Full Content is available to subscribers

Subscribe/Learn More  >

Optimization of a Real-Time Simulator Based on Recurrent Neural Networks for Compressor Transient Behavior Prediction

[+] Author Affiliations
M. Venturini

University of Ferrara, Ferrara, Italy

Paper No. GT2006-90117, pp. 543-555; 13 pages
  • ASME Turbo Expo 2006: Power for Land, Sea, and Air
  • Volume 5: Marine; Microturbines and Small Turbomachinery; Oil and Gas Applications; Structures and Dynamics, Parts A and B
  • Barcelona, Spain, May 8–11, 2006
  • Conference Sponsors: International Gas Turbine Institute
  • ISBN: 0-7918-4240-1 | eISBN: 0-7918-3774-2
  • Copyright © 2006 by ASME


In the paper, feed-forward Recurrent Neural Networks with a single hidden layer and trained by using a back-propagation learning algorithm are studied and developed for the simulation of compressor behavior under unsteady conditions. The data used for training and testing the RNNs are both obtained by means of a non-linear physics-based model for compressor dynamic simulation (simulated data) and measured on a multi-stage axial-centrifugal small size compressor (field data). The analysis on simulated data deals with the evaluation of the influence of the number of training patterns and of each RNN input on model response, both for data not corrupted and corrupted with measurement errors, for different RNN configurations and different values of the total delay time. For RNN models trained directly on experimental data, the analysis of the influence of RNN input combination on model response is repeated, as carried out for models trained on simulated data, in order to evaluate real system dynamic behavior. Then, predictor RNNs (i.e., those which do not include among the inputs the exogenous inputs evaluated at the same time step as the output vector) are developed and a discussion about their capabilities is carried out. The analysis on simulated data led to the conclusion that, to improve RNN performance, it is beneficial the adoption of a one-time delayed RNN, with an as low as possible total delay time (in the paper, 0.1 s) and trained with an as high as possible number of training patterns (at least 500). The analysis of the influence of each input on RNN response, conducted for RNN models trained on field data, showed that the single-step-ahead predictor RNN allowed very good performance, comparable with that of RNN models with all inputs (overall error for each single calculation equal to 1.3% and 0.9% for the two test cases considered). Moreover, the analysis of multi-step-ahead predictor capabilities showed that the reduction of the number of RNN calculations is the key factor for improving its performance over a significant time horizon. In fact, when a high test data sampling time is chosen (in the paper, 0.24 s), prediction errors were acceptable (lower than 1.9%).

Copyright © 2006 by ASME



Interactive Graphics


Country-Specific Mortality and Growth Failure in Infancy and Yound Children and Association With Material Stature

Use interactive graphics and maps to view and sort country-specific infant and early dhildhood mortality and growth failure data and their association with maternal

Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In