Stochastic Programming Model Integrating Pyrolysis Byproducts in the Design of Bioenergy Supply Chains
In search of alternative sources of energy to decrease our dependence on fossil fuels, biofuels have been presented as a promising sustainable solution. Biomass is an abundantly available resource for energy production , and as a result, biofuels derived from biomass have garnered much attention in recent years as a mainstream option to meet our increasing energy demands. Biofuels are also an environmentally friendly energy source; since it burns off cleaner than gasoline, it will reduce the emission of greenhouse gases .Biofuels are classified according to their feedstock. The first generation is produced from biomass that can be used as food for human consumption. The second-generation biofuels are derived from a wider range of feedstocks, which include municipal solid waste and lignocellulosic matter. Finally, the third-generation biofuels are derived from algal biomass . In the United States, lignocellulosic biomass is the most viable option given that it is not in direct competition with food intended for human consumption .More affordable gasoline prices derived from fossil fuels is one of the primary factors that has kept biofuels from becoming a viable energy source. As a result, biofuels require improvements in areas such as conversion technologies, genetic manipulation of feedstock, and optimization of the supply chain to become a competitive option. Certain operations in a biomass supply chain such as harvesting, handling, storing, and transporting are all potential areas of improvement that will have a significant impact in the biofuel yields. Biomass can be converted into biofuel through different thermal, biological, and physical processes. Pyrolysis has been one of the popular thermochemical processes in recent years. There are three main operating conditions that pyrolysis can be classified into: conventional, fast, and flash pyrolysis. However, this thesis will primarily focus on fast pyrolysis due to its advantages via storing, transporting, and versatility of the final product. Fast pyrolysis is the process of rapidly heating organic material at high temperatures in the absence of oxygen . In this thermochemical process, the biomass decomposes into three byproducts, which include bio-oil (60-75%), bio-char (15-25%), and syngas (10-20%) depending on the feedstock . Some of the basic characteristics of fast pyrolysis include high heat transfer and heating rate, and as a result the moisture content of the biomass is a property that affects the conversion yield. Since temperature is one of the main factors in the conversion process, high levels of moisture will require more energy to evaporate the water and produce high quality bio-oil . Therefore, the pyrolysis process initiates by drying the feedstock to the desired moisture levels. Additionally, the feedstock must undergo cutting and drying to ensure the desired particle size before being processed by the reactors to ensure a rapid reaction. Furthermore, another factor that can affect the conversion process is the ash content of the biomass. In the thermochemical conversion process, a high concentration of ash can impact the maintenance cost of the boiler by accumulating at a faster rate . Due to the different reactions that occur at different temperatures in the pyrolysis process, the yield of primary products obtained may vary. Low temperature and low heating rate process will lead to the maximum yield of char. Low temperature and a high heating rate with short gas residence will yield the most bio-oil. Biomass physical and chemical properties are important factors that must be considered to optimize a biomass supply chain. This thesis offers an extension to a proposed model by Aboytes et al. , in which the biomass quality variability, among other key operational factors are taken into consideration to produce pyrolysis byproducts, specifically, bio-ethanol. The extension includes two additional set of arcs in which the flow of the pyrolysis byproducts yields will be used to supply the bio-ethanol demand of the top 10 most populous cities in Texas and the bio-char demand of 17 power plants in a real case study across the state of Texas. The mathematical model proposed in this work aims to minimize investment, transportation and quality related costs given a selection of the conversion technologies, number of facilities (depots and biorefineries), their locations, and the production mix of byproducts to supply the demand of bio-char for 17 power plants and bio-ethanol for 10 cities in Texas.