Search

The paper presents the statistical data of the significant parameters of lightning flash, collected by many researchers over many years around the world. The significant parameters of a lightning flash are: peak current, waveshape and velocity of the return stroke, the total flash charge and /spl int/I/sup 2/dt. Negative first strokes have traditionally been considered to produce the worst stress on the system insulation. The subsequent negative strokes have significantly lower peak current but shorter wavefronts. This may stress the system insulation more. The positive strokes have about the same median current value as the negative first strokes and longer fronts, thus producing less stress. However, their duration is longer than that of the negative strokes. Therefore, the system insulation may be damaged because of the lower volt-time characteristic for long-duration waves. The positive strokes may also cause more thermal damage because of their significantly higher charge and /spl int/I/sup 2/dt. The relationship between the return-stroke velocity and the current peak is a significant parameter in estimating lightning-induced voltages and also in estimating the peak current from the radiated electromagnetic fields of the lightning channel. For better accuracy, the current and the velocity should be measured simultaneously. Better methods to measure the stroke current need to be developed. Correlation coefficient between various lightning parameters is another important parameter which will affect the analysis significantly. Lightning characteristics should be classified according to geographical regions and seasons instead of assuming these characteristics to be globally uniform.

Surge arresters were evaluated as a means of improving reliability by reduction in 13 kV distribution line flashovers; five feeders were extensively monitored for three years. Lightning-caused faults were detected with the US National Lightning Detection Network, magnetic-field and optical sensors attached to fault recorders, and lightning-activated camera systems. Three circuits had arresters added on all phases with spacings of 40 m (every pole), 200 m, and 400 m respectively. Two other control circuits had no additional arresters. There was no discernible difference in the lightning performance of the circuits outfitted with the different arrester spacings when line length and lightning flash exposure are considered. Evidence indicated that most of the faults were caused by direct strokes rather than induced voltage flashovers. The monitored circuits are in areas with relatively heavy tree coverage which should reduce the number of direct strokes but cause higher induced voltages. In a surprising result, the circuit with arresters installed on every pole had several lightning-caused faults.

Reliability and power quality are two of the most discussed topics in the utility industry today. Many groups have performed power quality studies with varying results. One large industrial customer (Champion International) in northern New York State, which experienced numerous process interruptions in its paper mill due to power system disturbances, is the focus of this study. Monitoring methods, results, inadequacies of standards, and unique mitigation techniques used to conquer their problems are discussed in this paper.

Rocket triggered lightning flashes were triggered near a de-energized experimental distribution line while simultaneous measurements of the induced phase to neutral voltage were performed. Electric field, magnetic field and lightning stroke current characteristics associated with the triggered flashes were also measured. Analysis of 63 stroke currents and corresponding induced voltages reveals that the peak induced voltages are roughly 63% larger than predicted by the commonly used Rusck theory but in fairly good agreement with modern coupling theory using measured electric and magnetic fields as inputs. The induced voltages had peak magnitudes between about 8 and 100 kV for stroke magnitudes ranging from a few kiloamperes up to about 44 kA. The vertical electric field resulting from the lightning flash was found to play a dominant role in the induction process, but the magnetic field also contributed significantly to the total induced voltage. These measurements suggest that induced voltages on distribution lines are larger than expected and may lead to more flashovers than expected.

The Niagara Mohawk Power Corporation (NMPC) Research and Development Department sponsored a major power quality study of two distribution feeders in the Buffalo, New York region. All levels of these systems, from the substation bus to the customer service entrance, were instrumented with monitoring equipment. A variety of measurements, encompassing both transient and steady state system behavior, were performed. The use of multiple monitoring locations allowed NMPC to assess the origins and scope of various disturbances. The study generated a database which can serve as a guide for assessing relative power quality of the NMPC system. The study also formulated suggestions on areas which the industry (both the power industry and consumer products industry) might address in the future on the effect of standard utility operation towards consumer appliances. >

This paper reports a study of distribution line lightning performance, using computer simulations of lightning overvoltages. The results of previous investigations are extended with a detailed model of induced voltages from nearby strokes, coupled into a realistic power system model. The paper also considers the energy duty of distribution-class surge arresters exposed to direct strokes. The principal result is that widely separated pole-top arresters can effectively protect a distribution line from induced-voltage flashovers. This means that nearby lightning strokes need not be a significant lightning performance problem for most distribution lines. >

The uncertainties involved in evaluating induced voltage lightning performance of distribution lines are reviewed, and two algorithms to facilitate computer simulation are outlined. These are a simplified surge impedance model of each wire section of the line, permitting the coupled electromagnetic fields to be replaced by simple algebraically computed forward-backward (FB) current pairs, easily processed by computer to handle current reflections and refractions from power transformers, arresters, taps and other discontinuities; and an algorithm for distribution line-induced voltage lightning performance calculations that integrates between a striking distance curve and a critical flashover-distance curve. >

A program, the distribution lightning program (DLP), for analyzing lightning flashovers is discussed. The goal is to improve protection of overhead distribution circuits from lightning. Various lightning protection schemes can be simulated. The program uses traveling wave calculations to analyze direct lightning strokes and induced voltages caused by nearby strokes. Shield wires and arresters, two line protection options that reduce momentary outages caused by lightning, are discussed. The analysis shows that enhanced insulation level, grounding, and conductor placement can improve flashover performance. Two examples are provided to show the flexibility of computer analysis in the design and selection of overhead distribution structures. DLP was written to analyze distribution designs, but the calculation techniques and some of the results can be applied to transmission lines as well. >

A method for obtaining transfer functions for power system stabilizer (PSS) design is presented. It results in an accurate reduced-order model of a generator connected to a large system. The method is based on Prony signal analysis and incorporates both local and interarea electromechanical oscillatory modes. Two PSS units are designed using the proposed method along with root-locus and sequential decentralized control techniques. The design procedure and results are illustrated using computer simulation of a 27 bus system with 16 generators. The test system was developed to exhibit interarea modes that are comparable to the western North American power system. This provides a new perspective on damping interarea oscillatory modes using PSS units. >

Three-phase reclosers with single-phase tripping are being more widely employed on utility distribution feeders. They have the advantages of increasing reliability for the customers on the feeders, because the majority of distribution feeder faults are single-phase. However, certain difficulties may occur when three-phase industrial and commercial loads, with three-phase transformers and motor loads are interrupted on one phase only. These include overheating of motors, ferroresonance of transformers, tripping of power electronic drives and backfeeding of distribution lines. This paper examines these effects, their severity and mitigation techniques.

Basic recommendations are provided for enhancing the reliability of new and existing capacitor controller installations by optimizing their resistance to surges. The sources and types of surges commonly experienced by capacitor controllers are discussed with consideration given to circuit parameters that can influence the magnitude of surges reaching the controller. Different controller surge protection configurations are examined including providing auxiliary secondary-side surge protection. The discussion culminates with some basic surge protection recommendations to enhance the lightning performance of capacitor controllers

This paper considers several approaches to use overcurrent protection to improve distribution reliability. Having more protective devices that are properly coordinated assures that the fewest customers possible are interrupted and makes fault-finding easier. Fusing is an inexpensive way to isolate faults, and this paper describes some often-overlooked fusing considerations. Applications of fuse blowing and fuse saving are also considered. Reclosing practices can be optimized to limit impacts on customers. In addition, single-phase protective devices can also limit customer outage time by reducing the number of customers affected

Historically, electrical distribution systems have been designed to deliver power from the distribution substations to the end users. The future evolution of the distribution system, as envisioned in EPRI's Advanced Distribution Automation (ADA) initiative, will merge power delivery with distributed intelligent control and communications systems. This paper covers the results of a survey of electric utilities and a follow-up seminar with survey participants. Survey responses were obtained from eleven US and Canadian utilities across a wide spectrum. Responders included large multi-state utilities, companies in large urban areas, publicly and privately owned utilities, and a small rural utility. These results will help to shape future planning as to the most effective pathways to move forward in the evolution of the distribution system of the future

Rpad is an open-source, interactive, Web-based analysis program. Rpad is an analysis package, a Web-page designer, and a GUI designer all wrapped in one. This paper covers the architecture of Rpad and how it draws on a number of open-source technologies. Demonstrations show how such technology has been used for data mining of outage databases. Open-source software is very good for piecing together technologies.

A description is given of a method of switched-capacitor control for damping of low-frequency power system oscillations. A form of self-tuning generalized predictive control with discrete constraints placed on the control signal is used. The control is calculated on the basis only of local network information. Simulation results are shown for a nonlinear model for a nine-bus power system. >

Momentary interruptions caused by lightning were a severe problem for a Champion International paper mill in upstate New York. Following a detailed investigation, the Niagara Mohawk Power Corporation (NMPC) installed line surge arresters to the supply circuit to eliminate lightning-caused interruptions. The paper mill is supplied by a 115-kV circuit which was characterized by low insulation level and poor grounding. Lightning arresters were found to be the best solution. Extensive line performance monitoring of the circuit performed before and after the arresters were installed has provided some valuable information. The arresters have successfully eliminated the lightning-caused interruptions.