The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
Model-Based State Estimation for Fault Detection under Disturbance
MetadataShow full item record
The measurement of process states is critical for process monitoring, advanced process control, and process optimization. For chemical processes where state information cannot be measured directly, techniques such as state estimation need to be developed. Model-based state estimation is one of the most widely applied methods for estimation of unmeasured states basing on a high-fidelity process model. However, certain disturbances or unknown inputs not considered by process models will generate model-plant mismatch. In this dissertation, different model-based process monitoring techniques are developed and applied for state estimation under uncertainty and disturbance. Case studies are performed to demonstrate the proposed methods. The first case study estimates leak location from a natural gas pipeline. Non-isothermal state equations are derived for natural gas pipeline flow processes. A dual unscented Kalman filter is used for parameter estimation and flow rate estimation. To deal with sudden process disturbance in the natural gas pipeline, an unknown input observer is designed. The proposed design implements a linear unknown input observer with time-delays that considers changes of temperature and pressure as unknown inputs and includes measurement noise in the process. Simulation of a natural gas pipeline with time-variant consumer usage is performed. New optimization method for detection of simultaneous multiple leaks from a natural gas pipeline is demonstrated. Leak locations are estimated by solving a global optimization problem. The global optimization problem contains constraints of linear and partial differential equations, integer variable, and continuous variable. An adaptive discretization approach is designed to search for the leak locations. In a following case study, a new design of a nonlinear unknown input observer is proposed and applied to estimate states in a bioreactor. The design of such an observer is provided, and sufficient and necessary conditions of the observer are discussed. Experimental studies of batch and fed-batch operation of a bioreactor are performed using Saccharomyces cerevisiae strain mutant SM14 to produce β-carotene. The state estimation of the process from the designed observer is demonstrated to alleviate the model-plant mismatch and is compared to the experimental measurements.
Pan, Xinghua (2016). Model-Based State Estimation for Fault Detection under Disturbance. Doctoral dissertation, Texas A & M University. Available electronically from