eduzhai > Applied Sciences > Engineering >

A Novel Method For Designing Transferable Soft Sensors And Its Application

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: In this paper, a new approach is proposed for designing transferable softsensors. Soft sensing is one of the significant applications of data-drivenmethods in the condition monitoring of plants. While hard sensors can be easilyused in various plants, soft sensors are confined to the specific plant theyare designed for and cannot be used in a new plant or even used in some newworking conditions in the same plant. In this paper, a solution is proposed forthis underlying obstacle in data-driven condition monitoring systems.Data-driven methods suffer from the fact that the distribution of the data bywhich the models are constructed may not be the same as the distribution of thedata to which the model will be applied. This ultimately leads to the declineof models accuracy. We proposed a new transfer learning (TL) based regressionmethod, called Domain Adversarial Neural Network Regression (DANN-R), andemployed it for designing transferable soft sensors. We used data collectedfrom the SCADA system of an industrial power plant to comprehensivelyinvestigate the functionality of the proposed method. The result reveals thatthe proposed transferable soft sensor can successfully adapt to new plants.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...