• Title/Summary/Keyword: neural network optimization

Search Result 818, Processing Time 0.03 seconds

A Study on the Relationship Between Welding Variables and Bead Width Using a Neural Network (신경회로망을 이용한 용접공정변수와 비드폭과의 상관관계에 관한 연구)

  • Kim, I. J.;Park, C. U.;Kim, I. S.;Park, S. Y.;Jeong, Y. J.;Lim, H.;Park, J. S.
    • Proceedings of the Korean Society of Precision Engineering Conference
    • /
    • 2000.11a
    • /
    • pp.699-702
    • /
    • 2000
  • The automation and control of robotic welding process is a very complex assignment because the system is affected by a number of variables which are very difficult to determine or predict in practice. Not only the optimization of the robotic welding process is considered from the point of view of the time and the cost of manufacturing. as well as quality of the weldment. the human factors of the production and many other factors must taken into consideration. hi order to determine the optimal parameters of robotic welding process, it is necessary to build a computer model representing all parameters influencing the welding process as well as the mutual dependence between them. This paper presents an approach to modeling the robotic welding process in which all parameters affecting the welding process are included using a neural network. A detailed analysis of the simulation results has been carried out to evaluate the proposed neural network model.

  • PDF

Determination of Optimum Heating Regions for Thermal Prestressing Method Using Artificial Neural Network (인공신경망을 이용한 온도프리스트레싱 공법의 적정 가열구간 설정에 관한 연구)

  • 김상효;김준환;김강미
    • Proceedings of the Computational Structural Engineering Institute Conference
    • /
    • 2003.04a
    • /
    • pp.327-334
    • /
    • 2003
  • Thermal Prestressing Method for continuous composite girder bridges is a new design and construction method developed to induce initial composite stresses in the concrete slab at negative bending regions. Due to the induced initial stresses, prevention of tensile cracks at concrete slab, reduction of steel girder section, and reduction of reinforcing bars are possible. Thus, economical and construction efficiency can be improved. Method for determining optimum heating region of Thermal Prestressing Method, has not been established although such method is essential for increasing efficiency of the designing process. Trial-and-error method used in previous studies is far from efficient and more rational method for computing optimal heating region is required. In this study, efficient method for determining optimum heating region in the use of Thermal Prestressing Method is developed based on artificial neural network algorithm, which is widely adopted to pattern recognition, optimization, diagnosis, and estimation problems in various fields. Back-propagation algorithm, which is commonly used as a learning algorithm in neural network problems, is used for training of the neural network. Through case studies of 2-span continuous and 3-span continuous composite girder bridges using the developed process, the optimal heating regions are obtained.

  • PDF

Neural Network Modeling of Ion Energy Impact on Surface Roughness of SiN Thin Films (신경망을 이용한 SiN 박막 표면거칠기에의 이온에너지 영향 모델링)

  • Kim, Byung-Whan;Lee, Joo-Kong
    • Journal of Surface Science and Engineering
    • /
    • v.43 no.3
    • /
    • pp.159-164
    • /
    • 2010
  • Surface roughness of deposited or etched film strongly depends on ion bombardment. Relationships between ion bombardment variables and surface roughness are too complicated to model analytically. To overcome this, an empirical neural network model was constructed and applied to a deposition process of silicon nitride (SiN) films. The films were deposited by using a pulsed plasma enhanced chemical vapor deposition system in $SiH_4$-$NH_4$ plasma. Radio frequency source power and duty ratio were varied in the range of 200-800 W and 40-100%. A total of 20 experiments were conducted. A non-invasive ion energy analyzer was used to collect ion energy distribution. The diagnostic variables examined include high (or) low ion energy and high (or low) ion energy flux. Mean surface roughness was measured by using atomic force microscopy. A neural network model relating the diagnostic variables to the surface roughness was constructed and its prediction performance was optimized by using a genetic algorithm. The optimized model yielded an improved performance of about 58% over statistical regression model. The model revealed very interesting features useful for optimization of surface roughness. This includes a reduction in surface roughness either by an increase in ion energy flux at lower ion energy or by an increase in higher ion energy at lower ion energy flux.

Graph neural network based multiple accident diagnosis in nuclear power plants: Data optimization to represent the system configuration

  • Chae, Young Ho;Lee, Chanyoung;Han, Sang Min;Seong, Poong Hyun
    • Nuclear Engineering and Technology
    • /
    • v.54 no.8
    • /
    • pp.2859-2870
    • /
    • 2022
  • Because nuclear power plants (NPPs) are safety-critical infrastructure, it is essential to increase their safety and minimize risk. To reduce human error and support decision-making by operators, several artificial-intelligence-based diagnosis methods have been proposed. However, because of the nature of data-driven methods, conventional artificial intelligence requires large amount of measurement values to train and achieve enough diagnosis resolution. We propose a graph neural network (GNN) based accident diagnosis algorithm to achieve high diagnosis resolution with limited measurements. The proposed algorithm is trained with both the knowledge about physical correlation between components and measurement values. To validate the proposed methodology has a sufficiently high diagnostic resolution with limited measurement values, the diagnosis of multiple accidents was performed with limited measurement values and also, the performance was compared with convolution neural network (CNN). In case of the experiment that requires low diagnostic resolution, both CNN and GNN showed good results. However, for the tests that requires high diagnostic resolution, GNN greatly outperformed the CNN.

A Method for Optimizing the Structure of Neural Networks Based on Information Entropy

  • Yuan Hongchun;Xiong Fanlnu;Kei, Bai-Shi
    • Proceedings of the Korea Inteligent Information System Society Conference
    • /
    • 2001.01a
    • /
    • pp.30-33
    • /
    • 2001
  • The number of hidden neurons of the feed-forward neural networks is generally decided on the basis of experience. The method usually results in the lack or redundancy of hidden neurons, and causes the shortage of capacity for storing information of learning overmuch. This research proposes a new method for optimizing the number of hidden neurons bases on information entropy, Firstly, an initial neural network with enough hidden neurons should be trained by a set of training samples. Second, the activation values of hidden neurons should be calculated by inputting the training samples that can be identified correctly by the trained neural network. Third, all kinds of partitions should be tried and its information gain should be calculated, and then a decision-tree correctly dividing the whole sample space can be constructed. Finally, the important and related hidden neurons that are included in the tree can be found by searching the whole tree, and other redundant hidden neurons can be deleted. Thus, the number of hidden neurons can be decided. In the case of building a neural network with the best number of hidden units for tea quality evaluation, the proposed method is applied. And the result shows that the method is effective

  • PDF

Optimal algorithm of part-matching process using neural network (신경 회로망을 이용한 부품 조립 공정의 최적화 알고리즘)

  • 오제휘;차영엽
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 1996.10b
    • /
    • pp.143-146
    • /
    • 1996
  • In this paper, we propose a hopfield model for solving the part-matching which is the number of parts and positions are changed. The goal of this paper is to minimize part-connection in pairs and net total path of part-connection. Therefore, this kind of problem is referred to as a combinatorial optimization problem. First of all, we review the theoretical basis for hopfield model to optimization and present two method of part-matching; Traveling Salesman Problem (TSP) and Weighted Matching Problem (WMP). Finally, we show demonstration through computer simulation and analyzes the stability and feasibility of the generated solutions for the proposed connection methods.

  • PDF

Efficient Fixed-Point Representation for ResNet-50 Convolutional Neural Network (ResNet-50 합성곱 신경망을 위한 고정 소수점 표현 방법)

  • Kang, Hyeong-Ju
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.22 no.1
    • /
    • pp.1-8
    • /
    • 2018
  • Recently, the convolutional neural network shows high performance in many computer vision tasks. However, convolutional neural networks require enormous amount of operation, so it is difficult to adopt them in the embedded environments. To solve this problem, many studies are performed on the ASIC or FPGA implementation, where an efficient representation method is required. The fixed-point representation is adequate for the ASIC or FPGA implementation but causes a performance degradation. This paper proposes a separate optimization of representations for the convolutional layers and the batch normalization layers. With the proposed method, the required bit width for the convolutional layers is reduced from 16 bits to 10 bits for the ResNet-50 neural network. Since the computation amount of the convolutional layers occupies the most of the entire computation, the bit width reduction in the convolutional layers enables the efficient implementation of the convolutional neural networks.

A study on the performance improvement of the quality prediction neural network of injection molded products reflecting the process conditions and quality characteristics of molded products by process step based on multi-tasking learning structure (다중 작업 학습 구조 기반 공정단계별 공정조건 및 성형품의 품질 특성을 반영한 사출성형품 품질 예측 신경망의 성능 개선에 대한 연구)

  • Hyo-Eun Lee;Jun-Han Lee;Jong-Sun Kim;Gu-Young Cho
    • Design & Manufacturing
    • /
    • v.17 no.4
    • /
    • pp.72-78
    • /
    • 2023
  • Injection molding is a process widely used in various industries because of its high production speed and ease of mass production during the plastic manufacturing process, and the product is molded by injecting molten plastic into the mold at high speed and pressure. Since process conditions such as resin and mold temperature mutually affect the process and the quality of the molded product, it is difficult to accurately predict quality through mathematical or statistical methods. Recently, studies to predict the quality of injection molded products by applying artificial neural networks, which are known to be very useful for analyzing nonlinear types of problems, are actively underway. In this study, structural optimization of neural networks was conducted by applying multi-task learning techniques according to the characteristics of the input and output parameters of the artificial neural network. A structure reflecting the characteristics of each process step was applied to the input parameters, and a structure reflecting the quality characteristics of the injection molded part was applied to the output parameters using multi-tasking learning. Building an artificial neural network to predict the three qualities (mass, diameter, height) of injection-molded product under six process conditions (melt temperature, mold temperature, injection speed, packing pressure, pacing time, cooling time) and comparing its performance with the existing neural network, we observed enhancements in prediction accuracy for mass, diameter, and height by approximately 69.38%, 24.87%, and 39.87%, respectively.

K-Means-Based Polynomial-Radial Basis Function Neural Network Using Space Search Algorithm: Design and Comparative Studies (공간 탐색 최적화 알고리즘을 이용한 K-Means 클러스터링 기반 다항식 방사형 기저 함수 신경회로망: 설계 및 비교 해석)

  • Kim, Wook-Dong;Oh, Sung-Kwun
    • Journal of Institute of Control, Robotics and Systems
    • /
    • v.17 no.8
    • /
    • pp.731-738
    • /
    • 2011
  • In this paper, we introduce an advanced architecture of K-Means clustering-based polynomial Radial Basis Function Neural Networks (p-RBFNNs) designed with the aid of SSOA (Space Search Optimization Algorithm) and develop a comprehensive design methodology supporting their construction. In order to design the optimized p-RBFNNs, a center value of each receptive field is determined by running the K-Means clustering algorithm and then the center value and the width of the corresponding receptive field are optimized through SSOA. The connections (weights) of the proposed p-RBFNNs are of functional character and are realized by considering three types of polynomials. In addition, a WLSE (Weighted Least Square Estimation) is used to estimate the coefficients of polynomials (serving as functional connections of the network) of each node from output node. Therefore, a local learning capability and an interpretability of the proposed model are improved. The proposed model is illustrated with the use of nonlinear function, NOx called Machine Learning dataset. A comparative analysis reveals that the proposed model exhibits higher accuracy and superb predictive capability in comparison to some previous models available in the literature.

Swarm-based hybridizations of neural network for predicting the concrete strength

  • Ma, Xinyan;Foong, Loke Kok;Morasaei, Armin;Ghabussi, Aria;Lyu, Zongjie
    • Smart Structures and Systems
    • /
    • v.26 no.2
    • /
    • pp.241-251
    • /
    • 2020
  • Due to the undeniable importance of approximating the concrete compressive strength (CSC) in civil engineering, this paper focuses on presenting four novel optimizations of multi-layer perceptron (MLP) neural network, namely artificial bee colony (ABC-MLP), grasshopper optimization algorithm (GOA-MLP), shuffled frog leaping algorithm (SFLA-MLP), and salp swarm algorithm (SSA-MLP) for predicting this crucial parameter. The used dataset consists of 103 rows of information concerning seven influential parameters (cement, slag, water, fly ash, superplasticizer, fine aggregate, and coarse aggregate). In this work, the best-fitted complexity of each ensemble is determined by a population-based sensitivity analysis. The GOA distinguished its self by the least complexity (population size = 50) and emerged as the second time-effective optimizer. Referring to the prediction results, all tested algorithms are able to construct reliable networks. However, the SSA (Correlation = 0.9652 and Error = 1.3939) and GOA (Correlation = 0.9629 and Error = 1.3922) performed more accurately than ABC (Correlation = 0.7060 and Error = 4.0161) and SFLA (Correlation = 0.8890 and Error = 2.5480). Therefore, the SSA-MLP and GOA-MLP can be promising alternatives to laboratorial and traditional CSC evaluative methods.