On-Conveyor Belt Analysis of Coal

The present report describes the design and plant-testing of a prototype commercial on-belt ash analyser at the Bengalla CHPP (Muswellbrook, NSW). This analyser uses the Neutron Inelastic Scatter (NIS) and Thermal Neutron Capture (TNC) Analysis (NITA) technique, which has been investigated  extensively in earlier ACARP projects C5051 (laboratory feasibility study)  and the previous stage of C9042 (development and lab testing of a field prototype and evaluate the suitability of the technique for specific energy and sulphur measurement). 

The Neutron Inelastic scatter and Thermal neutron capture Analyser (NITA) on-belt coal analysers will have the following characteristics and major benefits: 

• Minimal sensitivity to changes in mineral matter composition, resulting in much improved accuracy of ash analysis compared to on-belt Dual Energy Transmission (DUET) gauges.
• Minimal sensitivity to material segregation and changes in profile/loading, resulting in a similar level of accuracy to the commercial by-line Prompt Gamma Neutron Activation Analysis (PGNAA) gauges.
• The ability to provide simultaneous determination of specific energy, sulphur and elemental abundance in coal.
• The potential to handle a far greater range of stream depths than the commercial PGNAA gauges, which are limited to stream depths of ~300mm in coal.

The NITA technology has been demonstrated in a plant environment analysing a total of 77 samples from 8 different seams, with a composition range of 9.2-17.8 %ash and 6.0-18.1 %H2O (as determined using rapid analysis). Historical data and experience suggests that the mineral matter composition for these samples will be highly variable.

Through the investigations detailed in this report, the NITA analyser has been demonstrated to be capable of achieving:

• An accuracy of 0.46 %ash. Specifically, a measured value for total r.m.s. error of 0.77 %ash, measured r.m.s error includes various uncertainties (estimated to be 0.61 %ash) associated with the chemical analysis and sample collection procedures used to generate the chemical laboratory data provided for calibration. In our earlier laboratory work [2], it was demonstrated that a measured r.m.s error of 0.75 %ash corresponded to an actual accuracy of 0.53 %ash after much smaller laboratory sampling errors had been excluded.
• Accurate measurement of ash value in material which is expected to have significant variability in mineralogy, and in the past has not been amenable to analysis using the DUET technique.
• Accurate measurement of ash value in the presence of significant levels of segregation and changes in mineral matter composition. 

CSIRO seeks to communicate the results of this project more fully to the Australian coal preparation industry with the intention of implementing this technology where it may benefit the industry. The ongoing encouragement and support already offered by ACARP towards this endeavour is gratefully acknowledged.

CSIRO is collaborating with a commercial partner to demonstrate and complete the implementation of the NITA technology in industry.  A commercial prototype analyser (NITA II) is under development and a plant installation is expected to occur in mid 2008.