Search

“Propagation, which looks at spreading in complex networks, can be seen from many viewpoints; it is undesirable, or desirable, controllable, the mechanisms generating that propagation can be the topic of interest, but in the end all depends on the setting. This book covers leading research on a wide spectrum of propagation phenomenon and the techniques currently used in its modelling, prediction, analysis and control. Fourteen papers range over topics including epidemic models, models for trust inference, coverage strategies for networks, vehicle flow propagation, bio-inspired routing algorithms, P2P botnet attacks and defences, fault propagation in gene-cellular networks, malware propagation for mobile networks, information propagation in crisis situations, financial contagion in interbank networks, and finally how to maximize the spread of influence in social networks. The compendium will be of interest to researchers, those working in social networking, communications and finance and is aimed at providing a base point for further studies on current research. Above all, by bringing together research from such diverse fields, the book seeks to cross-pollinate ideas, and give the reader a glimpse of the breath of current research.”.

Weather information is an important factor in load forecasting models. Typically, load forecasting models are constructed and tested using actual weather readings. However, online operation of load forecasting models requires the use of weather forecasts, with associated weather forecast errors. These weather forecast errors inevitably lead to a degradation in model performance. This is an important factor in load forecasting but has not been widely examined in the literature. The main aim of this paper is to present a novel technique for minimizing the consequences of this degradation. In addition, a supplementary technique is proposed to model weather forecast errors to reflect current accuracy. The proposed technique utilizes a combination of forecasts from several load forecasting models (sub-models). The parameter estimation may thus be split into two parts: sub-model and combination parameter estimation. It is shown that the lowest PMSE corresponds to training the sub-models with actual weather but training the combiner with forecast weather.

This paper is concerned with the case of an exogenous system in which a model is required to forecast a periodic output time series using a causal input. A novel approach is developed in which the wavelet packet transform is taken of both the dependent time series and causal input. This results in two sets of basis dictionaries and requires two bases to be chosen. It is proposed that the best bases to choose are those which maximize the mutual information. Input selection is then implemented by eliminating those coefficients of the selected input basis with low mutual information. As an example, a model is constructed to forecast short-term electrical demand.

In this paper, an examination of the effect of including rainfall inputs in the forecasting of daily vehicular traffic volumes is undertaken. A case study is carried out at a busy intersection in Dublin city to examine if any reduction in forecasting error can be obtained by the incorporation of rainfall inputs. This paper also demonstrates the value of incorporating lessons learned from linear time series modelling to the non-linear analysis undertaken.

In a competitive electricity market environment, power producers and consumers need, on one hand, accurate load and/or electricity consumption forecasting tools. These tools will ensure an a-priori knowledge on the amount of energy needed for production. On the other hand, forecasting electricity prices, may play a very important role for producers and consumers when planning bidding strategies, in order to maximize their benefits and utilities, respectively. This paper will address both issues in a newly deregulated Algerian Market, with an analysis of the electricity national demand. An example of electricity demand model will be presented, as well as an attempt to introduce price modelling taking into account that there is only one electricity supplier on the market. Possibilities of price forecasting in a deregulated market may be projected

Electric utilities require short-term forecasts of electricity demand (load) in order to schedule generating plant up to several days ahead on an hourly basis. Errors in the forecasts may lead to generation plant operation that is not required or sub-optimal scheduling of generation plants. In addition, with the introduction of the Electricity Regulation Act 1999, a deregulated market structure has been introduced, adding increased impetus to reducing forecast error and the associated costs. This thesis presents a strategy for reducing costs from electrical demand forecast error using models designed specifically for the Irish system. The differences in short-term load forecasting models are examined under three independent categories: how the data is segmented prior to modelling, the modelling technique and the approach taken to minimise the effect of weather forecast errors present in weather inputs to the load forecasting models. A novel approach is presented to determine whether the data should be segmented by hour of the day prior to modelling. Several segmentation strategies are analysed and the one appropriate for Irish data identified. Furthermore, both linear and nonlinear techniques are compared with a view to evaluating the optimal model type. The effect of weather forecast errors on load forecasting models, though significant, has largely been ignored in the literature. Thus, the underlying issues are examined and a novel method is presented which minimises the effect of weather forecast errors.

Electrical demand is driven by economic and human activity, which has obvious daily, weekly and yearly cycles as well as a long-term trend and special periods such as bank holidays, Christmas etc., all of which are reflected in load data. These characteristics of electrical demand must inevitably be incorporated into any demand-forecasting model. However, with the exception of a few papers, the vast bulk of the literature on electrical demand forecasting is concerned with forecasting techniques. This paper proposes several methods with which to quantify the characteristics of Irish electrical load data prior to modelling.

Weather inputs are an important factor in load forecasting. The arrival of a weather front or introduction of weather forecast errors lead to a change in the significance of weather as input. A data fusion of individual models is introduced, which uses a priori information regarding the significance of weather inputs in these situations, to produce a superior forecast to any individual model.

The paper investigates whether a time series or a set of independent points is a more appropriate description of 24-hour Irish electrical load data. A set of independent points means that the load at each hour of the day is independent from the load at any other hour. The data is first split into 24 series, one for each hour of the day i.e. a 1am 2am 3am series etc. These are called parallel series. The linear cross-correlation's of the parallel series are used to indicate independence. While the loads at 9am and 6pm to 8pm appear independent the remaining loads are highly inter-correlated. This suggests that 24-hour electrical load data has a dual nature. Two techniques are used to test this hypothesis. The first technique models each parallel series using neural networks. This technique is found to be computationally expensive. The second technique uses a hybrid technique called the Multi Time Scale (MTS) technique. This models 24-hour electrical load data as a time series that can be adjusted by 5 parallel forecasts and a daily cumulative model. The results show that the MTS forecasts are superior to the parallel forecasts except for 9am and 6pm to 8pm. A composite model using neural networks for 9am and 6pm to 8pm and the MTS model elsewhere takes advantage of the dual nature of the data reducing error and computational expense.

This paper presents a comparison between parallel linear and parallel neural network models. Parallel models consist of 24 separate models, one for each hour of the day. Each parallel model decomposes the load into a linear Auto- Regressive (AR) part and a residual. Exogenous linear and neural network model performance is compared in predicting this residual. Three days or 72 hours of current and delayed weather variables are available as exogenous inputs for the residual models. Input selection comprises of testing the bootstrapped performance of a linear model. The inputs are ordered using 4 methods derived from a mix of the T-ratio of the linear coefficients and Principal Component Analysis (PCA). The neural network models are found to give superior results due to the non-linear AR nature of the residual

This paper presents a comparison between parallel linear and parallel neural network models. Parallel models consist of 24 separate models, one for each hour of the day. Each parallel model decomposes the load into a linear Auto- Regressive (AR) part and a residual. Exogenous linear and neural network model performance is compared in predicting this residual. Three days or 72 hours of current and delayed weather variables are available as exogenous inputs for the residual models. Input selection comprises of testing the bootstrapped performance of a linear model. The inputs are ordered using 4 methods derived from a mix of the T-ratio of the linear coefficients and Principal Component Analysis (PCA). The neural network models are found to give superior results due to the non-linear AR nature of the residual.