Method for setting model threshold of facility monitoring system

1. A method of setting a model threshold value for detecting an anomaly of a facility monitoring system, the method comprising:

acquiring sensor data output from each sensor;

extracting two or more feature values for each of variables included in the sensor data of each sensor;

acquiring output data by inputting input data which consists of a combination of the extracted feature values to a trained neural network model; and

comparing the input data and the output data and determining an initial threshold value in a case that there is no failure data based on comparison result values respectively including a Health Index (HI) indicating a device state index, and wherein the initial threshold value is determined based a largest comparison result value among the comparison result values.

2. The method of claim 1 , wherein the acquiring of the sensor data includes acquiring sensor data having at least one variable output from each sensor that senses an operation of a device for each sensor.

3. The method of claim 1 , wherein the extracting of the two or more feature values includes searching for device state information for the sensor data of each sensor and extracting the two or more feature values for the each of variables included in each sensor in a unit of the device state information.

4. The method of claim 1 , wherein the extracting of the two or more feature values includes generating model input values for the number of cases having a feature value for a device state in a unit of a variable included in the sensor data.

5. The method of claim 4 , wherein the extracting of the two or more feature values includes, when N device state types are set, generating a model input value for the number of cases having feature values corresponding to the N device state types in a unit of the variable.

6. The method of claim 5 , wherein the N device state types are set with user setting values or automatically set through machine learning based on time series data for a predetermined time.

7. The method of claim 1 , wherein the extracting of the two or more feature values includes extracting the feature value including an upper limit value corresponding to an Upper Control Limit (UCL) and a lower limit value corresponding to a Lower Control Limit (LCL) in a range of a control limit for the device state feature among the sensor data output from each sensor.

8. The method of claim 1 , wherein the extracting of the two or more feature values includes extracting the feature values of at least two of an upper limit value, a lower limit value, a mean, standard deviation, covariance for the sensor data output from each sensor.

9. The method of claim 1 , wherein the acquiring of the output data by inputting the input data including an extracted upper limit value and a lower limit value to the trained neural network model includes acquiring the output data by inputting the input data to the trained autoencoder-based neural network model.

10. The method of claim 1 , wherein the comparing of the input data and the output data and the determining of the initial threshold value for detecting an anomaly based on the comparison result value includes setting a user set value selected from the calculated comparison result values as the initial threshold value for detecting the anomaly.

11. The method of claim 10 , wherein the user setting value is any one of a maximum value, a mean, and a minimum value among the calculated comparison result values.

12. The method of claim 1 , wherein the comparison result value calculated by comparing the input data and the output data is calculated based on a reconstruction error of the input data and the output data.

13. The method of claim 1 , wherein the determined initial threshold value is set for each device.

14. A computer program stored in a computer readable storage medium, wherein when the computer program is executed by one or more processors, the computer program performs following operations for setting a model threshold value, the operations comprising:

acquiring sensor data output from each sensor;

extracting two or more feature values for each of variables included in the sensor data of each sensor;

acquiring output data by inputting input data which consists of a combination of the extracted feature values to a trained neural network model; and

comparing the input data and the output data and determining an initial threshold value for detecting an anomaly in a case that there is no failure data based on a comparison result values, respectively including a Health Index (HI) indicating a device state index, and wherein the initial threshold value is determined based a largest comparison result value among the comparison result values.

15. A computing device for providing a method of setting a model threshold value, the computing device comprising:

a processor including one or more cores; and

a memory,

wherein the processor acquires sensor data output from each sensor,

extracts two or more feature values for each of variables included in the sensor data of each sensor,

acquires output data by inputting input data which consists of a combination of the extracted feature value to a trained neural network model, and

compares the input data and the output data and determining an initial threshold value for detecting an anomaly in a case that there is no failure data based on comparison result values respectively including a Health Index (HI) indicating a device state index, and wherein the initial threshold value is determined based a largest comparison result value among the comparison result values.