Robert J. Marks

Electric load forecasting using an artificial neural network

An artificial neural network (ANN) approach is presented for electric load forecasting. The ANN is used to learn the relationship among past, current and future temperatures and loads. In order to provide the forecasted load, the ANN interpolates among the load and temperature data in a training data set. The average absolute errors of the 1 h and 24 h-ahead forecasts in tests on actual utility data are shown to be 1.40% and 2.06%, respectively. This compares with an average error of 4.22% for 24 h ahead forecasts with a currently used forecasting technique applied to the same data. >

The use of cone-shaped kernels for generalized time-frequency representations of nonstationary signals

Generalized time-frequency representations (GTFRs) which use cone-shaped kernels for nonstationary signal analysis are presented. The cone-shaped kernels are formulated for the GTFRs to produce good resolution simultaneously in time and frequency. Specifically, for a GTFR with a cone-shaped kernel, finite time support is maintained in the time dimension along with an enhanced spectrum in the frequency dimension, and the cross-terms are smoothed out. Experimental results on simulated data and real speech show the advantages of the GTFRs with the cone-shaped kernels through comparisons to the spectrogram and the pseudo-Wigner distribution. >

Introduction to Shannon sampling and interpolation theory

Regaining original signals transformed from analog to digital systems or assessing information lost in the process are the fundamental issues addressed by sampling and interpolation theory. This study attempts to understand, generalize and extend the cardinal series of Shannon sampling theory.

Query-based learning applied to partially trained multilayer perceptrons

An approach is presented for query-based neural network learning. A layered perceptron partially trained for binary classification is considered. The single-output neuron is trained to be either a zero or a one. A test decision is made by thresholding the output at, for example, one-half. The set of inputs that produce an output of one-half forms the classification boundary. The authors adopted an inversion algorithm for the neural network that allows generation of this boundary. For each boundary point, the classification gradient can be generated. The gradient provides a useful measure of the steepness of the multidimensional decision surfaces. Conjugate input pairs are generated using the boundary point and gradient information and presented to an oracle for proper classification. These data are used to refine further the classification boundary, thereby increasing the classification accuracy. The result can be a significant reduction in the training set cardinality in comparison with, for example, randomly generated data points. An application example to power system security assessment is given.

Swarm intelligence for routing in communication networks

Swarm intelligence, as demonstrated by natural biological swarms, exhibits numerous powerful features that are desirable in many engineering systems, such as communication networks. In addition, new paradigms for designing autonomous and scalable systems may result from analytically understanding and extending the design principles and operations inherent in intelligent biological swarms. A key element of future design paradigms will be emergent intelligence - simple local interactions of autonomous swarm members, with simple primitives, giving rise to complex and intelligent global behavior. Communication network management is becoming increasingly difficult due to the increasing network size, rapidly changing topology, and complexity. A new class of algorithms, inspired by swarm intelligence, is currently being developed that can potentially solve numerous problems of such networks. These algorithms rely on the interaction of a multitude of simultaneously interacting agents. A survey of such algorithms and their performance is presented here.

Minimum power broadcast trees for wireless networks: integer programming formulations

Wireless multicast/broadcast sessions, unlike wired networks, inherently reach several nodes with a single transmission. For omnidirectional wireless broadcast to a node, all nodes closer will also be reached. Heuristic algorithms for constructing the minimum power tree in wireless networks have been proposed by Wieselthier et al. and Stojmenovic et al. Recently, an evolutionary search procedure has been proposed by Marks et al. In this paper, we present three different integer programming models which can be used for an optimal solution of the minimum power broadcast/multicast problem in wireless networks. The models assume complete knowledge of the distance matrix and is therefore most suited for networks where the locations of the nodes are fixed.

Inversion of snow parameters from passive microwave remote sensing measurements by a neural network trained with a multiple scattering model

The inversion of snow parameters from passive microwave remote sensing measurements is performed with a neural network trained with a dense-media multiple-scattering model. The input-output pairs generated by the scattering model are used to train the neural network. Simultaneous inversion of three parameters, mean-grain size of ice particles in snow, snow density, and snow temperature from five brightness temperatures, is reported. It is shown that the neural network gives good results for simulated data. The absolute percentage errors for mean-grain size of ice particles and snow density are less than 10%, and the absolute error for snow temperature is less than 3 K. The neural network with the trained weighting coefficients of the three-parameter model is also used to invert SSMI data taken over the Antarctic region. >

Optimization of intensity modulated beams with volume constraints using two methods: cost function minimization and projections onto convex sets.

For accurate prediction of normal tissue tolerance, it is important that the volumetric information of dose distribution be considered. However, in dosimetric optimization of intensity modulated beams, the dose-volume factor is usually neglected. In this paper we describe two methods of volume-dependent optimization for intensity modulated beams such as those generated by computer-controlled multileaf collimators. The first method uses a volume sensitive penalty function in which fast simulated annealing is used for cost function minimization (CFM). The second technique is based on the theory of projections onto convex sets (POCS) in which the dose-volume constraint is replaced by a limit on integral dose. The ability of the methods to respect the dose-volume relationship was demonstrated by using a prostate example involving partial volume constraints to the bladder and the rectum. The volume sensitive penalty function used in the CFM method can be easily adopted by existing optimization programs. The convex projection method can find solutions in much shorter time with minimal user interaction.

Preliminary results on using artificial neural networks for security assessment (of power systems)

Artificial neural network (ANN) techniques are explored as a tool to assess the dynamic security of power systems. The basic role of ANNs is to provide assessment of the systems stability based on training examples from offline analysis. Such an assessment would be useful as an operations aid. In essence, ANNs interpolate among the planning analysis data. The authors present the results of a study to assess the capability of ANNs to learn from the offline stability analysis results and give accurate stability assessments when queried with data representing the current system status. The important feature of the result is that correct stability assessments are provided by the ANN not only when it is queried with an element of the training set of data but also under other operating conditions.<<ETX>>

A performance comparison of trained multilayer perceptrons and trained classification trees

The important differences between multilayer perceptrons and classification trees are considered. A number of empirical tests on three real-world problems in power-system load forecasting, power-system security prediction, and speaker-independent vowel recognition are presented. The load-forecasting problem, which is partially a regression problem, uses past trends to predict the critical needs of future power generation. The power-security problem uses the classifier as an interpolator of previously known states of the system. The vowel-recognition problem is representative of the difficulties in automatic speech recognition caused by variability across speakers and phonetic context. In all cases even with various sizes of training sets, the multilayer perceptron performed as well as or better than the trained classification trees. It is therefore concluded that there is not enough theoretical basis to demonstrate clear-cut superiority of one technique over the other. >