Institute for Systems Research Technical Reports
Permanent URI for this collectionhttp://hdl.handle.net/1903/4376
This archive contains a collection of reports generated by the faculty and students of the Institute for Systems Research (ISR), a permanent, interdisciplinary research unit in the A. James Clark School of Engineering at the University of Maryland. ISR-based projects are conducted through partnerships with industry and government, bringing together faculty and students from multiple academic departments and colleges across the university.
Browse
19 results
Search Results
Item Fixture-Based Design Similarity Measures for Variant Fixture Planning(1999) Balasubramanian, Sundar; Herrmann, Jeffrey W.; ISROne of the important activities in process planning is the design of fixtures to position, locate and secure the workpiece during operations such as machining, assembly and inspection. The proposed approach for variant fixture planning is an essential part of a hybrid process planning methodology.The aim is to retrieve, for a new product design, a useful fixture from a given set of existing designs and their fixtures. Thus, the variant approach exploits this existing knowledge.
However, since calculating each fixture's feasibility and then determining the necessary modifications for infeasible fixtures would require too much effort, the approach searches quickly for the most promising fixtures based on a surrogate design similarity measure. Then, it evaluates the definitive usefulness metric for those promising fixtures and identifies the best one for the new design.
Item Using Neural Networks to Generate Design Similarity Measures(1999) Balasubramanian, Sundar; Herrmann, Jeffrey W.; Herrmann, Jeffrey W.; ISRThis paper describes a neural network-based design similarity measure for a variant fixture planning approach. The goal is to retrieve, for a new product design, a useful fixture from a given set of existing designs and their fixtures. However, since calculating each fixture feasibility and then determining the necessary modifications for infeasible fixtures would require too much effort, the approach searches quickly for the most promising fixtures. The proposed approach uses a design similarity measure to find existing designs that are likely to have useful fixtures. The use of neural networks to generate design similarity measures is explored.This paper describes the back-propagation algorithm for network learning and highlights some of the implementation details involved. The neural network-based design similarity measure is compared against other measures that are based on a single design attribute.Item Commodity Trading Using Neural Networks: Models for the Gold Market(1997) Brauner, Erik; Dayhoff, Judith E.; Sun, Xiaoyun; ISREssential to building a good financial forecasting model is having a realistic trading model to evaluate forecasting performance. Using gold trading as a platform for testing we present a profit based model which we use to evaluate a number of different approaches to forecasting. Using novel training techniques we show that neural network forecasting systems are capable of generating returns for above those of classical regression models.Item Automated Network Fault Management(1996) Viswanathan, P.; ISR; CSHCNWith the recent growth of telecommunication networks, fault management has gained much importance. Since it is difficult for humans to manage large networks, automation of many of these functions has attracted much attention. Some of the ideas proposed for automating such functions include the use of artificial intelligence techniques. Neural help to analyze large volumes of numerical data. Expert systems help to analyze observed symptoms and identify the cause using a rule-based approach. However, research in artificial intelligence has shown that when either of these two methods is used alone, several weaknesses are observed in the resulting system. Thus, some other methodology would be required for tackling such large problems.In this thesis, an approach involving the use of a hybrid system involving both neural networks and expert systems for performing automated network fault management is investigated. Data networks using the X.25 protocol are considered. A minimum cost routing scheme is used for re-routing future calls given the occurrence of a fault. A method for partitioning the data (obtained from the X.25 network) between the neural network and the expert system is suggested. Radial basis function networks are used as the neural network architecture for performing fault classification using performance data. Queries are provided for the expert system to determine the type of fault that occurred using the results of the neural network, together with alarms, SNMP traps, and X.25 SNMP statistics.
Item In Process Machine Tool Vibration Cancellation Using Electreostrictive Actuators(1996) Eshete, Zelalem; Zhang, G.; ISRItem Dynamic Attractors and Basin Class Capacity in Binary Neural Networks(1995) Dayhoff, Judith E.; Palmadesso, Peter J.; ISRThe wide repertoire of attractors and basins of attraction that appear in dynamic neural networks not only serve as models of brain activity patterns but create possibilities for new computational paradigms that use attractors and their basins. To develop such computational paradigms, it is first critical to assess neural network capacity for attractors and for differing basins of attraction, depending on the number of neurons and the weights. In this paper we analyze the attractors and basins of attraction for recurrent, fully-connected single layer binary networks. We utilize the network transition graph - a graph that shows all transitions from one state to another for a given neural network - to show all oscillations and fixed-point attractors, along with the basins of attraction. Conditions are shown whereby pairs of transitions are possible from the same neural network. We derive a lower bound for the number of transition graphs possible 2n2- n , for an n-neuron network. Simulation results show a wide variety of transition graphs and basins of attraction and sometimes networks have more attractors than neurons. We count thousands of basin classes - networks with differing basins of attraction - in networks with as few as five neurons. Dynamic networks show promise for overcoming the limitations of static neural networks, by use of dynamic attractors and their basins. We show that dynamic networks have high capacity for basin classes, can have more attractors than neurons, and have more stable basin boundaries than in the Hopfield associative memory.Item Target Discrimination with Neural Networks(1995) Lin, Daw-Tung; Dayhoff, Judith E.; Resch, C.L.; ISRThe feasibility of distinguishing multiple type components of exo-atmospheric targets is demonstrated by applying the Time Delay Neural Network (TDNN) and the Adaptive Time-Delay Neural Network (ATNN). Exo-atmospheric targets are especially difficult to distinguish using currently available techniques because all target parts follow the same spatial trajectory. Thus classification must be based on light sensors that record signal over time. Results have demonstrated that the trained neural networks were able to successfully identify warheads from other missile parts on a variety of simulated scenarios, including differing angles and tumbling. The network with adaptive time delays (the ATNN) performs highly complex mapping on a limited set of training data and achieves better generalization to overall trends of situations compared to the TDNN, which includes time delays but adapts only its weights. The ATNN was trained on additive noisy data and it is shown that the ATNN possesses robustness to environment variations.Item Analysis and Synthesis of Distributed Systems(1994) Zhuang, Y.; Baras, J.S.; ISRWe first model and analyze distributed systems including distributed sensors and actuators. We then consider identification of distributed systems via adaptive wavelet neural networks (AWNNs) by taking advantage of the multiresolution property of wavelet transforms and the parallel computational structure of neural networks. A new systematic approach is developed in this dissertation to construct an optimal discrete orthonormal wavelet basis with compact support for spanning the subspaces employed for system identification and signal representation. We then apply a backpropagation algorithm to train the network to approximate the system. Filter banks for parameterizing wavelet systems are studied. An analog VLSI implementation architecture of the AWNN is also given in this dissertation. This work is applicable to signal representation and compression under optimal orthonormal wavelet bases in addition to progressive system identification and modeling. We anticipate that this work will find future applications in signal processing and intelligent systems.Item A Population-Based Search from Genetic Algorithms through Thermodynamic Operation(1994) Sun, Ray-Long; Dayhoff, Judith E.; Weigand, William A.; ISRThe guided random search techniques, genetic algorithms and simulated annealing, are very promising strategies, and both techniques are analogs from physical and biological systems. Through genetic algorithms, the simulation of evolution for the purposes of parameter optimization has generally demonstrated itself to be a robust and rapid optimization technique. The simulated annealing algorithm often finds high quality candidate solutions. Limitations, however, occur in performance because optimization may take large numbers of iterations or final parameter values may be found that there are not at global minimum (or maximum) points. In this paper we propose a population-based search algorithm that combines the approaches from genetic algorithms and simulated annealing. The combined approach, called GASA, maintains a population of individuals over a period of generations. In the GASA technique, simulated annealing is used in choices regarding a subset of individuals to undergo crossover and mutation. We show that the GASA technique performs superior to a genetic algorithm on the Bohachevsky function, an objective function with m any local minima. The methodology and the test results on function optimization are given and compared with classical genetic algorithms.Item Identification of Infinite Dimensional Systems Via Adaptive Wavelet Neural Networks(1993) Zhuang, Y.; Baras, John S.; ISRWe consider identification of distributed systems via adaptive wavelet neural networks (AWNNs). We take advantage of the multiresolution property of wavelet systems and the computational structure of neural networks to approximate the unknown plant successively. A systematic approach is developed in this paper to find the optimal discrete orthonormal wavelet basis with compact support for spanning the subspaces employed for system identification. We then apply backpropagation algorithm to train the network with supervision to emulate the unknown system. This work is applicable to signal representation and compression under the optimal orthonormal wavelet basis in addition to autoregressive system identification and modeling. We anticipate that this work be intuitive for practical applications in the areas of controls and signal processing.