Theoretical Analysis And Simulation Application Of Kinetic Model For Thermal Conversion Of Biomass

The comprehensive understanding of the thermal conversion kinetics of lignocellulosic biomass plays an important role in the numerical simulation,especially in computational fluid dynamics（CFD）simulation for biomass thermal conversion processes.The selection of the kinetic model and its parameters affects the numerical simulation results,the design of reactors and the optimization of conversion processes.Therefore,it is essential to systematically study the kinetic models for biomass thermal conversion.The distributed activation energy model（DAEM）is one of the most accurate and comprehensive models and the Gaussian distribution function is usually used to represent the activation energy distribution for biomass thermal conversion processes.And it is not sufficiently accurate for addressing the activation energy distribution in the initial and final stages of the biomass thermal decomposition reactions.Besides,the DAEM with a single activation energy distribution cannot describe those reactions well since some biomass thermal conversion processes normally involve two or more sub-processes.Since the Logistic distribution has slightly longer tails which is more suitable to represent the activation energies for biomass thermal conversion,the double Logistic DAEM has been developed in this thesis.The equations of the double logistic DAEM with different forms,numerical calculation method,physical meaning of the model parameters have been presented.The typical biomass thermal conversion kinetic processes have been systematically analyzed.In the first part of this thesis,the sensitivity analysis of the first and nth-order kinetic models is conducted in order to analyze the influence of model parameters on the model output,which is fundamental for the model parameter estimation and its practical application.In the third part,the double Logistic DAEM was coupled with the biomass particle pyrolysis model,and the releasing characteristics of pyrolysis volatiles and the change of temperature distribution of biomass particles are obtained.The above studies will provide theoretical support for the mechanism investigation and industrial application of biomass thermal conversion.1)The local sensitivities of the conversion and derivative conversion with respect to the frequency factor,the logarithm of the frequency factor,the activation energy and the reaction order for the first and nth order kinetic models were calculated by using the central difference method.The optimal values of parametric changes were obtained:ΔE=0.2 kJ mol^-1,ΔA=10¹² s^-1,ΔlnA=0.02 andΔn=0.04.The results have shown that the sensitivities of the first and nth order kinetic models with respect to the activation energy and the logarithm of the frequency factor are significant and that the frequency factor and the reaction order affects the nth order kinetic model slightly.Compared with the frequency factor,the natural logarithm of the frequency factor is a better choice in the parameter estimation of the first and nth order kinetic models.2)The numerical calculation of the double logistic DAEM can be performed by the following method:the inner dT integral can be expressed in the form of the exponential integral,and the outer d E integral can be numerically solved by using the Simpson’s rule.The double Logistic DAEM can be used to represent three typical thermal conversion processes,namely separated,overlapped and partially overlapped processes.It is found that,for the partially overlapped process,the form of the minor peak（overlapped peak or peak shoulder）depends on the values of the frequency factor and heating rate.Considering the simulated processes and related examples from literature,the double Logistic DAEM has been remarked as a more reliable tool with abundant flexibility to explain the thermal decomposition of lignocellulosic biomass and other complex solid fuels.3)The double logistic DAEM was successfully coupled into the biomass particle pyrolysis model for the numerical simulation of pyrolysis processes.The coupled DAEM pyrolysis model was numerical solved based on the finite volume method.It was found that the intra-particle temperature gradient is more obvious and the predicted heating up time is longer for larger biomass particle.At the beginning of biomass pyrolysis,the heating rate of biomass particle is high(up to 3540 K s^-1),then rapidly decreases.According to the obtained mass loss rate curves of biomass particle pyrolysis,the pyrolysis process includes two sub-processes:drying and devolatilization.The conversion rate curve of the devolatilization process is consistent with the partially overlapped process described by the double Logistic DAEM,which corresponds to the decomposition of pseudo-components（pseudo-hemicellulose and pseudo-cellulose）.Compared with TGA tests,the simulation results with the double Logistic DAEM coupled pyrolysis model are closer to the real biomass pyrolysis process.