A REAL TIME COAL CONTENT ORE GRADE (C2OG) SENSOR

Swanson, Dr. Rand

National Energy Technology Laboratory (U.S.)

关键词: Platinum;    Spectrometers;    On-Line Measurement Systems;    Classification;    01 Coal, Lignite, And Peat;   

DOI  :  10.2172/791709
RP-ID  :  FC26-01NT41057--02
RP-ID  :  FC26-01NT41057
RP-ID  :  791709

美国|英语

来源: UNT Digital Library

【 摘 要 】

The overall approach of this effort is to spectrally image ore or coal, and then use the spectral content (i.e., the particular colors of the ore or coal) to differentiate between the ore or coal grades. Currently, experts with practiced eyes do just this to identify the grade of platinum/palladium ore from the Stillwater Mine in south-central Montana. Additionally, trained eyes can identify high-sulfur and high-ash coal visually. The premise of this effort is that machine vision can accomplish this same differentiation. During the first quarter, machine vision results using a digital color camera did not correlate as well with assay results for platinum/palladium ore as would be required for a commercial device. One of the possible reasons for this is that the digital camera did not provide enough spectral information to obtain good differentiation between the sulfides associated with high-grade platinum/palladium ore and background interference, most notably yellow grease that contaminates some of the sample and green colored rock. The second quarter efforts have largely been devoted to implementing an imaging spectrometer for machine vision. In brief, modifying an imaging spectrometer that was designed for remote sensing from a Remotely Controlled (RC) airplane has done this. The imaging spectrometer provides 320 spectral channels, allowing for much better spectral resolution that can be obtained with a digital color camera, which provides 3 spectral channels. Preliminary results, as discussed below in more detail, are encouraging. The technical portion of the report below is organized into subsections as dictated by the DoE contract for this effort. These sections are: Experimental Apparatus, Experimental and Operating Data, Data Reduction, and Hypothesis and Conclusions. Partners in this effort are: Montana Tech of the University of Montana, Stillwater Mining Co., Western Syncoal, and the Montana Board of Research and Commercialization.