Version 1.1, part of Release 2018b, includes the following enhancements:

  • Feature Selection Metrics: Evaluate features to determine best indicators of system degradation and improve accuracy of remaining useful life predictions
  • Features for Rotating Machinery: Extract the residual, difference, and regular signals from a time-synchronous averaged signal to generate diagnostic feature
  • fileEnsembleDatastore Object: Read all variable types from ensemble member while loading file only once
  • Ensemble Datastore Objects: Read multiple ensemble members in one operation
  • fileEnsembleDatastore Object: Create ensembles of files with multiple file extensions

See the Release Notes for details.

Version 1.0, part of Release 2018a, includes the following enhancements:

  • Survival, similarity, and time-series models for remaining useful life (RUL) estimation
  • Time, frequency, and time-frequency domain feature extraction methods for designing condition indicators
  • Managing and labeling of sensor data imported from local files, Amazon S3, Windows Azure Blob Storage, and Hadoop Distributed File System
  • Managing and labeling of simulated machine data from Simulink models
  • Examples for developing predictive maintenance algorithms for motors, gearboxes, batteries, and other machines

See the Release Notes for details.