Anomaly detection device, plant system, feature information generation device, anomaly detection method, feature information generation method and program

An anomaly detection device comprising:

• a process data obtaining unit that obtains process data of a process having a plurality of process characteristics that represents a tendency of the process data;
• and
• an anomaly detection unit that detects an anomalous state of the process based on a degree of anomaly that indicates a distance in the obtained process data from a hyperplane that indicates a boundary of a space that represents a normal state of the process generated from feature information that represents a feature of normal process data for each of the process characteristics,
• wherein the feature information that represents the normal state of the process is optimum feature information for each of the process characteristics found by evaluating a candidate of feature information based on an error between the normal process data and estimated process data calculated using the candidate of feature information, the candidate of feature information being generated for each of the process characteristics from a plurality of items of normal process data obtained when the process is in the normal state.

A feature information generation device comprising:

• a feature information generation unit that generates, from a plurality of items of normal process data obtained when a process having a plurality of process characteristics that represents a tendency of the process data is in a normal state, a candidate of feature information that represents a feature of the normal process data for each of the process characteristics;
• and
• a search unit that searches for optimum feature information for each of the process characteristics by evaluating the candidate of feature information based on an error between the normal process data and estimated process data calculated using the candidate of feature information,
• wherein a space that represents the normal state of the process is generated from the feature information.

An anomaly detection method comprising:

• a process data obtaining step of obtaining process data of a process having a plurality of process characteristics that represents a tendency of the process data;
• and
• an anomaly detection step of detecting an anomalous state of the process based on a degree of anomaly that indicates a distance in the obtained process data from a hyperplane that indicates a boundary of a space that represents a normal state of the process generated from feature information that represents a feature of normal process data for each of the process characteristics,
• wherein the feature information that represents the normal state of the process is optimum feature information for each of the process characteristics found by evaluating a candidate of feature information based on an error between the normal process data and estimated process data calculated using the candidate of feature information, the candidate of feature information being generated for each of the process characteristics from a plurality of items of normal process data obtained when the process is in the normal state.

A feature information generation method comprising:

• a feature information generation step of generating, from a plurality of items of normal process data obtained when a process having a plurality of process characteristics that represents a tendency of the process data is in a normal state, a candidate of feature information that represents a feature of the normal process data for each of the process characteristics;
• and
• a search step of searching for optimum feature information for each of the process characteristics by evaluating the candidate of feature information based on an error between the normal process data and estimated process data calculated using the candidate of feature information,
• wherein a space that represents the normal state of the process is generated from the feature information.

A non-transitory computer-readable recording medium that includes a program recorded thereon, the program including instructions that cause a computer to perform:

• obtaining process data of a process having a plurality of process characteristics that represents a tendency of the process data;
• and
• detecting an anomalous state of the process based on a degree of anomaly that indicates a distance in the obtained process data from a hyperplane that indicates a boundary of a space that represents a normal state of the process generated from feature information that represents a feature of normal process data for each of the process characteristics,
• wherein the feature information that represents the normal state of the process is optimum feature information for each of the process characteristics found by evaluating a candidate of feature information based on an error between the normal process data and estimated process data calculated using the candidate of feature information, the candidate of feature information being generated for each of the process characteristics from a plurality of items of normal process data obtained when the process is in the normal state.

A non-transitory computer-readable recording medium that includes a program recorded thereon, the program including instructions that cause a computer to perform:

• generating, from a plurality of items of normal process data obtained when a process having a plurality of process characteristics that represents a tendency of the process data is in a normal state, a candidate of feature information that represents a feature of the normal process data for each of the process characteristics;
• and
• searching for optimum feature information for each of the process characteristics by evaluating the candidate of feature information based on an error between the normal process data and estimated process data calculated using the candidate of feature information,
• wherein a space that represents the normal state of the process is generated from the feature information.