Smart motor data analytics with real-time algorithm

A computer program comprising instructions which, when the program is executed by a computer, cause the computer to:

• continuously receive samples of input data based on at least one physical quantity over time of a rotating machine;
• deriving, in real-time, one or more samples of an envelope of the input data based on the samples of the input data by a rectification followed by an asynchronous complex In-phase Quadrature (IQ) envelope detector, or by a biquad filter approach;
• update in real-time one or more accumulator variables based on the one or more samples including a current sample and at least one preceding sample of the input data;
• compute in real-time one or more spectral features based on the one or more accumulator variables and one or more supplemental variables;
• and
• determine a condition of the rotating machine based on the one or more spectral features;
• wherein the one or more accumulator variables are updated in real-time based on the one or more samples including the current sample, a first preceding sample and a second preceding sample of the one or more samples and the one or more spectral features are computed in real-time by a Goertzel Algorithm;
• wherein the totality of the one or more accumulator variables is sufficient to determine the condition of the rotating machine, and wherein the m supplemental variables are temporarily needed for computing the one or more spectral features and the one or more supplemental variables are not based on the received samples of the input data.

A system comprising:

• a rotating machine;
• at least one sensor configured to measure at least one physical quantity over time of the rotating machine;
• and
• a data processing system communicatively connected to the at least one sensor, the data processing system comprising:
• an interface configured to receive samples of input data based on at least one physical quantity over time of the rotating machine;
• and
• a processor configured to:
• continuously receive samples of input data based on at least one physical quantity over time of a rotating machine;
• derive, in real-time, one or more samples of an envelope of the input data based on the samples of the input data by a rectification followed by an asynchronous complex In-phase Quadrature (IQ) envelope detector, or by a biquad filter approach;
• update in real-time one or more accumulator variables based on samples including a current sample and at least one preceding sample of the one or more samples;
• compute in real-time one or more spectral features based on the one or more accumulator variables and one or more supplemental variables;
• and
• determine a condition of the rotating machine based on the one or more spectral features;
• wherein the one or more accumulator variables are updated in real-time based on the one or more samples including the current sample, a first preceding sample, and a second preceding sample of the one or more samples and the one or more spectral features are computed in real-time by a Goertzel Algorithm;
• wherein the totality of the one or more accumulator variables is sufficient to determine the condition of the rotating machine, and wherein the m supplemental variables are temporarily needed for computing the one or more spectral features and the one or more supplemental variables are not based on the received samples of the input data;
• wherein the at least sensor is further configured to provide the measured at least one physical quantity as samples of input data to the data processing system.