• 한국공간구조학회논문집
• /
• 제21권1호
• /
• pp.79-86
• /
• 2021

Control performance of a smart tuned mass damper (TMD) mainly depends on control algorithms. A lot of control strategies have been proposed for semi-active control devices. Recently, machine learning begins to be applied to development of vibration control algorithm. In this study, a reinforcement learning among machine learning techniques was employed to develop a semi-active control algorithm for a smart TMD. The smart TMD was composed of magnetorheological damper in this study. For this purpose, an 11-story building structure with a smart TMD was selected to construct a reinforcement learning environment. A time history analysis of the example structure subject to earthquake excitation was conducted in the reinforcement learning procedure. Deep Q-network (DQN) among various reinforcement learning algorithms was used to make a learning agent. The command voltage sent to the MR damper is determined by the action produced by the DQN. Parametric studies on hyper-parameters of DQN were performed by numerical simulations. After appropriate training iteration of the DQN model with proper hyper-parameters, the DQN model for control of seismic responses of the example structure with smart TMD was developed. The developed DQN model can effectively control smart TMD to reduce seismic responses of the example structure.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부 B
• /
• pp.648-650
• /
• 1998

This paper proposes a constructive algorithm for generating the Takagi-Sugeno type fuzzy model through the sequential learning from training data set. The proposed algorithm has a two-stage learning scheme that performs both structure and parameter learning simultaneously. The structure learning constructs fuzzy model using two growth criteria to assign new fuzzy rules for given observation data. The parameter learning adjusts the parameters of existing fuzzy rules using the LMS rule. To evaluate the performance of the proposed fuzzy modeling approach, well-known benchmark is used in simulation and compares it with other modeling approaches.

• 수산해양교육연구
• /
• 제6권2호
• /
• pp.198-216
• /
• 1994

The problem of promoting instructional effect using reorganizing the content of textbook is one of the major concerns of many education theorists and teachers. The results of many researches about above problem reveal that reorganizing the content of textbook promotes the ability of recall and problem solving of learners. The content structure of current navigation textbook revealed a categorical structure as its basic framework, though it seems to be a poor one. A categorical structure is known as providing an inferior information processing mechanism for learners than a learning hierarchy content structure is. Furthermore current content structure hasn't given any considerations to navigation in practice, spatial contexts and sequential events of ships from a harbor to another harbor. The learning hierarchy content structure has an advantage of giving learners more systematic and stronger knowledge networks than a categorical structure.

• Transactions on Control, Automation and Systems Engineering
• /
• 제3권3호
• /
• pp.170-175
• /
• 2001

Q-learning is a kind of reinforcement learning where the agent solves the given task based on rewards received from the environment. Most research done in the field of Q-learning has focused on discrete domains, although the environment with which the agent must interact is generally continuous. Thus we need to devise some methods that enable Q-learning to be applicable to the continuous problem domain. In this paper, an extended fuzzy rule is proposed so that it can incorporate Q-learning. The interpolation technique, which is widely used in memory-based learning, is adopted to represent the appropriate Q value for current state and action pair in each extended fuzzy rule. The resulting structure based on the fuzzy inference system has the capability of solving the continuous state about the environment. The effectiveness of the proposed structure is shown through simulation on the cart-pole system.

• 컴퓨터교육학회논문지
• /
• 제15권4호
• /
• pp.61-77
• /
• 2012

본 연구는 이러닝 학습자의 자기조절학습에 영향을 미치는 학업정서와 선행요인들 간의 관계를 규명하고자 하였다. 통제-가치이론 및 선행연구 고찰을 통해 자기조절학습에 영향을 미치는 선행요인으로 지각된 과제가치, 자기효능감, 수업에서 강조되는 목표구조, 학업정서 중 즐거움, 두려움, 지루함을 선정하고 모형을 상정하였다. 상정된 모형은 국내 대학의 이러닝 수강생 535명의 자료를 바탕으로 검증되었다. 분석 결과, 상정된 모형은 타당하였고, 지각된 과제가치, 자기효능감과 수업에서의 성취목표구조는 학습자의 세부 학업정서에 영향을 미치는 것으로 확인되었다. 또한 학업정서 중 즐거움만이 자기조절학습에 유의한 영향을 미쳤으며, 수업성취목표구조에 따라 학업정서와 자기조절학습에 각기 다른 영향력을 나타내었다. 본 연구결과를 바탕으로 시사점을 논의하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 B
• /
• pp.386-389
• /
• 1997

Error-backpropagation has been used in the bulk of Practical applications for neural networks. While an emulator, a multilayered neural network, learns to identify the system's dynamic characteristics. There is, however, no concrete theoretical results about the structure of a plant and the structure of a multilayered neural network and the learning rate. The paper investigates the relation between structure of a plant and a multilayered network and learning rate. Simulation study shows that the plant signal with a short period and a fast sam time is preferable for learning of the network emulator.

• 한국멀티미디어학회논문지
• /
• 제24권4호
• /
• pp.519-527
• /
• 2021

Alzheimer's disease is one of the challenges to tackle in the coming aging era and is attempting to diagnose and predict through various biomarkers. While the application of various deep learning-based technologies as powerful imaging technologies has recently expanded across the medical industry, empirical design is not easy because there are various deep earning neural networks architecture and categorical hyperparameters that rely on problems and data to solve. In this paper, we show the possibility of optimizing a deep learning neural network structure and hyperparameters for Alzheimer's disease classification in amyloid brain images in a representative deep earning neural networks architecture using genetic algorithms. It was observed that the optimal deep learning neural network structure and hyperparameter were chosen as the values of the experiment were converging.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제11권3호
• /
• pp.135-142
• /
• 2011

Unlike using the sequence-based representation for a chromosome in previous genetic algorithms for Bayesian structure learning, we proposed a matrix representation-based genetic algorithm. Since a good chromosome representation helps us to develop efficient genetic operators that maintain a functional link between parents and their offspring, we represent a chromosome as a matrix that is a general and intuitive data structure for a directed acyclic graph(DAG), Bayesian network structure. This matrix-based genetic algorithm enables us to develop genetic operators more efficient for structuring Bayesian network: a probability matrix and a transpose-based mutation operator to inherit a structure with the correct edge direction and enhance the diversity of the offspring. To show the outstanding performance of the proposed method, we analyzed the performance between two well-known genetic algorithms and the proposed method using two Bayesian network scoring measures.

• 전자공학회논문지S
• /
• 제36S권2호
• /
• pp.28-37
• /
• 1999

재 강화 학습 방법을 다수의 제어입력을 가진 시스템에 대한 자율적 제어 기법 습득에 활용하기 위해서 차등책임 적응비평 학습구조를 제안하였다. 재 강화 학습은 여러 단계의 제어동작 끝에 얻어지는 최종 비평값을 활용하여 그 전에 행해졌던 제어 동작을 강화 혹은 약화 학습하는 자율적 학습방법이다. 대표적인 재강화학습 방법은 적응비평학습 구조를 이용하는 방법인데 비평모듈과 동작모듈을 이용하여 외부 비평 값을 최대로 활용함으로써 학습효과를 극대화시키는 방법이다. 이 학습방법에서는 단일한 제어입력을 갖는 시스템으로만 적용이 제한된다는 단점이 있다. 제안한 차등책임 적응비평 학습 구조에서는 비평함수를 제어 입력 인자의 함수로 구축한 다음 제어인자에 대한 차별 화된 비평 값을 부분미분을 통하여 산출함으로써 다수의 제어입력을 가진 시스템의 제어기술 학습이 가능하게 하였다. 제안한 학습제어 구조는 학습속도가 빠른 CMAC 신경회로망을 이용하여 구축하였으며 2개의 제어입력을 갖는 2-D Cart-Pole 시스템과 3 개의 제어입력을 갖는 인간구조 로봇시스템의 앉는 동작의 학습제어 시뮬레이션을 통하여 효용성을 확인하였다.

• International Journal of Advanced Culture Technology
• /
• 제8권1호
• /
• pp.199-206
• /
• 2020

I suggested the design learning environment based on affordance concept approach for supporting and improving learners' behavior and outcome for convergent design education in this study. The design learning space should be applied teaching and learning activity, especially learners' behavior, physical space condition to support the design thinking process. The design learning space needs openness, individuality and connectivity to support the learners' behavioral to immerse, participate, cooperate, understand, think and fulfill the design thinking process. The composition principles of the learning environment for convergent design education supports communication and collaboration among members for independence and interaction. The spaces for design research and teaching needs high privacy while facilitating visual communications through special materials and wall structure design. Also, for connectivity to improve the learners' physical and visual contact, the environment of the classrooms requires flexibility and mobility by providing an open space integrating unit cells for realizing learning purpose. These are provided by formed of an open structure for inducing visual communication and physical contact to involve the design activities and the mutual interchange.