Abstract
The learning process of neural network in the impedance CT for void distribution measurement was improved. A void distribution function to express void distribution only by three parameters was introduced, so that 81 of output layer units in the previous network, was decreased to 3 units. Input layer units were reduced by eliminating duplicated current data and numbers of hidden layer units were minimized. The best learning parameters were recalculated about the slimed network by using the error back propagation technique. As the results, the learning speed of the impedance CT was about 30 times faster than that of the previous paper. Two types of experiments were conducted using a square duct. One used acrylic rods as dummy voids, the other used air-water two phase flow. Comparison between experiment and prediction indicated that the impedance CT method was able to predict void distributions with good accuracy.
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 callMore From: TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series B
Disclaimer: 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.
