Scalable system and engine for forecasting wind turbine failure
An example method utilizing different pipelines of a prediction system, comprises receiving event and alarm data from event logs, failure data, and asset data from Supervisory Control and Data Acquisition (SCADA) system(s), retrieving patterns of events, receiving historical sensor data from sensors of components of wind turbines, training a set of models to predict faults for each component using the patterns of events and historical sensor data, each model of a set having different observation time windows and lead time windows, evaluating each model of a set using standardized metrics, comparing evaluations of each model of a set to select a model with preferred lead time and accuracy, receiving current sensor data from the sensors of the components, applying the selected model(s) to the current sensor data to generate a component failure prediction, comparing the component failure prediction to a threshold, and generating an alert and report based on the comparison to the threshold.
1. A non-transitory computer readable medium comprising executable instructions, the executable instructions being executable by one or more processors to perform a method, the method comprising:
receiving log data from one or more Supervisory Control and Control and Data Acquisition (SCADA) systems that monitor any number of wind turbines, the log data being generated during a first period of time;
receiving historical wind turbine component failure data and wind turbine asset data from the one or more SCADA systems, the historical wind turbine component failure data and wine turbine asset data being generated during the first period of time;
creating cohort instances based on the wind turbine failure data and wind turbine asset data, each cohort representing a subset of the wind turbines, the subset of the wind turbines including a same type of controller and a similar geographical location, the geographical location of the wind turbines of the subset of wind turbines being within the wind turbine asset data;
generating a feature matrix, the feature matrix including a unique feature identifier for each feature of the log data in the feature matrix;
extracting patterns of events from the feature matrix based on the cohort instances;
receiving first historical sensor data of the first time period, the first historical sensor data including sensor data from one or more sensors of one or more components of renewable energy assets, the first historical sensor data indicating at least one first failure associated with the one or more components of the renewable energy asset during the first time period;
generating a first set of failure prediction models using the first historical sensor data and the patterns of events, each of the first set of failure prediction models being trained using different amounts of first historical sensor data based on different observation time windows and different lead time windows, each observation time window including a time period during which first historical data is generated, each lead time window including a period of time before a predicted failure;
selecting a first selected failure prediction model from the first set of failure prediction models based on the observation time windows and lead time windows, the first selected failure prediction model including the lead time window;
receiving first current sensor data of a second time period, the first current sensor data including sensor data from the one or more sensors of the one or more components of the renewable energy asset;
applying the first selected failure prediction model to the first current sensor data to generate a first failure prediction on a failure of at least one component of the one or more components;
comparing the first failure prediction to a trigger criteria; and
generating and transmitting a first alert based on the comparison of the failure prediction to the trigger criteria, the alert indicating the at least one component of the one or more components and information regarding the failure prediction.
2. The non-transitory computer readable medium of claim 1 , the method further comprises cleaning of the log data from the event logs, the cleaning comprising discarding event data missing expected information.
3. The non-transitory computer readable medium of claim 1 , wherein extracting patterns of events from the feature matrix comprises counting a number of event codes of events that occurred during a time interval using the feature matrix and sequence the event codes to include dynamics of events in a longitudinal time dimension.
4. The non-transitory computer readable medium of claim 3 , wherein the time interval is the time period, the first historical sensor data being generated by the one or more sensors during the time period.
5. The non-transitory computer readable medium of claim 1 , wherein events of the patterns of events occur during the time period.
6. The non-transitory computer readable medium of claim 1 , wherein the renewable energy asset is a wind turbine.
7. The non-transitory computer readable medium of claim 1 , wherein the first historical sensor data is the historical wind turbine component failure data.
8. The non-transitory computer readable medium of claim 1 , wherein the method further comprises retrieving the trigger criteria from a datastore including a plurality of trigger criteria, the trigger criteria being retrieved based at least in part on the at least one component of the one or more components.
9. The non-transitory computer readable medium of claim 1 , wherein the method further comprises receiving operational signals from the one or more SCADA systems and extracting features from the operational signals, wherein generating the first set of failure prediction models uses the first historical sensor data, the patterns of events, and extracted features from the operational signals.
10. A component failure prediction system, comprising:
at least one processor; and
memory containing instructions, the instructions being executable by the at least one processor to:
receive log data from one or more Supervisory Control and Control and Data Acquisition (SCADA) systems that monitor any number of wind turbines, the log data being generated during a first period of time;
receive historical wind turbine component failure data and wind turbine asset data from the one or more SCADA systems, the historical wind turbine component failure data and wine turbine asset data being generated during the first period of time;
create cohort instances based on the wind turbine failure data and wind turbine asset data, each cohort representing a subset of the wind turbines, the subset of the wind turbines including a same type of controller and a similar geographical location, the geographical location of the wind turbines of the subset of wind turbines being within the wind turbine asset data;
generate a feature matrix, the feature matrix including a unique feature identifier for each feature of the log data in the feature matrix;
extracting patterns of events from the feature matrix based on the cohort instances;
receive first historical sensor data of the first time period, the first historical sensor data including sensor data from one or more sensors of one or more components of renewable energy assets, the first historical sensor data indicating at least one first failure associated with the one or more components of the renewable energy asset during the first time period;
generate a first set of failure prediction models using the first historical sensor data and the patterns of events, each of the first set of failure prediction models being trained using different amounts of first historical sensor data based on different observation time windows and different lead time windows, each observation time window including a time period during which first historical data is generated, each lead time window including a period of time before a predicted failure;
select a first selected failure prediction model from the first set of failure prediction models based on the observation time windows and lead time windows, the first selected failure prediction model including the lead time window;
receive first current sensor data of a second time period, the first current sensor data including sensor data from the one or more sensors of the one or more components of the renewable energy asset;
apply the first selected failure prediction model to the first current sensor data to generate a first failure prediction on a failure of at least one component of the one or more components;
compare the first failure prediction to a trigger criteria; and
generate and transmit a first alert based on the comparison of the failure prediction to the trigger criteria, the alert indicating the at least one component of the one or more components and information regarding the failure prediction.
11. The system of claim 10 , wherein the instructions being further executable by the at least one processor to: clean the log data from the event logs, the cleaning comprising discarding event data missing expected information.
12. The system of claim 10 , wherein extract patterns of events from the feature matrix comprises counting a number of event codes of events that occurred during a time interval using the feature matrix and sequence the event codes to include dynamics of events in a longitudinal time dimension.
13. The system of claim 12 , wherein the time interval is the time period, the first historical sensor data being generated by the one or more sensors during the time period.
14. The system of claim 10 , wherein events of the patterns of events occur during the time period.
15. The system of claim 10 , wherein the renewable energy asset is a wind turbine.
16. The system of claim 10 , wherein the first historical sensor data is the historical wind turbine component failure data.
17. The system of claim 10 , wherein the method further comprises retrieving the trigger criteria from a datastore including a plurality of trigger criteria, the trigger criteria being retrieved based at least in part on the at least one component of the one or more components.
18. The system of claim 10 , wherein the method further comprises receiving operational signals from the one or more SCADA systems and extracting features from the operational signals, wherein generating the first set of failure prediction models uses the first historical sensor data, the patterns of events, and extracted features from the operational signals.
19. A method comprising:
receive log data from one or more Supervisory Control and Control and Data Acquisition (SCADA) systems that monitor any number of wind turbines, the log data being generated during a first period of time;
receive historical wind turbine component failure data and wind turbine asset data from the one or more SCADA systems, the historical wind turbine component failure data and wine turbine asset data being generated during the first period of time;
create cohort instances based on the wind turbine failure data and wind turbine asset data, each cohort representing a subset of the wind turbines, the subset of the wind turbines including a same type of controller and a similar geographical location, the geographical location of the wind turbines of the subset of wind turbines being within the wind turbine asset data;
generate a feature matrix, the feature matrix including a unique feature identifier for each feature of the log data in the feature matrix;
extracting patterns of events from the feature matrix based on the cohort instances;
receive first historical sensor data of the first time period, the first historical sensor data including sensor data from one or more sensors of one or more components of renewable energy assets, the first historical sensor data indicating at least one first failure associated with the one or more components of the renewable energy asset during the first time period;
generating a first set of failure prediction models using the first historical sensor data and the patterns of events, each of the first set of failure prediction models being trained using different amounts of first historical sensor data based on different observation time windows and different lead time windows, each observation time window including a time period during which first historical data is generated, each lead time window including a period of time before a predicted failure;
selecting a first selected failure prediction model from the first set of failure prediction models based on the observation time windows and lead time windows, the first selected failure prediction model including the lead time window;
receiving first current sensor data of a second time period, the first current sensor data including sensor data from the one or more sensors of the one or more components of the renewable energy asset;
applying the first selected failure prediction model to the first current sensor data to generate a first failure prediction on a failure of at least one component of the one or more components;
comparing the first failure prediction to a trigger criteria; and
generating and transmitting a first alert based on the comparison of the failure prediction to the trigger criteria, the alert indicating the at least one component of the one or more components and information regarding the failure prediction.