Infinity Supercritical

Supercritical Fluid Extraction Search Engine - Infinity Supercritical CO2 Extraction Publications Search Engine - Cannabis Industry - Publications On Demand - Real Time Browser Display

This SDR - Spinning Disc Reactor and Cavitation Reactor search was updated real-time via Filemaker on:

SDR - Spinning Disc Reactor and Cavitation Reactor Contents List

Previous Page View | Next Page View

Search Completed. Publication Name:

tp-44-pap.pdf

Page Number: 001

PDF Text:

Indian Society for Non-Destructive Testing Hyderabad Chapter

Proc. National Seminar on Non-Destructive Evaluation Dec. 7 - 9, 2006, Hyderabad

Neural Network Based Methodology for Cavitation Detection in Pressure Dropping Devices of PFBR

P.K. Gupta1, P.A. Kumar1, A. Kaul2, G.K. Pandey1, G. Padmakumar1, V. Prakash1 and C. Anandbabu1

1Fast Reactor Technology Group, Indira Gandhi Centre for Atomic Research, Kalpakkam-603 102, pkgupta@igcar.gov.in

2Netaji Subash Institute of Technology, New Delhi

Abstract

This paper discusses a reliable, low cost and non-intrusive method of detecting the occurrence of cavitation in the pressure drop devices used for flow zoning in PFBR core. This involves classification of a pressure drop device, under given operating conditions, as cavitating or noncavitating using an adequately trained neural network. In this work, two models are assimilated, namely a process dependent ANN model and a generalized ANN model. The RMS of the wideband [10-100 kHz] acoustic signal and skewness of the acoustic time signal acquired from an accelerometer installed downstream of the test section are fed as features to the process model and generalized model respectively, for training. The networks are trained using feed forward back propagation training algorithm till the test error reduces to a predetermined level. The networks are tested using data not under the training domain and classification errors are derived. This paper discusses the measurements and instrumentation employed, experimental methodology followed, training and testing of the network and discussion on the results obtained.

Keywords: Cavitation detection, Artificial Neural Network (ANN), Root Mean Square (RMS), Feed forward back propagation, Network training, Testing, Classification error, Dependence function

1. Introduction

The PFBR core is divided into 15 flow zones to regulate flow in proportion to the heat generated in the Fuel Subassembly (FSA) [1]. This flow zoning has been achieved by means of installing the pressure drop devices at the foot of the SA. These pressure drop devices should meet the pressure drop requirement without any cavitation. The cavitation free performance of the device must be ensured because it can lead to reactivity perturbations, positive sodium void coefficient, dimensional changes due to erosion etc. There is an ongoing experimental program

in place for assessing the hydraulic performance of flow zoning devices in Fast Reactor Technology Group (FRTG). In this development program, many orifice configurations have been qualified for cavitation free performance, for use in the different flow zones of PFBR. To supplement the experimental program, an ANN based approach for cavitation detection is investigated in which an adequately trained and tested ANN is used to qualify a pressure drop device as cavitating or non-cavitating under given operating conditions so that cavitation characteristics of any device could be

NDE-2006

PDF Image:

 Supercritical Fluid Extraction tp-44-pap.pdf Page 001
SUPERCRITICAL CO2 EXTRACTOR FOR SALE: 5L, 10L and 20L - 2,000 psi Complete Supercritical Fluid Extraction System - closed loop. Our systems are made at our fabrication shop in the state of Washington. Systems available: 5L, 10L, 20L, 100L, and larger custom. Typical build time is 2 weeks. This is not a Apeks Supercritical or Waters Supercritical CO2 Fluid Extraction System. We believe it is a better system, since we use 1/2 inch tubing for the CO2 flow, which is 4 times the flow rate of a Apeks, and up to 64 times the flow rate of a Waters system. We believe the result is faster processing time and better yield from increased exposure to CO2 solvent. Perfect for extraction of terpenes, trichomes, and Cannabinoids. Supercritical CO2 Fluid Extraction is a art, and the operator must have experience and training to obtain results. Go to website

Search Engine Contact: greg@infinitysupercritical.com