Knowledge extraction from medium voltage load diagrams to support the definition of electrical tariffs

With the electricity market liberalization, distribution and retail companies are looking for better market strategies based on adequate information upon the consumption patterns of its electricity customers. In this environment all consumers are free to choose their electricity supplier. A fair insight on the customer´s behaviour will permit the definition of specific contract aspects based on the different consumption patterns. In this paper Data Mining (DM) techniques are applied to electricity consumption data from a utility client’s database. To form the different customer´s classes, and find a set of representative consumption patterns, we have used the Two-Step algorithm which is a hierarchical clustering algorithm. Each consumer class will be represented by its load profile resulting from the clustering operation. Next, to characterize each consumer class a classification model will be constructed with the C5.0 classification algorithm.

Symbolic Knowledge Extraction from Trained Neural Networks Governed by Lukasiewicz Logics

This work describes a methodology to extract symbolic rules from trained neural networks. In our approach, patterns on the network are codified using formulas on a Lukasiewicz logic. For this we take advantage of the fact that every connective in this multi-valued logic can be evaluated by a neuron in an artificial network having, by activation function the identity truncated to zero and one. This fact simplifies symbolic rule extraction and allows the easy injection of formulas into a network architecture. We trained this type of neural network using a back-propagation algorithm based on Levenderg-Marquardt algorithm, where in each learning iteration, we restricted the knowledge dissemination in the network structure. This makes the descriptive power of produced neural networks similar to the descriptive power of Lukasiewicz logic language, minimizing the information loss on the translation between connectionist and symbolic structures. To avoid redundance on the generated network, the method simplifies them in a pruning phase, using the "Optimal Brain Surgeon" algorithm. We tested this method on the task of finding the formula used on the generation of a given truth table. For real data tests, we selected the Mushrooms data set, available on the UCI Machine Learning Repository.

Histogram-based DNA analysis for the visualization of chromosome, genome and species information

We describe a novel approach to explore DNA nucleotide sequence data, aiming to produce high-level categorical and structural information about the underlying chromosomes, genomes and species. The article starts by analyzing chromosomal data through histograms using fixed length DNA sequences. After creating the DNA-related histograms, a correlation between pairs of histograms is computed, producing a global correlation matrix. These data are then used as input to several data processing methods for information extraction and tabular/graphical output generation. A set of 18 species is processed and the extensive results reveal that the proposed method is able to generate significant and diversified outputs, in good accordance with current scientific knowledge in domains such as genomics and phylogenetics.

A data-mining-based methodology for transmission expansion planning

In recent decades, all over the world, competition in the electric power sector has deeply changed the way this sector’s agents play their roles. In most countries, electric process deregulation was conducted in stages, beginning with the clients of higher voltage levels and with larger electricity consumption, and later extended to all electrical consumers. The sector liberalization and the operation of competitive electricity markets were expected to lower prices and improve quality of service, leading to greater consumer satisfaction. Transmission and distribution remain noncompetitive business areas, due to the large infrastructure investments required. However, the industry has yet to clearly establish the best business model for transmission in a competitive environment. After generation, the electricity needs to be delivered to the electrical system nodes where demand requires it, taking into consideration transmission constraints and electrical losses. If the amount of power flowing through a certain line is close to or surpasses the safety limits, then cheap but distant generation might have to be replaced by more expensive closer generation to reduce the exceeded power flows. In a congested area, the optimal price of electricity rises to the marginal cost of the local generation or to the level needed to ration demand to the amount of available electricity. Even without congestion, some power will be lost in the transmission system through heat dissipation, so prices reflect that it is more expensive to supply electricity at the far end of a heavily loaded line than close to an electric power generation. Locational marginal pricing (LMP), resulting from bidding competition, represents electrical and economical values at nodes or in areas that may provide economical indicator signals to the market agents. This article proposes a data-mining-based methodology that helps characterize zonal prices in real power transmission networks. To test our methodology, we used an LMP database from the California Independent System Operator for 2009 to identify economical zones. (CAISO is a nonprofit public benefit corporation charged with operating the majority of California’s high-voltage wholesale power grid.) To group the buses into typical classes that represent a set of buses with the approximate LMP value, we used two-step and k-means clustering algorithms. By analyzing the various LMP components, our goal was to extract knowledge to support the ISO in investment and network-expansion planning.

Bid definition method for electricity markets based on an adaptive multiagent system

Electricity markets are complex environments with very particular characteristics. MASCEM is a market simulator developed to allow deep studies of the interactions between the players that take part in the electricity market negotiations. This paper presents a new proposal for the definition of MASCEM players’ strategies to negotiate in the market. The proposed methodology is multiagent based, using reinforcement learning algorithms to provide players with the capabilities to perceive the changes in the environment, while adapting their bids formulation according to their needs, using a set of different techniques that are at their disposal. Each agent has the knowledge about a different method for defining a strategy for playing in the market, the main agent chooses the best among all those, and provides it to the market player that requests, to be used in the market. This paper also presents a methodology to manage the efficiency/effectiveness balance of this method, to guarantee that the degradation of the simulator processing times takes the correct measure.

Problem-Based Learning e suas implicações: breve revisão teórica

A ESTSP-IPP implementou em 2008-2009 um novo modelo pedagógico, o PBL, em três licenciaturas. Este modelo tem sido considerado capaz de promover a aquisição de conhecimentos mas também o desenvolvimento de competências transversais valorizadas no mercado de trabalho; orienta-se em torno de problemas significativos da realidade profissional, trabalhados segundo a metodologia dos sete passos, destacando-se a aprendizagem através de pesquisa individual e trabalho de grupo; e visa ainda desenvolver processos cognitivos e metacognitivos como levantar hipóteses, comparar, analisar, interpretar e avaliar. Neste artigo, caracterizamos brevemente o modelo e respectivas implicações, justificando o interesse em investigar as repercussões da sua implementação.

Power systems reliability calculation based on fuzzy data mining

This paper presents a methodology supported on the data base knowledge discovery process (KDD), in order to find out the failure probability of electrical equipments’, which belong to a real electrical high voltage network. Data Mining (DM) techniques are used to discover a set of outcome failure probability and, therefore, to extract knowledge concerning to the unavailability of the electrical equipments such us power transformers and high-voltages power lines. The framework includes several steps, following the analysis of the real data base, the pre-processing data, the application of DM algorithms, and finally, the interpretation of the discovered knowledge. To validate the proposed methodology, a case study which includes real databases is used. This data have a heavy uncertainty due to climate conditions for this reason it was used fuzzy logic to determine the set of the electrical components failure probabilities in order to reestablish the service. The results reflect an interesting potential of this approach and encourage further research on the topic.

Ontologies for reusing data cleaning knowledge

The emergence of new business models, namely, the establishment of partnerships between organizations, the chance that companies have of adding existing data on the web, especially in the semantic web, to their information, led to the emphasis on some problems existing in databases, particularly related to data quality. Poor data can result in loss of competitiveness of the organizations holding these data, and may even lead to their disappearance, since many of their decision-making processes are based on these data. For this reason, data cleaning is essential. Current approaches to solve these problems are closely linked to database schemas and specific domains. In order that data cleaning can be used in different repositories, it is necessary for computer systems to understand these data, i.e., an associated semantic is needed. The solution presented in this paper includes the use of ontologies: (i) for the specification of data cleaning operations and, (ii) as a way of solving the semantic heterogeneity problems of data stored in different sources. With data cleaning operations defined at a conceptual level and existing mappings between domain ontologies and an ontology that results from a database, they may be instantiated and proposed to the expert/specialist to be executed over that database, thus enabling their interoperability.

Modeling group decision meeting participants with an Agent-based approach

This paper aims to present a multi-agent model for a simulation, whose goal is to help one specific participant of multi-criteria group decision making process.This model has five main intervenient types: the human participant, who is using the simulation and argumentation support system; the participant agents, one associated to the human participant and the others simulating the others human members of the decision meeting group; the directory agent; the proposal agents, representing the different alternatives for a decision (the alternatives are evaluated based on criteria); and the voting agent responsiblefor all voting machanisms.At this stage it is proposed a two phse algorithm. In the first phase each participantagent makes his own evaluation of the proposals under discussion, and the voting agent proposes a simulation of a voting process.In the second phase, after the dissemination of the voting results,each one ofthe partcipan agents will argue to convince the others to choose one of the possible alternatives. The arguments used to convince a specific participant are dependent on agent knowledge about that participant. This two-phase algorithm is applied iteratively.

A model updating technique based on FRFs for damped structures

Model updating methods often neglect that in fact all physical structures are damped. Such simplification relies on the structural modelling approach, although it compromises the accuracy of the predictions of the structural dynamic behaviour. In the present work, the authors address the problem of finite element (FE) model updating based on measured frequency response functions (FRFs), considering damping. The proposed procedure is based upon the complex experimental data, which contains information related to the damped FE model parameters and presents the advantage of requiring no prior knowledge about the damping matrix structure or its content, only demanding the definition of the damping type. Numerical simulations are performed in order to establish the applicability of the proposed damped FE model updating technique and its results are discussed in terms of the correlation between the simulated experimental complex FRFs and the ones obtained from the updated FE model.

Factors influencing innovation decision making in Portuguese firms

There is an undeniable positive effect of innovation for both firms and the economy, with particular regards to the financial performance of firms. However, there is an important role of the decision making process for the allocation of resources to finance the innovation process. The aim of this paper is to understand what factors explain the decision making process in innovation activities of Portuguese firms. This is an empirical study, based on the modern theoretical approaches, which has relied on five key aspects for innovation: barriers, sources, cooperation, funding; and the decision making process. Primary data was collected through surveys to firms that have applied for innovation programmes within the Portuguese innovation agency. Univariate and multivariate statistical techniques were used. Our results suggest that the factors that mostly influence the Portuguese firms’ innovation decision-making processes are economical and financial (namely those related to profit increase and labour costs reduction).

A exportação do mel português: um estudo exploratório sobre as motivações, barreiras e estratégias

Dissertação de Mestrado apresentado ao Instituto de Contabilidade e Administração do Porto para a obtenção do grau de Mestre em Empreendedorismo e Internacionalização, sob orientação de Professor Doutor António Correia de Barros

Prevalence of alcohol abuse and associated factors in a population-based study

OBJECTIVE: To estimate the prevalence of alcohol abuse/dependence and identify associated factors among demographic, family, socioeconomic and mental health variables. METHODS: A household survey was carried out in the urban area of Campinas, southeastern Brazil, in 2003. A total of 515 subjects, aged 14 years or more were randomly selected using a stratified cluster sample. The Self-Report Questionnaire and the Alcohol Use Disorder Identification Test were used in the interview. Prevalences were calculated, and univariate and multivariate logistic analyses performed by estimating odds ratios and 95% confidence intervals. RESULTS: The estimated prevalence of alcohol abuse/dependence was 13.1% (95% CI: 8.4;19.9) in men and 4.1% (95% CI: 1.9;8.6) in women. In the final multiple logistic regression model, alcohol abuse/dependence was significantly associated with age, income, schooling, religion and illicit drug use. The adjusted odds ratios were significantly higher in following variables: income between 2,501 and 10,000 dollars (OR=10.29); income above 10,000 dollars (OR=10.20); less than 12 years of schooling (OR=13.42); no religion (OR=9.16) or religion other than Evangelical (OR=4.77); and illicit drug use during lifetime (OR=4.47). Alcohol abuse and dependence patterns were different according to age group. CONCLUSIONS: There is a significantly high prevalence of alcohol abuse/dependence in this population. The knowledge of factors associated with alcohol abuse, and differences in consumption patterns should be taken into account in the development of harm reduction strategies.

Health-related behavior and quality of life among the elderly: a population-based study

OBJECTIVE: To assess the association between health-related behaviors and quality of life among the elderly. METHODS: A population-based cross-sectional study was carried out including 1,958 elderly living in four areas in the state of São Paulo, southeastern Brazil, 2001/2002. Quality of life was assessed using the Medical Outcomes Study SF-36-Item Short Form Health Survey instrument. This instrument's eight subscales and two components were the dependent variables. Independent variables were physical activity, weekly frequency of alcohol consumption and smoking. Multiple linear regression models were used to control for the effect of gender, age, schooling, work, area of residence and number of chronic conditions. RESULTS: Physical activity was positively associated with the eight SF-36 subscales. The stronger associations were found for role-physical (β=11.9), physical functioning (β=11.3) and physical component. Elderly individuals who consumed alcohol at least once a week showed a better quality of life than those did not consume alcohol. Compared to non-smokers, smokers had a poorer quality of life for the mental component (β=-2.4). CONCLUSIONS: The study results showed that physical activity, moderate alcohol consumption and no smoking are positively associated with a better quality of life in the elderly.

Traditional knowledge for promotion of socioeconomic inclusion of local communities

This paper discusses the key role played by public research institutes for promoting socioeconomic inclusion of local communities based on traditional knowledge and traditional medicine. Nongovernmental organizations and cooperatives have had an important role in raising financial resources, being involved with advocacy of local communities and advancing legislation changes. But strict best manufacturing practices regulations imposed by the Brazilian National Health Surveillance Agency on the requirements for approval and commercialization of drugs based on herbal medicine products call for the involvement of strong public research institutes capable of supporting community-based pharmacies. Thus, public research institutes are pivotal as they can conduct scientific research studies to evidence the efficacy of herbal medicine products and help building the capacity of local communities to comply with current regulations.