ACARP Project Number: C3091        Published: October 96

This project examines the ability of CQIM (version 1.2) to predict the performance of Australian coals in some of the many processes that occur in a modern coal-fired power station. The continued growth in the world thermal coal market is resulting in greater diversity in thermal coal suppliers. The operators of coal-fired power stations are looking at computer software to assist in the evaluation of the impact of coal properties on the operating and maintenance costs.

One such computer software is EPRI's Coal Quality Impact Model (CQIM) developed by Black & Veatch. CQIM is a computer model designed to facilitate the prediction of performance and cost impacts of firing alternative coals at an existing or proposed power generating station. This project examines the ability of CQIM (version 1.2) to predict the performance of Australian coals in some of the many processes that occur in a modern coal-fired power station. The processes examined in this project were limited to ones where full or pilot scale data was available for comparison.

The processes that were examined in this project were:

The full scale data used in this evaluation was supplied by Pacific Power, and pertained to equipment at Eraring Power Station.

Eraring Power Station operates four 660 MW IHI designed generating units. These units are single furnace, twin-drum type using natural circulation with a divided back pass and balanced draught. Each unit is fed by seven tube mills.

The pilot scale data was taken from the testing of 38 coals in ACIRL's 150kW pilot scale boiler simulation furnace. Eleven of the coals were blends.

The adjustment, based on one coal's performance, of the mill performance curves used by CQIM is a satisfactory approach to predict mill capacity for the coals being evaluated. CQIM does not attempt to predict the size distribution of the coal or mineral matter.

To determine the combustion performance of a coal, CQIM allows the user to select either a simple model or detailed model.

The simple unburned carbon model is based on empirical correlations. The detailed unburned carbon model uses estimates of the char kinetic parameters with a detailed determination of the time temperature history of the particle to estimate the unburned carbon loss.

CQIM's simple unburned carbon model gives a good estimate of the carbon burnout for a wide range of coals as shown by the comparison with pilot scale data. The detailed burnout model suffers from instability problems which could reduce its accuracy.

The use of indices, determined from the bulk ash properties, to predict slagging and fouling characteristics of a coal does not fully account for the impact of individual minerals or the influence of fluid dynamics. CQIM failed to predict the reduced heat absorption by the waterwall that was recorded during one of the tests at Eraring. This was due to the coal's unusually low ash absorptivity.

The power of CQIM is its ability to relate coal quality to fuel costs, maintenance costs and the availability of a power station.

This ability of CQIM to model the large number of interacting processes that occur within a power station depends on the use of correlations relating these processes to coal properties.

CQIM reduces the possible errors in using empirical correlations by using the known performance of one coal to calibrate its predictions for other coals fired in the same power station.

This means the user has to be careful about the choice of the calibration coal, ensuring that this calibration coal is similar to the other coals that are to be evaluated, the completeness of its performance data.

With the improvements that are planned for future versions of CQIM, it is likely that its use by power station operators to evaluate potential coal supplies will increase. Australian producers should continue to evaluate future versions of CQIM.