IRANIAN JOURNAL OF ELECTRICAL AND COMPUTER ENGINEERING (IJECE)
Year:2003 | Volume:2 | Issue:1|Papers Count:11

Power utilities must understand how customers use their products and monitoring/survey analysis of customer usage patterns is the best way to understand how customers use electricity. This is essential to formulate any long term planning, forecasting or marketing strategy. However, only customer’s service level loads (total loads) are available. As the collection and analysis of customer usage data is very costly and time consuming, most power utilities do not allocate the necessary resources required to properly address this issue. That’s why such an important necessity has been left unattended. If we could decompose the service level loads into end-use load profiles, then we are able to analyze them. In this study, a multi-agent system is applied to analyze the residential customer usage patterns in a cost-effective way. For this study, a power utility has recorded the service level load of residential customers in 15-minute intervals over 3 months. A residential survey collected by the same utility is also used. The seven-channel end-use meters are installed as a part of a pilot program to study the validity of end-use load research. In this research, every agent represents an appliance. Also, an artificial neural network (ANN) is assigned for each agent. Back-propagation (BP) learning algorithm is used to learn about the environment and communication among agents. We analyzed the manner in which the BP learning algorithm can be used in such a system. We also discovered several problems discussed in the paper in applying BP learning to multi-agent systems. Here for the sake of simplicity, only one household is taken into account for simulation. The simulation shows that the present study can provide detail information that currently does not exist. This project is currently being implemented.

The paper deals with the methodology of designing a complete fuzzy model of a DC drive based on a suitable database of measured input–output values. This methodology covers the entire range of possible drive inputs, without requiring any information about the drive structure and parameters. The presented modeling method is verified by simulation by means of the MATLAB® Programming package.

A novel integral variable structure control (VSC) system and a preview control system to control the speed of a DC drive and obtain maximum efficiency over the whole control range are proposed in this paper. The reduced dynamic equation of the DC (RDE-DC) drive system is derived based on maximum efficiency. This is done to simplify the complexity of the control system and hence to reduce the execution time of the microprocessor. An integral action is introduced into the VSC to improve the transient response, minimize the steady state tracking error and reduce the rise and settling times of the drive system. An augmented system utilizing a pure integrator (filter) is introduced into the VSC system to mitigate the input chattering problem of the drive system. A preview controller is also synthesized and implemented with the RDE-DC drive system to maximize the efficiency and control the drive speed. This controller utilizes few future values of the desired signal and disturbance signal. The desired signal is the desired drive speed while the load torque is considered as a disturbance signal. The design procedures and comparisons between the different VSC control systems are made. MATLAB® simulation studies are carried-out to investigate the feasibility, tracking performance and robustness of the control system with changing the speed; torque and parameters of the RDE-DC drive system.  

In this paper a global Artificial Neural Network (ANN), algorithm for on-line speed control of a three-phase induction motor (IM), is proposed. This algorithm is based on the optimal preview controller. It comprises a novel error system and vector control of the IM. The IM model includes three input variables, which are the stator angular frequency and the two components of the stator space voltage vector, and three output variables, which are the rotor angular velocity and the two components of the stator space flux linkage. The objective of the proposed algorithm is to achieve rotor speed control, field orientation control and constant flux control. In order to emulate the characteristic of the optimal preview controller within global and accurate performance system, a neural network-based technique for the on-line purpose of speed control of IM, is implemented. This technique is utilized based on optimizing the speed control problem using the optimal preview control law. The numerical solution is used to train a feed ANN using the radial basis method. Successive trained data is utilized to obtain global stability operation for the IM over the whole control intervals. This data includes, several desired speed trajectories and different load torque operations in addition to the motor parameter variations. Digital computer simulation results have been carried-out to demonstrate the feasibility, reliability and effectiveness of the proposed global ANN algorithm.

This paper deals with results of leakage current measurements on 20 kV porcelain insulators. The measurements were performed in a time frame of one period (50 Hz) and the insulators were contaminated from a compound of salt and kaolin. The study was made via I-U characteristic curves plotted for one cycle of voltage application. The investigation of the observed phenomena showed that the response to the applied voltage is linear at first and thereafter takes the form of current pulses. Partial discharges (streamers), short and long duration arcs are the most likely physical phenomena detected experimentally. This assumption is convincingly supported from calculations of the average energy per pulse, which was found to be close to values reported in the related literature.

This paper presents a simple and systematic approach to steer an underwater vehicle model by considering two different cases: (i) when all actuators are functional, and (ii) when one actuator is not working. In first case, the model of an underwater vehicle is steered by employing a Lie bracket extension of the original system and the resulting feedback law is as a composition of a standard stabilizing feedback control for the extended system and a periodic continuation of a parameterized solution to an open loop, finite horizon control problem stated in the logarithmic coordinates of flows. In second case (which represents a physical example where second level Lie bracket is necessary for controllability), the original system is decomposed into two subsystems; one subsystem, which is fifth dimensional, steered by a similar approach used in case (i) and the second subsystem, which is one dimensional, steered by using sinusoidal inputs. The mixture of both type of control is utilized to steer the actual system. The synthesis method is general, in that it applies to a large class of drift free, completely controllable systems, for which the associated controllability Lie algebra is locally nilpotent.

Weighted median (WM) filters, a nonlinear filter class based on a median operator and a weight vector associated with samples inside the filter window, take their popularity from the robust order-statistic theory, the noise attenuation capability and the degree of the freedom related to filter design. In order to adapt a filter behavior for a variety of statistics describing the desired signal and the noise distribution, there were developed some optimization algorithms based on estimation and structural approaches. In this paper, we focus on optimal weighted median filters based on the estimation approach. Besides well-known WM optimization algorithms that utilize linear and sigmoidal approximation of a sign function, we test and analyze a genetic approach that outperforms others optimal WM algorithms especially in terms of the signal-detail preservation.

In this paper, we have considered the problem of maintaining connected components in quadtree representation of binary images when a small portion of the image undergoes change. The batch approach to recalculate the connected components information is very expensive. Our algorithms update the quadtree and the connected components labeling when a homogeneous region in the quadtree is changed. Our algorithms visit less number of nodes as compared to the batch approach. For small changes in the image, the proposed algorithms save time to update the quadtree and take less time to update the labels of the components on the average.

The direct-sequence spread-spectrum (DS-SS) transmission system offers a promising solution to an overcrowded frequency spectrum amid growing demand for mobile and personal communication systems. Spreadspectrum system by its very nature is an interference-tolerant modulation. However, there are situations where the processing gain is inadequate and interference suppression techniques must be employed. In this paper, it will be shown how linear or non-linear estimators inserted into DS-SS receivers can be applied for interference suppression. In our consideration, Wiener filters (WF) and Volterra filters (VF) will be used as the estimators. In order to demonstrate the performance properties of discussed DS-SS receivers, a number of computer experiments have been done. The results of the experiments will show that in the case of wideband, narrowband or joined wideband and narrowband interference the VF application in DS-SS receivers can outperform the WF application or DS-SS receivers based on simple application of matched filters (MF).

The electromagnetic coupling of incident plane wave through rectangular apertures perforating two parallel infinite conducting planes is analyzed using the characteristic modes theory based on the application of image theory and the equivalence principle. First, the integral equations for the parallel planes problem are established. Then, the integral equations are discretized into matrix equivalence. Next, the characteristic modes theory is applied to solve the matrix equations and to obtain the equivalent magnetic currents over the apertures surfaces in the planes. Finally, numerical results for the equivalent magnetic currents and for the output radiation pattern are given for the case of three apertures distributed over the two planes.

This paper presents a new recursive formulation for Walsh-Hadamard Transform (WHT) that allows the generation of higher order (longer size) multidimensional (m-d) WHT architectures from m 2 lower order (shorter sizes) WHT architectures. The objective of our work is to derive a unified framework and a design methodology that allows direct mapping of the proposed algorithms into modular VLSI architectures. Our methodology is based on manipulating tensor product forms so that they can be mapped directly into modular parallel architectures. The resulting WHT circuits have very simple modular structure and regular topology.