Analysis of time series sensor measurements in physical systems

1. A method for analyzing time series data of a physical system, the method comprising:

a) storing a process graph representing the physical system, wherein all nodes of the process graph represent devices of the physical system and all edges of the process graph represent causal physical influences between the devices, wherein the devices comprise physical assets and sensors, where the sensors are configured to measure physical quantities;

b) storing physics rules representing relations between quantities;

c) storing a set of multiple physics models, wherein each of the multiple physics models comprises a physics rule selected from the stored physics rules and an assignment of quantities of the physics rule to properties of the physical system;

d) storing sensor time series data, wherein the sensor time series data comprises: i) a sequence of values of a physical quantity measured by one of the sensors paired with ii) a corresponding sequence of times when the values of the sequence of values were measured;

e) retrieving sensor data streams from the time series data, wherein each of the sensor data streams comprises a sequence of time-value pairs, wherein each of the sensor data streams is associated with a sensor identifier, a non-zero time offset, and a sampling period;

f) producing a metric data stream from the retrieved sensor data streams in accordance with a physics model selected from the stored set of multiple physics models;

wherein producing the metric data stream comprises:

i) synchronizing the sensor data streams by adjusting non-zero time offsets of the sensor data streams and adding interpolated values and times to the sensor data streams to produce synchronized sensor data streams with equal sampling periods: wherein the synchronized data streams have a common period given by a greatest common denominator of periods of the sensor data streams; and

ii) performing a point-wise computation over values of the synchronized sensor data streams in accordance with the physics model.

2. The method of claim 1 further comprising:

defining a machine learning model stream by specifying a type of machine learning model, a training data stream, a time window, and a period;

applying the machine learning model stream to data streams to produce a model output metric data stream.

3. The method of claim 2 further comprising:

determining a support value for the training data stream, where the support value is indicative of the fraction of missing values in the training data stream; and

determining whether the support value exceeds a predetermined threshold.

4. The method of claim 2 wherein the machine learning model is a model selected from the group consisting of linear regression, isolation forests, lightweight on-line detector of anomalies, logistic regression, random forests, and gradient-boosted trees.

5. The method of claim 1 further comprising producing from the synchronized sensor data streams interval streams, and/or key-value pairs of metadata.

6. The method of claim 1 wherein properties of the physical system comprise physical properties derived from a set of sensor nodes of the process graph.

7. The method of claim 1 wherein properties of the physical system comprise statistical properties derived from a set of sensor nodes of the process graph.

8. The method of claim 1 wherein the relations between quantities are relations between measurable physical quantities.

9. The method of claim 1 wherein the relations between quantities are relations between quantities derived from measurable physical quantities.