Impact of Feedstock Attributes on the Performance of Processing in Cellulosic Ethanol Applications

Feedstock supply system predominately utilized in the production of cellulosic biofuels, is typically designed around a single biomass type available within a small supply radius. While the limited supply area reduces transportation cost, there is little opportunity to buffer against the wide variability of physical and chemical properties encountered during processing and conversion. Feedstock variability is a major contributor to delays and disruptions in the preprocessing and conversion systems. Classical approaches to eliminating delays and disruptions utilize a sequential approach, which leads to solutio0ns that do not consider interactions across the system. To examine the impact of feedstock variability on processing and conversion performance, Discrete Event Simulation (DES) models were developed for preprocessing operations of a herbaceous feedstock supply system, including feedstock property-based equipment performance, equipment failure modes (biomass properties causing equipment failures), and the expected time to repair or correct these failures. Initial models were developed to represent a commercial scale biorefinery (focusing on preprocessing and infeed to conversion reactor), with failure and repair information based on industrial operational experience. Initial runs of the models resulted in a capacity utilization of 29.1% of design. Following this, conversion performance data were also included to integrate aspects of both chemical and physical performance into the model. The performance of the physical system was based on pilot-scale data, which produced similar results as the initial simulations and had a capacity utilization factor of 28.9%. Measured conversion yield, based on the feedstock variation, resulted in an average yield of 88.1% of design. This methodology provides a way to dynamically examine biorefinery operational performance in relation to feedstock variability. A new metric, termed “Operational Reliability,” is defined to capture operational and yield impacts and provide an overall picture of how well the system operates; for the system modeled, modeled operational reliability was 25.5%.