An apparatus may include a processor that may be caused to access a plurality of measurements of a device. The processor may provide the plurality of measurements as an input to an autoencoder, the autoencoder being trained based on measurements of devices in working condition and access an output of the autoencoder, the output comprising a reconstruction of the input based on decoding an encoded version of the input. The processor may further be caused to determine whether the device will fail based on the output.

A predictive maintenance system and method for intelligent manufacturing equipment is provided. The predictive maintenance system includes a first-stage predictive maintenance module, a second-stage predictive maintenance module, and a maintenance decision module. The first-stage predictive maintenance module includes an acquisition module, a human-computer interaction module, a calculation module, and a storage module. The second-stage predictive maintenance module includes a communication module, a setting module, and a prediction module. The maintenance decision module is configured to receive a first-stage remaining service life calculated by the first-stage predictive maintenance module and a second-stage remaining service life predicted by the second-stage predictive maintenance module, and determine a predictive maintenance strategy of the intelligent manufacturing equipment according to the first-stage remaining service life and the second-stage remaining service life. The present disclosure may reduce unexpected shutdown, reduce the costs of operation and maintenance, and improve the efficiency of operation and maintenance.

The present invention provides a machine abnormality marking and abnormality prediction system connected with a factory host and including a parameter streaming unit connected with the machines, an abnormality reporting unit, a prediction analysis unit, and a neural network classifier. The parameter streaming unit and the abnormal reporting unit collect data of each machine in the factory, and the collected data are compared and analyzed with historical records by the prediction analysis unit. The generated parameter values can be continuously compared with the collected data to predict the state of each machine in the factory, and provide an early warning of possible abnormality or need of maintenance, so that the personnel in the factory can arrange production line maintenance or capacity adjustment in advance or adjust the machine of the factory production line, to avoid occasional shutdown and reduce factory losses.

An example embodiment involves rules related to repairing software programs, wherein the rules associate indications of software program failures with repair applications that are configured to correct the software program failures. One or more processors are configured to: (i) receive, by a predictive model, a representation of an execution history of a particular software program, wherein the predictive model has been trained on a corpus of execution histories of the software programs; (ii) generate, by the predictive model and from the execution history, a failure prediction for the particular software program; (iii) receive, by an automated repair controller application, the failure prediction from the predictive model; (iv) based on applying the rules to the failure prediction, determine, by the automated repair controller application, a repair application from the repair applications; and (v) cause, by the automated repair controller application, the repair application to be executed within the network.

Methods and systems are provided for monitoring a health of a vehicle component. In one embodiment, a method is provided, comprising dividing a population of vehicles of a connected vehicle population into a plurality of vehicle classes; for each vehicle class of the plurality of vehicle classes, training a class-specific model of the vehicle class to predict a health status variable of a vehicle component included in the vehicle class based on labelled data from historic databases and calibration data; and for each vehicle class of the plurality of vehicle classes, using a first Federated Learning strategy to request local model data from each vehicle of a plurality of vehicles of the vehicle class; receive the local model data from the plurality of vehicles; update the class-specific model based on the received local model data; and send updated parameters of the class-specific model to vehicles included in the vehicle class.

An analysis device according to an aspect of the present disclosure: acquires a plurality of pieces of premise information and measurement data which relate to states of a plurality of mechanisms which configure a production line; identifies causal relationships among the plurality of mechanisms by statistically analyzing the plurality of pieces of measurement data under constraint conditions imposed by the premise information; outputs causal relationship information indicating the identified causal relationships; accepts a revision to the causal relationships indicated by the outputted causal relationship information; revises the premise information so as to impose the constraint conditions which comport with the revised causal relationships; and saves the revised premise information.

A computer-readable medium may include instructions that may cause a processor to perform operations that may include receiving audio data representative of sound waves generated by industrial devices and extracting features from the audio data. The features may be representative of a portion of the audio data. The operations may also include identifying a subset of the features based on distances between each of the plurality of features in an information space. The information space may include known clusters. The operations may then include determining that the subset of the features corresponds to an unknown cluster in the information space, performing a constrained classification operation based on each feature of the subset of the features to identify a new known cluster for the information space, and modifying operations of the industrial devices based on the new known cluster.

A modular analysis engine provided classification of variables and data in an industrial automation environment. The module may be instantiated upon receipt of an input data structure, such as containing annotated data for any desired variables related to the machine or process monitored and/or controlled. The data may be provided in a batch or the engine may operate on streaming data. The output of the module may be a data structure that can be used by other modules, such as for modeling, optimization, and control. The classification may allow for insightful analysis, such as for textual classification of alarms provided in the automation setting.

Methods and systems for predicting failure in a cyber-physical system include determining a prediction index based on a comparison of input time series, from respective sensors in a cyber-physical system, to failure precursors. A failure precursor is detected in the input time series, responsive to a comparison of the prediction index to a threshold. A subset of the sensors associated with the failure precursor is determined, based on a gradient of the prediction index. A corrective action is performed responsive to the determined subset of sensors.

A method for identifying a fault of at least one mechanical machine, including causing a first plurality of sensors coupled to a corresponding first plurality of mechanical machines to acquire a first plurality of sets of signals emanating from the first plurality of mechanical machines, the first plurality of mechanical machines sharing at least one characteristic, supplying at least the first plurality of sets of signals of the first plurality of mechanical machines to a pre-existing fault classifier previously trained to automatically identify faults of a second plurality of mechanical machines based on signals emanating therefrom and previously acquired by a second plurality of sensors, the second plurality of sensors being of a different type than the first plurality of sensors, the second plurality of mechanical machines sharing the at least one characteristic, modifying the pre-existing fault classifier by employing transfer learning, based at least on the first plurality of sets of signals of the first plurality of mechanical machines, thereby providing a modified fault classifier, applying the modified fault classifier to at least one additional set of signals acquired by at least one sensor of the first plurality of sensors and emanating from at least one given mechanical machine sharing the at least one characteristic, the modified fault classifier being configured to automatically identify at least one fault of the at least one given mechanical machine based on the at least one additional set of signals, and providing a human sensible output, by an output device, including at least identification of the fault of the at least one given mechanical machine, at least one of a repair or maintenance operation being performed based on the human sensible output.