Multivariate statistical monitoring of process operating performance

Process computers routinely collect hundreds to thousands of pieces of data from a multitude of plant sensors every few seconds. This has caused a “data overload” and due to the lack of appropriate analyses very little is currently being done to utilize this wealth of information. Operating personnel typically use only a few variables to monitor the plant's performance. However, multivariate statistical methods such as PLS (Partial Least Squares or Projection to Latent Structures) and PCA (Principal Component Analysis) are capable of compressing the information down into low dimensional spaces which retain most of the information. Using this method of statistical data compression a multivariate monitoring procedure analogous to the univariate Shewart Chart has been developed to efficiently monitor the performance of large processes, and to rapidly detect and identify important process changes. This procedure is demonstrated using simulations of two processes, a fluidized bed reactor and an extractive distillation column.     

Operations optimization and control design for a petroleum distillation process

Process automation involves both steady-state optimization and feedback control. In this paper, the economic optimum steady-state operating conditions for a petroleum crude distillation process are determined using a tray-to-tray model. This model is used to evaluate the steady-state operation achieved with several multivariable process control designs. Only one controller design was found to result in essentially optimum steady-state operation; others deviated significantly. The method applied in this paper relies on a knowledge of the key disturbances occurring in the process; good economic performance may not be obtained should other disturbance types occur. The results demonstrate the importance of implementing the operations optimization through a control strategy which will respond properly to disturbances.     

Hydrodynamics and recovered papers deinking in an ozone flotation column

This work was aimed at investigating the potential use of ozone flotation for recovered papers deinking. The mixing characteristics and the bubble size of a Venturi aerated laboratory flotation column were studied in the presence of simplified model systems and of an industrial pulp slurry. Experimental results showed that surfactants dissolved in the pulp slurry stabilized air bubbles while cellulose fibers promoted coalescence. Moreover, bubble buoyancy was sufficient to fluidize the fiber suspension generating perfect mixing. The gas–liquid transfer coefficient of ozone (k_La) estimated from gas hold-up, air bubble size and k_La measured in water was high enough (0.84 1/s) for the complete transfer of ozone in the pulp slurry and the generation of ozone-free gas effluent. With O₃ dosage of 0.8 mg O₃/mg COD, ozone flotation allowed increasing chemical oxygen demand removal from 41 to 63% with no effect on the ink flotation efficiency, which remained close to that obtained with air flotation, i.e. ～92%.     

Study of oil soluble polymers as drag reducers

This investigation was undertaken to find the most effective material which would reduce the friction coefficient in turbulent flow when added in small quantities to oil pipelines. For this purpose, a series of oil-soluble polymers, namely homopolymers and copolymers of alkyl methacrylates, alkyl acrylates, and alkyl styrenes were synthesized. Emulsion polymerization techniques were used. Commercially available alkyl methacrylate and alkyl acrylate monomers were used in the synthesis. Monomeric alkyl styrenes were synthesized and structures established prior to polymerization. Intrinsic viscosities were measured and viscosity average molecular weights were calculated for several of the homopolymers synthesized in this study. Reduction of factional drag and resistance to shear degradation were measured by pumping a solution of the polymer in a hydrocarbon solvent through a pipe and recording the pressure drop across the pipe. Drag-reducing properties of several of the polymers were correlated in terms of their viscosity average molecular weights. Drag reduction of poly (isodecyl methacrylate) was studied in various hydrocarbon solvents. Drag-reducing behavior of polymers prepared in this study exhibited a strong dependence on molecular weight; increasing the molecular weight increased the drag reduction for a given polymer concentration and pipe size. Several of these polymers were found to be superior to commercially available polyisobutylene as drag reducers, especially in terms of shear stability.     

Application of the transportation model to a large-scale “Districting” problem

The “Application of the Transportation Model to a Large-Scale ‘Districting’ Problem” describes a method of partitioning an area containing many geographical locations, each with an associated activity or workload, into districts called “tours.” The objective is to assign each location to a tour in such a way that the total workload assigned to each tour falls within specified limits and that the total cost to service the locations from a specified set of tour “centers” is minimized. The present approach is compared and contrasted with other approaches to the problem in several contexts. The actual situation motivating the present work and the derivation of the transportation linear programming model is described. Special attention is given to features which aid implementation and which more realistically model the actual situation by handling the user's implicit constraints, i.e. those not explicitly contained in the transportation formulation. Finally, the results of an actual large-scale problem are presented.     

Physiological responses of horses to a treadmill simulated speed and endurance test in high heat and humidity before and after humid heat acclimation

Food movements during complete feeding sequences on soft and hard foods (8 g of chicken spread, banana, and hard cookie) were investigated in 10 normal subjects; 6 of these subjects also ate 8 g peanuts. Foods were coated with barium sulfate. Lateral projection videofluorographic tapes were analyzed, and jaw and hyoid movements were established after digitization of records for 6 subjects. Sequences were divided into phases, each involving different food management behaviors. After ingestion, the bite was moved to the postcanines by a pull-back tongue movement (Stage I transport) and processed for different times depending on initial consistency. Stage II transport of chewed food through the fauces to the oropharyngeal surface of the tongue occurred intermittently during jaw motion cycles. This movement, squeeze-back, depended on tongue-palate contact. The bolus accumulated on the oropharyngeal surface of the tongue distal to the fauces, below the soft palate, but was cycled upward and forward on the tongue surface, returning through the fauces into the oral cavity. The accumulating bolus spread into the valleculae. The total oropharyngeal accumulation time differed with initial food consistency but could be as long as 8-10 sec for the hard foods. There was no predictable tongue-palate contact at any time in the sequence. A new model for bolus formation and deglutition is proposed.     

Integrating process and controller models for the design of self-optimizing control

The steady-state behavior of an existing plant depends on the independent input variables, process equipment and process controllers. This paper presents a method for formulating models that represent the effects of controllers when they are included within a steady-state process flowsheet. The method replaces the controller equations with the equivalent stationarity conditions representing the relationship between the controlled variables and the implemented manipulated variables at steady state. The method is demonstrated for the centralized multivariable Dynamic Matrix Control algorithm applied to two processes, binary distillation and gasoline blending. The integrated process and control system simulation is used to design controllers that improve the profitability of processes without extensive real-time calculations; this is sometimes termed self-optimizing control. For both processes, controllers were designed that yielded higher profit than standard control methods and that approached the highest possible profit achieved by frequent real-time optimization.     

FREE CONVECTION EXPERIMENTS OF ATMOSPHERIC AIR IN VERTICAL OPEN ANNULI

Experiments were conducted for free convection of atomospheric air in open annuli (annulus ratios of 1.155, 1.33, 1.63 and 12.0) with an isoflux inner wall and an adiabatic outer wall for surface temperatures up to 856 K. The experiments established that Rayleigh number increased as the film temperature of air increased, reaching a maximum at approximately 375 K, regardless of the annulus ratio. Beyond this temperature, the Rayleigh number decreased as the film temperature increased. For film temperatures of air less than about 375 K, the Nusselt number increased proportionally to the Rayleigh number raised to the one-fourth power. For higher film temperatures, however, the Nusselt number was inversely proportional to the Rayleigh number.