Stochastic Simulation of Pulverized Coal (PC) Processes

Abstract

An increasing population and electricity demand in the U.S. require capacity expansion of power systems. The National Energy Technology Laboratory (NETL), U.S. Department of Energy (DOE), has invested considerable efforts on research and development to improve the design and simulation of these power plants. Incorporation of novel process synthesis techniques and realistic simulation methodologies yield optimal flowsheet configurations and accurate estimation of their performance parameters. To provide a better estimation of such performance indicators, simulation models should predict the process behavior based on not only deterministic values of well-known input parameters but also uncertain variables associated with simulation assumptions. In this work, the stochastic simulation of a load-following pulverized coal (PC) power plant takes into account the variation of three input variables, namely, atmospheric air temperature, atmospheric air humidity, and generation load. These uncertain variables are characterized with probability density functions (pdfs) obtained from available atmospheric and electrical energy generation data. The stochastic simulation is carried out by obtaining a sample of values from the pdfs that generates a set of scenarios under which the model is run. An efficient sampling technique [Hammersley sequence sampling (HSS)] guarantees a set of scenarios uniformly distributed throughout the uncertain variable range. Then, each model run generates results on performance parameters as cycle efficiency, carbon emissions, sulfur emissions, and water consumption that are statistically analyzed after all runs are completed. Among these parameters, water consumption is of importance because an increasing demand has been observed mostly in arid regions of the country and, therefore, constrains the operability of the processes. This water consumption is significantly affected by atmospheric uncertainties. The original deterministic process model simulation was designed in Aspen Plus, and a CAPE-OPEN compliant stochastic simulation capability is employed to run the uncertainty analysis. Initially, the influences of atmospheric conditions and load change on the performance parameters are analyzed separately to understand their individual influences on the process, and then their simultaneous variation is analyzed to generate more realistic estimations of the process performance.

Monetizing shale gas to polymers under mixed uncertainty: Stochastic modeling and likelihood analysis

“…As mentioned earlier, uncertainty is an inherent characteristic in a shale gas monetizing process, which leads to significant challenges in handling variability and/or stochasticity in parameter values. Typically, uncertainty can be characterized by probability density functions, which have been calculated on the basis of the real‐world data and appropriate assumptions . However, the overwhelming majority of the shale gas‐based processes are designed to handle deterministic feedstock .…”

Minimization of Water Consumption under Uncertainty for a Pulverized Coal Power Plant

“…NETL’s systems analysis reports on PC plant water consumption make several assumptions that could become important sources of uncertainty since they are directly related to the water consumption and can be fairly variable. − Air temperature and humidity continuously fluctuate throughout the US with the changing seasons, and their influence on water consumption was mentioned above. The effect of the variability of these atmospheric conditions on some performance parameters including water consumption for PC plants was studied through a stochastic simulation …”

“…Data for the air temperature and humidity can be obtained from DOE’s energy efficiency and renewable energy program (EERE) Web site () as described in previous works . In ref , to perform stochastic simulations the authors organized data in four seasons, starting at September 2005 and finishing at August 2007 and histograms of the frequency distributions were fitted to the appropriate probability density functions as log-normal and uniform* (segmented uniform distribution) distributions to represent the air conditions variability of an average midwestern urban center during each of the four seasons of a year. Results of the characterization of the uncertainty are included in the Supporting Information.…”

“…The effect of the variability of these atmospheric conditions on some performance parameters including water consumption for PC plants was studied through a stochastic simulation. 15 The main source of information for cooling tower design conditions, including air humidity considerations, is the text by Hensley on cooling tower fundamentals. 12 The author suggests using average measurements of the wet-bulb temperature to determine the temperature range and approach temperature that guarantee the appropriate cold water conditions at peak demand.…”