• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.2
• /
• pp.13-21
• /
• 1996

The necessity and usefulness of higher-order neural networks have been well-known since early days of neurocomputing. However the explosive number of terms has hampered the design and training of such networks. In this paper we present an evolutionary learning method for efficiently constructing problem-specific higher-order neural models. The crux of the method is the neural tree representation employing both sigma and pi units, in combination with the use of an MDL-based fitness function for learning minimal models. We provide experimental results in classification and prediction problems which demonstrate the effectiveness of the method. I. Introduction topology employs one hidden layer with full connectivity between neighboring layers. This structure has One of the most popular neural network models been very successful for many applications. However, used for supervised learning applications has been the they have some weaknesses. For instance, the fully mutilayer feedforward network. A commonly adopted connected structure is not necessarily a good topology unless the task contains a good predictor for the full *d*dWs %BH%W* input space.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.05a
• /
• pp.309-320
• /
• 2007

The data imbalance problem which can be uncounted in data mining classification problems typically means that there are more or less instances in a class than those in other classes. It causes low prediction accuracy of the minority class because classifiers tend to assign instances to major classes and ignore the minor class to reduce overall misclassification rate. In order to solve the data imbalance problem, there has been proposed a number of techniques based on resampling with replacement, adjusting decision thresholds, and adjusting the cost of the different classes. In this paper, we study the feasibility of the combination usage of the techniques previously proposed to deal with the data imbalance problem, and suggest a combination method using genetic algorithm to find the optimal combination ratio of the techniques. To improve the prediction accuracy of a minority class, we determine the combination ratio based on the F-value of the minority class as the fitness function of genetic algorithm. To compare the performance with those of single techniques and the matrix-style combination of random percentage, we performed experiments using four public datasets which has been generally used to compare the performance of methods for the data imbalance problem. From the results of experiments, we can find the usefulness of the proposed method.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.5
• /
• pp.1012-1020
• /
• 2013

In this work, Self-adaptive Differential Evolutionary (SaDE) algorithm is proposed to solve Optimal Location and Size of Capacitor (OLSC) problem in radial distribution networks. To obtain the SaDE algorithm, two improvements have been applied on control parameters of mutation and crossover operators. To expand the study, three load conditions have been considered, i.e., constant, varying and effective loads. Objective function is introduced for the load conditions. The annual cost is fitness of problem, in addition to this cost, CPU time, voltage profile, active power loss and total installed capacitor banks and their related costs have been used for comparisons. To confirm the ability of each improvements of SaDE, the improvements are studied both in separate and simultaneous conditions. To verify the effectiveness of the proposed algorithm, it is tested on IEEE 10-bus and 34-bus radial distribution networks and compared with other approaches.

• Industrial Engineering and Management Systems
• /
• v.10 no.1
• /
• pp.54-64
• /
• 2011

The purpose of this study is to optimally allocate the human resources to tasks while minimizing the total daily human resource costs and smoothing the human resource usage. The human resource allocation problem (hRAP) under consideration contains two kinds of special constraints, i.e. operational precedence and skill constraints in addition to the ordinary constraints. To deal with the multiple objectives and the special constraints, first we designed this hRAP as a network problem and then proposed a Pareto multistage decisionbased genetic algorithm (P-mdGA). During the evolutionary process of P-mdGA, a Pareto evaluation procedure called generalized Pareto-based scale-independent fitness function approach is used to evaluate the solutions. Additionally, in order to improve the performance of P-mdGA, we use fuzzy logic controller for fine-tuning of genetic parameters. Finally, in order to demonstrate the applicability and to evaluate the performance of the proposed approach, P-mdGA is applied to solve a case study in a hotel, where the managers usually need helpful automatic support for effectively allocating hotel staff to hotel tasks.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.1322-1327
• /
• 2007

Many automotive companies have tried to apply the aluminum alloy sheet to car body because reducing the car weight can improve the fuel efficiency of vehicle. In order to do that, sheet materials require of weldablity, formability, productivity and so on. Aluminum alloy was not easy to join these metals due to its material properties. Thus, the laser is good heat source for aluminum alloy welding because of its high heat intensity. However, the welding quality was not good by porosity, underfill, and magnesium loss in welded metal for AA5182 aluminum alloy. In this study, Nd:YAG laser welding of AA 5182 with filler wire AA 5356 was carried out to overcome this problem. The weldability of AA5182 laser welding with AA5356 filler wire was investigated in terms of tensile strength and Erichsen ratio. For full penetration, mechanical properties were improved by filler wire. In order to optimize the process parameters, model to estimate tensile strength by artificial neural network was developed and fitness function was defined in consideration of weldability and productivity. Genetic algorithm was used to search the optimal point of laser power, welding speed, and wire feed rate.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.6 no.2
• /
• pp.125-132
• /
• 2003

Uzi fly (Exorista bombycis Louis) is one of the major larval endo-parasitoid of silkworm (Bombyx mori). It causes extensive damage to sericulture industry. The application of synthetic organic pesticides has tremendous impact on minimizing pest population but their overuse and frequent misuse and high sensitivity towards the silkworms, has forced the entomologists to search for alternatives to chemical control, which is safe to silkworm, environment and farm workers. Biological control continues to offer exciting possibilities for the control of fly pest population. It is environmentally safe alternative to chemical control and offering a long-term protection. Several potential hymenopteran parasitoids have been screened. Among successful natural enemies, Nesolynx thymus, Trichomalopsis apanteloctena, Trichopria sp., Brachymeria lasus, Pediobius sp., Spalangia sp., Spilomicrus karnatakensis and Dhirhinus sp. are important. It is essential to predict accurately the efficacy of these natural enemies in a new habitat prior to its introduction. The important desirable attributes of these potential parasitoids viz., host searching capacity, specificity, power of increase and fitness and adaptability of the parasitoid in new environment has been recorded. Results of the host parasitoid interaction indicate that the aging of the host function as a factor that influence the host finding efficiency of the parasitoid. It is highly scored with 15-20 hrs old pupa of the host. However, aging of the parasitoid does not significantly affect it. The sex ratio is female biased which is advantageous from biological control point of view, Biological suppression methods involving conservation and utilization of natural enemies have been discussed in detail.

• Advances in Energy Research
• /
• v.4 no.4
• /
• pp.255-274
• /
• 2016

This paper presents a novel symbiotic organisms search (SOS) algorithm to optimize both real power loss (RPL) and voltage stability limit (VSL) of a transmission network by controlling the variables such as unified power flow controller (UPFC) location, UPFC series injected voltage magnitude and phase angle and transformer taps simultaneously. Mathematically, this issue can be formulated as nonlinear equality and inequality constrained multi objective, multi variable optimization problem with a fitness function integrating both RPL and VSL. The symbiotic organisms search (SOS) algorithm is a nature inspired optimization method based on the biological interactions between the organisms in ecosystem. The advantage of SOS algorithm is that it requires a few control parameters compared to other meta-heuristic algorithms. The proposed SOS algorithm is applied for solving optimum control variables for both single objective and multi-objective optimization problems and tested on New England 39 bus test system. In the single objective optimization problem only RPL minimization is considered. The simulation results of the proposed algorithm have been compared with the results of the algorithms like interior point successive linear programming (IPSLP) and bacteria foraging algorithm (BFA) reported in the literature. The comparison results confirm the efficacy and superiority of the proposed method in optimizing both single and multi objective problems.

• Journal of Computing Science and Engineering
• /
• v.9 no.2
• /
• pp.39-50
• /
• 2015

Genetic algorithm (GA) and particle swarm optimization (PSO) are two excellent approaches to multimodal optimization problems. However, slow convergence or premature convergence readily occurs because of inappropriate and inflexible evolution. In this paper, a novel optimization algorithm with a flexible grid optimization (FGO) is suggested to provide adaptive trade-off between exploration and exploitation according to the specific objective function. Meanwhile, a uniform agents array with adaptive scale is distributed on the gird to speed up the calculation. In addition, a dominance centroid and a fitness center are proposed to efficiently determine the potential guides when the population size varies dynamically. Two types of subregion division strategies are designed to enhance evolutionary diversity and convergence, respectively. By examining the performance on four benchmark functions, FGO is found to be competitive with or even superior to several other popular algorithms in terms of both effectiveness and efficiency, tending to reach the global optimum earlier. Moreover, FGO is evaluated by applying it to a parameter decision in a least squares support vector machine (LS-SVM) to verify its practical competence.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1161-1164
• /
• 1996

In this paper, a parameter optimization technique for a mobile robot navigation is discussed. Authors already have proposed a navigation algorithm for mobile robots with sonar sensors using fuzzy decision making theory. Fuzzy decision making selects the optimal via-point utilizing membership values of each via-point candidate for fuzzy navigation goals. However, to make a robot successfully navigate through an unknown and cluttered environment, one needs to adjust parameters of membership function, thus changing shape of MF, for each fuzzy goal. Furthermore, the change in robot configuration, like change in sensor arrangement or sensing range, invokes another adjusting of MFs. To accomplish an intelligent way to adjust these parameters, we adopted a genetic algorithm, which does not require any formulation of the problem, thus more appropriate for robot navigation. Genetic algorithm generates the fittest parameter set through crossover and mutation operation of its string representation. The fitness of a parameter set is assigned after a simulation run according to its time of travel, accumulated heading angle change and collision. A series of simulations for several different environments is carried out to verify the proposed method. The results show the optimal parameters can be acquired with this method.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1355-1360
• /
• 2005

In this paper, we approach the problem of image filter design automation using a kind of intrinsic evolvable hardware architecture. For the purpose of implementing the intrinsic evolution process in a common FPGA chip and evolving a complicated digital circuit system-image filter, the design automation system employs the reconfigurable circuit architecture as the reconfigurable component of the EHW. The reconfigurable circuit architecture is inspired by the Cartesian Genetic Programming and the functional level evolution. To increase the speed of the hardware evolution, the whole evolvable hardware system which consists of evolution algorithm unit, fitness value calculation unit and reconfigurable unit are implemented by a commercial FPGA chip. The Celoxica RC1000 card which is fitted with a Xilinx Virtex xcv2000E FPGA chip is employed as the experiment platform. As the result, we conclude the terms of the synthesis report of the image filter design automation system and hardware evolution speed in the Celoxica RC1000 card. The evolved image filter is also compared with the conventional image filter form the point of filtered image quality.