Recent emphasis has been placed on merging regional forest inventory data with satellite-based information both to improve the efficiency of estimates of population totals, and to produce regional maps of forest variables. There are numerous ways in which forest class and structure variables may be modeled as functions of remotely sensed variables, yet surprisingly little work has been directed at surveying modem statistical techniques to determine which tools are best suited to the tasks given multiple objectives and logistical constraints. Here, a series of analyses to compare nonlinear and nonparametric modeling techniques for mapping a variety of forest variables, and for stratification of field plots, was conducted using data in the Interior Western United States. The analyses compared four statistical modeling techniques for predicting two discrete and four continuous forest inventory variables. The modeling techniques include generalized additive models (GAMs), classification and regression trees (CARTs), multivariate adaptive regression splines (MARS), and artificial neural networks (ANNs). Alternative stratification schemes were also compared for estimating population totals. The analyses were conducted within six ecologically different regions using a variety of satellite-based predictor variables. The work resulted in the development of an objective modeling box that automatically models spatial response variables as functions of any assortment of predictor variables through the four nonlinear or nonparametric modeling techniques. In comparing the different modeling techniques, all proved themselves workable in an automated environment, though ANNs were more problematic. When their potential mapping ability was explored through a simple simulation, tremendous advantages were seen in use of MARS and ANN for prediction over GAMs, CART, and a simple linear model. However, much smaller differences were seen when using real data. In some instances, a simple linear approach worked virtually as well as the more complex models, while small gains were seen using more complex models in other instances. In real data runs, MARS performed (marginally) best most often for binary variables, while GAMs performed (marginally) best most often for continuous variables. After considering a subjective "ease of use" measure, computing time and other predictive performance measures, it was determined that MARS had many advantages over other modeling techniques. In addition, stratification tests illustrated cost-effective means to improve precision of estimates of forest population totals. Finally, the general effect of map accuracy on the relative precision of estimates of population totals obtained under simple random sampling compared to that obtained under stratified random sampling was established and graphically illustrated as a tool for management decisions.

No description available.

New reactor designs and the license extensions of the current reactors has created new condition monitoring challenges. A major challenge is the creation of a data-based model for a reactor that has never been built or operated and has no historical data. This is the motivation behind the creation of a hybrid modeling technique based on first principle models that adapts to include operating reactor data as it becomes available. An Adaptive Non-Parametric Model (ANPM) was developed for adaptive monitoring of small to medium size reactors (SMR) but would be applicable to all designs. Ideally, an adaptive model should have the ability to adapt to new operational conditions while maintaining the ability to differentiate faults from nominal conditions. This has been achieved by focusing on two main abilities. The first ability is to adjust the model to adapt from simulated conditions to actual operating conditions, and the second ability is to adapt to expanded operating conditions. In each case the system will not learn new conditions which represent faulted or degraded operations. The ANPM architecture is used to adapt the model's memory matrix from data from a First Principle Model (FPM) to data from actual system operation. This produces a more accurate model with the capability to adjust to system fluctuations. This newly developed adaptive modeling technique was tested with two pilot applications. The first application was a heat exchanger model that was simulated in both a low and high fidelity method in SIMULINK. The ANPM was applied to the heat exchanger and improved the monitoring performance over a first principle model by increasing the model accuracy from an average MSE of 0.1451 to 0.0028 over the range of operation. The second pilot application was a flow loop built at the University of Tennessee and simulated in SIMULINK. An improvement in monitoring system performance was observed with the accuracy of the model improving from an average MSE of 0.302 to an MSE of 0.013 over the adaptation range of operation. This research focused on the theory, development, and testing of the ANPM and the corresponding elements in the surveillance system.

No description available.