There are two main threads of this thesis, namely, an unknown (unmeasurable) input reconstruction and fault detection and diagnosis. The developed methods are in the form of parity equations, i.e. finite impulse response filters of the available input and output measurements. In the first thread the design of parity equations for the purpose of an unknown input reconstruction of linear, time-invariant, discrete-time, stochastic systems is taken into consideration. An underlying assumption is that both measurable system inputs as well as the outputs can be subjected to noise, which leads to an errors-in-variables framework. The main contribution of the scheme is accommodation of the Lagrange multiplier method in order to minimise the influence of the noise on the unknown input estimate. Two potential applications of the novel input reconstruction method are proposed, which are a control enhancement of a hot strip steel rolling mill and an estimation of a pollutant level in a river. Furthermore, initial research is conducted in the field of the unknown input recon- struction for a class of nonlinear systems, namely, Hammerstein-Wiener systems, where a linear dynamic block is preceded and followed by a static nonlinear function. Many man-made as well as naturally occurring systems can be accurately described using Hammerstein-Wiener models. However, it is considered that not much attention has been paid to Hammerstein-Wiener systems in the errors-in-variables framework and in this thesis it is aimed to narrow this gap. The second thread considers a problem of robust (disturbance decoupled) fault de- tection as well as fault isolation and identification. Unmeasurable external stimuli, parameter variations or discrepancies between the system and the model act as distur- bances, which can obstruct the fault detection process and lead to false alarms. Thus, a fault detection filter needs to be decoupled from the disturbances. In this thesis the right eigenstructure assignment method used for the robust fault detection filter design is extended to systems with unstable invariant zeros. Another contribution re- gards the design of robust parity equations of any arbitrary order using both left and right eigenstructure assignment. Furthermore, a parity equation-based fault isolation and identification filter is designed which provides an estimate of the fault. A simple method for the calculation of thresholds whose violation indicates a fault occurrence is also proposed for the errors-in-variables framework.

published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

Predictive process control (PPC) is the use of predictive, physical models as the basis for process control [1]. In contrast, conventional control algorithms utilize statistical models that are derived from repetitive process trials. PPC employs in-process monitoring and control of manufacturing processes. PPC algorithms are very promising approaches for welding of small lots or customized products with rapid changes in materials, geometry, or processing conditions. They may also be valuable for welding high value products for which repeated trials and waste are not acceptable. In this research, small-lot braze-welding of UNS C22000 commercial bronze with gas metal arc welding (GMAW) technology is selected as a representative application of PPC. Thermal models of the welding process are constructed to predict the effects of changes in process parameters on the response of temperature measurements. Because accurate thermal models are too computationally expensive for direct use in a control algorithm, metamodels are constructed to drastically reduce computational expense while retaining a high degree of accuracy. Then, the feasibility of PPC of welding applications is analyzed with regard to uncertainties and time delays in an existing welding station and thermal metamodels of the welding process. Lastly, a qualitative residual stress model is developed to nondestructively assess weld quality in end-user parts. / text

This study uses Monte Carlo simulation to examine the performance of a variable frequency sampling cumulative sum control chart scheme for controlling the mean of a normal process. The study compares the performance of the method with that of a standard fixed interval sampling cumulative sum control chart scheme. The results indicate that the variable frequency sampling cumulative sum control chart scheme is superior to the standard cumulative sum control chart scheme in detecting a small to moderate shift in the process mean.

No description available.

Approved for public release; distribution is unlimited / This study models DoD suicides as a Poisson process to detect departures from usual variation using a self-starting control chart scheme. Methods are implemented in a Microsoft Excel spreadsheet with Visual Basic macros for ease of use. Persistent shifts in the process mean are detected in the following months for each service component. Army: August 1985 (increase), September 1987 (decrease), April 1991 (increase), November 1997 (decrease), and September 2001 (decrease). Navy: December 1990 (decrease), January 1993 (increase), May 1994 (decrease), July 1995 (increase), and March 1996 (decrease). Marine Corps: January 1993 (increase) and March 1998 (decrease). Air Force: January 1988 (increase), April 1990 (decrease), November 1994 (increase), November 1998 (decrease), and April 1999 (decrease). / Commander, United States Navy

Process Management (PM) is deemed as one of the most important managerial innovations of the last 20 years. However, the current concept of PM, which mainly comprises the perspective of process control and incremental process improvement, may not be able to adequately address the increasingly rapid-changing environment. Moreover, rigorous effort to examine PM's concept, to establish measures and to understand its performance effect, is surprisingly inadequate. / To bridge this gap, this study strives to address the question: what should PM entail and how it affects operations performances. PM is reconceptualized and operationalized by integrating radical process improvement as one of its key components. This instills strong theoretical underpinning that radical process change has become a normative activity rather than an abnormal remedy to organizations. Special effort is devoted in the scale development of PM. A theoretically sound and psychometrically valid scale has been established in this thesis. The results show that it is reliable and valid for use in the following studies. This thesis also advances the understanding of PM's effect on operations performances and its fit with process types. Finding confirms that organizations implementing integrated PM in general perform better than those executing individual dimensions. It further reveals that PM must fit process type otherwise operations performances will suffer. The extent of integrated PM applied seems to hinge on the complexity of the operations. In other words, organizations should not blindly adopt a one-for-all PM strategy when tackling different process types. Finally, this study is also one of the first attempts to investigate PM from a multidimensional perspective. Previous studies tend to treat it as unidimensional and disregard the relevant internal dynamics occurs among the PM dimensions. This thesis reveals that the complexity of their interaction may be far beyond the normal expectation. At the end, several noticeable avenues deserve further research efforts are highlighted. / Ng, Chi Hung. / "March 2008." / Advisers: T. S. Lee; Xiande Zhao; Leo Sin. / Source: Dissertation Abstracts International, Volume: 70-03, Section: A, page: 0940. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (p. 174-189). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest dissertations and theses, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Typescript (photocopy). / Digitized by Kansas Correctional Industries

This thesis examines the existing theories and applications of Multivariate Statistical Process Control, outlines areas of difficulty and proposes a new technique of multivariate process control chart with input-output relationship for optimal process control. The process control techniques developed up to the present time focused on the fast detection of out-of-control signals, and achieved considerable success in that respect. The techniques reported on multivariate process control thus far include extensions of univariate process control charts to their multivariate counterparts, ranging from classical Shewharts charts to modern Cumulative Sum Process Control charts. Alternative approaches in this area include Principal Component Approach, Partial Regression approach, Baysian modelling and sequential tests on detection of change points. Although each method has its own limitation, these new developments have significantly contributed to the achievement of a constant high quality of products and services. The techniques of process control are yet incomplete. They require continuous attention, as production and service technologies are being continuously developed.In particular, the level of automation, re-engineering of production processes and ever demanding economic optimality of technology demand the re-engineering of statistical process control. The CFM chart developed in this thesis would open the door to this area of science and lays a critical foundation for future research / Doctor of Philosophy (PhD)

Process control has been playing an increasingly important role in many industrial applications as an effective way to improve product quality, process costeffectiveness and safety. Simple linear dynamic models are used extensively in process control practice, but they are limited to the type of process behavior they can approximate. It is well-documented that simple nonlinear models can often provide much better approximations to process dynamics than linear models. It is evident that there is a potential of significant improvement of control quality through the implementation of the model-based control procedures. However, such control applications are still not widely implemented because mathematical process models in model-based control could be very difficult and expensive to obtain due to the complexity of those systems and poor understanding of the underlying physics. The main objective of this thesis is to develop new approaches to modeling and control of nonlinear processes. In this thesis, the multivariable nonlinear processes are approximated using a model with a static nonlinearity and a linear dynamics. In particular, the Hammerstein model structure, where the nonlinearity is on the input, is used. Cardinal spline functions are used to identify the multivariable input nonlinearity. Highlycoupled nonlinearity can also be identified due to flexibility and versatility of cardinal spline functions. An approach that can be used to identify both the nonlinearity and linear dynamics in a single step has been developed. The condition of persistent excitation has also been derived. Nonlinear control design approaches for the above models are then developed in this thesis based on: (1) a nonlinear compensator; (2) the extended internal model control (IMC); and (3) the model predictive control (MPC) framework. The concept of passivity is used to guarantee the stability of the closed-loop system of each of the approaches. In the nonlinear compensator approach, the passivity of the process is recovered using an appropriate static nonlinearity. The non-passive linear system is passified using a feedforward system, so that the passified overall system can be stabilized by a passive linear controller with the nonlinear compensator. In the extended IMC approach, dynamic inverses are used for both the input nonlinearity and linear dynamics. The concept of passive systems and the passivity-based stability conditions are used to obtain the invertible approximations of the subsystems and guarantee the stability of the nonlinear closed-loop system. In the MPC approach, a numerical inverse is implemented. The condition for which the numerical inversion is guaranteed to converge is derived. Based on these conditions, the input space in which the numerical inverse can be obtained is identified. This constitutes new constraints on the input space, in addition to the physical input constraints. The total input constraints are transformed into linear input constraints using polytopic descriptions and incorporated in the MPC design.