Coal blending is an important tool for optimizing power plant operations and fuel costs. KEMA has extensive experience studying and optimizing coal blending in Dutch power plants. Some coal properties, like ash content and sulfur, can be estimated through arithmetic averages of the blended coals. However, other properties like grindability, ash behavior, and NOx emissions are determined by the individual coals and cannot be simply estimated. KEMA has developed models and relationships based on extensive testing to predict how different coal blends will perform.
Coal blending is an important tool for optimizing power plant operations and fuel costs. KEMA has extensive experience studying and optimizing coal blending in Dutch power plants. Some coal properties, like ash content and sulfur, can be estimated through arithmetic averages of the blended coals. However, other properties like grindability, ash behavior, and NOx emissions are determined by the individual coals and cannot be simply estimated. KEMA has developed models and relationships based on extensive testing to predict how different coal blends will perform.
Coal blending is an important tool for optimizing power plant operations and fuel costs. KEMA has extensive experience studying and optimizing coal blending in Dutch power plants. Some coal properties, like ash content and sulfur, can be estimated through arithmetic averages of the blended coals. However, other properties like grindability, ash behavior, and NOx emissions are determined by the individual coals and cannot be simply estimated. KEMA has developed models and relationships based on extensive testing to predict how different coal blends will perform.
Coal blending is an important tool for optimizing power plant operations and fuel costs. KEMA has extensive experience studying and optimizing coal blending in Dutch power plants. Some coal properties, like ash content and sulfur, can be estimated through arithmetic averages of the blended coals. However, other properties like grindability, ash behavior, and NOx emissions are determined by the individual coals and cannot be simply estimated. KEMA has developed models and relationships based on extensive testing to predict how different coal blends will perform.
Coal quality and power plant operation The quality of coals can vary over a wide range due to coalification history, mining and upgrading of a coal after mining. A range of parameters determines coal quality, since power plant operations cover a wide range of processes. Coal quality criteria therefore range from basics like the lower heating value via moisture and sulphur content to grindability and slagging propensity. The table below specifies a number of characteristic coal quality parameters and their relevance for the operation of a coal-fired power plant.
Parameter Unit Relevant for Lower heating value MJ/kg Power and load change rate Moisture % or g/MJ Mill capacity, coal handling Ash % or g/MJ Mill capacity, wear, ignition, fly-ash and bottom- ash processing, fly-ash quality Volatiles % or as fuel ratio Ignition, NO x -formation, unburned carbon Hardgrove grindability HGI units Mill capacity and unburned carbon Sulphur % or g/MJ FGD capacity, SO 2 emission Chlorine % Corrosion risk Ash fusion behaviour C Slagging and fouling propensity Ash composition % Quality of fly-ash and bottom-ash, slagging and fouling propensity, precipitator efficiency
Coal blending in the Netherlands In the Netherlands only internationally traded coals are utilised for power generation. This means that for more than thirty years different coals from all over the world are utilised in the Netherlands. The quality of the imported coals in terms of lower heating value (LHV), ash, volatile matter, moisture, sulphur and grindability (HGI) varied to a large extend. The table below specifies the ranges over which these parameters varied in the last 10 years.
These wide ranges of coal quality could only be fired due to blending facilities available for all the power stations in The Netherlands. In close co-operation with the power stations KEMA has performed since 1988 a large number of studies and experiments to optimise coal blending in relation to power plant operation. Most of these studies has been performed in full scale practice using all seven coal-fired boiler installations in The Netherlands.
It appears that coal blending is a powerful tool to optimise between fuel costs and power plant operation. Striking examples are the successful results obtained at the Maasvlakte power station at the Rotterdam harbour. This power station had to combat several problems at their plant operation, which was symbolised by the so-called Diabolic Triangle. In order to fulfil simultaneously the constraints for thermal efficiency, NO x -emission and unburned carbon blending proved to be very effective means to meet these constraints at acceptable fuel costs. From systematic analysis of power plant results for more than thirty different coals and even more coal blends a number of relations has been formulated. Using these relations the power plant was able to effectively compose coal blends within the existing constraints.
Extensive experience has been gained by KEMA in full-scale measurements with element distributions in a power plant. This applies to the main elements in the coal and the ash as well as the trace elements. It has resulted in a large database with the results of these measurements obtained from firing single coals as well as coal blends. On the basis of these data a model has been developed in order to predict element distributions over ashes (fly- ash, bottom ash), gaseous emissions, FGD by-products and water effluent emissions.
-3- 01-09-04
Difficulties in prediction of coal blend properties For some coal parameters simple optimisation by coal blending is possible based on the calculation of the arithmetic mean of the parameters from the individual coal in the blend. This applies for example for total ash amount from the mean ash content or SO 2 emission from the mean sulphur content. The relative amounts of fly-ash and bottom ash, however, depend on the composition of the ash and this will influence slagging and fouling. These parameters can not simply be calculated for a blend as arithmetic means. The behaviour of individual ash particles in a blend is a decisive factor for these phenomenons. This means that blending of coal with different ashes will not always result in a blend with a behaviour of the mean but a behaviour that is dominated by one or two of the coals. Based on full-scale experience KEMA has found that installation specific blending rules related to such operational effects can however be developed to enable optimised use of coal blends.
Another important property of the coal is the Free Swelling Index (FSI). This parameter gives information on the plasticity of coal during the initial heating stage of the coal. Several studies have indicated that a too high contribution (in the blend) of a coal with a high FSI value may lead to burner blockage and pf duct blockage leading an increased risk of fires in the pf ducts. Also for the FSI the property of individual particles is the determining factor in the behaviour of the blend. Therefore the calculation of the arithmetic mean of the FSI for a blend does not give a relevant prediction for the actual behaviour of the blend in the boiler.
Literature example of the differences in actual HGI and calculated HGI for coal blends.
-4- 01-09-04
Also with respect to other operational effects such as grindability or NO x -emission the behaviour of a blend can not always be predicted by applying the arithmetic means of the coal parameters of a blend. When in a blend coals are used having differences in grindability that go beyond a defined range the grindability of the blend (and the resulting throughput and coal fineness) will be determined by the coal with the lowest HGI-value. We have experienced that the defined range of HGI values where this effect manifests itself, depends on the type of coal mill as well as the absolute HGI values.
Example of test runs where blending in the boiler was investigated
The behaviour of coal blends in relation to NO x -emission and burnout has been extensively investigated by KEMA. In modern low-NO x combustion systems NO x -emission and burnout appears to depend on several coal parameters, such as volatile matter, nitrogen content, distribution of nitrogen over volatile matter and fixed carbon and ash content. Furthermore the relative importance of these factors is also determined by the configuration of the combustion system. This was investigated by KEMA in five power stations with different low- NO x configurations. Over thirty full-scale trials have been performed. These trials ranged from single coals and 50/50 blends, to blends with five different coal types, as well as blending in coals in the boiler (firing different coal types at different burner levels). Installation specific relations between volatile content, ash content, nitrogen content and nitrogen distribution over volatile matter and fixed carbon on the one hand and NO x -emission and burnout on the other hand have been developed.
-5- 01-09-04
Example of NO x -results for individual coals and blends from these coals
Coal blending and fly-ash application As in The Netherlands no land filling with fly-ash or bottom ash is allowed, a large amount of effort has been given over the years to determine the effects of coal quality on the quality of fly-ash and bottom-ash. The quality is related to industrial applications such as cement and concrete manufacturing, road building and also to environmental properties (leaching behaviour). This work has also been conducted in relation to the coal blending studies and the studies on element distributions. The results have shown that in most cases, coal blending leads to ash qualities that can be predicted on the basis of the individual coal.
KEMA publications related to coal blending C.M. Rozendaal, C.H. Gast, H.N. van Vliet, and K.R.G. Hein, 1992. Coal blending and its effects on NOx and unburned carbon in large scale power plants in The Netherlands. DVV- Kolloquium "Kohlestaub-Aufbereitung: ein Betrag zur NOx-minderung. December 1992, Essen, Germany.
C.M. Rozendaal, C.H. Gast, H.N. van Vliet, 1993. The diabolic triangle in modern low-NOx coal firing. Proceedings of the 11th International Conference on Power Stations. September 1993, Liege, Belgium.
-6- 01-09-04
C.M. Rozendaal, J.G. Witkamp, H.N. van Vliet, A.M.C. Vissers, 1997. Impact of coal quality and coal blending on NOx emissions for two pulverized coal fired units. EPRI-DOE-EPA combined utility air pollutant control symposium, "The Mega Symposium. August, 1997, Washington DC, USA.
C.M. Rozendaal, A.J.C. Korthout, A. Meerkerk, 1997. Coal blending and the performance of the low-NOx pulverized coal fired unit Amer 9 in the Netherlands. VGB-Fachtagung "Feuerungen 1997". Essen, Germany.
MEIJ, R., 1997. Prediction of environmental quality of by-products of coal-fired power plants; elemental composition and leaching. In: Proceedings of the international conference WASCON '97, held in Houthem St. Gerlach, The Netherlands, June 4-7, 1997. The proceedings are published in "Studies in Environmental Science 71, Waste Materials in Construction; Putting Theory into Practice". edited by J.J.J.M. Goumans, G.J. Senden and H.A. van der Sloot, published by Elsevier, Amsterdam, 1997, ISBN 0-444-82771-4.
C.M. Rozendaal, 1999. Impact of coal quality on NO x -emissions from power plants. Thesis Delft University of Technology, The Netherlands. November 1999.
MEIJ, R., 1999. Mass balance studies of trace elements at coal-fired power plants including co-combustion of waste and biomass. In: Proceedings of CEM 99, International Conference on Emissions Monitoring, held at the University of Warwick, Coventry, UK, September 6-8, 1999. The paper was presented at the TraceElements Workshop organised by IEA on 9 September 1999.
MEIJ, R., WINKEL, B.H. te and LINDEMAN, J.H.W., 2001. Co-firing and ash quality. Prediction of the elemental composition of the ash. In: Proceedings of Morella 2001 International Workshop on Novel Products form Combustion Residues: Opportunities and Limitations, organised by EC Thematic Network PROGRES and held at Morella, Spain June 608, 2001.
J.G. Witkamp, C.H. Gast, C.M. Rozendaal, C.K. Man and J.R. Gibbins, 2002. Char nitrogen and NO x -emissions. World Coal, January 2002, pp. 43 - 46.
C.H. Gast, M.C. Bouwmeester, A.J.C. Korthout, 2002. What is the real price for hard coal? VGB Power Tech 9, pp 44-46, 2002
MEIJ, R. and TE WINKEL, B.H., 2003. The Emissions of SO2, NOx, PM, Trace Elements and other HAPS of Coal-Fired Power Stations in the Netherlands. In the proceedings of the 12th International Conference on Coal Science, Cairns, Australia, 2nd - 6th November 2003.
-7- 01-09-04
References EON Benelux, Maasvlakte Power Station, two boilers of 520 MWe, tangentially fired combustion system with low-NO x combustion.
Essent, Amer Power Station, two boilers of 600 MWe, tangentially fired combustion system. One boiler with first generation low-NO x combustion. One boiler with modern low-NO x
burners and low-NO x air distribution.
Delta, Borssele Power Station, one boiler of 400 MWe, tangentially fired combustion system with low-NO x combustion.
Nuon Power, Hemweg Power Station, one boiler of 600 MWe, opposed wall firing with modern low-NO x burners and low-NO x air distribution.
Electrabel, Gelderland Power Station, one boiler of 600 MWe, opposed wall firing with modern low-NO x burners.