Two Layer Multiquadric-Biharmonic Artificial Neural Network for Area Quasigeoid Surface Approximation with GPS-Levelling Data

Xingsheng Deng,Xinzhou Wang 한국항해항만학회 2006 한국항해항만학회 학술대회논문집 Vol.2 No.-

The geoidal undulations are needed for determining the orthometric heights from the Global Positioning System GPS-derived ellipsoidal heights. There are several methods for geoidal undulation determination. The paper presents a method employing a simple architecture Two Layer Multiquadric-Biharmonic Artificial Neural Network (TLMB-ANN) to approximate an area of 4200 square kilometres quasigeoid surface with GPS-levelling data. Hardy’s Multiquadric-Biharmonic functions is used as the hidden layer neurons’ activation function and Levenberg-Marquardt algorithm is used to train the artificial neural network. In numerical examples five surfaces were compared: the gravimetric geometry hybrid quasigeoid, Support Vector Machine (SVM) model, Hybrid Fuzzy Neural Network (HFNN) model, Traditional Three Layer Artificial Neural Network (ANN) with tanh activation function and TLMB-ANN surface approximation. The effectiveness of TLMB-ANN surface approximation depends on the number of control points. If the number of well-distributed control points is sufficiently large, the results are similar with those obtained by gravity and geometry hybrid method. Importantly, TLMB-ANN surface approximation model possesses good extrapolation performance with high precision.

공유가치창출(CSV)이 사회적성과와 협력기업성과에 미치는 파급효과에 관한 연구 : 인공신경망(ANN) 분석방법을 중심으로

방원석(Won-Seok Ba),N. S. Reddy,신재익(, Jae-Ik Shin) 한국자료분석학회 2021 Journal of the Korean Data Analysis Society Vol.23 No.3

본 논문은 기업의 지속적인 경쟁우위 요소로 관심을 받고 있는 공유가치창출(CSV)과 사회적 성과, 협력기업성과에 미치는 파급효과를 인공신경망(Artificial Neural Network) 분석방법을 중심으로 분석하고자 한다. 연구대상은 항공제조기업의 근무자를 대상으로 설문 조사를 실시, 234부를 최종분석에 사용하였다. 분석방법으로 기존의 구조방정식(AMOS)의 경로분석과 인공신경망(ANN) 분석방법을 비교하고 이에 따른 ANN 분석방법이 주는 의미와 시사점을 제시하였다. 분석결과, CSV의 하위요인인 경제적가치, 환경적가치, 사회적가치가 사회적성과 및 협력기업성과에 미치는 영향을 AMOS를 통한 경로분석결과와 ANN분석결과 CSV의 하위요인과 사회적성과 및 협력기업성과 간에는 유의미한 긍정적 관계가 있는 것으로 밝혀졌다. 결과적으로 AMOS를 활용한 분석방법의 선형적 연구결과와 달리, ANN분석방법을 통한 연구결과는 비선형적 연구결과를 보여주었으며, 이러한 연구논문의 결과를 바탕으로, 기업이 CSV활동을 통해 사회적성과와 협력기업성과를 달성할 수 있는 동반성장전략임을 규명하고 있으며, 의미 있는 학문적, 실무적 시사점을 제시하고 있다. This paper aims to analyze the ripple effect on the creation of shared value (CSV), social performance, and partner company performance, which are attracting attention as an element of continuous competitive advantage of companies, focusing on the analysis method of artificial neural networks. The subject of this study was a questionnaire survey for workers in aviation manufacturing companies, and 234 copies were used for the final analysis. As an analysis method, path analysis of the existing structural equations and artificial neural networks (ANNs) analysis methods were compared, and the meaning and implications of the ANNs analysis method were presented. As a result of the analysis, the impact of the sub-factors of CSV, such as economic value, environmental value, and social value, on social performance and partner business performance, was determined between the results of path analysis through AMOS and ANNs analysis results. It was found that there was a significant positive relationship. As a result, based on the results of this research paper derived through the ANN analysis method, it has been identified as a win-win growth strategy that enables companies to achieve social and partner business performance through CSV activities, and has meaningful academic and practical implications. Are presented.

유전자 적응형 시간 지연 신경망(GATDNN)을 이용한 바덴포(Bardenpho) 고도하수처리 공정의 성능 예측

Yoonseok Timothy Hong,백병천 한국도시환경학회 2018 한국도시환경학회지 Vol.18 No.3

Wastewater treatment systems are characterized by large temporal variability of inflow, variable concentrations of components in the incoming wastewater to the plant, and highly variable biological reactions within the process. The behavior of observed process variables within a wastewater treatment plant (WWTP) at a certain time instant is the combined effect of various processes initiated at different moments in the past. This is called a time-delay effect in the system. Due to the nature of strong nonlinear mapping, neural networks provide advantages as a modeling and identification tool over a structure-based model. However, the determination of the architecture of the artificial neural networks (ANNs) and the selection of key input variables with a time delay is not easy. In our research, a genetic adapted time-delay neural network (GATDNN), which is a combination of time-delay neural network (TDNN) and genetic algorithms (GAs), was developed and applied to the full-scale Bardenpho advanced sewage treatment process. In a GATDNN, a three-step modelling procedure was performed: (1) selection of significant input variables to maximise the predictive accuracy for each specific output; (2) finding a suitable network topology for the ANN-based process estimator;(3) sensitivity analysis. The results demonstrate that the modelling technique presented using a GATDNN provides a valuable tool for predicting the outputs with high levels of accuracy and identifying key operating variables. This work will permit the development of a reliable control strategy thus reducing the burden of the process engineer. 하폐수처리공정은 복잡한 생물학적 처리과정과 많은 입력변수들의 시간에 따른 강력한 동적인 변동성으로 인해서 이전의 운전 상황이 차후의 운전상황에 많은 영향을 미친다. 이런한 현상을 시간지연 효과라 한다. 강력한 비선형 매핑(mapping) 특성 때문에 신경망은 구조 기반 모델에 대한 모델링 및 식별 도구로서의 이점을 제공하지만, 일반 인공신경망(ANN)의 경우 시스템 설계방식(architecture)의 결정 및 시스템에서의 시간 지연과 관련된 주요 입력 변수의 선택은 쉽지 않다. 그래서 본 연구에서는 시간 지연 신경망과 유전자 알고리즘을 결합한 유전자 적응형 시간 지연 신경망(GATDNN)을 개발하여 바덴포(Bardenpho) 고도하수처리 공정 모델링에 적용하였다. GATDNN에서 3 단계 모델링 절차가 수행되었다. (1) 각 특정 출력에 대한 예측 정확도를 극대화하기 위해 중요한 입력 변수 선택 (2) ANN 기반 프로세스추정기에 적합한 네트워크 토폴로지(topology)를 찾기 (3) 민감도 분석. 모델링 결과 GATDNN을 사용하여 제시된 모델링 기술이 높은 정확도로 출력을 예측하고 주요 작동 변수를 식별하는 데 유용한 도구임을 보여주었다. 이 작업은 고도하수처리장 운전을 위한 신뢰할 수 있는 제어 전략의 개발을 가능하게 하여 엔지니어의 부담을 줄여 줄 수 있다.

Damage detection in structures using Particle Swarm Optimization combined with Artificial Neural Network

L. Nguyen-Ngoc,H. Tran-Ngoc,T. Bui-Tien,A. Mai-Duc,M. Abdel Wahab,Huan X. Nguyen,G. De Roeck 국제구조공학회 2021 Smart Structures and Systems, An International Jou Vol.28 No.1

In this paper, a novel approach to damage identification in structures using Particle Swarm Optimization (PSO) combined with Artificial neural network (ANN) is proposed. With recent substantial advances, ANN has been extensively utilized in a wide variety of fields. However, because of the application of backpropagation algorithms based on gradient descent techniques, ANN may be trapped in local minima when seeking the best solution. This may reduce the accuracy of ANN. Therefore, we propose employing an evolutionary algorithm, namely PSO to deal with the local minimum problems of ANN. PSO is employed to improve the training parameters of ANN consisting of weight and bias ratios by reducing the deviation between calculated and desired results. These training parameters are then used to train the network. Since PSO applies global search techniques to look for the best solution, it can assist the network in avoiding local minima by looking for a beneficial starting point. In order to assess the effectiveness of the proposed approach, both numerical and experimental models with different damage scenarios are employed. The results show that ANN -PSO not only significantly reduces computational time compared to PSO but also possibly identifies damages in the considered structures more accurately than ANN and PSO separately.

중소하천유역에서 유출량 산정을 위한 NEURAL NETWORK의 적용

최윤영 ( Yun Young Choi ) 한국수처리학회 2003 한국수처리학회지 Vol.11 No.1

In order to resolve the rainfall-runoff forecast model s uncertainty of model parameters and to increase the model s output, the study utilized Neural Networks model such as ANN and GRNN model. which mathematically interpret human thought processes. In order to calibrate and verify the runoff forecast model, seven flood events(1997-2002) observed at the Kumho water level gage station located on the midstream of Kumho river were chosen, of which five were used as calibration data and two as that for verification. First, as a result of applying ANN model, of the ten models, the ANN-8-8(10 input layer nodes, 8 nodes in the first hidden layer, 8 nodes in the second hidden layer, 1 output layer node) model was found to be the more suitable model in an actual hydro-event. Second, as a result of applying GRNN model, of the nine models, the GN-4a(l0 input layer nodes, 4 nodes in the pattern layer, spread σ=0.05, 1 output layer node) model was considered suitable. In addition, the numbers of pattern layer node increased, but it didn`t increase the learning rate. Finally, according to the results of the statistical analysis of the ANN and GRNN model, it was shown that ANN model was better. However, since there are numerous difficulties in determining the superiority of a model based on data only from the seven heavy rainfall events used in this study, it is judged that continued application and study based on more quantitative and qualitative data are necessary.

Multiple Network-on-Chip Model for High Performance Neural Network

Dong, Yiping,Li, Ce,Lin, Zhen,Watanabe, Takahiro The Institute of Electronics and Information Engin 2010 Journal of semiconductor technology and science Vol.10 No.1

Hardware implementation methods for Artificial Neural Network (ANN) have been researched for a long time to achieve high performance. We have proposed a Network on Chip (NoC) for ANN, and this architecture can reduce communication load and increase performance when an implemented ANN is small. In this paper, a multiple NoC models are proposed for ANN, which can implement both a small size ANN and a large size one. The simulation result shows that the proposed multiple NoC models can reduce communication load, increase system performance of connection-per-second (CPS), and reduce system running time compared with the existing hardware ANN. Furthermore, this architecture is reconfigurable and reparable. It can be used to implement different applications of ANN.

An Artificial Neural Network for Predicting the Physiochemical Properties of Fish Oil Microcapsules Obtained by Spray Drying

Mortaza Aghbashlo,Hossien Mobli,Shahin Rafiee,Ashkan Madadlou 한국식품과학회 2013 Food Science and Biotechnology Vol.22 No.3

The aim of this work was to develop an artificial neural network (ANN) to predict the physiochemical properties of fish oil microcapsules obtained by spray drying method. The relation amongst inlet-drying air temperature, outlet-drying air temperature, aspirator rate,peristaltic pump rate, and spraying air flow rate with 5performance indices, namely capsules’ residual moisture content, particle size, bulk density, encapsulation efficiency,and peroxide value was bridged by using ANN. A multilayer perceptron ANN was developed to predict the performance indices based on the input variables. The optimal ANN model was found to be a 5-10-5 structure with tangent sigmoid transfer function, Levenberg-Marquardt error minimization algorithm, and 1,000 training epochs. This optimal network was capable to predict the outputs with R2 values higher than 0.87. It was concluded that ANN is a useful tool to investigate, approximate, and predict the encapsulation characteristics of fish oil.

Development of Automatic Ship Berthing System Using Artificial Neural Network and Distance Measurement System

Van-Suong Nguyen,Van-Cuong Do,Nam-Kyun Im 한국지능시스템학회 2018 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.18 No.1

In the studies on an berthing control of ship, an artificial neural network (ANN) model is commonly employed as the main controller to control the rudder and the propeller. The existing ANN controllers that use the parameters consisting of the ship position and the ship heading as inputs cannot be applied to control automatically the ship into berth in different ports. To deal with this problem, the parameters, such as relative bearing and distance from ship to berth calculated by radar can be used as inputs for the controller. However, the calculation of these factors is not accurate because some errors arise on using radar for berthing process. This leads to the lack of confidence in ship berthing system using the parameters determined by radar. In this research, the neural network based-automatic berthing system is developed for ship by using the parameters which are measured by distance measurement system. By this proposed system, the ship is brought automatically into berth in different ports without retraining the neural network. In addition, this system guarantees that the parameters used for inputs of the neural network is measured exactly and continually. To validate the proposed algorithm, numerical simulations are carried out to two imaginary ports and a real port, and result showed the good performance of the proposed system for automatic ship berthing.

SEM-Artificial Neural Network 2단계 접근법에 의한 클라우드 스토리지 서비스 이용의도 영향요인에 관한 연구

Guangbo Jiang,권순동 한국데이터전략학회 2023 Journal of information technology applications & m Vol.30 No.6

This study aims to identify the influencing factors of intention to use cloud services using the SEM-ANN two-step approach. In previous studies of SEM-ANN, SEM presented R² and ANN presented MSE(mean squared error), so analysis performance could not be compared. In this study, R² and MSE were calculated and presented by SEM and ANN, respectively. Then, analysis performance was compared and feature importances were compared by sensitivity analysis. As a result, the ANN default model improved R² by 2.87 compared to the PLS model, showing a small Cohen's effect size. The ANN optimization model improved R² by 7.86 compared to the PLS model, showing a medium Cohen effect size. In normalized feature importances, the order of importances was the same for PLS and ANN. The contribution of this study, which links structural equation modeling to artificial intelligence, is that it verified the effect of improving the explanatory power of the research model while maintaining the order of importance of independent variables. .

ANN Based Direct Modeling of Permanent Magnet DC Tachogenerator Sensor

Ashu Gautam,Nidhi Sharma 보안공학연구지원센터 2015 International Journal of Hybrid Information Techno Vol.8 No.10

Non-linearity coupled by all of the sensors and transducers gives set to either difficulties for direct digital readout, on-chip interface, testing, calibration and control. Also, the performance of a transducer is affected adversely by variations in working environments over them. Under the sovereignty of ANN based transducer modeling, the use of single layer ANN has been proposed in two separate studies with quite affluent results. The first existing model is based on the architecture of an adaptive linear (ADALINE) network trained with Widrow-Hoff’s learning algorithm. The other is based on the concept of Functional Link Artificial Neural Network (FLANN) designed on the architecture of a single layer linear ANN trained with Gradient-descent with momentum based learning algorithm. To have an optimal solution, it is proposed to amalgamate the direct model of the transducers using the concept of a Polynomial-ANN trained with BFGS Quasi-Newton Learning algorithm. The proposed Polynomial-ANN oriented transducer model is implemented based on the topology of a single-layer feed-forward back-propagation network. The proposed transducer modeling technique provides an extremely fast convergence speed with increased accuracy for the assessment of static input-output characteristics of the transducers and also for the solution of linearizing the non-linear transducers.