In this paper, a robust formulation for the structure-control design of mechatronic systems is developed. The proposed robust approach aims at minimization of the sensitivity of the nominal design objectives with resp...
详细信息
In this paper, a robust formulation for the structure-control design of mechatronic systems is developed. The proposed robust approach aims at minimization of the sensitivity of the nominal design objectives with respect to uncertain parameters. The robust integrated design problem is established as a nonlinear multiobjective dynamic optimization one, which in order to consider synergetic interactions uses mechanical and control nominal design objectives. A planar parallel robot and its controller are simultaneously designed with the proposed approach when the nominal design objectives are the tracking error and the manipulability measure. The payload at the end-effector is considered as the uncertain parameter. Experimental results show that a robustly designed parallel robot presents lower sensitivity of the nominal design objectives under the effects of changes at the payload than a nonrobustly designed one.
An inverse heat transfer problem is discussed to estimate simultaneously the unknown position and timewise varying strength of a heat source by utilizing differentialevolution approach. A two dimensional enclosure wi...
详细信息
An inverse heat transfer problem is discussed to estimate simultaneously the unknown position and timewise varying strength of a heat source by utilizing differentialevolution approach. A two dimensional enclosure with isothermal and black boundaries containing non-scattering, absorbing and emitting gray medium is considered. Both radiation and conduction heat transfer are included. No prior information is used for the functional form of timewise varying strength of heat source. The finite volume method is used to solve the radiative transfer equation and the energy equation. In this work, instead of measured data, some temperature data required in the solution of the inverse problem are taken from the solution of the direct problem. The effect of measurement errors on the accuracy of estimation is examined by introducing errors in the temperature data of the direct problem. The prediction of source strength and its position by the differentialevolution (DE) algorithm is found to be quite reasonable. (C) 2013 Elsevier Ltd. All rights reserved.
In this paper, we propose an extreme learning machine (ELM) with tunable activation function (TAF-ELM) learning algorithm, which determines its activation functions dynamically by means of the differentialevolution a...
详细信息
In this paper, we propose an extreme learning machine (ELM) with tunable activation function (TAF-ELM) learning algorithm, which determines its activation functions dynamically by means of the differential evolution algorithm based on the input data. The main objective is to overcome the problem dependence of fixed slop of the activation function in ELM. We mainly considered the issue of processing of benchmark problems on function approximation and pattern classification. Compared with ELM and E-ELM learning algorithms with the same network size or compact network configuration, the proposed algorithm has improved generalization performance with good accuracy. In addition, the proposed algorithm also has very good performance in the TAF neural networks learning algorithms.
Frequency response masking (FRM) technique along with the Canonic Signed Digit (CSD) representation is a good alternative for the design of a computationally efficient, sharp transition width, high speed finite impuls...
详细信息
Frequency response masking (FRM) technique along with the Canonic Signed Digit (CSD) representation is a good alternative for the design of a computationally efficient, sharp transition width, high speed finite impulse response (FIR) filter. This paper proposes two novel approaches for the joint optimization of an FRM FIR digital filter in the CSD space. The first approach uses the recently emerged Artificial Bee Colony (ABC) algorithm and the second approach uses the differentialevolution (DE) algorithm. In this paper, both the algorithms are modified in such a way that, they are suitable for the solution of the optimization problem posed, in which the search space consists of integers and the objective function is nonlinear. The optimization variables are encoded such that they permit the reduction in computational cost. The salient feature of the above approaches is the reduced computational complexity while obtaining good performance. Simulation results show that the ABC based design technique performs better than that using DE, which in turn outperforms the one using integer coded genetic algorithm (GA). The proposed optimization approaches can be extended to the solution of integer programming problems in other engineering disciplines also. (C) 2012 Elsevier Ltd. All rights reserved.
This paper proposes an optimization-based model for generic document summarization. The model generates a summary by extracting salient sentences from documents. This approach uses the sentence-to-document collection,...
详细信息
This paper proposes an optimization-based model for generic document summarization. The model generates a summary by extracting salient sentences from documents. This approach uses the sentence-to-document collection, the summary-to-document collection and the sentence-to-sentence relations to select salient sentences from given document collection and reduce redundancy in the summary. To solve the optimization problem has been created an improved differential evolution algorithm. The algorithm can adjust crossover rate adaptively according to the fitness of individuals. We implemented the proposed model on multi-document summarization task. Experiments have been performed on DUC2002 and DUC2004 data sets. The experimental results provide strong evidence that the proposed optimization-based approach is a viable method for document summarization. (C) 2012 Elsevier Ltd. All rights reserved.
This work presents a comparison of three evolutionary algorithms, the particle swarm optimization, the differential evolution algorithm and a hybrid algorithm derived from the previous, when applied to the generation ...
详细信息
This work presents a comparison of three evolutionary algorithms, the particle swarm optimization, the differential evolution algorithm and a hybrid algorithm derived from the previous, when applied to the generation and demand dispatch problem. An optimization problem is formulated in the context of a small grid with partially flexible demand that can be shifted along a time horizon. It is assumed that grid operator dispatches generation and flexible demand along the time horizon aiming at minimizing generation costs. Consumption restrictions associated with flexible demand are modeled by equality and inequality energy constraints. Power flow equality constraints and inequality constraints due to operational limits for each dispatch interval are represented. The paper discusses a methodology for evolutionary algorithms performance assessment and states the importance of using statistical tools. The comparison is initially conducted using the IEEE 30-bus test system. Problem dimension effect is addressed considering different number of dispatch intervals in the time horizon. Moreover, the algorithms are applied to the 192-bus system of a Brazilian distribution utility, in the particular context of a load management program for large consumers of the company. In this application, the quality of the near-optimal solution obtained with the stochastic algorithms is evaluated by comparing with an analytical optimization algorithm solution. (c) 2012 Elsevier Ltd. All rights reserved.
In this work, the goal is to find the optimum working point for intelligent wells that is equipped with down-hole chokes and producing from a multi-layer reservoir under pressure maintenance process. Two evolutionary ...
详细信息
In this work, the goal is to find the optimum working point for intelligent wells that is equipped with down-hole chokes and producing from a multi-layer reservoir under pressure maintenance process. Two evolutionary optimization algorithms are used and results are compared from different points of view. Therefore, the differentialevolution technique and genetic algorithm has been used in conjugation with a reservoir simulator. The overall optimization technique is applied to heterogeneous example problems involving smart and multi segment wells. Improvement of predicted performance, which is as high as 7% in one case is demonstrated for all of the examples studied.
In this paper, a novel method degrading the combined effect of four-wave mixing (FWM) and amplified spontaneous emission (ASE) noise of the amplifier on the most heavily affected channel in an equally channel spaced w...
详细信息
In this paper, a novel method degrading the combined effect of four-wave mixing (FWM) and amplified spontaneous emission (ASE) noise of the amplifier on the most heavily affected channel in an equally channel spaced wavelength division multiplexing (WDM) system containing in-line optical amplifiers is proposed. FWM effect is directly related to input powers of channels. So, FWM effect can be degraded by controlling channel input powers. In the proposed method, varying the input power of each channel in an optical fiber, the output optical signal to noise ratio (OSNR) values are evaluated and input powers of all channels are optimized in order to maximize the OSNR value of the channel having the lowest OSNR. To interpret the results obtained, output OSNR values for the minimum optical input power launched to the system by each channel are also computed. Being compared to the computed results for minimum optical input powers, the lowest output OSNR value among all channels for optimized input powers shows a 5.1867 dB increase in a 5-channel system, a 3.5988 dB increase in a 9-channel system, a 3.0855 dB increase in a 15-channel system and a 1.6795 dB increase for a 21-channel system. Furthermore, output OSNR values of all channels exhibit a significant increase. (c) 2013 Elsevier GmbH. All rights reserved.
Wavelength detection problem is addressed in a serial wavelength division multiplexing (WDM) fiber Bragg grating (FBG) sensor network, where the spectra of the FBGs are partially or fully overlapped. Specifically, an ...
详细信息
Wavelength detection problem is addressed in a serial wavelength division multiplexing (WDM) fiber Bragg grating (FBG) sensor network, where the spectra of the FBGs are partially or fully overlapped. Specifically, an optimization model utilizing the transfer matrix method to analyze the reflection spectra is presented. The search space of this problem indicates that the optimal solution and the sub-optimal solution are completely symmetric. Considering the peculiarity of the problem, an improved differentialevolution (DE) algorithm, named Swap-DE, is used to calculate the Bragg wavelengths. Experimental results show that the wavelength detection error is typically smaller than 0.2 pm. (C) 2013 Elsevier B.V. All rights reserved.
We study and calibrate a cohort-based model which captures the characteristics of a mortality surface with a parsimonious, continuous-time factor approach. The model allows for imperfect correlation of the mortality i...
详细信息
We study and calibrate a cohort-based model which captures the characteristics of a mortality surface with a parsimonious, continuous-time factor approach. The model allows for imperfect correlation of the mortality intensity across generations. It is implemented on UK data for the period 1900-2008. Calibration by means of stochastic search and the differentialevolution optimization algorithm proves to yield robust and stable parameters. We provide in-sample and out-of-sample, deterministic as well as stochastic forecasts. Calibration confirms that correlation across generations is smaller than one. (c) 2013 Elsevier B.V. All rights reserved.
暂无评论