Probabilistic modeling of innovative clean coal technologies: Implications for technology evaluation and research planning. (Volumes I and II)

A probabilistic approach is developed which allows the explicit and quantitative representation of the uncertainties inherent in innovative technologies. Probabilistic analyses provide insights into the uncertainties in process performance and cost not possible with conventional deterministic or sensitivity analysis. Applications of the approach are illustrated via analyses of the performance and cost of the fluidized bed copper oxide process, an advanced technology for the control of SO{dollar}sb2{dollar} and NO{dollar}sb{lcub}rm x{rcub}{dollar} emissions from coal-fired power plants, and three integrated gasification combined cycle (IGCC) systems. Engineering performance and cost models of conceptual commercial-scale systems for each technology provides the basis for the analysis.; For each technology evaluated, uncertainties in performance and cost parameters of the engineering models were explicitly characterized using probability distributions. Estimates of uncertainty were based on literature review, data analysis, and elicitation of the expert judgment of process engineers involved in technology development.; The engineering models were exercised in probabilistic modeling environments to characterize the uncertainties in key measures of process performance and cost. The resulting uncertainties in performance and cost provide a quantitative measure of the risk of either poor performance or high cost associated with innovative process technologies. The key input uncertainties that drive uncertainty in performance and cost can be identified and prioritized. Thus, probabilistic analysis has direct implications for cost estimating, risk assessment, and research planning.; Competing technologies are compared probabilistically to obtain quantification of the probability that an advanced technology will have higher performance and lower cost than conventional technology. Additional research is assumed to reduce the uncertainty in key input parameters. Therefore, the expected pay-off from additional research is evaluated using alternative assumptions regarding uncertainties. Engineering model results are used as inputs to decision models, to gain further insights regarding technology selection and research strategies.; For most of the analyses considered here, the probabilistic approach is found to yield higher estimates of cost and lower estimates of plant performance than obtained from traditional deterministic approaches to technology evaluation. A key benefit from probabilistic analysis is the explicit characterization of skewed uncertainties in innovative technologies, which are a key source of cost growth often overlooked.