Meta-heuristics tunning using CBR for dynamic scheduling

This paper addresses the problem of Biological Inspired Optimization Techniques (BIT) parameterization, considering the importance of this issue in the design of BIT especially when considering real world situations, subject to external perturbations. A learning module with the objective to permit a Multi-Agent Scheduling System to automatically select a Meta-heuristic and its parameterization to use in the optimization process is proposed. For the learning process, Casebased Reasoning was used, allowing the system to learn from experience, in the resolution of similar problems. Analyzing the obtained results we conclude about the advantages of its use.

Optimization techniques for power distribution planning with uncertainties: a comparative study

Important research effort has been devoted to the topic of optimal planning of distribution systems. The non linear nature of the system, the need to consider a large number of scenarios and the increasing necessity to deal with uncertainties make optimal planning in distribution systems a difficult task. Heuristic techniques approaches have been proposed to deal with these issues, overcoming some of the inherent difficulties of classic methodologies. This paper considers several methodologies used to address planning problems of electrical power distribution networks, namely mixedinteger linear programming (MILP), ant colony algorithms (AC), genetic algorithms (GA), tabu search (TS), branch exchange (BE), simulated annealing (SA) and the Bender´s decomposition deterministic non-linear optimization technique (BD). Adequacy of theses techniques to deal with uncertainties is discussed. The behaviour of each optimization technique is compared from the point of view of the obtained solution and of the methodology performance. The paper presents results of the application of these optimization techniques to a real case of a 10-kV electrical distribution system with 201 nodes that feeds an urban area.

Self-optimizing through CBR learning

In this paper, we foresee the use of Multi-Agent Systems for supporting dynamic and distributed scheduling in Manufacturing Systems. We also envisage the use of Autonomic properties in order to reduce the complexity of managing systems and human interference. By combining Multi-Agent Systems, Autonomic Computing, and Nature Inspired Techniques we propose an approach for the resolution of dynamic scheduling problem, with Case-based Reasoning Learning capabilities. The objective is to permit a system to be able to automatically adopt/select a Meta-heuristic and respective parameterization considering scheduling characteristics. From the comparison of the obtained results with previous results, we conclude about the benefits of its use.

Aspectos de aprendizagem em optimização

A optimização e a aprendizagem em Sistemas Multi-Agente são consideradas duas áreas promissoras mas relativamente pouco exploradas. A optimização nestes ambientes deve ser capaz de lidar com o dinamismo. Os agentes podem alterar o seu comportamento baseando-se em aprendizagem recente ou em objectivos de optimização. As estratégias de aprendizagem podem melhorar o desempenho do sistema, dotando os agentes da capacidade de aprender, por exemplo, qual a técnica de optimização é mais adequada para a resolução de uma classe particular de problemas, ou qual a parametrização é mais adequada em determinado cenário. Nesta dissertação são estudadas algumas técnicas de resolução de problemas de Optimização Combinatória, sobretudo as Meta-heurísticas, e é efectuada uma revisão do estado da arte de Aprendizagem em Sistemas Multi-Agente. É também proposto um módulo de aprendizagem para a resolução de novos problemas de escalonamento, com base em experiência anterior. O módulo de Auto-Optimização desenvolvido, inspirado na Computação Autónoma, permite ao sistema a selecção automática da Meta-heurística a usar no processo de optimização, assim como a respectiva parametrização. Para tal, recorreu-se à utilização de Raciocínio baseado em Casos de modo que o sistema resultante seja capaz de aprender com a experiência adquirida na resolução de problemas similares. Dos resultados obtidos é possível concluir da vantagem da sua utilização e respectiva capacidade de adaptação a novos e eventuais cenários.

Multilayer Perceptron Neural Networks Training Through Charged System Search and its Application for Non-Technical Losses Detection

The non-technical loss is not a problem with trivial solution or regional character and its minimization represents the guarantee of investments in product quality and maintenance of power systems, introduced by a competitive environment after the period of privatization in the national scene. In this paper, we show how to improve the training phase of a neural network-based classifier using a recently proposed meta-heuristic technique called Charged System Search, which is based on the interactions between electrically charged particles. The experiments were carried out in the context of non-technical loss in power distribution systems in a dataset obtained from a Brazilian electrical power company, and have demonstrated the robustness of the proposed technique against with several others natureinspired optimization techniques for training neural networks. Thus, it is possible to improve some applications on Smart Grids.

The Salmon Algorithm - A New Population Based Search Metaheuristic

This thesis introduces the Salmon Algorithm, a search meta-heuristic which can be used for a variety of combinatorial optimization problems. This algorithm is loosely based on the path finding behaviour of salmon swimming upstream to spawn. There are a number of tunable parameters in the algorithm, so experiments were conducted to find the optimum parameter settings for different search spaces. The algorithm was tested on one instance of the Traveling Salesman Problem and found to have superior performance to an Ant Colony Algorithm and a Genetic Algorithm. It was then tested on three coding theory problems - optimal edit codes, optimal Hamming distance codes, and optimal covering codes. The algorithm produced improvements on the best known values for five of six of the test cases using edit codes. It matched the best known results on four out of seven of the Hamming codes as well as three out of three of the covering codes. The results suggest the Salmon Algorithm is competitive with established guided random search techniques, and may be superior in some search spaces.

Abordagem de construção de arquitetura homogênea para comitês via meta-aprendizagem

In the world we are constantly performing everyday actions. Two of these actions are frequent and of great importance: classify (sort by classes) and take decision. When we encounter problems with a relatively high degree of complexity, we tend to seek other opinions, usually from people who have some knowledge or even to the extent possible, are experts in the problem domain in question in order to help us in the decision-making process. Both the classification process as the process of decision making, we are guided by consideration of the characteristics involved in the specific problem. The characterization of a set of objects is part of the decision making process in general. In Machine Learning this classification happens through a learning algorithm and the characterization is applied to databases. The classification algorithms can be employed individually or by machine committees. The choice of the best methods to be used in the construction of a committee is a very arduous task. In this work, it will be investigated meta-learning techniques in selecting the best configuration parameters of homogeneous committees for applications in various classification problems. These parameters are: the base classifier, the architecture and the size of this architecture. We investigated nine types of inductors candidates for based classifier, two methods of generation of architecture and nine medium-sized groups for architecture. Dimensionality reduction techniques have been applied to metabases looking for improvement. Five classifiers methods are investigated as meta-learners in the process of choosing the best parameters of a homogeneous committee.

Genetic algorithm, MIP and improvement heuristic applied to the MLCLP with backlogging

The present paper solves the multi-level capacitated lot sizing problem with backlogging (MLCLSPB) combining a genetic algorithm with the solution of mixed-integer programming models and the improvement heuristic fix and optimize. This approach is evaluated over sets of benchmark instances and compared to methods from literature. Computational results indicate competitive results applying the proposed method when compared with other literature approaches. © 2013 IEEE.

Otimização multiobjetivo na análise da integração de geração distribuída às redes de distribuição

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Algoritmo tabu search especializado para o problema de planejamento da expansao de sistemas de transmissão

Pós-graduação em Engenharia Elétrica - FEIS

A nature-inspired approach to speed up optimum-path forest clustering and its application to intrusion detection in computer networks

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

PSO and neural networks: Optimal combination to solve non linear complex problems

Swarm colonies reproduce social habits. Working together in a group to reach a predefined goal is a social behaviour occurring in nature. Linear optimization problems have been approached by different techniques based on natural models. In particular, Particles Swarm optimization is a meta-heuristic search technique that has proven to be effective when dealing with complex optimization problems. This paper presents and develops a new method based on different penalties strategies to solve complex problems. It focuses on the training process of the neural networks, the constraints and the election of the parameters to ensure successful results and to avoid the most common obstacles when searching optimal solutions.

Statin use and survival in colorectal cancer: Results from a population-based cohort study and an updated systematic review and meta-analysis

BACKGROUND: The aim of this study was to investigate the association between statin use and survival in a population-based colorectal cancer (CRC) cohort and perform an updated meta-analysis to quantify the magnitude of any association.

METHODS: A cohort of 8391 patients with newly diagnosed Dukes' A-C CRC (2009-2012) was identified from the Scottish Cancer Registry. This cohort was linked to the Prescribing Information System and the National Records of Scotland Death Records (until January 2015) to identify 1064 colorectal cancer-specific deaths. Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for cancer-specific mortality by statin use were calculated using time dependent Cox regression models. The systematic review included relevant studies published before January 2016. Meta-analysis techniques were used to derive combined HRs for associations between statin use and cancer-specific and overall mortality.

RESULTS: In the Scottish cohort, statin use before diagnosis (HR=0.84, 95% CI 0.75-0.94), but not after (HR=0.90, 95% CI 0.77-1.05), was associated with significantly improved cancer-specific mortality. The systematic review identified 15 relevant studies. In the meta-analysis, there was consistent (I(2)=0%,heterogeneity P=0.57) evidence of a reduction in cancer-specific mortality with statin use before diagnosis in 6 studies (n=86,622, pooled HR=0.82, 95% CI 0.79-0.86) but this association was less apparent and more heterogeneous (I(2)=67%,heterogeneity P=0.03) with statin use after diagnosis in 4 studies (n=19,152, pooled HR=0.84, 95% CI 0.68-1.04).

CONCLUSION: In a Scottish CRC cohort and updated meta-analysis there was some evidence that statin use was associated with improved survival. However, these associations were weak in magnitude and, particularly for post-diagnosis use, varied markedly between studies.

The incidence of schizophrenia: A systematic review

There has been a debate on whether or not the incidence of schizophrenia varies across time and place. In order to optimise the evidence upon which this debate is based, we have undertaken a systematicsystematic review of the literature. In this paper we provide an overview of the methods of the review and a preliminary analysis of the studies identified to date. Electronic databases (Medline, Psychlnfo, Embase, LILAC) were systematically searched for articles published between January 1965 and December 2001. The search terms were: (schizo* OR psycho*)AND (incidence OR prevalence). References were also identified from review articles, reference list and by writing to authors. To date we have identified 137 papers drawn from 33 nations. 37 papers in language other than English await translation. The currently included papers have generated 1413 different items of rate information data. In order to analyze these data we have undertaken several sequential filters in order to identify (a) non-overlapping data, (b) birth cohort study versus noncohort studies, (c) overall and sex-specific rates, (d) diagnostic criteria, (e) age ranges, (f) epoch of study, and (g) data on migrant or other special interest groups. In addition, we will examine the impact of urbanicity of site, age and/or sex standardization, and quality score on the incidence rates. The various discrete incidence rates will be presented graphically and the impact of various filters on these rates will be inspected using meta-analytic techniques. The use of meta-analysis may help elucidate the epidemiological landscape with respect to the incidence of schizophrenia and aid in the generation of new hypothesis. Acknowledgements: The Stanley Medical Research Institute supported project

Long-term price range forecast applied to risk management using regression models

Long-term contractual decisions are the basis of an efficient risk management. However those types of decisions have to be supported with a robust price forecast methodology. This paper reports a different approach for long-term price forecast which tries to give answers to that need. Making use of regression models, the proposed methodology has as main objective to find the maximum and a minimum Market Clearing Price (MCP) for a specific programming period, and with a desired confidence level α. Due to the problem complexity, the meta-heuristic Particle Swarm Optimization (PSO) was used to find the best regression parameters and the results compared with the obtained by using a Genetic Algorithm (GA). To validate these models, results from realistic data are presented and discussed in detail.