A system and a method for monitoring a device including the steps of obtaining operation information from a device, wherein the operation information is associated with the condition of the device in operation; and processing the operation information with a device modelling engine to determine one or more operation conditions of the device.

A method for detecting at least one machine defect provides defining from the machine kinematic data at least one condition indicator reflecting its condition, recording operating condition data of the machine and condition monitoring data of the machine during a predetermined period when the machine is operating normally, determining condition indicator values using condition monitoring data, and for determining current condition indicator values from the at least one condition indicator and the current condition monitoring data, a machine learning algorithm, predicting condition indicator values with respect to the current operating condition data, training the machine learning algorithm to establish a relation between the operating condition data and condition indicator values, and comparing the current condition indicator values and the predicted condition indicator values, and for determining if the machine is presumed to operate normally or not according to the result of the comparison.

A method of operating a chiller to avoid future surge events, the method comprises applying chiller operating data associated with a chiller as an input to one or more machine learning models; and generating a threshold for a controllable chiller variable to prevent a future chiller surge event from occurring based on an output of the one or more machine learning models, further comprising affecting operation of the chiller based on the threshold to prevent the future chiller surge event from occurring. The method enables automatic control of a chiller to avoid future chiller surge events.

Disclosed herein are methods, systems, and devices for utilizing distributed reinforcement learning and consensus control to most effectively generate and utilize energy. In some embodiments, individual turbines within a wind farm may communicate to reach a consensus as to the desired yaw angle based on the wind conditions.

Systems and methods to monitor a wind turbine azimuth drivetrain. Azimuth variation characteristics data are accumulated from wind turbines over a period of time. Clusters of values within the azimuth variation characteristics data are identified and a respective condition of the main drivetrain is associated with different clusters of values. After the associating, a measured set of azimuth variation characteristics data is received. A cluster corresponds to values in the measured set of azimuth variation characteristics data is determined and a condition associated with that cluster is determined to be a condition associated with the subject main drivetrain. That condition is then reported.

The present invention provides a hydraulic turbine cavitation acoustic signal identification method based on big data machine learning. According to the method, time sequence clustering based on multiple operating conditions under the multi-output condition of the hydraulic turbine set is performed by utilizing an neural network, characteristic quantities of the hydraulic turbine set under a steady condition in a healthy state is screened; a random forest algorithm is introduced to perform feature screening of multiple measuring points under steady-state operation of the hydraulic turbine set, optimal feature measuring points and optimal feature subsets are extracted, finally a health state prediction model is constructed by using gated recurrent units; whether incipient cavitation is present in the equipment is judged. The present invention can effectively identify the occurrence of incipient cavitation in the hydraulic turbine set, reducing unnecessary shutdown of the equipment and prolonging the service life.

A method and apparatus for monitoring formation of ice on a wind turbine blade are provided. The method includes: capturing an image of the blade through a camera; detecting a region of the blade from the captured image; clearing image information of a background region from the captured image, which is in the captured image except for the region of the blade, to obtain a blade image; and inputting the obtained blade image into a recognition model of ice on a blade obtained by training on a sample set, to determine whether ice is on the captured blade, wherein the sample set comprises a plurality of blade images indicating that ice is on blades.

A method of operating a chiller to avoid future surge events, the method comprises applying chiller operating data associated with a chiller as an input to one or more machine learning models; and generating a threshold for a controllable chiller variable to prevent a future chiller surge event from occurring based on an output of the one or more machine learning models, further comprising affecting operation of the chiller based on the threshold to prevent the future chiller surge event from occurring. The method enables automatic control of a chiller to avoid future chiller surge events.

A method of correcting turbine underperformance includes calculating a power production curve using monitored data, detecting changes between the monitored data and a baseline power production curve, generating operability curves for paired operational variables from the monitored data, detecting changes between the operability curves and corresponding baseline operability curves, comparing the changes to a respective predetermined metric, and if the change exceeds the metric, providing feedback to a turbine control system identifying at least one of the paired operational variables for each paired variable in excess of the metric. A system and a non-transitory computer-readable medium are also disclosed.

The present invention relates to a method of operating a wind turbine with an operational parameter where values of the operational parameter are obtained by different sensors and compared to determine the validity of the value. A first value and a second value of the operational parameter are obtained different sensors and validated by comparing the two values. The wind turbine being operated using a validated value as the operational parameter. The two sensors are selected among a trained machine learning model, a reference sensor and a computerized physical model.