• Smart Structures and Systems
• /
• v.29 no.5
• /
• pp.677-691
• /
• 2022

Water cycle algorithm (WCA) has been a very effective optimization technique for complex engineering problems. This study employs the WCA for simultaneous prediction of heating load (L_H) and cooling load (L_C) in residential buildings. This algorithm is responsible for optimally tuning a neural network (NN). Utilizing 614 records, the behavior of the L_H and L_C is explored and the captured knowledge is then used to predict for 154 unanalyzed building conditions. Since the WCA is a population-based algorithm, different numbers of the searching agents were tested to find the most optimum configuration. It was observed that the best solution is discovered by 500 agents. A comparison with five newly-developed benchmark optimizers, namely equilibrium optimizer (EO), multi-tracker optimization algorithm (MTOA), slime mould algorithm (SMA), multi-verse optimizer (MVO), and electromagnetic field optimization (EFO) revealed that the WCANN predicts the desired parameters with considerably larger accuracy. Obtained root mean square errors (1.4866, 2.1296, 2.8279, 2.5727, 2.5337, and 2.3029 for the L_H and 2.1767, 2.6459, 3.1821, 2.9732, 2.9616, and 2.6890 for the L_C) indicated that the most reliable prediction was presented by the proposed model. The EFONN, however, provided a more time-effective solution. Lastly, an explicit predictive formula was elicited from the WCANN.

• Steel and Composite Structures
• /
• v.49 no.6
• /
• pp.633-643
• /
• 2023

In this study, the natural frequencies of Functional Graded Materials (FGM) plates are predicted using Artificial Neural Network (ANN). A model based on Third-order Shear Deformation Theory (TSDT) and FEM is used to train the ANN model. Different training methods are tested to simulate input and output dependency. As this is a parametric model, several architectures and optimization algorithms were tested. The proposed model allows us to minimize the CPU time to evaluate candidate material properties for FGM plate material selection and demonstrate their influence on dynamic behavior. Consequently, the time required for the FGM design process (candidate materials for material selection) and the geometric optimization of the FGM structure would remain reasonable. The ANN model can help industries to produce FGM plates with good mechanical properties of the selected materials. I addition, this model can be used to directly predict vibration behavior by testing a large number of FGM plates, representing all possible combinations of metals and ceramics in today's industry, without having to solve any eigenvalue problems.

• The Korean Journal of Applied Statistics
• /
• v.26 no.5
• /
• pp.847-856
• /
• 2013

We propose a modified learning process for generalized neural network using a learning algorithm by Liu et al. (2001). We consider the effect of initial weights, training results and learning errors using a modified learning process. We employ an incremental training procedure where training patterns are learned systematically. Our algorithm starts with a single training pattern and a single hidden layer neuron. During the course of neural network training, we try to escape from the local minimum by using a weight scaling technique. We allow the network to grow by adding a hidden layer neuron only after several consecutive failed attempts to escape from a local minimum. Our optimization procedure tends to make the network reach the error tolerance with no or little training after the addition of a hidden layer neuron. Simulation results with suitable initial weights indicate that the present constructive algorithm can obtain neural networks very close to minimal structures and that convergence to a solution in neural network training can be guaranteed. We tested these algorithms extensively with small training sets.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.3
• /
• pp.175-183
• /
• 2008

In wireless sensor networks with ad hoc networking capability, sensor nodes are battery operated and are usually disposable once deployed. As a result, each sensor node senses and communicates with limited energy and, thus, energy efficiency has been studied as a key design factor which determines lifetime of a wireless sensor network, and it is more improved recently by using so-called cross-layer optimization technique. In this paper, we propose and implement a new energy saving mechanism that reduces energy consumption during data collection by controlling transmission power at sensor nodes and then measure its performance in terms of lifetime improvement for the wireless sensor network platform ZigbeX. When every sensor node transmits sensed data to its clusterhead, it controls its transmission power down to as low level as communication is possible, resulting in energy saving. Each sensor node controls its transmission power based on RSSI(Received Signal Strength Indicator) of the packet received from its clusterhead. In other words, the sensor node can save energy by controlling its transmission power down to an appropriate level that its clusterhead safely receives the packet it transmits. According to the repetitive experiment of the proposed scheme on the ZigbeX platform using the packet analyzer developed by us, it is observed that the network lifetime is prolonged by up to 21.9% by saying energy during the data collection occupying most amount of network traffic.

• Journal of Korea Water Resources Association
• /
• v.32 no.1
• /
• pp.71-81
• /
• 1999

The objective of this study is to develop a model which can design an optimal pipe network system of least cost while satisfying all the design constraints including hydraulic constraints using a genetic algorithm technique. Hydraulic constraints interfaced with the simulation program(KYPIPE) checked feasible solution region. Genetic algorithm(GA) technique is a relatively new optimization technique. The GA is known as a very powerful search and optimization technique especially when solving nonlinear programming problems. The model developed in this study selects optimal pipe diameters in the form of commercial discrete sizes using the pipe diameters and the pumping powers as decision variables. The model not only determines the optimal diameters and pumping powers of pipe network system but also satisfies the discharge and pressure requirements at demanding nodes. The model has been applied to an imaginary and an existing pipe network systems. One system is adopted from journal papers which has been used as an example network by many other researchers. Comparison of the results shows compatibility of the model developed in this study. The model is also applied to a system in Goyang city in order to check the model applicability to finding of optimal pumping powers. It has been found that the developed model can be successfully applied to optimal design of pipe network systems in a relatively simple manner.

• Journal of KIISE:Databases
• /
• v.36 no.1
• /
• pp.8-21
• /
• 2009

A wireless sensor network is a computer network which consists of spatially distributed devices, called sensor nodes. In wireless sensor networks, energy efficiency is a key issue since sensor nodes must resides upon limited energy. To retrieve sensor information without dealing with the network issues, a sensor network is treated as conceptual database on which query can be requested. When multiple queries are requested for processing in a wireless sensor network, energy consumption can be significantly reduced if common partial results among similar queries can be effectively shared. In this paper, we propose an energy efficient multi-query processing technique based on the coverage relationship between multiple queries. When a new query is requested, our proposed technique derives an equivalent query from queries running at the moment, if it is derivable. Our technique first computes the set of running queries that may derive a partial result of the new query and then test if this set covers all the result of the new query attribute-wise and tuple-wise. If the result of the new query can be derived from the results of executing queries, the new query derives its result at the base station instead of being executed in the sensor network.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.6
• /
• pp.1589-1600
• /
• 1997

The construction of optical access network costs upper 40% in total installation cost of total optical network. Optimization of access network therefore is core in optical network. Advanced countries include EU, Japan and USA already have researched access network. This paper presents analysis of three broad-band fiber-optics subscriber loop architectures(HFC, ATM-PON, Super PON). The analyses focus on the specific demonstrated architectures and use component cost projections to estimate future network costs on a per-subscriber basis. We use TITAN(Tool for Introduction Scenarios and Techno-Economic Evaluation of The Access Network) model. We find that ATM-PON can deliver voice and ISDN data at installed first costs than the other architectures. This is due to the sharing bandwidth among a cluster of subscribers within Curb. This work intends to support establishing guidelines for strategic decisions regarding the development of the access network alternatives of different operators.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.1
• /
• pp.78-89
• /
• 2006

In development of network-enabled consumer electronics, much of the time and effort is spent analyzing and solving network performance problems. In this paper, we define an instance of such problems discovered while developing a commercial home network gateway. We then analyze its cause and propose a solution mechanism. Our home network gateway uses art asymmetric link (ADSL) and suffers from an undesirable phenomenon where downlink traffic interferes with upload speed. We call this phenomenon the network performance interference problem. While this problem can easily be confused with receive livelock caused by packet contention at the input queue, we and that this is not the case. By performing extensive experiments and analysis, we reveal that our problem is caused by packet contention at the output queue and certain intrinsic characteristics of TCP. We devise an ACK-separation queuing mechanism for this problem and implement it in the home network gateway Our experiments show that it effectively solves the problem.

• Convergence Security Journal
• /
• v.11 no.1
• /
• pp.11-18
• /
• 2011

Wireless mesh network technology with transmission speeds similar to wired and wireless technology means to build, compared with wired networks, building a more efficient network to provide convenience and flexibility. The wireless mesh network router nodes in the energy impact of the mobility is less constrained and has fewer features entail. However, the characteristics of various kinds due to network configuration settings and the choice of multiple paths that can occur when the system overhead and there are many details that must be considered. Therefore, according to the characteristics of these network routing technology that is reflected in the design and optimization of the network is worth noting. In this paper, a multi-path setting can be raised in order to respond effectively to the problem of the router node data loss and bandwidth according to traffic conditions and links to elements of the hop count evaluation by using a genetic algorithm as a workaround for dynamic routing the routing metric for wireless mesh network scheme is proposed.

• Journal of KIISE:Information Networking
• /
• v.29 no.4
• /
• pp.386-396
• /
• 2002

This paper presents a neural network-based near-optimal routing algorithm. It employs a modified Hopfield Neural Network (MHNN) as a means to solve the shortest path problem. It uses every piece of information that is available at the peripheral neurons in addition to the highly correlated information that is available at the local neuron. Consequently, every neuron converges speedily and optimally to a stable state. The convergence is faster than what is usually found in algorithms that employ conventional Hopfield neural networks. Computer simulations support the indicated claims. The results are relatively independent of network topology for almost all source-destination pairs, which nay be useful for implementing the routing algorithms appropriate to multi -hop packet radio networks with time-varying network topology.