A machine learning apparatus that learns an alarm factor in a motor drive device includes a state observation unit that obtains a feature amount as a state variable from the motor drive device and an alarm factor as label data, the alarm factor corresponding to the feature amount, and a learning unit that generates a learning model for inferring a new alarm factor corresponding to a new feature amount, from a dataset created on a basis of a combination of the state variable and the label data. The feature amount includes at least one of a detected current value detected from the motor, a speed command value specifying a rotational speed of the motor, an output voltage value output to the motor, an estimated speed value of the motor, and a detected speed value of the motor.

An abnormality determination system includes a state quantity obtaining circuit and an abnormality determination circuit. The state quantity obtaining circuit is configured to obtain a state quantity associated with a mechanical system. The abnormality determination circuit is configured to, according to a learning content obtained in a machine learning process and based on the state quantity, determine as to at least one of an occurrence of an abnormality in the mechanical system, an occurrence position of the abnormality, and a cause of the abnormality.

A closed loop control system for controlling a plant comprises a controller includes an input arranged to receive a feedback signal from the plant. The controller is arranged to compare a value of the feedback signal to a set point value x(s) and to produce an error signal ε(s) that is at least partially dependent on a difference between the value of the feedback signal and the set point value. The controller also includes an output arranged to supply the error signal ε(s) to the plant. A monitor is arranged to compare a value of the error signal ε(s) produced by the controller to a threshold value and to produce a warning signal when the value of the error signal ε(s) exceeds the threshold value for a period of time greater than a predetermined period.

A method for measuring health of a motor includes: (a) measuring, by an electrocardiogram sensor, vibration of the motor to obtain at least two electrical signals with each of the at least two electrical signals representing a harmonic of the vibration of the motor; (b) comparing each of the at least two electrical signals with a corresponding baseline; (c) based on the comparison, determining whether any one of the at least two electrical signals includes one or more artifacts wherein an artifact in a respective one of the at least two electrical signals is a deviation from a respective one of the corresponding baseline; and (d) based on any one of the at least two electrical signals including the one or more artifacts, providing an estimated time to failure for the motor.

A processing apparatus acquires a first rise transition that is a rise transition of the temperature of a winding and a second rise transition that is a rise transition of the temperature detected by a temperature sensor in a state where voltage application for raising the temperature of the winding to a predetermined temperature is performed, determines a predetermined temperature characteristic model by calculating a predetermined parameter on the basis of the second rise transition, and further determines a winding temperature characteristic model by calculating a winding related parameter on the basis of the first rise transition.

An abnormality diagnosis system performs an abnormality diagnosis of a plurality of motor systems that include a motor for moving a moving body. The abnormality diagnosis system identifies a comparison target system that is the motor system, among the motor systems, that is a comparison target in relation to a diagnosis target system that is the motor system to be a target of the abnormality diagnosis. The abnormality diagnosis system acquires a state-related value that is a value related to an operation state of the motor from each of the diagnosis target system and the comparison target system. The abnormality diagnosis system performs a comparison of the state-related value that is acquired from the diagnosis target system and the state-related value that is acquired from the comparison target system, and diagnoses a presence or absence of an abnormality in the diagnosis target system using a result of the comparison.

A technique for performing condition monitoring of a plurality of components of a plurality of wind turbines is disclosed. The technique includes, for each of the plurality of wind turbines, measuring at least one time-series electrical output signal of the wind turbine generator, each time-series electrical output signal representing an aggregate of a plurality of subset signals. Such condition monitoring further includes mathematically extracting one or more such subset signals from the at least one time-series electrical output signal of the generator, and associating one or more such subset signals with one or more wind turbine components. A fault threshold value for each such one or more subset signals may be determined, and the one or more such subset signals are compared to such fault threshold value. Based on such association and comparison, the condition of such a wind turbine component of the plurality of wind turbines is determined.

Methods and systems for determining an alternator condition in a motor vehicle are provided. The method includes receiving a maximum cranking voltage and a maximum cranking voltage time stamp from the motor vehicle over an asset interface of the telematics device; receiving a maximum device voltage and a maximum device voltage time stamp from the motor vehicle over the asset interface, and determining a potential alternator undercharging condition if a duration between the maximum cranking voltage time stamp and the maximum device voltage time stamp is greater than an undercharging indicator duration threshold. Advantageously, an alternator may be repaired or replaced before it fails thus averting having the motor vehicles inoperable.

An abnormality detection method according to one aspect of the present disclosure is a method of detecting an abnormality in an AC signal to be input from an AC power supply. The method includes, where an ideal AC signal is represented as V.sub.0 sin ωt (V.sub.0: amplitude, co: angular frequency, t: time), calculating an arithmetic value including a value represented by sin.sup.2ωt+cos.sup.2ωt and determining that the AC signal is abnormal when the arithmetic value is out of a threshold range.

An HVAC system includes a blower, a motor drive, and a controller. A benchmark rate of the flow of air provided by the blower and a corresponding benchmark power output of the motor drive associated with operation of the blower at a test condition are received. The controller determines a first motor drive frequency at which the motor drive is operating. Based on the benchmark rate and a comparison of the first motor drive frequency to the predefined motor drive frequency, a first rate of the flow of air provided by the blower is determined. At a later time, a current power output of the motor drive is determined during operation of the blower at the test condition. Based on a comparison of the current benchmark power output to the benchmark power output, an updated benchmark rate of the flow of air provided by the blower is determined.