Preprocessor of abnormality sign diagnosing device and processing method of the same
A preprocessor includes a sensor data storage part that is connected to an abnormality sign diagnosing device and stores multi-dimensional sensor data, a data analysis processing part that calculates a variable value by using the multi-dimensional sensor data stored in the sensor data storage part, an analysis data storage part that stores the variable value calculated by the data analysis processing part, and an analysis data addition processing part that combines the multi-dimensional sensor data stored at the sensor data storage part and the variable value stored in the analysis data storage part and outputs a combined result to the abnormality sign diagnosing device.
1. A preprocessor comprising:
a first memory that is connected to an abnormality sign diagnosing device and stores multi-dimensional sensor data;
a first processor that calculates a correlation coefficient between two pieces of sensor data selected from the multi-dimensional sensor data stored in the first memory, extracts a specimen correlation coefficient value from the calculated correlation coefficient, selects the correlation coefficient based on whether either one of an average value and a standard deviation of the specimen correlation coefficient value or a correlation strength order represented by the specimen correlation coefficient value exceeds a threshold value, and calculates a variable value of the selected correlation coefficient;
a second memory that stores the variable value calculated by the first processor; and
a second processor that combines the multi-dimensional sensor data stored in the first memory and the variable value stored in the second memory, and outputs the combined result to the abnormality sign diagnosing device.
2. The preprocessor according to claim 1 , wherein
the multi-dimensional sensor data and the variable value that are combined by the second processor have an identical form.
3. The preprocessor according to claim 2 , wherein,
when at least one of the multi-dimensional sensor data and data on the variable value that have the identical form and are combined by the second processor is missing, the data corresponding to a time when the data is missing is supplemented or removed.
4. A processing method of a preprocessor that is connected to an abnormality sign diagnosing device and has a first memory and a second memory, the preprocessor performing the processing method comprising:
calculating a correlation coefficient between two pieces of sensor data selected from the multi-dimensional sensor data stored in the first memory,
extracting a specimen correlation coefficient value from the calculated correlation coefficient,
selecting the correlation coefficient as a variable by using either one of an average value and a standard deviation of the specimen correlation coefficient value or a correlation strength order represented by the specimen correlation coefficient value,
calculating a variable value of the selected variable,
storing the variable value in the second memory, and
combining the multi-dimensional sensor data stored in the first memory and the variable value stored in the second memory, and outputs the combined result to the abnormality sign diagnosing device.
5. The processing method of the preprocessor according to claim 4 , wherein
the multi-dimensional sensor data and the variable value that are combined by the preprocessor have an identical form.
6. The processing method of the preprocessor according to claim 5 , wherein,
when at least one of the multi-dimensional sensor data and data on the variable value that are combined by the preprocessor having the identical form is missing, the data corresponding to a time when the data is missing is supplemented or removed.