Integrated Computer-Aided Engineering - Volume 19, issue 2

Authors: Chabuk, Timur | Reggia, James | Lohn, Jason | Linden, Derek

Article Type: Research Article

Abstract: In recent years, evolutionary computation has been successfully used to solve problems involving engineering design and invention, sometimes producing results that are qualitatively different than previous traditionally-designed solutions. However, while evolutionary methods appear to be a promising tool for supporting design, their usefulness is substantially limited by their computational expense and inability to integrate expert knowledge with evolutionary search. Here we develop and evaluate methods for causally-guided evolutionary design based on expert-supplied cause-effect relations that guide how genetic operators are applied (in contrast to conventional genetic operations which are carried out blindly and randomly), using these methods for antenna array …design. To our knowledge, this is the first study that biases genetic operations in response to the specific performance characteristics of the individuals to which they are applied, and the first to use explicit cause-effect relations to guide this process. Our experimental evaluation compares using evolutionary systems with and without causal guidance to design directional dipole antenna arrays that meet pre-specified performance criteria. We find that causally-guided systems produce optimal solutions with significantly greater frequency and significant computational savings, suggesting that this approach may substantially improve the use of evolutionary computation in engineering design. Show more

Keywords: Causal knowledge, antenna array design, knowledge-guided evolution, causally-guided evolution

DOI: 10.3233/ICA-2012-0395

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 111-124, 2012

Authors: Rodríguez-Seda, Erick J. | Stipanović, Dušan M. | Spong, Mark W.

Article Type: Research Article

Abstract: Consensus and coordination of multiple teleoperated agents has been a control topic of growing interest over the last decade due, in part, to its many potential applications and related complex challenges. One of these challenges is to guarantee stability and motion coordination in the presence of inherent communication and input delays as well as system nonlinearities in the multi-agent system. Addressing this challenge herein, we report on a model reference robust control framework that guarantees stability, motion coordination, and formation control of N output strictly passive, nonlinear Lagrangian systems with arbitrarily large, nonuniform control input delays. The control framework is …comprised of a reference model coupled to the nonlinear systems via the use of N modified scattering transformation blocks. It is shown that the overall control architecture is stable for any large constant delay, robust to system uncertainties, and that the control parameters are delay-independent. A numerical example with four nonlinear planar manipulators and another example with six ground vehicles are finally presented to illustrate the performance of the proposed controller. Show more

Keywords: Teleoperation, networked control systems, multi-agent systems, time delay systems, robotics

DOI: 10.3233/ICA-2012-0396

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 125-136, 2012

Authors: Juhola, Martti | Siermala, Markku

Article Type: Research Article

Abstract: We designed an algorithm in order to examine the importance of variables in data sets for variable evaluation and weighting. In particular, it is designated for the evaluation whether a data set includes such information that is useful for the separation of classes in classification and prediction. Such an evaluation can be performed for an entire data set or separately classes or variables. The scatter method is based on traversing through a data set as near neighbour cases and counting class changes, i.e., when the classes of near cases are changed. The fewer the changes, the more compact the classes …are in a variable space so that they are possible to separate with high classification accuracy. We tested the method with different data sets of medical origin. Their results showed that the scatter method can be used to explore how separable the classes in these data sets were. This is useful for variable evaluation and weighting. Show more

Keywords: Data analysis, variable evaluation, variable weighting, importance of variables, separation of classes

DOI: 10.3233/ICA-2011-0385

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 137-149, 2012

Authors: Al-Mubaid, Hisham | Moazzam, Dawood

Article Type: Research Article

Abstract: Radiation shielding has been an active subject of research in the space industry for many years. We propose a new model for mining material properties for radiation shielding. This work represents an effective way of using learning and feature selection for selecting the material properties that most affect the shielding effectiveness of materials. The methodology relies on machine learning as a measure for the identified subsets of material properties for radiation shielding. This is a new direction in working with radiation shielding using purely computational techniques with machine learning. The experimental results showed that the approach is quite effective in …eliminating redundant features and identifying the most significant properties related to radiation shielding capability of materials. For example, we have identified some material properties, besides Density, like Heat of Fusion, Atomic Number, X-ray Absorption Edge, Electrical Resistivity, and Specific Heat Capacity that are highly related to the radiation shielding as they have been proved computationally. The evaluation results also show that all machine learning algorithms can induce more robust separating models for the subsets of reduced number of features that are highly significant in the domain of radiation shielding. Show more

DOI: 10.3233/ICA-2011-0386

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 151-163, 2012

Authors: Kim, Jaewoo | Jeon, HahnEarl | Lee, Jaiyong

Article Type: Research Article

Abstract: Network Management is the process of managing, monitoring, and controlling the network. Conventional network management was based on wired network which is heavy and unsuitable for resource constrained Wireless Sensor Networks (WSNs). WSNs can be large scale networks and it is impossible to manage each node individually. Also, the polling mechanism of Simple Network Management Protocol (SNMP) imposes heavy management traffic overhead. Since management messages consume resources of WSNs, it can affect the performance of the network. Therefore, it is necessary for WSNs to perform energy efficient network management. Moreover, the management of WSNs should support sustainability. In this paper, …we propose a network management framework for WSNs to reduce management traffic. Based on the cluster-based network management architecture, we classify the Management Information Base (MIB) and define the management messages and message exchange operation for each kind of MIB. The analysis result indicates that the proposed framework can reduce management traffic compared to the polling mechanism and there exists an optimal number of clusters for a given number of nodes. We also propose a lifetime estimation method as an application of the proposed framework. The proposed method uses messages of the proposed framework and estimate the network lifetime from the received management information. Simulation result shows that the proposed method can evaluate the network lifetime and reduce the number exchanged message by using the cluster-based architecture. Show more

Keywords: Network management, wireless sensor network (WSN), management information base (MIB), sustainability, network lifetime

DOI: 10.3233/ICA-2012-0397

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 165-178, 2012

Authors: Grzenda, Maciej | Bustillo, Andres | Quintana, Guillem | Ciurana, Joaquim

Article Type: Research Article

Abstract: Surface roughness generation is influenced by many complex and interrelated factors. Moreover, in real industrial conditions many different milling tools have to be used to create a final product. Hence, the acquisition of experimental data used to set up artificial intelligence models of individual tools is a complicated task. The aim of this paper is to present a new strategy to improve the artificial intelligence models used to predict surface roughness with small datasets. The new strategy proposed in this paper relies on dimensionality reduction to minimise the number of experiments required to train the models. Unlike in …most other approaches, the dimensionality reduction is not applied to set an a priori determined dimension. In the proposed approach, the scale of dimensionality reduction is controlled by the quality of roughness prediction models created with the transformed data. This strategy has been tested on high torque milling operations using multilayer perceptrons i.e. the most frequently used artificial intelligence models for this task. Experiments were conducted to obtain data to train the models. Finally, a comparison was made of the models' performance with and without dimensionality reduction based on Principal Component Analysis, which confirmed the merits of the proposed technique. Show more

Keywords: Dimensionality reduction, roughness modelling, principal component analysis, multilayer perceptron, rough mill

DOI: 10.3233/ICA-2012-0398

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 179-197, 2012

Authors: Yan, Li | Ma, Z.M.

Article Type: Research Article

Abstract: Aiming at distinguishing entities, the concept of entity comparison plays an important role in object management in object-oriented databases as well as object-oriented knowledge bases. In order to model uncertain data and complex-valued attributes as well as complex relationships among objects/classes, fuzzy object-oriented databases have been introduced. In this paper, we identify two levels of entities in fuzzy object-oriented databases, which are fuzzy classes and fuzzy objects. Based on the semantic measure of fuzzy data, we identify several possible relationships between fuzzy entities being compared, and propose a more general way to compare the objects and classes with fuzzy data …types in the fuzzy object-oriented databases. We introduce the notion of equivalence degree of two fuzzy entities. We apply this notion to compare two fuzzy entities so that we know to which extent two objects or two classes refer to the same real-world entity. Show more

Keywords: Fuzzy object-oriented databases, fuzzy object, fuzzy class, entity comparison, equivalence degree

DOI: 10.3233/ICA-2012-0399

Citation: Integrated Computer-Aided Engineering, vol. 19, no. 2, pp. 199-212, 2012