We discuss data symbolization as a tool for identifying temporal patterns in complex measurement signals. We describe the basic concepts involved and illustrate their application for the analysis of gas-bubble injection data. Specific issues addressed include selection of symbolization parameters, construction of symbol-sequence histograms, and statistical characterization and comparison of these histograms. We demonstrate that symbolsequence statistics can reveal unique information about deterministic patterns. Such information may be useful for developing flow diagnostics and comparing computational models with experiments.

We propose that data-symbolization methods derived from nonlinear dynamics and chaos theory can be useful for characterizing and monitoring patterns in fluidized-bed measurement signals. Data symbolization involves the discretization of a measurement signal into a limited set of values. In this discretized form, the measurements can be processed very efficiently to detect dynamic patterns that signify various types of physical phenomena, including bubbling, slugging, and transitions between fluidization states. Besides computational efficiency, symbolic methods are also robust when noise is present. Using various types of measurements from experimental beds, we illustrate specific examples of how symbolization can be applied to fluidization diagnostics. We also suggest directions for future research.

We describe a PC-based computer model which simulates the 3-dimensional dynamic behavior of multiple interacting bubbles in bubbling and slugging fluidized beds. We show that this model can predict important performance-related features such as bubble diameter, slug length, bubble coalescence rate, and global circulation patterns. A useful characteristic of this model is that it can be used to simulate bed behavior in near-real time.

Rhythmic {open_quotes}whooshing{close_quotes} sounds associated with rising bubbles are a characteristic feature of many fluidized beds. Although clearly distinguishable to the ear, these sounds are rather complicated in detail and seem to contain a large background of apparently irrelevant stochastic noise. While it is clear that these sounds contain some information about bed dynamics, it is not obvious how this information can be interpreted in a meaningful way. In this presentation we describe a technique for processing bed sounds that appears to work well for beds with large particles operating in a slugging or near-slugging mode. We find that our processing algorithm allows us to determine important bubble/slug features from sound measurements alone, including slug location at any point in time, the average bubble frequency and frequency variation, and corresponding dynamic pressure drops at different bed locations. We also have been able to correlate a portion of the acoustic signal with particle impacts on surfaces and particle motions near the grid. We conclude from our observations that relatively simple sound measurements can provide much diagnostic information and could be potentially used for bed control. 5 refs., 4 figs.

The Morgantown Energy Technology Center has development programs in a number of fossil energy technologies which use fluidized beds as reactors to carry out combustion, gasification, and desulfurization. The diagnosis of operating problems and control of bed behavior is critical to the efficient performance of these systems. The overall goal of the present work is to develop novel techniques to improve diagnosis and control of these systems. Chaotic time series analysis has recently been used with pressure drop, void fraction, and heat transfer data to characterize fluidized bed dynamics. Unique chaos parameters derived from this analysis can distinguish among various fluidization conditions within gas fluidized systems. In this paper, a basis for understanding the physical meaning of several chaotic parameters is developed and illustrated.

This paper reports on the application of mutual information theory to the analysis of transient differential pressure and temperature signals from a fluidized bed. The signals were recorded around a heat transfer tube which was placed horizontally into a bubbling fluidized bed. The heat transfer tube was instrumented with fast response surface thermocouples and differential pressure sensors. Mutual information theory was used to identify the periodicity and the predictability of the local instantaneous differential pressure and temperature signals. It was also used to interpret the bubble-particle packet dynamics around the instrumented heat transfer tube. As theoretical and limiting cases, purely periodic and random signals were observed. The conventional signal processing tools such as autocorrelation, cross-correlation and fast Fourier transformation (FFT) were used as preliminary tools to analyze data. The qualitative similarities between the mutual information function and the autocorrelation function are shown. The first minimum of the mutual information function is used to reconstruct the phase portrait from the one-dimensional time series. It is suggested that if the first minimum of the mutual information function exists, then using the time derivative of the measured one-dimensional signal with the least number of bins provides a better time delay τ than using the measured signal directly.

The objective of the Morgantown Energy Technology Center (METC) high pressure combustion facility is to provide a mid-scale facility for combustion and cleanup research to support DOE`s advanced gas turbine, pressurized, fluidized-bed combustion, and hot gas cleanup programs. The facility is intended to fill a gap between lab scale facilities typical of universities and large scale combustion/turbine test facilities typical of turbine manufacturers. The facility is now available to industry and university partners through cooperative programs with METC. Currently two combustion rigs are operating and one additional project is under construction for the facility. Space is available in the test cells for at least one additional test rig. A pressurized pulsed combustor began operating in July of 1993. The combustor will carry out pulsed combustion of natural gas at pressures up to 10 atmospheres. A high pressure steady flow rig is currently completely fabricated. The objective of this rig is to test novel, steady-flow, pressurized combustors that produce very low NO{sub x} and other emissions. An evaporation rig currently is in startup. This rig will test the concept of water injection in an externally fired cycle. The specific technical issue that the unit will address is evaporation rates of water droplets in high pressure flows.

Recent applications of chaotic time series analysis to pressure-drop and voidage measurements from gas-fluidized beds have demonstrated that such data exhibit many of the signature characteristics of deterministic chaos. Research is now being sponsored by the US Department of Energy to determine if knowledge and control of chaotic structure in fluidized beds can be used to improve the performance of fossil energy conversion processes using fluidized beds. In this paper we discuss the use of chaotic time series analysis for evaluating fluidization quality. In particular, we show that certain chaotic features in pressure-drop measurements can be used to determine the type of fluidization (e.g., slugging, smoothly bubbling, turbulent) both locally and globally in the bed. We also suggest examples of how this information can be used for implementing closed-loop control of fluidized-bed hydrodynamics.

Recent applications of chaotic time series analysis to gas fluidized beds have demonstrated that substantial information about fluidization conditions within the. bed can be extracted from voidage and pressure drop data. In this paper, a technique is presented to characterize fluidized bed behavior based on the crossings of the phase space trajectory through the principal component planes. Starting with either pressure drop or void fraction versus time data, time series embedding and principal component analysis is used to construct a phase space trajectory for the data. This trajectory characterizes the dynamical state of the bed. The technique presented decomposes the trajectory by sorting the orbits into types characteristic of different modes of bed behavior, such as emulsion phase fluctuations, bubbling, slugging, bubble coalescence, and de-fluidization. The basis for the method and the analysis of data from experiments in several fluidized beds will be presented. The overall goal of these studies is to improve the diagnostic and control of fossil energy fluidized bed processes.

The Morgantown Energy Technology Center has developed a unique method for imaging voidage distributions within fluidized beds. This system allow high-speed three-dimensional imaging of the voidage distribution in a bed to be recorded. From these imaging data a variety of visualizations can be constructed and quantitative information extracted. Five materials with differing particles shapes and sizes were fluidized and imaged in a 15.24 cm diameter bed over a range of superficial velocities. The imaging revealed a variety of bubble sizes and shapes depending on the material and velocity. Bubble properties including a frontal diameter, length, velocity and spacing were determined. A correlation of bubble rise velocity developed from an earlier study was modified to include the bubble length as the primary dimension defining the bubble. The imaging also revealed detailed voidage distributions around bubbles and slugs. These images show round-nose bubbles and slugs for the fine materials examined. With the coarser materials, a very blunt slug was common. The technique has the potential to substantially improve design and scale-up of fluidized beds and other gas-solid systems by providing a detailed understanding of voidage distributions in these systems.

Atmospheric fluidized-bed combustion (AFBC), pressurized fluidized-bed combustion (PFBC), and integrated gasification combined-cycle (IGCC) systems, near-term, coal-based technology options for new, base-load capacity additions are being demonstrated in projects currently underway. Longer-term technology options can be envisioned that potentialy will have lower capital, operating, and maintenance costs particularly for small increments of new capacity, higher efficiencies, the ability to economically meet increasingly stringent environmental standards, shorter construction times, higher reliability, improved load-response characteristics, tolerance to a wide range of coal feed-stocks, and infrastructure acceptability. Candidate longer-term technologies include gas turbine-based systems using air-blown, entrained flow gasifiers coupled with novel cleanup processes; PFBC systems utilizing a topping combustor; coal gasification/fuel cell systems; and coal-fueled gas turbines. This paper discusses the advantages and market niches of these longer-term technology options.

Integrated coal gasification combined-cycle (IGCC) power systems offer the potential of superior efficiency and environmental performance over power plants using pulverized coal-fired boilers with scrubbers to generate electricity in the United States. The Cool Water plant is demonstrating the feasibility of an IGCC system using an entrained-bed gasifier and ''cold'' gas cleanup technology. Technology is now being developed to simplify the IGCC system, increase its efficiency and reduce its capital costs. Hot gas sulfur and particulate cleanup is the most promising technology option for the gas supply block. Improved performance is also available from the power island by use of high-efficiency aircraft derivative turbines. Progress in these technologies and the exceptional match of these IGCC systems to the projected needs of the utility industry is presented.