Systems and methods for determining operational relationships in building automation and control networks
Techniques for determining point type, equipment type, equipment instance, and equipment relationship for different points associated with pieces of equipment located at a building is described. The techniques may include obtaining data corresponding to point(s) associated with one or more pieces of equipment located at a building controlled by a building control network, and using the data and statistical model(s) to determine point type(s), equipment type(s), and/or equipment instance(s) for the point(s). The techniques may include determining operational relationships between different pieces of equipment located at the building.
1 . A computer-implemented method, comprising:
using at least one hardware processor to perform:
obtaining first data for a first piece of equipment located at a building controlled by a building control network, the first data including time series data corresponding to one or more first measurement points associated with the first piece of equipment;
determining, using the first data and at least one statistical model trained using training data indicating a plurality of measurement point types and a plurality of equipment types for different pieces of equipment, at least one measurement point type for the one or more first measurement points,
wherein the at least one statistical model includes an encoder, and determining the at least one measurement point type for the one or more first measurement points further comprises determining, using the encoder and the first data, at least one feature of the first data, wherein the at least one feature of the first data includes a ratio of consecutive data values in the first data that are substantially similar to a total number of data values in the first data; and
monitoring operation of equipment in the building based, at least in part, on the at least one measurement point type determined for the one or more first measurement points, wherein monitoring operation of equipment in the building comprises:
identifying, based at least in part on data values associated with a first measurement point of the one or more first measurement points and the measurement point type determined for the first measurement point, one or more pieces of equipment in the building is in need of service; and
outputting a recommendation to calibrate or replace the first piece of equipment and/or a second piece of equipment having an operational relationship with the first piece of equipment in response to determining one or more pieces of equipment in the building is in need of service.
2 . The method of claim 1 , further comprising:
obtaining second data for a second piece of equipment located at the building, the second data including time series data corresponding to one or more second measurement points associated with the second piece of equipment; and
determining, using the second data and the at least one statistical model, at least one measurement point type for the one or more second measurement points and an equipment type for the second piece of equipment.
3 . The method of claim 1 , wherein the at least one measurement point type includes one or more measurement point types selected from the group consisting of: a sensor, an actuator, a setpoint, and an alarm.
4 . The method of claim 1 , wherein the method further comprises determining, using the first data and the at least one statistical model, an equipment type for the first piece of equipment.
5 . The method of claim 4 , wherein the equipment type is selected from the group consisting of: an air handling unit, a variable-air-volume box, a boiler, a chiller, a fan, a filter, and a thermostat.
6 . The method of claim 1 , wherein determining the at least one measurement point type for the one or more first measurement points further comprises providing the first data as an input to the encoder and obtaining an output indicating the at least one feature.
7 . The method of claim 1 , wherein the at least one feature of the first data includes at least one selected from the group consisting of: a mean value for the first data, a median value for the first data, a standard deviation value for the first data, a kurtosis value for the first data, a skewness value for the first data, a minimum value for the first data, a maximum value for the first data, a median absolute deviation value for the first data, a mean absolute deviation value for the first data, and an interquartile range value for the first data, an autocorrelation value for the first data.
8 . The method of claim 1 , wherein the first data corresponds to one measurement point associated with the first piece of equipment, and the at least one feature of the first data includes a correlation value between the first data and second data corresponding to another measurement point of the first piece of equipment.
9 . The method of claim 1 , wherein the at least one feature of the first data includes a percentage of data values in the first data containing a decimal point.
10 . The method of claim 1 , wherein the at least one feature of the first data includes a correlation value between the time series data of the first data and outdoor temperature for a location of the building.
11 . The method of claim 1 , wherein the at least one feature of the first data includes at least one amplitude of one or more Fourier harmonics of the first data.
12 . The method of claim 1 , wherein the at least one feature of the first data includes at least frequency corresponding to one or more Fourier harmonics of the first data.
13 . The method of claim 1 , wherein the at least one statistical model includes at least one classifier, and determining the at least one measurement point type for the one or more first measurement points further comprises determining, using the at least one classifier and the at least one feature, the at least one measurement point type for the one or more first measurement points.
14 . The method of claim 13 , wherein determining the at least one measurement point type for the one or more first measurements points further comprises providing the at least one feature as an input to the at least one classifier and obtaining an output indicating the at least one measurement point type.
15 . The method of claim 14 , wherein the output includes values corresponding to a plurality of measurement point types, and wherein determining the at least one measurement point type further comprises selecting, based on the values corresponding to the plurality of measurement point types, the at least one measurement point type for the one or more first measurement points from among the plurality of measurement point types.
16 . The method of claim 14 , wherein the output includes values corresponding to a plurality of equipment types, and the method further comprises determining an equipment type for the first piece of equipment at least in part by selecting, based on the values corresponding to the plurality of equipment types, the equipment type for the first piece of equipment from among the plurality of equipment types.
17 . The method of claim 13 , wherein the at least one classifier comprises one or more classifiers selected from the group consisting of: a support vector machine classifier, a logistic regression classifier, a gradient boosted classifier, a decision tree classifier, a Bayesian classifier, a Bayesian network classifier, a random forest classifier, a k-nearest neighbors classifier, a neural network classifier, and an extremely randomized trees classifier.
18 . The method of claim 1 , wherein determining the at least one measurement point type for the one or more first measurement points further comprises providing the first data as an input to the at least one statistical model and obtaining an output indicating the at least one measurement point type as an output.
19 . The method of claim 18 , wherein the at least one statistical model comprises at least one neural network.
20 . The method of claim 19 , wherein the at least one neural network comprises a 1-dimensional convolutional neural network.
21 . The method of claim 19 , further comprising:
performing a normalization process on the first data to obtain normalized first data,
wherein determining the at least one measurement point type for the one or more first measurement points further comprises:
providing the normalized first data as an input to the at least one neural network; and
obtaining an output indicating the at least one measurement point type and the equipment type as an output.
22 . The method of claim 19 , further comprising:
training the at least one neural network using the training data,
wherein training the at least one neural network further comprises performing layer normalization.
23 . The method of claim 19 , further comprising:
applying a multidimensional transformation process to the first data to obtain transformed first data,
wherein determining the at least one measurement point type for the one or more first measurement points further comprises:
providing the transformed first data as an input to the at least one neural network; and
obtaining an output indicating the at least one measurement point type as an output.
24 . The method of claim 23 , wherein the transformed first data includes a multidimensional matrix.
25 . The method of claim 23 , wherein the multidimensional transformation process is a geometrical transformation process.
26 . The method of claim 23 , wherein the multidimensional transformation process is a probabilistic transformation process.
27 . The method of claim 1 , further comprising:
determining, based, at least in part, on the at least one measurement point type for the one or more first measurement points that the second piece of equipment has the operational relationship with the first piece of equipment.
28 . A computer-implemented method, comprising:
using at least one hardware processor to perform:
obtaining first data for a first piece of equipment located at a building controlled by a building control network, the first data including time series data corresponding to one or more first measurement points associated with the first piece of equipment;
determining, using the first data and at least one statistical model trained using training data indicating a plurality of measurement point types and a plurality of equipment types for different pieces of equipment, at least one measurement point type for the one or more first measurement points, wherein the first data further comprises text data corresponding to the one or more first measurement points associated with the first piece of equipment, and the at least one statistical model further comprises a first statistical model and a second statistical model;
providing the time series data as an input to the first statistical model to obtain a first output;
providing the text data as an input to the second statistical model to obtain a second output,
wherein determining the at least one measurement point type for the one or more first measurement points comprises determining, using the first output and the second output, the at least one measurement point type for the one or more first measurement points; and
monitoring operation of equipment in the building based, at least in part, on the at least one measurement point type determined for the one or more first measurement points, wherein monitoring operation of equipment in the building comprises:
identifying, based at least in part on data values associated with a first measurement point of the one or more first measurement points and the measurement point type determined for the first measurement point, one or more pieces of equipment in the building is in need of service; and
outputting a recommendation to calibrate or replace the first piece of equipment and/or a second piece of equipment having an operational relationship with the first piece of equipment in response to determining one or more pieces of equipment in the building is in need of service.
29 . The method of claim 28 , wherein the at least one statistical model further comprises a third statistical model, and determining the at least one measurement point type for the one or more first measurement points further comprises providing the first output and the second output as inputs to the third statistical model to obtain an output indicating the at least one measurement point type.
30 . A system comprising:
at least one hardware processor; and
at least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by the at least one hardware processor, cause the at least one hardware processor to:
obtain first data for a first piece of equipment located at a building controlled by a building control network, the first data including time series data corresponding to one or more first measurement points associated with the first piece of equipment;
determine, using the first data and at least one statistical model trained using training data indicating a plurality of measurement point types and a plurality of equipment types for different pieces of equipment, at least one measurement point type for the one or more first measurement points,
wherein the at least one statistical model includes an encoder, and determining the at least one measurement point type for the one or more first measurement points further comprises determining, using the encoder and the first data, at least one feature of the first data, wherein the at least one feature of the first data includes a ratio of consecutive data values in the first data that are substantially similar to a total number of data values in the first data; and
monitor operation of equipment in the building based, at least in part, on the at least one measurement point type determined for the one or more first measurement points, wherein monitoring operation of equipment in the building comprises:
identifying, based at least in part on data values associated with a first measurement point of the one or more first measurement points and the measurement point type determined for the first measurement point, one or more pieces of equipment in the building is in need of service; and
outputting a recommendation to calibrate or replace the first piece of equipment and/or a second piece of equipment having an operational relationship with the first piece of equipment in response to determining one or more pieces of equipment in the building is in need of service.
31 . At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one hardware processor, cause the at least one hardware processor to:
obtain first data for a first piece of equipment located at a building controlled by a building control network, the first data including time series data corresponding to one or more first measurement points associated with the first piece of equipment;
determine, using the first data and at least one statistical model trained using training data indicating a plurality of measurement point types and a plurality of equipment types for different pieces of equipment, at least one measurement point type for the one or more first measurement points, wherein the first data further comprises text data corresponding to the one or more first measurement points associated with the first piece of equipment, and the at least one statistical model further comprises a first statistical model and a second statistical model;
provide the time series data as an input to the first statistical model to obtain a first output;
provide the text data as an input to the second statistical model to obtain a second output, wherein determining the at least one measurement point type for the one or more first measurement points comprises determining, using the first output and the second output, the at least one measurement point type for the one or more first measurement points; and
monitor operation of equipment in the building based, at least in part, on the at least one measurement point type determined for the one or more first measurement points, wherein monitoring operation of equipment in the building comprises:
identifying, based at least in part on data values associated with a first measurement point of the one or more first measurement points and the measurement point type determined for the first measurement point, one or more pieces of equipment in the building is in need of service; and
outputting a recommendation to calibrate or replace the first piece of equipment and/or a second piece of equipment having an operational relationship with the first piece of equipment in response to determining one or more pieces of equipment in the building is in need of service.