Taiwan is a small island state surrounded by the oceans but with large population. With limited land space, it would be worthwhile considering how to stabilize the existing coast or to create stable artificial beaches. Under the onslaught of storm surge and large wave from typhoons, beach erosion would occur accompanying by formation of a submerged bar beyond the surf zone with the sand removed from the beach. After the storm, the bar material maybe transport back by the swell and predominant waves which helps recover the original beach, thus producing a beach profile in dynamic equilibrium. The main purpose of this research is to use the back-propagation neural network¡]BPNN¡^, which trains a sample model and creates a system for the estimation, prediction, decision making and verification of an anticipated event. By the BPNN, we can simulate the key characteristic parameters for the storm beach profile resulting from typhoon action. Source data for training and verification are taken from the experimental results of beach profile change observed in large-scale wave tank¡]LWT¡^conducted by Coastal Engineering Research Center¡]CERC¡^in the USA in the 1960s and that from the Central Research Institute of Electric Power Industry in Japan in the 1980s. Some of the data are used as training pairs and others for verification and prediction of the key parameters of berm erosion and bar formation. Through literature review and simulation on the related parameters for storm beach profile, methodology for the prediction of the beach profile and bar/berm characteristics can be established.

In many practical applications, it is well known that data collected inevitably contain one or more anomalous outliers; that is, observations that are well separated from the majority or bulk of the data, or in some fashion deviate from the general pattern of the data. The occurrence of outliers may be due to misplaced decimal points, recording errors, transmission errors, or equipment failure. These outliers can lead to erroneous parameter estimation and consequently affect the correctness and accuracy of the model inference. In order to solve these problems, three robust fuzzy neural networks (FNNs) will be proposed in this dissertation. This provides alternative learning machines when faced with general nonlinear learning problems. Our emphasis will be put particularly on the robustness of these learning machines against outliers. Though we consider only FNNs in this study, the extension of our approach to other neural networks, such as artificial neural networks and radial basis function networks, is straightforward. In the first part of the dissertation, M-estimators, where M stands for maximum likelihood, frequently used in robust regression for linear parametric regression problems will be generalized to nonparametric Maximum Likelihood Fuzzy Neural Networks (MFNNs) for nonlinear regression problems. Simple weight updating rules based on gradient descent and iteratively reweighted least squares (IRLS) will be derived. In the second part of the dissertation, least trimmed squares estimators, abbreviated as LTS-estimators, frequently used in robust (or resistant) regression for linear parametric regression problems will be generalized to nonparametric least trimmed squares fuzzy neural networks, abbreviated as LTS-FNNs, for nonlinear regression problems. Again, simple weight updating rules based on gradient descent and iteratively reweighted least squares (IRLS) algorithms will be provided. In the last part of the dissertation, by combining the easy interpretability of the parametric models and the flexibility of the nonparametric models, semiparametric fuzzy neural networks (semiparametric FNNs) and semiparametric Wilcoxon fuzzy neural networks (semiparametric WFNNs) will be proposed. The corresponding learning rules are based on the backfitting procedure which is frequently used in semiparametric regression.

The adult Drosophila central nervous system is produced by two phases of neurogenesis: the first phase occurs during embryonic development where the larval brain is formed and the second occurs during larval development to form the adult brain. Neurogenesis in both phases is caused by the activation of neural stem cell division and subsequent progenitor cell division and terminal differentiation. Proper activation of neural stem cell division in the larval brain is essential for proper patterning and functionality of the adult central nervous system. Initiation of neural stem cell proliferation requires signaling from the Fibroblast Growth Factor (FGF) homolog Branchless (Bnl) and by the Hedgehog (Hh) growth factor. I have focused on the interactions between both of these signaling pathways with respect to post-embryonic neural stem cell proliferation using the Drosophila larval brain. Using proliferation assays and quantitative real-time PCR, I have shown that Bnl and Hh signaling is inter-dependent in the 1st instar larval brain and activates neural stem cell proliferation. I have also shown that overexpression of bnl can rescue signaling and neuroblast proliferation in a hh mutant. However, overexpression of hh does not rescue signaling or neuroblast proliferation in a bnl mutant, suggesting that Bnl is the signaling output of the Bnl-Hh feedback loop and that all central brain and optic lobe neural stem cells require Bnl signaling to initiated proliferation.

Neural conduction delay is a serious issue for organisms that need to act in real time. Perceptual phenomena such as the flash-lag effect (FLE, where the position of a moving object is perceived to be ahead of a brief flash when they are actually colocalized) suggest that the nervous system may perform extrapolation to compensate for delay. However, the precise neural mechanism for extrapolation has not been fully investigated. The main hypothesis of this dissertation is that facilitating synapses, with their dynamic sensitivity to the rate of change in the input, can serve as a neural basis for extrapolation. To test this hypothesis, computational and biologically inspired models are proposed in this dissertation. (1) The facilitatory activation model (FAM) was derived and tested in the motion FLE domain, showing that FAM with smoothing can account for human data. (2) FAM was given a neurophysiological ground by incorporating a spike-based model of facilitating synapses. The spike-based FAM was tested in the luminance FLE domain, successfully explaining extrapolation in both increasing and decreasing luminance conditions. Also, inhibitory backward masking was suggested as a potential cellular mechanism accounting for the smoothing effect. (3) The spike-based FAM was extended by combining it with spike-timing-dependent plasticity (STDP), which allows facilitation to go across multiple neurons. Through STDP, facilitation can selectively propagate to a specific direction, which enables the multi-neuron FAM to express behavior consistent with orientation FLE. (4) FAM was applied to a modified 2D pole-balancing problem to test whether the biologically inspired delay compensation model can be utilized in engineering domains. Experimental results suggest that facilitating activity greatly enhances real time control performance under various forms of input delay as well as under increasing delay and input blank-out conditions. The main contribution of this dissertation is that it shows an intimate link between the organism-level problem of delay compensation, perceptual phenomenon of FLE, computational function of extrapolation, and neurophysiological mechanisms of facilitating synapses (and STDP). The results are expected to shed new light on real-time and predictive processing in the brain, and help understand specific neural processes such as facilitating synapses.

Water production is a challenge in production operations because it is generally costly to produce, treat, and it can hamper hydrocarbon production. This is especially true for gas wells in unconventional reservoirs like shale because the relatively low gas rates increase the economic impact of water handling costs. Therefore, we have considered the following questions regarding water production from shale gas wells: (1) What is the effect of water production on gas production? (2) What are the different water producing mechanisms? and (3) What is the water production potential of a new well in a given gas shale province. The first question was answered by reviewing relevant literature, highlighting observed deficiencies in previous approaches, and making recommendations for future work. The second question was answered using a spreadsheet based Water-Gas-Ratio analysis tool while the third question was investigated by using artificial neural networks (ANN) to decipher the relationship between completion, fracturing, and water production data. We will consequently use the defined relationship to predict the average water production for a new well drilled in the Barnett Shale. This study also derived additional insight into the production trends in the Barnett shale using standard statistical methods. The following conclusions were reached at the end of the study: 1) The observation that water production does not have long term deleterious effect on gas production from fractured wells in tight gas sands cannot be directly extended to fractured wells in gas shales because the two reservoir types do not have analogous production mechanisms. 2) Based on average operating conditions of well in the Barnett Shale, liquid loading was found to be an important phenomenon; especially for vertical wells. 3) A neural network was successfully used to predict average water production potential from a well drilled in the Barnett shale. Similar methodology can be used to predict average gas production potential. Results from this work can be utilized to mitigate risk of water problems in new Barnett Shale wells and predict water issues in other shale plays. Engineers will be provided a tool to predict potential for water production in new wells.

This thesis is to study the distribution system loss by applying artificial neural networks(ANN). To enhance the efficiency of loss analysis, the distribution system network has been obtained by retrieving that component information for the automated mapping and facility management system (AM/FM). The topology process and node reduction has also been applied to identify the network configuration and the input data for load flow analysis. The load survey study is used to derive the typical load patterns of various customer losses. The monthly energy consumption of customers by each transformer, which has been retrieved for the customer information system(CIS), is used to derive the hourly loading of each distribution transformer. The three phase load flow analysis has been performed for different types of distribution feeders to solve feeder loss to generate the data set for the training and testing of neural networks. The ANN for distribution loss analysis, which has been obtained after network training, can solve the distribution system loss very efficiently according to the feeder load demand, length, transformer capacity and voltage level. With short feeder length and voluminous customers served by the distribution feeders in urban area, the transformer core loss and secondary line loss contribute most of the distribution feeder loss. On the other hand, the line loss of rural distribution feeder is more significant because of the longer distribution lines to serve more scattering customers. With the neural based distribution system loss modeling, the distribution system loss can be estimated very easily, which can provide Taipower a good reference to enhance the operation efficiency of distribution system.

The thesis illustrates the development of DSP-based systems-¡§DSP Based Face Characteristic Extraction and Identity Recognition System¡¨.The principal system consists of three major subsystems and two kinds of structure of recognition algorithms.Three major subsystems are Image Acquisition System.Image Preprocessing System,and face characteristic extraction individually.Two kinds of structure are Competitive Neural Network and Gaussian mixture model respectively In actual proving,we adopt colored half-length face image alone only face image,and simulate on PC.In order to acquire the characteristic parameter with the different parts to the people faces , and then achieve the purpose that the identity discerns.Finally implant it to DSP .Shown by the experimental result,this system can really reach the anticipative goal,and gain good recognition and efficiency.

The thesis is composed of two topics : a low-noise instru-mentation amplifier (IA) for neural signal sensing and a low-power implantable bladder pressure monitor SOC (system-on-chip). A low-noise instrumentation amplifier for bio-medical appli-cations is proposed in the first topic. It is designed for sampling vague neural signals thanks to its high gain, high CMRR in a pre-defined bandwidth. A low-power implantable bladder pressure monitor system is presented in the next topic. The system contains several parts : a commercial pressure sensor, an IA, an analog to digital converter (ADC), a parallel to serial converter (PtoS), an RF transmitter and a sleep controller. The IA with 1-atm canceling is designed for high resolution and linearity in the pre-defined bladder pressure range. For low power and low speed applications, a successive approximation ADC (SA ADC) is employed in the system. A clear flag is added to the PtoS to enhance reliability. Our chip saves a great portion of power to extend the processing time owing to the novel sleep controller.

An electric power system consists of the generating stations, the transmission lines, and the distribution systems. Transmission lines are the connecting links between the generating stations and the distribution systems. With the rapid growth of economy and technology, the demand for large blocks of power, power quality and increased reliability suggested the interconnection of neighboring systems. Transmission lines are elements of a network which connects the generating plants to the distribution systems, and could extend hundreds of miles . Because of the long distances traversed by transmission lines over open area, they tend to fade by natural and artificial calamity imposed on the power system. It maybe easy to discover the fault with sufficient information in the populous region. When fault occurs in the remote region, it is difficult to identify the outage location. An efficient and reliable technique is thus desirable to resolve the problem. This dissertation presents the fault location for high voltage lines with Artificial Neural Network( ANN ) method. Beside the fault location, this research also improve the problem further by considering the fault resistance. The fault resistance may not remain the same due to the variation of environmental factors. The fault location may involve errors owing to the fault resistance. An algorithms has been developed in this dissertation to calculate fault resistance and revise the ANN training data for three-phase fault, double line-to-ground fault, single line-to-ground fault, and line-to-line fault. To verify the effectiveness of the method, practical transmission lines were used for tests. The results proved that the method could be used to identify the fault location effectively and help dispatchers determine a reference distance.

­^¤å´£­n¡G A utility function model of individual credit card holder based on their spending behavior is constructed in this research. An accumulation of the individual utility of three different risk attitudes of cardholders may be useful for promoting the profits of credit card business for banks. Due to the privacy of cardholders and the lack of real data, a questionnaire sampling is used to collect data for this study. A result of this experimental study indicates that credit card holders with a different sex, age, level of education, asset condition, seniority, and occupation have different risk tendency. Based on 249 effective samples in this research, credit card holders who belong to females, teenagers, relatively low educated, without real estate, middle seniority, and relatively volatile occupation are more risk seeking. Relatively risk seeking credit card holders have the tendency to make use of their revolving credit and to borrow cash or to buy financial products with their credit cards. For those with three different risk attitudes, their default of credit card loans are not significantly different. The finding indicates relatively risk seeking cardholders may contribute more profits to the credit card business for banks. A risk attitude classification model built by artificial neural network has also been developed. The model may assist banks' administrators using their applicants' demographics to distinguish their risk attitude for approving an appropriate credit limit for a cardholder's expenditure to promote the total credit card profit for banks.