Management of uncertainty in sensor validation, sensor fusion, and diagnosis of mechanical systems using soft computing techniques

This dissertation provides means to deal with uncertainty in complex sensor driven systems through sensor validation, sensor fusion, and diagnosis. These means include probability theory, neural network theory, and fuzzy logic. In particular, this thesis introduces means for fuzzy sensor validation and fusion which were compared with a probabilistic data association scheme. The fuzzy sensor validation and fusion approach uses non-symmetric validation regions in which sensors readings are assigned confidence values. Each sensor has its own dynamic validation curve which is shaped according to sensor characteristics, taking into account the range, external factors affecting the sensor, reliability of the sensor, etc. The curves have their maximum value at the predicted value which is arrived at using fuzzy exponential weighted moving average time series predictor. Confidence curves attain minima at the boundaries of the validation gate which are determined by the maximum physically possible change a system can undergo in one time sample. Since readings outside the gate are implausible, they are discarded. Fusion is performed using a weighted average of sensor readings and confidence values, the predicted value scaled by the operating condition, and--if available--functionally redundant values calculated from sensors other than the directly redundant ones. Each method performs best in the presence of certain types of noise and recommendations are made as to which approach is more appropriate under various conditions. Several applications from extant systems (intelligent vehicle highway systems, gas turbine power plants, milling machines) show the feasibility of the approaches developed.; Another aspect of this dissertation is to provide a tool for diagnosis in the presence of vague symptoms. This is achieved though fuzzy abduction which can diagnose crisp as well as soft faults. This means that faults can be diagnosed if they occur to some degree. The proposed algorithm computes a closeness measure taking into account the distance from an observed symptom set to the modeled symptom set for all failure combinations. It then ranks the failure sets according to maximum closeness measure and minimum cover, i.e., number of faults. As an extension, a framework for fuzzy influence diagrams is provided which uses this closeness measure.