Spelling suggestions: "subject:"electric power plants - load"" "subject:"electric power plants - road""
11 |
Enhanced voltage regulation in lightly-loaded, meshed distribution networks using a phase shifting transformerSithole, Frederick Silence 03 June 2013 (has links)
M.Ing. (Electrical and Electronic Engineering) / Long transmission lines in power system require high line loading in order to lower voltage limits due to line losses. For relatively long lines, line charging is high and thus higher voltage limits reached at low loading. It follows then that it is a challenge to maintaining the voltages between the acceptable limits for relatively long lines. This dissertation highlights the problems experienced when load varying from very low to very high is supplied by very long parallel lines of different impedance characteristic. When the load is extremely high, there are low voltages experienced which are solved by use of shunt capacitors and/or adding more lines. When the load is extremely low, there are high voltages experienced which are solved by use of shunt reactors and/or switching some of the lines off. The type of solutions to this two loading extremes as indicated above, can be problematic, in that; new lines requires servitudes which can take too long, shunt capacitors and reactors in this type of the network is not desirable since the introduction of too many of these devices have maintenance implications and they would require continuous switching to maintain acceptable voltages, resulting in complicated operation of the network. This research proposes the use of a phase shifting transformer located on one of two parallel corridors supplying power to a load located remotely from the rest of the system. The transformer is able to rearrange the active power flows to vary loadings of the corridors and the improvements in voltage regulation can be realised during both low and high load conditions.
|
12 |
Short term load forecasting by means of neural networks and programmable logic devices for new high electrical energy usersManuel, Grant 09 April 2014 (has links)
D.Phil. (Electrical and Electronic Engineering) / Load forecasting is a necessary and an important task for both the electrical consumer and electrical supplier. Whilst many studies emphasize the importance of determining the future demand, few papers address both the forecasting algorithm and computational resources needed to offer a turnkey solution to address the load forecasting problem. The major contribution that, this paper identified is a turnkey load forecasting algorithm. A turnkey forecasting solution is defined by a comprehensive solution that incorporates both the algorithm and processing elements needed to execute the algorithm in the most effective and efficient manner. An electrical consumer, namely the operator of a rapid railway system was faced with a problem of having to forecast the notified network demand and energy consumption. The forecast period was expected to be between a very short term window for maintenance reasons and long term for the requirements warranted by the electrical supplier. The problem was addressed by firstly reviewing the most common forms of load forecasting for which there are two types. These are statistically based methods and methods based upon artificial intelligence. The basic principle of a statistical approach is to approximate or define a curve that best defines the relationship between the load and its parameters. Regression and similar day approach methods use the defined correlation of past values in order to forecast the future behaviour. In other words the future load forecast is forecasted by observing the behaviour of the factors that influenced the load behaviour in the past. The underlying factors that influence the final load may be identified by means of a top down drill down approach. In this way both the load factors and influential variables may be identified. This paper makes use of relevance trees to create a structure of load and influential variables. For a regression forecasting model, the behaviour of the load is modelled according to weather and non-weather variables. The load may be stochastic or deterministic, linear or nonlinear. One of the biggest problems with statistical models is the lack of generality. One model may yield more acceptable results over another model simply because of the sensitivity of the model to one load element that defines the model significantly. Regression type forecast models are an example of this where the elements that define the load are broadly divided into weather and non-weather elements. It is important that the correlation curve reflects the true correlation between the load and its elements. The recursive properties of a statistical based techniques (Kalman filter) allows that the relationship be refined. For methods such as neural networks, the relationship between the load elements that define the future load behaviour is learnt by presenting a series of patterns and then a forecast model is derived. Rigorous mathematical equations are replaced with an artificial neural network where the load curve is learnt. Unlike a statistical based approach (ARMA models), the load does not first need to be defined as a stochastic or deterministic series. In terms of a stochastic approach (non stationery process), the load first would have to be brought to a stationery process. For artificial neural networks, such processes are eliminated and the future forecast is derived faster in terms of a turnkey approach (tested solution). Artificial Neural Networks (ANN) has gained momentum since the eighties. Specifically in the area of forecasting, neural networks have become a common application. In this thesis, data from a railway operator was used to train the neural network and then future data is forecasted. Two embedded processing elements were then evaluated in terms of speed, memory and ability to execute complex mathematical functions (libraries). These were namely a Complex Programmable Logic Device (CPLD) and microcontroller (MCU). The ANN forecasting algorithm was programmed on both a MCU and PLD and compared by means of timing models and hardware platform testing. The most ideal turnkey solution was found to be the ANN algorithm residing on a PLD. The accuracy and speed results surpassed that of a MCU.
|
13 |
Elements of load forecasting and generation planningGithinji, John N. January 1983 (has links)
M.S.
|
14 |
Symbolic and connectionist machine learning techniques for short-term electric load forecastingRajagopalan, Jayendar 22 August 2009 (has links)
This work applies connectionist neural network learning techniques and symbolic machine learning techniques to the problem of short-term electric load forecasting. The short-term electric load forecasting problem considered here is the prediction of bus loads one day ahead. The forecast quantities of interest are average integrated daily load and daily peak load.
The primary objectives of this work are two-fold: to determine the forces driving the load demand and produce a human intelligible model, and use of this model to forecast load for new, unseen scenarios.
In the first part of this work, connectionist techniques for modeling bus load is presented. Critical design issues for neural network modeling and implementation such as neural network architecture, training database creation, training dataset selection, training data normalization are presented in context of nonlinear modeling in general and electric load forecasting in particular. Local function approximation and nearest neighbor norms techniques are applied to this task. Simulations are performed for forecast of average bus loads of the town of Blacksburg, Virginia, U.S.A; the connectionist model is able to forecast integrated average daily load with an accuracy of about 2.5%. Connectionist neural network knowledge acquisition algorithms are however, not mature enough, presently, to handle complex real world problems such as knowledge extraction from large databases. Presence of symbolic along with numeric data in input and output poses problems for data pre-processing for neural network training. Only at the time of completion of this thesis are researchers discussing the possibility of using special techniques to present symbolic data for neural networks. Also, multilayer feedforward networks trained by the backpropagation algorithm perform poorly in forecasting chaotic patterns such as those encountered in peak load demand.
Symbolic machine learning techniques are powerful concept acquisition techniques that extract underlying knowledge from large databases. They are sufficiently powerful to accept symbolic and numeric data. Inductive learning algorithms employing a statistical 72 test as the splitting criterion are applied to extract load dependency information. The extracted patterns are expressed as graphic decision trees and equivalent human intelligible high level language if-then rules. Implementation details of the statistical decision algorithm are discussed and simulations are performed to construct decision trees. Using this model, new cases are forecast. This algorithm is capable of forecasting holiday and weekend loads too. The proposed algorithm is robust enough to handle raw, unprocessed databases which contain missing data. The peak load forecasting problem is solved using a simple methodology that combines the robustness of decision trees and the numerical accuracy of connectionist models.
The two paradigms, connectionist and symbolic learning techniques are compared from a knowledge acquisition and forecasting perspective and directions for further work suggested. / Master of Science
|
15 |
Real time data acquisition for load managementGhosh, Sushmita 15 November 2013 (has links)
Demand for Data Transfer between computers has increased ever since the introduction of Personal Computers (PC). Data Communicating on the Personal Computer is much more productive as it is an intelligent terminal that can connect to various hosts on the same I/O hardware circuit as well as execute processes on its own as an isolated system.
Yet, the PC on its own is useless for data communication. It requires a hardware interface circuit and software for controlling the handshaking signals and setting up communication parameters. Often the data is distorted due to noise in the line. Such transmission errors are imbedded in the data and require careful filtering.
The thesis deals with the development of a Data Acquisition system that collects real time load and weather data and stores them as historical database for use in a load forecast algorithm in a load management system. A filtering technique has been developed here that checks for transmission errors in the raw data. The microcomputers used in this development are the IBM PC/XT and the AT&T 3B2 supermicro computer. / Master of Science
|
16 |
Knowledge-based and statistical load forecast model development and analysisMoghram, Ibrahim Said January 1989 (has links)
Most of the techniques that have been applied to the short-term load forecasting problem fall within the time series approaches. The exception to this has been a new approach based on the application of expert systems. Recently several techniques have been reported which apply the rule-based (or expert systems) approach to the short-term load forecasting problem. However, the maximum lead time used for these forecasts has not gone beyond 48 hours, even though there is a significant difference between these algorithms in terms of their data base requirements (few weeks to 10 years).
The work reported in this dissertation deals with two aspects. The first one is the application of rule-based techniques to weekly load forecast. A rule-based technique is presented that is capable of issuing a 168-hour lead-time load forecast. The second aspect is the development of a comprehensive load forecasting system that utilizes both the statistical and rule-based approaches. This integration overcomes the deficiencies that exist in both of these modeling techniques.
The load forecasting technique is developed using two parallel approaches. In the first approach expert information is used to identify weather variables, day types and diurnal effects that influence the electrical utility load. These parameters and hourly historical loads are then selectively used for various statistical techniques (e.g., univariate, transfer function and linear regression). A weighted average load forecast is then produced which judiciously combines the forecasts from these three techniques. The second approach, however, is free of any significant statistical computation, and is based totally on rules derived from electric utility experts. The data base requirement for any of these approaches do not extend more than four weeks ol hourly load, dry bulb and dew point temperatures. When the algorithms are applied to generate seven-day ahead load forecasts for summer (August) and winter (February) the average forecast errors for the month come under 3%. / Ph. D.
|
17 |
Optimal allocation of reactive power to mitigate fault delayed voltage recoveryMadan, Sandhya 09 July 2010 (has links)
The Masters Thesis research focuses on reactive power and voltage control during and following major power system disturbances such as faults and subsequent loss of transmission line(s) or generator(s), voltage recovery phenomena following successful fault clearing, dynamic swings of power systems and local voltage suppression, etc. During these events, load and other system dynamics may cause reactive power deficiencies and system voltage issues such as delayed voltage recovery. These phenomena may lead to secondary events such as tripping of loads and/or circuits. Dynamic VAr sources such as generators, static VAr compensators (SVCs), STATCOMs etc and to a lesser degree static VAr sources such as capacitor or reactor banks, can help the system recover from these contingencies by providing fast modulation of the reactive power. Because of the higher cost of dynamic VAr resources, it is important to optimize the deployment of these devices by minimizing the total installed capacity of dynamic VAR resources while meeting the technical requirement and achieving the necessary performance of the system. We refer to this problem as the optimal allocation of dynamic VAR sources (OAODVARS). OAODVARS has been addressed with traditional analytic methods as well as with Artificial Intelligence methods such as genetic algorithms and Tabu search using mostly power flow type models. Both type of methods, as reported in the literature, have not provided satisfactory solutions because they ignore system dynamics and especially load dynamics, in other words they are based on power flow type models. In addition the AI methods have been proved to be extremely inefficient. We propose a new approach that has the following two advantages: (a) it is based on a realistic model that captures system dynamics and (b) it is based on the efficient successive approximation dynamic programming. The solution is provided as a sequence of planning decisions over the planning horizon. The proposed method will be demonstrated on the IEEE 24-bus reliability test system.
|
18 |
Probabilistic low voltage distribution network design for aggregated light industrial loadsVan Rhyn, Pierre 25 February 2015 (has links)
D.Ing. / This thesis initially reviews current empirical and probabilistic electrical load models available to distribution design engineers today to calculate voltage regulation levels in low voltage residential, commercial and light industrial consumer networks. Although both empirical and probabilistic techniques have extensively been used for residential consumers in recent years, it has been concluded that commercial and light industrial consumer loads have not been a focus area of probabilistic load study for purposes of low voltage feeder design. However, traditional empirical techniques, which include adjustments for diversity to accommodate non-coincidental electrical loading conditions, have generally been found to be applied using in-house design directives with only a few international publications attempting to address the problem. This work defines the light industrial group of consumers in accordance with its international Standard Industrial Classification (SIC) and presents case studies on a small group of three different types of light industrial sub-classes, It is proposed and proved that the electrical load models can satisfactorily be described as beta-distributed load current models at the instant of group or individual maximum power demand on typical characteristic 24-hour load cycles. Characteristic mean load profiles were obtained by recording repetitive daily loading of different sub-classes, ensuring adequate sample size at all times. Probabilistic modelling of light industrial loads using beta-distributed load current at maximum demand is a new innovation in the modelling of light industrial loads. This work is further -complemented by the development of a new probabilistic summation algorithm in spreadsheet format. This algorithm adds any selected number of characteristic load current profiles, adjusted for scale, power factor, and load current imbalance, and identifies the combined instant of group or system maximum demand. This spreadsheet also calculates the characteristic beta pdf parameters per phase describing the spread and profile of the combined system loading at maximum demand. These parameters are then conveniently used as input values to existing probabilistic voltage regulation algorithms to calculate voltage regulation in single-, bi- and three-phase low voltage distribution networks.
|
19 |
Area COI-based slow frequency dynamics modeling, analysis and emergency control for interconnected power systemsDu, Zhaobin, 杜兆斌 January 2008 (has links)
published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy
|
20 |
Developing a neural network model to predict the electrical load demand in the Mangaung municipal areaNigrini, Lucas Bernardo January 2012 (has links)
Thesis (D. Tech. (Engineering: Electric)) -- Central University of technology, 2012 / Because power generation relies heavily on electricity demand, consumers are required to wisely manage their loads to consolidate the power utility‟s optimal power generation efforts. Consequently, accurate and reliable electric load forecasting systems are required.
Prior to the present situation, there were various forecasting models developed primarily for electric load forecasting. Modelling short term load forecasting using artificial neural networks has recently been proposed by researchers.
This project developed a model for short term load forecasting using a neural network. The concept was tested by evaluating the forecasting potential of the basic feedforward and the cascade forward neural network models.
The test results showed that the cascade forward model is more efficient for this forecasting investigation. The final model is intended to be a basis for a real forecasting application. The neural model was tested using actual load data of the Bloemfontein reticulation network to predict its load for half an hour in advance.
The cascade forward network demonstrates a mean absolute percentage error of less than 5% when tested using four years of utility data. In addition to reporting the summary statistics of the mean absolute percentage error, an alternate method using correlation coefficients for presenting load forecasting performance results are shown.
This research proposes that a 6:1:1 cascade forward neural network can be trained with data from a month of a year and forecast the load for the same month of the following year. This research presents a new time series modeling for short term load forecasting, which can model the forecast of the half-hourly loads of weekdays, as well as of weekends and public holidays. Obtained results from extensive testing on the Bloemfontein power system network confirm the validity of the developed forecasting approach. This model can be implemented for on-line testing application to adopt a final view of its usefulness.
|
Page generated in 0.0619 seconds