Anomaly detection with correlation coefficients

A system, comprising:

• a processor;
• and
• a memory having instructions stored thereon, which, when executed by the processor, performs an operation for detecting an anomaly is sensor data generated in a substrate processing apparatus, the operation comprising:
• receiving a plurality of data sets, each data set corresponding to a sensor in a plurality of sensors disposed in a substrate processing apparatus, wherein each sensor is configured to monitor a respective substrate processing metric;
• selecting a first data set corresponding to a first substrate processing metric monitored by the first sensor disposed in the substrate processing apparatus;
• selecting a second data set corresponding to a second substrate processing metric monitored by the second sensor disposed in the substrate processing apparatus, wherein the first sensor and the second sensor are defined as a sensor pair;
• for each of the first and second data sets, generating a reference correlation by selecting a subset of values in each data set;
• z-normalizing the difference of correlation to the reference correlation;
• filtering the z-normalized data set to remove one or more outliers from the z-normalized data set to generate a filtered data set;
• identifying one or more anomalies in the filtered data set;
• and
• adjusting process parameters of the substrate processing apparatus, based on the one or more identified anomalies from the filtered data set.

A non-transitory computer readable medium having instructions stored thereon, which, when executed by a processor, cause the processor to perform a method of detecting an anomaly is sensor data generated in a substrate processing apparatus, comprising:

• receiving a plurality of data sets, each data set corresponding to a sensor in a plurality of sensors disposed in a substrate processing apparatus, wherein each sensor is configured to monitor a respective substrate processing metric;
• selecting a first data set corresponding to a first substrate processing metric monitored by the first sensor disposed in the substrate processing apparatus;
• selecting a second data set corresponding to a second substrate processing metric monitored by the second sensor disposed in the substrate processing apparatus, wherein the first sensor and the second sensor are defined as a sensor pair;
• for each of the first and second data sets, generating a reference correlation by selecting a subset of values in each data set;
• normalizing the difference of correlation to the reference correlation;
• filtering the z-normalized data set to remove one or more outliers from the z-normalized data set to generate a filtered data set;
• identifying one or more anomalies in the filtered data set;
• and
• adjusting process parameters of the substrate processing apparatus, based on the one or more identified anomalies from the filtered data set.