Failure probability evaluation device and failure probability evaluation method

1 . A failure probability evaluation device that evaluates a failure probability of a mechanical system, comprising one or more memory devices having a program stored thereon that, when executed by one or more processors, cause the one or more processors to:

acquire time-series data of stress and use a rainflow method for the time-series data to generate an occurrence frequency distribution of a probability variable, wherein the probability variable is at least one of a stress amplitude when evaluating high-cycle fatigue within an elastic deformation range of a component of the mechanical system or a strain amplitude when evaluating low-cycle fatigue including plastic deformation of a component of the mechanical system;

store a failure model that computes the failure probability of the mechanical system, and the occurrence frequency distribution of the probability variable to be used for the failure model;

acquire measurement data from one or more sensors of the mechanical system, the measurement data to be used for the failure model;

estimate a probability density function of an end portion of the occurrence frequency distribution based on an extreme value statistical model, estimate a probability density function of a portion other than the end portion of the occurrence frequency distribution, and use the probability density function of the end portion and the probability density function of the portion other than the end portion to estimate an entire probability density function of the occurrence frequency distribution;

compute the failure probability of the mechanical system based on the entire probability density function and the failure model;

display a shape of the entire probability density function of the probability variable and the failure probability;

compute, using a first probability variable, a second probability variable to be used for the failure model, and said second probability variable is stored in the one or more memory devices;

perform a combination of Monte Carlo simulation using a probability density function of the first probability variable and a computation of the second probability variable from the first probability variable to calculate a probability density function of the second probability variable at a point in time when a predetermined time elapses from the time of the computation, and compute the failure probability at a point of time when the predetermined time elapses from the time of the computation; and

display at least one of a probability of failure due to a static load, a probability that fatigue damage occurs at the point in time when the predetermined time elapses, or a reproduction level indicating that damage occurs once on average at the point in time when the predetermined time elapses;

wherein the first probability variable indicates the stress amplitude or the strain amplitude, and the second probability variable indicates the stress amplitude or the strain amplitude and cumulative damage computed from a fatigue life curve stored in the one or more memory devices.

2 . The failure probability evaluation device according to claim 1 , wherein the one or more processors are further configured to estimate the probability density function of the end portion of the occurrence frequency distribution using a general Pareto distribution.

3 . The failure probability evaluation device according to claim 2 , wherein the one or more processors are further configured to estimate the probability density function of the portion other than the end portion of the occurrence frequency distribution using a kernel method.

4 . The failure probability evaluation device according to claim 1 , wherein at least one probability variable that is among probability variables to be used for the failure model and is not a probability variable used by the one or more processors to estimate the entire probability density function is given based on a single parametric probability distribution.

5 . The failure probability evaluation device according to claim 4 , wherein the probability variable used by the one or more processors to estimate the entire probability density function indicates the stress of the mechanical system, and the probability variable given based on the parametric probability distribution indicates strength of the mechanical system.

6 . The failure probability evaluation device according to claim 1 , wherein the one or more processors are configured to compute the failure probability based on a primary reliability theory or a secondary reliability theory.

7 . The failure probability evaluation device according to claim 1 , wherein a reproduction level and a parameter of an extreme value model of the entire probability density function estimated by the one or more processors are displayed by the one or more processors.

8 . A failure probability evaluation method comprising:

acquiring time-series data of stress and use a rainflow method for the time-series data to generate an occurrence frequency distribution of a probability variable, wherein the probability variable is at least one of a stress amplitude when evaluating high-cycle fatigue within an elastic deformation range of a component of a mechanical system or a strain amplitude when evaluating low-cycle fatigue including plastic deformation of a component of the mechanical system;

storing a failure model that computes a failure probability of the mechanical system, and an occurrence frequency distribution of the probability variable to be used for the failure model;

acquiring measurement data from one or more sensors of the mechanical system, the measurement data to be used for the failure model;

estimating, based on an extreme value model, a probability density function of an end portion of the occurrence frequency distribution that computes the failure probability of the mechanical system;

estimating a probability density function of a portion other than the end portion of the occurrence frequency distribution;

using the probability density function of the end portion and the probability density function of the portion other than the end portion to estimate an entire probability density function of the occurrence frequency distribution;

computing the failure probability of the mechanical system based on the entire probability density function and the failure model;

displaying a shape of the entire probability density function and the failure probability;

using a first probability variable to compute a second probability variable to be used for the failure model stored in one or more memory devices;

performing a combination of Monte Carlo simulation using a probability density function of the first probability variable and a computation of the second probability variable from the first probability variable to calculate a probability density function of the second probability variable at a point in time when a predetermined time elapses from the time of the computation, and computes the failure probability at a point in time when the predetermined time elapses from the time of computation; and

displaying at least one of a probability of failure due to a static load, a probability that fatigue damage occurs at the point in time when the predetermined time elapses, or a reproduction level indicating that damage occurs once on average at the point in time when the predetermined time elapses;

wherein the first probability variable indicates the stress amplitude or the strain amplitude, and the second probability variable indicates the stress amplitude or the strain amplitude and cumulative damage computed from a fatigue life curve stored in the one or more memory devices.