• Smart Structures and Systems
• /
• v.31 no.3
• /
• pp.247-257
• /
• 2023

In structural health monitoring of large-scale structures, optimal sensor placement plays an important role because of the high cost of sensors and their supporting instruments, as well as the burden of data transmission and storage. In this study, a vibration sensor placement algorithm based on deep reinforcement learning (DRL) is proposed, which can effectively solve non-convex, high-dimensional, and discrete combinatorial sensor placement optimization problems. An objective function is constructed to estimate the quality of a specific vibration sensor placement scheme according to the modal assurance criterion (MAC). Using this objective function, a DRL-based algorithm is presented to determine the optimal vibration sensor placement scheme. Subsequently, we transform the sensor optimal placement process into a Markov decision process and employ a DRL-based optimization algorithm to maximize the objective function for optimal sensor placement. To illustrate the applicability of the proposed method, two examples are presented: a 10-story braced frame and a sea-crossing bridge model. A comparison study is also performed with a genetic algorithm and particle swarm algorithm. The proposed DRL-based algorithm can effectively solve the discrete combinatorial optimization problem for vibration sensor placements and can produce superior performance compared with the other two existing methods.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.10 no.2
• /
• pp.193-200
• /
• 2002

Simple spring-damper-mass models have been widely used to investigate whole-body vortical biodynamic response characteristics of the seated vehicle driver. Most previous models have not considered the effect of wobbling masses; i.e. heart, lungs, liver, intestine, etc. In this study, 4 -DOF seated driver model including one non-rigid mass representing wobbling visceral mass, 5-DOF model including intestine, and 10-DOF model including five lumbar vertebral masses were proposed. The model parameters were identified by a combinatorial optimization technique. simulated annealing method. The objective function was chosen as the sum of error between model response of seat-to-head transmissibility and driving point mechanical impedance and those of experimental data for subjects seated erect without backrest support. The model response showed a good agreement with the experimental response characteristics. Using a 10-DOF model, calculated resonance frequency of lumbar spine at 4Hz was matched well with experimental results of Panjabi et al.

• School Mathematics
• /
• v.11 no.1
• /
• pp.17-37
• /
• 2009

Concept and model of multiplication is not single. Concepts of multiplication can be classified into three cases: repeated addition, times idea, pairs set. Models of multiplication can be classified into four cases: measurement, rectangular pattern, combinatorial problem, number line. Among diverse cases of multiplication's concept and model, which case does elementary mathematics education lay stress on? This question is a controvertible didactical point. In this thesis, (1) mathematical and didactical analysis of multiplication's concept and model is performed, (2) a concrete program of teaching multiplication which is based on times idea is contrived, (3) With this new program, the teaching experiment is performed and its result is analyzed. Through this study, I obtained the following results and suggestions. First, the degree of testee's understanding of times idea is not high. Secondly, a sort of test problem which asks the testee to find times value is more easy than the one to find multiplicative resulting value. Thirdly, combinatorial problem can be handled as an application of multiplication. Fourthly, the degree of testee's understanding of repeated addition is high. In conclusion, I observe the fact that this new program which is based on times idea could be a alternative program of teaching multiplication which could complement the traditional method.

• Computers and Concrete
• /
• v.25 no.5
• /
• pp.401-409
• /
• 2020

Artificial neural networks are used as a useful tool in distinct fields of civil engineering these days. In order to control long-term quality of Self-Compacting Semi-Lightweight Concrete (SCSLC), the 90 days compressive strength is considered as a key issue in this paper. In fact, combined artificial neural networks are used to predict the compressive strength of SCSLC at 28 and 90 days. These networks are able to re-establish non-linear and complex relationships straightforwardly. In this study, two types of neural networks, including Radial Basis and Multilayer Perceptron, were used. Four groups of concrete mix designs also were made with two water to cement ratios (W/C) of 0.35 and 0.4, as well as 10% of cement weight was replaced with silica fume in half of the mixes, and different amounts of superplasticizer were used. With the help of rheology test and compressive strength results at 7 and 14 days as inputs, the neural networks were used to estimate the 28 and 90 days compressive strengths of above-mentioned mixes. It was necessary to add the 14 days compressive strength in the input layer to gain acceptable results for 90 days compressive strength. Then proper neural networks were prepared for each mix, following which four existing networks were combined, and the combinatorial neural network model properly predicted the compressive strength of different mix designs.

• Industrial Engineering and Management Systems
• /
• v.11 no.4
• /
• pp.310-330
• /
• 2012

Scheduling is an important tool for a manufacturing system, where it can have a major impact on the productivity of a production process. In manufacturing systems, the purpose of scheduling is to minimize the production time and costs, by assigning a production facility when to make, with which staff, and on which equipment. Production scheduling aims to maximize the efficiency of the operation and reduce the costs. In order to find an optimal solution to manufacturing scheduling problems, it attempts to solve complex combinatorial optimization problems. Unfortunately, most of them fall into the class of NP-hard combinatorial problems. Genetic algorithm (GA) is one of the generic population-based metaheuristic optimization algorithms and the best one for finding a satisfactory solution in an acceptable time for the NP-hard scheduling problems. GA is the most popular type of evolutionary algorithm. In this survey paper, we address firstly multiobjective hybrid GA combined with adaptive fuzzy logic controller which gives fitness assignment mechanism and performance measures for solving multiple objective optimization problems, and four crucial issues in the manufacturing scheduling including a mathematical model, GA-based solution method and case study in flexible job-shop scheduling problem (fJSP), automatic guided vehicle (AGV) dispatching models in flexible manufacturing system (FMS) combined with priority-based GA, recent advanced planning and scheduling (APS) models and integrated systems for manufacturing.

• Macromolecular Research
• /
• v.16 no.4
• /
• pp.320-328
• /
• 2008

We established a thermodynamic framework of group contribution method based on modified double lattice (MDL) model. The proposed model included the long-range interaction contribution caused by the Coulomb electrostatic forces, the middle-range interaction contribution from the indirect effects of the charge interactions and the short-range interaction from modified double lattice model. The group contribution method explained the combinatorial energy contribution responsible for the revised Flory-Huggins entropy of mixing, the van der Waals energy contribution from dispersion, the polar force, and the specific energy contribution from hydrogen bonding. We showed the solvent activities of various polymer solution systems in comparison with theoretical predictions based on experimental data. The proposed model gave a very good agreement with the experimental data.

• Journal of Korean Institute of Industrial Engineers
• /
• v.34 no.2
• /
• pp.181-189
• /
• 2008

Survivability of a network is one of the most important issues in designing present-day communication networks. For the past few decades, a lot of researches have proposed the mathematical models to evaluate the survivability of networks. In this paper, we attempt to survey such researches and classify them based on how these researches measure the survivability of a network.

• Proceedings of the PSK Conference
• /
• 2002.10a
• /
• pp.92-93
• /
• 2002

With the recent development of combinatorial chemistry, recombinant biotechnology and rational drug design, millions of compounds are being produced in the laboratories of pharmaceutical companies. These new drug candidates are evaluated their efficacy and toxicity through in vivo animal model studies which is very important in drug development. From these studies, very successful drug candidates are selected. (omitted)

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.69.2-69
• /
• 2002

Process measurements contain random and gross errors and the size estimation of gross errors is required for production accounting. Mixed integer programming technique had been applied to identify and estimate the gross errors simultaneously. However, the compensate model based on mixed integer programming used all measured variables or spanning tree as gross error candidates. This makes gross error estimation problem combinatorial or computationally expensive. Mixed integer programming with test statistics is proposed for computationally inexpensive gross error identification /estimation. The gross error candidates are identified by measurement test and the set of gross error candidates are...

• Proceedings of the KSME Conference
• /
• 2000.11a
• /
• pp.381-386
• /
• 2000

Simple spring-damper-mass models have been widely used to understand whole-body vertical biodynamic response characteristics of the seated vehicle driver. However, most previous models have not considered about the non-rigid masses(wobbling masses). A simple mechanical model of seated human body developed in this study included the torso represented by a rigid and a wobbling mass. Within the 0.5-20Hz frequency range and for excitation amplitudes maintained below $5ms^{-2}$, this 4-degree-of-freedom driver model is proposed to satisfy the measured vertical vibration response characteristics defined from a synthesis of published data for subjects seated erect without backrest support. The parameters are identified by using the combinatorial optimization technique, simulated annealing method. The model response was found to be provided a closer agreement with the response characteristics than previously published models.