weibull distribution. The method realizes accurate description of heterogeneous fault data by constructing a data-driven hybrid Weibull distribution model, introduces a competitive risk theory and a first fault contribution index, quantitatively evaluates the dynamic contribution of each subsystem to the failure of the whole machine, and constructs a fault risk quadrant graph to realize the visual decision of a maintenance strategy, thereby guiding machine tool manufacturers and users to formulate scientific and differentiated preventive maintenance plans and improving the operation reliability of the whole machine. In order to achieve the purpose, the invention adopts the technical means that: A numerical control machining center reliability modeling and risk analysis method based on mixed Weibull distribution comprises the following steps: step S1, data acquisition and subsystem division, acquiring field fault operation and data of a numerical control machining center, calculating fault interval Time (TBF), and dividing the whole machine into Each functional subsystem establishes a fault sample set of each subsystem; S2, constructing a mixed Weibull distribution model, and constructing a mixed Weibull distribution model with fault data of each subsystemThe mixed Weibull distribution model of the in