• Journal of Information Processing Systems
• /
• 제16권5호
• /
• pp.1223-1230
• /
• 2020

Recently, extensive studies have been conducted to apply deep learning to reinforcement learning to solve the state-space problem. If the state-space problem was solved, reinforcement learning would become applicable in various fields. For example, users can utilize dance-tutorial systems to learn how to dance by watching and imitating a virtual instructor. The instructor can perform the optimal dance to the music, to which reinforcement learning is applied. In this study, we propose a method of reinforcement learning in which the action space is dynamically adjusted. Because actions that are not performed or are unlikely to be optimal are not learned, and the state space is not allocated, the learning time can be shortened, and the state space can be reduced. In an experiment, the proposed method shows results similar to those of traditional Q-learning even when the state space of the proposed method is reduced to approximately 0.33% of that of Q-learning. Consequently, the proposed method reduces the cost and time required for learning. Traditional Q-learning requires 6 million state spaces for learning 100,000 times. In contrast, the proposed method requires only 20,000 state spaces. A higher winning rate can be achieved in a shorter period of time by retrieving 20,000 state spaces instead of 6 million.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1990년도 추계학술대회 논문집 학회본부
• /
• pp.352-356
• /
• 1990

In this paper proposed a new analysis method that can be controlled DC separately excited motor using DC chopper. An analysis method can be broadly divided the state variables method and the state space averaging method. The state variable method is largely used for analysis method in the time area, but it is complicated analysis of the nonlinear circuit and modeling of the system. Therefore a boundary of the current continuous mode and discontinuous mode can be definited by the state space averaging method. Also this paper proposed a new approximation analysis method using state space averaging method in the discontinuous mode.

• 한국소음진동공학회논문집
• /
• 제14권5호
• /
• pp.424-431
• /
• 2004

An implicit direct-time integration method for obtaining transient responses of general dynamic systems is described. The conventional Newmark method cannot be directly applied to state-space first-order differential equations, which contain no explicit acceleration terms. The method proposed here is the state-space Newmark method that incorporates the average velocity concept, and can be applied to an analysis of general dynamic systems that are expressed by state-space first-order differential equations. It is also readily coded into a program. Stability and accuracy analyses indicate that the method is numerically unconditionally stable like the conventional Newmark method, and has a period error of 2nd-order accuracy for small damping and 4th-order for large damping and an amplitude error of 2nd-order, regardless of damping. In addition, its utility and validity are confirmed by two application examples. The results suggest that the proposed state-space Newmark method based on average velocity be generally applied to the analysis of transient responses of general dynamic systems with a high degree of reliability with respect to stability and accuracy.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2003년도 추계학술대회논문집
• /
• pp.465-470
• /
• 2003

In this study, the state-space Newmark method based on average velocity is presented to analyse the transient dynamic response for general dynamic system. The conventional Newmark method based on average acceleration cannot he directly to the first-order state-space differential equations introducing the state-space vector. To overcome this problem, the time-step integration algorithm, based on average velocity concept, suitable for the first-order state-space differential equations is proposed In results, the proposed method has %he numerical stability and order of accuracy, which is proved analytically, equal to those of the conventional Newmark method based on average acceleration. Also, the formulation for numerical solution is very simple and the calculation time Is nearly equal to that of the conventional Newmark method based on average acceleration in spite of an increase of two times over matrix size. This method will be look forward to applying the general dynamic system to calculate the transient dynamic response.

• Steel and Composite Structures
• /
• 제22권1호
• /
• pp.161-182
• /
• 2016

A state space differential reproducing kernel (DRK) method is developed for the three-dimensional (3D) analysis of functionally graded material (FGM) axisymmetric circular plates with simply-supported and clamped edges. The strong formulation of this 3D elasticity axisymmetric problem is derived on the basis of the Reissner mixed variational theorem (RMVT), which consists of the Euler-Lagrange equations of this problem and its associated boundary conditions. The primary field variables are naturally independent of the circumferential coordinate, then interpolated in the radial coordinate using the early proposed DRK interpolation functions, and finally the state space equations of this problem are obtained, which represent a system of ordinary differential equations in the thickness coordinate. The state space DRK solutions can then be obtained by means of the transfer matrix method. The accuracy and convergence of this method are examined by comparing their solutions with the accurate ones available in the literature.

• Structural Engineering and Mechanics
• /
• 제14권4호
• /
• pp.491-504
• /
• 2002

A state-space method is proposed to analyze the aerodynamically coupled flutter problems of long-span bridges based on the modal coordinates of structure. The theory about complex modes is applied in this paper. The general governing equation of the system is converted into a complex standard characteristic equation in a state space format, which contains only two variables. The proposed method is a single-parameter searching method about reduced velocity, and it need not choose the participating modes beforehand and has no requirement for the form of structure damping matrix. The information about variations of system characteristics with reduced velocity and wind velocity can be provided. The method is able to find automatically the lowest critical flutter velocity and give relative amplitudes, phases and energy ratios of the participating modes in the flutter motion. Moreover, the flutter analysis of Jiangyin Yangtse suspension bridge with 1385 m main span is performed. The proposed method has proved reliable in its methodology and efficient in its use.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.214-218
• /
• 1992

In this paper, we propose a method for obtaining state-space realization form of two-dimensional transfer function matrices (2DTFM). It contains free parameters. And, we perform various consideration about it. Moreover, we present the conditions so that the state-space realization form exists.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1998년도 제13차 학술회의논문집
• /
• pp.244-249
• /
• 1998

In this paper, the hybrid state space self-tuning control technique Is studied within the framework of fuzzy systems and dual-rate sampling control theory. We show that fuzzy modeling techniques can be used to formulate chaotic dynamical systems. Then, we develop the hybrid state space self-tuning fuzzy control techniques with dual-rate sampling for digital control of chaotic systems. An equivalent fast-rate discrete-time state-space model of the continuous-time system is constructed by using fuzzy inference systems. To obtain the continuous-time optimal state feedback gains, the constructed discrete-time fuzzy system is converted into a continuous-time system. The developed optimal continuous-time control law is then convened into an equivalent slow-rate digital control law using the proposed digital redesign method. The proposed technique enables us to systematically and effective]y carry out framework for modeling and control of chaotic systems. The proposed method has been successfully applied for controlling the chaotic trajectories of Chua's circuit.

• Nuclear Engineering and Technology
• /
• 제49권1호
• /
• pp.134-140
• /
• 2017

A well-performed core power control to track load changes is crucial in pressurized water reactor (PWR) nuclear power stations. It is challenging to keep the core power stable at the desired value within acceptable error bands for the safety demands of the PWR due to the sensitivity of nuclear reactors. In this paper, a state-space model predictive control (MPC) method was applied to the control of the core power. The model for core power control was based on mathematical models of the reactor core, the MPC model, and quadratic programming (QP). The mathematical models of the reactor core were based on neutron dynamic models, thermal hydraulic models, and reactivity models. The MPC model was presented in state-space model form, and QP was introduced for optimization solution under system constraints. Simulations of the proposed state-space MPC control system in PWR were designed for control performance analysis, and the simulation results manifest the effectiveness and the good performance of the proposed control method for core power control.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2002년도 추계학술대회논문초록집
• /
• pp.320.1-320
• /
• 2002

In a hard disk drive (HDD) control system, a state-space controller/observer design is popularly adopted fur its advantages such as effective filtering of position and velocity, use of estimation error to handle servo defects, etc. In this report, a systematic method is proposed to accommodate the transport delay in the plant dynamics into the state estimator. (omitted)