Even though biomass gasification remains a promising technology regarding de-centralized sustainable energy supply, its main limitations, namely the issues of unsteady operation, tar formation in after-treatment systems, and consequential high maintenance requirements, have never been fully overcome. In order to tackle the latter two deficiencies and to increase the understanding of thermodynamic and thermokinetic producer gas phase phenomena within the after-treatment zones, a numerical system dynamic model has been created. Thereby, naphthalene has been chosen to represent the behavior of tars. The model has been validated against a wide variety of measured and simulated producer gas compositions. This work particularly focuses on the investigation and minimization of tar formation within after-treatment systems at low pressures and decreasing temperatures. Model-based analysis has led to a range of recommended measures, which could reduce the formation tendency and thus the condensation of tars in those zones. These recommendations are (i) to decrease gas residence time within pipes and producer gas purification devices, (ii) to increase temperatures in low-pressure zones, (iii) to increase hydrogen to carbon ratio, and (iv) to increase oxygen to carbon ratio in the producer gas. Furthermore, the numerical model has been included in the cloud-computing platform KaleidoSim. Thus, a wider range of process parameter combinations could be investigated in reasonable time. Consequentially, a simulation-based sensitivity analysis of producer gas composition with respect to process parameter changes was conducted and the validity basis of the above recommendations was enlarged.

尽管生物质气化在分散式可持续能源供应方面仍是一项很有前途的技术，但其主要局限性（即不稳定的运行，后处理系统中焦油的形成以及随之而来的高维护要求）从未得到完全克服。为了解决后两个缺陷，并增加对后处理区内热力学和热动力学产生器气相现象的理解，已创建了一个数值系统动力学模型。因此，已选择萘来代表焦油的行为。该模型已针对各种测量和模拟的生产气成分进行了验证。这项工作特别着重于在低压和低温下对后处理系统中焦油形成的研究和最小化。基于模型的分析导致了一系列建议措施，这些措施可以减少形成趋势，从而减少这些地区的焦油凝结。这些建议是（i）减少管道和生产气体净化装置中的气体停留时间；（ii）升高低压区的温度；（iii）增加氢碳比，以及（iv）增加氧碳煤气中的比例。此外，数值模型已包含在云计算平台KaleidoSim中。因此，可以在合理的时间内研究更广泛的过程参数组合。因此，