Abstract
In recent years, the data-driven soft sensor has achieved unprecedented development, which can update the model in real-time and save many costs. Most traditional methods are only suitable for single-mode processes, but the existing working conditions are generally multi-modal. In this paper, we propose an improved JITL method to modeling the complex industrial process. First, we apply AE to reduce the data dimensionality, exploring the intrinsic structure of data. Then a JITL strategy is adopted to select similar samples on the low-dimensional data and use them for the local model. Finally, PLS is developed for each model and predicts the variables. A three-phase flow case verifies the effectiveness of the proposed method. We find that compared with other methods, the proposed method has the highest accuracy.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
