Title:
Development of Novel Modelling Approaches for Agglomeration Prediction
Author(s):
Gatternig, B., Karl, J.
Document(s):
Paper
Abstract:
The utilization of novel biomass fuels such as agro-industrial residues or energy crops shows a large potential as a sustainable source of renewable energy. Particularly its combustion in fluidized beds seems ideal thanks to the inherent advantages of the process, such as a wide applicable fuel spectrum, tolerance to high ash fuels and low combustion temperatures. On the other hand the specific composition of biomass ashes, especially its high content of potassium, can lead to operational problems. One of those is the agglomeration of the bed material due to the formation of sticky ash layers around the particles. These layers cause the individual particles to form large agglomerates, which can no longer be fluidized. The consequence are unscheduled plant shut downs and long revision periods. The likelihood of a given fuel to cause such phenomena is determined by its composition, in particular its content and distribution of mineral matter. An abundance of Alkali metals, as well as Silicon – in the fuel ash or the bed material – causes low melting temperatures and thus the formation of agglomerates. These three variables – composition, interaction with the bed material and applicable temperature range – along with the operational parameters particle size and fluidization velocity are used to develop a novel method for discerning the agglomeration temperature of a given fuel. This article describes three different modeling approaches based on analytical, (semi-) empirical and statistical principles. An in-depth description of agglomeration mechanisms is given to explain the choice of modeling parameters. The validation of the models was performed using results from a lab-scale fluidized bed reactor and agglomeration temperatures published in literature. A comparison of the models revealed good results for the statistical approach with standard deviations of approximately 50K and an R2>0.7. These values can be further improved, as more experimental results of different biomass fuels and bed materials are incorporated into the model. The empirical and analytical approaches did not reach the high quality of the statistical results with deviations of ±10-15%. The analytical model is still being optimized and as the knowledge of the specific mechanisms for agglomeration increases, modifications can easily be implemented in the calculations.
Keywords:
fluidized bed combustion, ash agglomeration, high ash biomass fuels
Topic:
R&D on Biomass Conversion Technologies for Heating, Electricity and Chemicals
Subtopic:
Solid biofuel combustion in large utilities
Event:
21st European Biomass Conference and Exhibition
Session:
2DO.3.3
Pages:
652 - 666
ISBN:
978-88-89407-53-0
Paper DOI:
10.5071/21stEUBCE2013-2DO.3.3
Price:
FREE
