Search

This paper presents study about Dynamic Matrix Control (DMC) controller applied to speed control of DC motor. DMC controller parameters (prediction horizon, control horizon and damping rate of reference) are obtained through optimization methods employing heuristic, deterministic and hybrid strategies. The use of advanced control technique combined with using of optimization methods aims to achieve highly efficient control, reducing the transient state period and variations in steady state. These methods were applied on a simulation model in order to verify which one provides better control results. Index Terms—Predictive Control, Deterministic Optimization, Heuristic Optimization, Hybrid Optimization, DC motor.

TEZ9575 Tez (Yüksek Lisans) -- Çukurova Üniversitesi, Adana, 2014. Kaynakça (s. 57-61) var. ix, 63 s. : res., tablo ; 29 cm. Bu çalışmada, sınıflandırma işlemlerindeki nitelik seçme problemi için Yapay Arı Kolonisi Optimizasyon Tekniği ile Diferansiyel Gelişim Algoritmasını birleştiren bir hibrit yöntem önerilmektedir. Önerilen hibrit yöntem sınıflandırma problemlerinde sıklıkla kullanılan, UCI Veri Kaynağı’ndan elde edilen on beş veri kümesi üzerinde test edilmiştir. Aynı zamanda, önerilen hibrit yöntem performans değerlendirmesi için popüler üç nitelik seçme tekniği olan Ki-kare (ChiSquare), Bilgi Kazancı (Information Gain) ve Korelasyon Nitelik Seçimi (Correlation Feature Selection) ile kıyaslanmıştır. Bu çalışmanın amacı çalışma zamanı performansını ve sınıflandırıcının doğruluğunu iyileştirmek için sınıflandırma işlemi sırasında kullanılan nitelik sayısını azaltmaktır. Bu çalışmada elde edilen deney sonuçları geliştirdiğimiz hibrit yöntemin sınıflandırma işlemleri için iyi nitelikleri seçebildiğini göstermiştir. In this study, a hybrid method which combines Artificial Bee Colony Optimization Technique with Differential Evolution Algorithm is proposed for feature selection problem of classification tasks. The developed hybrid method was experimented on fifteen datasets from the UCI Repository which are commonly used in classification problems. The proposed hybrid feature selection method was also compared with the three most popular feature selection techniques that are Information Gain, ChiSquare and Correlation Feature Selection to evaluate its performance. The aim of this study is to reduce the number of features to be used during the classification process to improve run-time performance and accuracy of the classifier. The experimental results of this study showed that our developed hybrid method was able to select good features for classification tasks.

In this paper, we propose a challenging research direction for Constraint Programming and optimization techniques in general. We address problems where decisions to be taken affect and are affected by complex systems, which exhibit phenomena emerging from a collection of interacting objects, capable to self organize and to adapt their behaviour according to their history and feedback. Such systems are unfortunately impervious to modeling efforts via state-of-the-art combinatorial optimization techniques. We provide some hints on approaches to connect and integrate decision making and optimization technology with complex systems via machine learning, game theory and mechanism design. In the first case, the aim is to extract modeling components to express the relation between global decisions and observables emerging from the real system, or from an accurate predictive model (e.g. a simulator). In the second case, the idea is to exploit game theory, mechanism design and distributed decision making to drive the process toward realistic equilibrium points avoiding globally optimal, but unrealistic, configurations. We conclude by observing how dealing with the complexity of the considered problems will require to greatly extend the capabilities of state of the art solvers: in this context, we identify some key issues and highlight future research directions.

This paper investigated the different gear ratio matching effect on the series hybrid electric bus especially for the 90 kW &amp, 150kW traction motors on the fuel economy. The 90 kW &amp, 150 kW traction motors make the system use more efficient operating points, thus, the system is more efficient than when the 240 kW single unit is used. Furthermore, matching the different gear ratios allows the nonlinear characteristics in the traction motor efficiency to be used more efficiently, thus the standard for how to select the gear ratios was proposed. The fuel economy selection was optimized by the hybrid optimization methodology using RSM and the univariate search method, both of which are well known for their straightforward concepts and adequate performances, thus, the new suggested matching theory noticeably improved the system efficiency. The work presented here has profound implications for future studies for the design of the hybrid electric vehicle, as well as the plug-in hybrid and the electric vehicles and was carried out on the AMEsim-Simulink Co-simulation platform.

In this thesis, the basic principles and concepts of single and multi-objective Genetic Algorithms (GA) are reviewed. Two algorithms, one for single objective and the other for multi-objective problems, which are believed to be more efficient are described in details. The algorithms are coded with MATLAB and applied on several test functions. The results are compared with the existing solutions in literatures and shows promising results. Obtained pareto-fronts are exactly similar to the true pareto-fronts with a good spread of solution throughout the optimal region. Constraint handling techniques are studied and applied in the two algorithms. Constrained benchmarks are optimized and the outcomes show the ability of algorithm in maintaining solutions in the entire pareto-optimal region. In the end, a hybrid method based on the combination of the two algorithms is introduced and the performance is discussed. It is concluded that no significant strength is observed within the approach and more research is required on this topic. For further investigation on the performance of the proposed techniques, implementation on real-world engineering applications are recommended.

International audience; This paper presents a system-level method to decrease the power consumption of integrated audio Class-G amplifiers for mobile phones by using the same implementation of the level detector, but by changing the parameters of the switching algorithm. This method uses an optimization based on a simplified model simulation to quickly find the best power supply switching strategy in order to decrease the losses of the internal Class-AB amplifier. Using a few relevant equations of Class-G on the electrical level and by reducing the number of calculation points, this model can dramati-cally reduce the calculation time to allow power consumption evaluation in realistic case conditions compared to the currently available tools. This simplified model also evaluates the audio quality reproduction thanks to a psy-cho-acoustic method. The model has been validated by comparing model results and practical measurements on two industrial circuits. This proposed model is used by an optimizer based on a genetic algorithm associated with a pattern search algorithm to find the best power supply switching strategy for the internal Class-AB amplifier. The optimization results improve life-time performance by saving at least 25% in power consumption for typical use-case (1mW) com-pared to the industrial circuit studied and without losses in audio quality.

International audience; This paper deals with hybrid optimization schemes that combine meta-heuristics and exact optimization methods. In particular, Genetic Algorithms (GAs) and the Branch-and-Bound algorithm (B&B). Three parallel hybridization schemes combining GAs and the B&B algorithm are proposed: a parallel relay model where the two algorithms are executed in a pipeline mode, a low level hybrid scheme where a given operator of the GA is replaced by a B&B solver, and a cooperative hybrid scheme where the two algorithms cooperate and exchange information about: best-found solution, promising regions to exploit by the B&B and unexplored regions to explore by the GA.

A new approach for developing a Nusselt number correlation, in terms of relevant non-dimensional parameters, for turbulent forced convection flows in vertical channels using a judicious combination of transient cooling experiments with a hybrid optimization technique is reported. The temperature-time history, during the cooling of a heated plate, idealized as a lumped capacity heat transfer model, is recorded using a PC based data acquisition system. A numerically computed temperature-time history of the plate is then compared with the experimentally known temperature-time history to estimate the residual. The minimization of sum of the squares of the residual is done using a hybrid numerical optimization technique, i.e. a combination of Genetic Algorithm and the Levenberg-Marquardt method, in order to obtain the coefficient and the exponents of the pertinent non-dimensional parameters in the Nusselt number correlation. The parameters in the correlation are also retrieved using another global optimization technique, the Simulated Annealing (SA), for evaluating the consistency in parameter estimation. As a validation exercise, Nusselt number values estimated using the proposed correlation are compared with steady state experimental results and a good agreement of results endorses the efficacy of this approach. � 2008 Elsevier Ltd. All rights reserved.

In the paper a 2D joint inversion method is presented, which is applicable for the simultaneous determination of layer thickness variation and petrophysical parameters by processing well-logging data acquired in several boreholes along the profile. The so-called interval inversion method is tested on noisy synthetic data sets generated on hydrocarbon-bearing reservoir models. Numerical experiments are performed to study the convergence and stability of the inversion procedure. Data and model misfit, function distance related to layer thickness fitting are measured as well as estimation errors and correlation coefficients are computed to check the accuracy and reliability of inversion results. It is shown that the actual inversion procedure is stable and highly accurate, which arises from the great over-determination feature of the inverse problem. Even a case study is attached to the paper in which interval inversion procedure is applied for processing of multi-borehole logging data acquired in Hungarian hydrocarbon exploratory wells in order to determine petrophysical parameters and lateral changes of layer thicknesses.

In this paper, we present an improved hybrid optimization algorithm, which was applied to the hard combinatorial optimization problem, the quadratic assignment problem (QAP). This is an extended version of the earlier hybrid heuristic approach proposed by the author. The new algorithm is distinguished for the further exploitation of the idea of hybridization of the well‐known efficient heuristic algorithms, namely, simulated annealing (SA) and tabu search (TS). The important feature of our algorithm is the so‐called “cold restart mechanism”, which is used in order to avoid a possible “stagnation” of the search. This strategy resulted in very good solutions obtained during simulations with a number of the QAP instances (test data). These solutions show that the proposed algorithm outperforms both the “pure” SA/TS algorithms and the earlier author's combined SA and TS algorithm. Key words: hybrid optimization, simulated annealing, tabu search, quadratic assignment problem, simulation Patobulintas hibridinis optimizavimo algoritmas kvadratinio paskirstymo uždaviniui Santrauka Šiame straipsnyje pasi ulytas patobulintas hibridinis euristinis optimizavimo algoritmas gerai žinomam, sudetingam kombinatorinio optimizavimo uždaviniui, b utent, kvadratinio paskirstymo (KP) uždaviniui. Tai‐pagerinta autoriaus ankstesnio hibridinio algoritmo versija. Naujasis algoritmas pasižymi tuo, jog čia išpletota efektyviu euristiku (atkaitinimo modeliavimo (AM) (angi. simulated annealing) ir tabu paieškos (TP) (angi. tabu search) “hibridizacijos” ideja. “Hibridizacija” remiasi vadinamaja iteracine schema: TP algoritmas panaudojamas kaip post‐analizes proced ura AM algoritmo gautajam sprendiniui, savo ruožtu, AM algoritmas taikomas sprendiniu sekai, gautai sprendiniu diversifikavimo/generavimo keliu. Svarbi pasi ulyto algoritmo savybe yra ir ta, kad jame realizuotas vadinamasis “šaltojo pakartotinio starto” principas, kurio paskirtis padeti išvengti galimu paieškos “stagnacijos” situaciju. Naujasis algoritmas išbandytas su KP uždavinio duomenimis iš testiniu pavyzdžiu bibliotekos QAPLIB. Gauti eksperimentu rezultatai liudija, jog nagrinetiems KP uždavinio pavyzdžiams si ulomas algoritmas yra pranašesnis už ankstesnius atkaitinimo modeliavimo ir tabu paieškos algoritmus, taip pat už ankstesni autoriaus hibridini algoritma. First Published Online: 14 Oct 2010

Visual data mining via the construction of virtual reality spaces for the representation of data and knowledge, involves the solution of optimization problems. This paper introduces a hybrid technique based on particle swarm optimization (PSO) combined with classical optimization methods. This approach is applied to very high dimensional data from microarray gene expression experiments in order to understand the structure of both raw and processed data. Experiments with data sets corresponding to Alzheimer's disease show that high quality visual representations can be obtained by combining PSO with classical optimization methods. The behaviour of some of the parameters controlling the swarm evolution is also studied., IASTED International Joint Conference on Artificial Intelligence and Soft Computing (ASC'2004), September 1-3, 2004, Marbella, Spain

The inversion of vertical electrical sounding (VES) is normally performed considering a stratified medium formed by homogeneous, isotropic and horizontal layers. The simplicity of this geophysical model makes the inversion simple and computationally fast, and together with the main characteristics of the electroresistivity method, it was greatly responsible to make VES one of the most popular geophysical method for groundwater exploration and engineering geophysics. However, even in a sedimentary basin where the geology is more conform, the assumption of horizontal and homogeneous layers is not necessarily valid, limiting the reliability of the inversion results. In this paper we present a fast and robust 2D resistivity modeling and inversion algorithm for the interpretation of sets of VES. We consider three inversion algorithms: the Gauss-Newton method of linearized inversion (LI), the genetic algorithm (GA), and a hybrid approach (GA-LI) that uses LI to improve the best model at the end of each step of the GA. The medium parametrization consists of the partition of the domain into fixed homogeneous rectangular blocks such that their resistivities are the only free parameters. The apparent resistivity is evaluated by an iterative scheme that is derived from a finite-difference discretization of the potential differential equation. We enhance the convergence rate of the scheme by adopting an incomplete Cholesky preconditioner. Numerical results using synthetic and real 2D apparent resistivity data formed by sets of VES for the Schlumberger configuration illustrate the performance of the hybrid GA-LI algorithm. The VES field data were acquired near Conceição do Coité, state of Bahia, Brazil. We compare the performance of the LI, GA and GA-LI algorithms. A inversão de uma sondagem elétrica vertical (SEV) normalmente assume que o meio é estratifcado e formado por camadas horizontais homogêneas e isotrópicas. A simplicidade deste modelo geofísico torna a inversão simples e com reduzido custo computacional. Esta simplicidade, junto às principais qualidades do método de eletroresistividade, foi responsável por tornar a SEV um dos métodos geofísicos mais populares nos trabalhos de exploração de águas subterrâneas e geofísica aplicada à engenharia. Porém, mesmo em bacias sedimentares, onde a geologia é mais conforme, a hipótese de camadas planas e homogêneas não é válida, o que limita a confiabilidade dos resultados da inversão. Apresentamos neste artigo um algoritmo rápido e robusto de modelagem e inversão eletroresistiva para a interpretação de conjuntos de SEVs. Consideramos três algoritmos de inversão: o método de inversão linearizada de Gauss-Newton (LI), o algorítmo genético (GA), e uma abordagem híbrida (GA-LI) que usa a inversão linearizada para aprimorar o melhor modelo obtido ao final de cada geração do algoritmo genético. A parametrização do meio consiste na partição do dommínio em blocos retangulares e homogêneos, de modo que a resistividade de cada bloco é um parâmetro do modelo. A resistividade aparente é calculada com um método iterativo baseado numa aproximação por diferenças finitas da equação do potencial elétrico. Um precondicionamento do tipo Cholesky incompleto é utilizado para acelerar a convergência do método. Avaliamos a performance do método híbrido por meio de experimentos numéricos com perfis de eletroresistividades reais e sintéticos, formados por conjuntos de SEVs obtidas com o arranjo Schlumberger. Os dados de campo foram coletados nas proximidades de Conceição do Coité, estado da Bahia, Brasil.

Submitted by Santiago Fabio (fabio.ssantiago@hotmail.com) on 2011-10-13T18:31:06Z No. of bitstreams: 1 55555.pdf: 878136 bytes, checksum: 9687a521def08ade280c32b51afe7a01 (MD5) Made available in DSpace on 2011-10-13T18:31:06Z (GMT). No. of bitstreams: 1 55555.pdf: 878136 bytes, checksum: 9687a521def08ade280c32b51afe7a01 (MD5) Previous issue date: 2003 The inversion of vertical electrical sounding (VES) is normally performed considering a stratified medium formed by homogeneous, isotropic and horizontal layers. The simplicity of this geophysical model makes the inversion simple and computationally fast, and together with the main characteristics of the electroresistivity method, it was greatly responsible to make VES one of the most popular geophysical method for groundwater exploration and engineering geophysics. However, even in a sedimentary basin where the geology is more conform, the assumption of horizontal and homogeneous layers is not necessarily valid, limiting the reliability of the inversion results. In this paper we present a fast and robust 2D resistivity modeling and inversion algorithm for the interpretation of sets of VES. We consider three inversion algorithms: the Gauss-Newton method of linearized inversion (LI), the genetic algorithm (GA), and a hybrid approach (GA-LI) that uses LI to improve the best model at the end of each step of the GA. The medium parametrization consists of the partition of the domain into fixed homogeneous rectangular blocks such that their resistivities are the only free parameters. The apparent resistivity is evaluated by an iterative scheme that is derived from a finite-difference discretization of the potential differential equation. We enhance the convergence rate of the scheme by adopting an incomplete Cholesky preconditioner. Numerical results using synthetic and real 2D apparent resistivity data formed by sets of VES for the Schlumberger configuration illustrate the performance of the hybrid GA-LI algorithm. The VES field data were acquired near Conceição do Coité, state of Bahia, Brazil. We compare the performance of the LI, GA and GA-LI algorithms. São Paulo

Background Plasmodium is the causal parasite of malaria, infectious disease responsible for the death of up to one million people each year. Glycerophospholipid and consequently membrane biosynthesis are essential for the survival of the parasite and are targeted by a new class of antimalarial drugs developed in our lab. In order to understand the highly redundant phospholipid synthethic pathways and eventual mechanism of resistance to various drugs, an organism specific kinetic model of these metabolic pathways need to be developed in Plasmodium species. Results Fluxomic data were used to build a quantitative kinetic model of glycerophospholipid pathways in Plasmodium knowlesi. In vitro incorporation dynamics of phospholipids unravels multiple synthetic pathways. A detailed metabolic network with values of the kinetic parameters (maximum rates and Michaelis constants) has been built. In order to obtain a global search in the parameter space, we have designed a hybrid, discrete and continuous, optimization method. Discrete parameters were used to sample the cone of admissible fluxes, whereas the continuous Michaelis and maximum rates constants were obtained by local minimization of an objective function.The model was used to predict the distribution of fluxes within the network of various metabolic precursors. The quantitative analysis was used to understand eventual links between different pathways. The major source of phosphatidylcholine (PC) is the CDP-choline Kennedy pathway. In silico knock-out experiments showed comparable importance of phosphoethanolamine-N-methyltransferase (PMT) and phosphatidylethanolamine-N-methyltransferase (PEMT) for PC synthesis. The flux values indicate that, major part of serine derived phosphatidylethanolamine (PE) is formed via serine decarboxylation, whereas major part of phosphatidylserine (PS) is formed by base-exchange reactions. Sensitivity analysis of CDP-choline pathway shows that the carrier-mediated choline entry into the parasite and the phosphocholine cytidylyltransferase reaction have the largest sensitivity coefficients in this pathway, but does not distinguish a reaction as an unique rate-limiting step. Conclusion We provide a fully parametrized kinetic model for the multiple phospholipid synthetic pathways in P. knowlesi. This model has been used to clarify the relative importance of the various reactions in these metabolic pathways. Future work extensions of this modelling strategy will serve to elucidate the regulatory mechanisms governing the development of Plasmodium during its blood stages, as well as the mechanisms of action of drugs on membrane biosynthetic pathways and eventual mechanisms of resistance.

This study aims at optimizing the dynamics of a fixed – tank three – axle vehicle, with respect to both its vertical and lateral performance. This is accomplished by developing a hybrid optimization algorithm, which combines the conventional Evolution Strategy with well – known deterministic methods, such as line – search and/or trust – region. By utilizing a Monte Carlo analysis, the stiffness and damping characteristics of vehicle's suspensions are optimized with respect to the pitch and roll angles of the sprung mass, taking under consideration the design limitations, the geometrical constraints of the vehicle, as well as the lateral acceleration at overturn.

While the need for more efficient and integrated processes for the separation of non-ideal and azeotropic mixtures mandates model-based design methods, the resulting mixed-integer problems are highly nonlinear and particularly hard to solve without proper initialization. In order to handle such complex design problems, two hybrid optimization methods are presented in the current contribution. Both methods rely on a nested approach, which integrates a polylithic modeling and solution strategy, based on the solution of successively refined nonlinear programming problems, with an upper level metaheuristic for the initialization and optimization of discrete design decisions. Both methods are evaluated for two case studies. The optimization of a simple column design problem allows for the direct comparison with an available global deterministic optimization solver for identifying the global optimal solution. The optimization of an extractive distillation process with simultaneous solvent selection and energy integration illustrates the potential complexity that can be handled effectively.

Solvent-based separation processes, such as extractive distillation, show large potential for the separation of azeotropic mixtures. However, these processes are rather complex to design and optimize since the overall process performance depends strongly on the choice and amount of solvent and can only be evaluated for a process flowsheet with closed recycles. It is important to note that the potential for heat integration also depends strongly on the solvent choice. Consequently, a successive selection of a suitable solvent followed by process design and optimization and finally energy integration likely results in suboptimal choices. In order to allow for direct optimization of an extractive distillation process, including solvent selection and different means for energy integration, the current study proposes the use of a hybrid evolutionary-deterministic algorithm. The application is demonstrated for the separation of an azeotropic acetone-methanol mixture, considering six solvent candidates and up to four alternative means for energy integration. The results illustrate the existence of a multitude of suboptimal local solutions and demonstrate the capability of the proposed method to effectively overcome these limitations.