Information processing apparatus, information processing method, and, recording medium

1. An information processing apparatus for detection of an abnormality in a plant system, comprising:

a memory with instructions stored therein; and

one or more processors configured to execute the instructions, said instructions configured to, upon execution by the one or more processors, cause the processors to:

store a first model that is a model for one point in time, said first model indicating a relationship among a plurality of metrics of the plant system,

monitor the plant system using the first model to detect for an abnormality resulting from the first model,

upon detection of the abnormality resulting from the first model, generate a second model that is a model for a point in time at which the abnormality occurred, and

in an event where the second model indicates the relationship among the metrics among which the first model indicates the relationship as a result of the comparison, determine that the abnormality is caused by aged deterioration and output a result of the determination via an output device,

wherein the model indicates the relationship among the plurality of metrics with one or more relation expressions each between a value of one metric and a value of another metric among the plurality of metrics,

wherein the instructions are further configured to:

determine that the second model indicates the relationship among the metrics among which the first model indicates the relationship in a case where the second model includes one or more relation expressions for same pairs as pairs of metrics included in the first model, and

determine that the relationship among metrics in the plant system has changed in a case where the second model includes one or more relation expressions for the same pairs as pairs of metrics included in the first model, and a number of relation expressions changed between the first model and the second model among the relation expressions for the same pairs is equal to or more than a predetermined threshold value,

and wherein the relation expression is expressed by

y

⁡

(

t

)

=

∑

i

=

1

n

⁢

a

i

⁢

y

⁡

(

t

-

i

)

+

∑

j

=

1

m

⁢

b

j

⁢

x

⁡

(

t

-

k

-

j

)

+

c

,

in which

x (t) is a value of the one metric at a time t,

y (t) is a value of the another metric at the time t,

n, m, and k are parameters wherein

n is an integer equal to or more than one,

m is an integer equal to or more than one, and

k is an integer equal to or more than zero, and

a i (1≤i≤n), b j (1≤j≤m), and c are coefficients, and

the instructions are further configured to:

determine that the relation expression for the same pair has changed in a case where, between the first model and the second model, a difference in any one of the coefficients a i , b j , and c of the relation expression for the same pair is equal to or more than a predetermined threshold value or any one of the parameters n, m, k of the relation expression for the same pair has changed.

2. The information processing apparatus according to claim 1 , wherein the instructions are further configured to:

determine that the abnormality is caused by a failure of the plant system in a case where the second model does not indicate the relationship among the metrics among which the first model indicates the relationship.

3. The information processing apparatus according to claim 1 , wherein the instructions are further configured to:

output the changed relation expressions of the first model and the second model.

4. The information processing apparatus according to claim 2 , wherein

the model indicates the relationship among the plurality of metrics with one or more relation expressions each between a value of one metric and a value of another metric among the plurality of metrics, and

the instructions are further configured to:

determine that the second model indicates the relationship among the metrics among which the first model indicates the relationship in a case where the second model includes one or more relation expressions for same pairs as pairs of metrics included in the first model.

5. An information processing method for detection of an abnormality in a plant system, comprising steps of:

storing a first model that is a model for one point in time, said first model indicating a relationship among a plurality of metrics of the plant system;

monitoring the plant system using the first model to detect for an abnormality resulting from the first model;

upon detection of the abnormality resulting from the first model, generating a second model that is a model for a point in time at which the abnormality occurred;

in an event where the second model indicates the relationship among the metrics among which the first model indicates the relationship, determining that the abnormality is caused by aged deterioration and outputting a result of said determining via an output device,

the model indicating the relationship among the plurality of metrics with one or more relation expressions each between a value of one metric and a value of another metric among the plurality of metrics;

determining that the second model indicates the relationship among the metrics among which the first model indicates the relationship in a case where the second model includes one or more relation expressions for same pairs as pairs of metrics included in the first model; and

determining that the relationship among metrics in the plant system has changed in a case where the second model includes one or more relation expressions for the same pairs as pairs of metrics included in the first model, and a number of relation expressions changed between the first model and the second model among the relation expressions for the same pairs is equal to or more than a predetermined threshold value,

wherein the relation expression is expressed by

y

⁡

(

t

)

=

∑

i

=

1

n

⁢

a

i

⁢

y

⁡

(

t

-

i

)

+

∑

j

=

1

m

⁢

b

j

⁢

x

⁡

(

t

-

k

-

j

)

+

c

,

in which

x (t) is a value of the one metric at a time t,

y (t) is a value of the another metric at the time t,

n, m, and k are parameters wherein

n is an integer equal to or more than one,

m is an integer equal to or more than one, and

k is an integer equal to or more than zero, and

a i (1≤i≤n), b j (1≤j≤m), and c are coefficients,

and wherein the method further comprises a step of determining that the relation expression for the same pair has changed in a case where, between the first model and the second model, a difference in any one of the coefficients a i , b j , and c of the relation expression for the same pair is equal to or more than a predetermined threshold value or any one of the parameters n, m, k of the relation expression for the same pair has changed.

6. A non-transitory computer readable storage medium having recorded thereon a program that, upon execution by a computer, causes the computer to perform steps for detection of an abnormality in a plant system comprising:

storing a first model that is a model for one point in time, said first model indicating a relationship among a plurality of metrics of the plant system;

monitoring the plant system using the first model to detect for an abnormality resulting from the first model;

upon detection of the abnormality resulting from the first model, generating a second model that is a model for a point in time at which the abnormality occurred;

in an event where the second model indicates the relationship among the metrics among which the first model indicates the relationship, determining that the abnormality is caused by aged deterioration and outputting a result of said determining via an output device,

the model indicating the relationship among the plurality of metrics with one or more relation expressions each between a value of one metric and a value of another metric among the plurality of metrics;

determining that the second model indicates the relationship among the metrics among which the first model indicates the relationship in a case where the second model includes one or more relation expressions for same pairs as pairs of metrics included in the first model; and

determining that the relationship among metrics in the plant system has changed in a case where the second model includes one or more relation expressions for the same pairs as pairs of metrics included in the first model, and a number of relation expressions changed between the first model and the second model among the relation expressions for the same pairs is equal to or more than a predetermined threshold value,

wherein the relation expression is expressed by

y

⁡

(

t

)

=

∑

i

=

1

n

⁢

a

i

⁢

y

⁡

(

t

-

i

)

+

∑

j

=

1

m

⁢

b

j

⁢

x

⁡

(

t

-

k

-

j

)

+

c

,

in which

x (t) is a value of the one metric at a time t,

y (t) is a value of the another metric at the time t,

n, m, and k are parameters wherein

n is an integer equal to or more than one,

m is an integer equal to or more than one, and

k is an integer equal to or more than zero, and

a i (1≤i≤n), b j (1≤j≤m), and c are coefficients,

and wherein the program further causes the computer to perform a step of determining that the relation expression for the same pair has changed in a case where, between the first model and the second model, a difference in any one of the coefficients a i , b j , and c of the relation expression for the same pair is equal to or more than a predetermined threshold value or any one of the parameters n, m, k of the relation expression for the same pair has changed.