• Korean Chemical Engineering Research
• /
• 제45권1호
• /
• pp.103-108
• /
• 2007

HAZOP 기법은 전문가의 경험과 지식을 기초로 브레인스토밍(brainstorming) 방식을 적용함으로써 체계적으로 잠재 위험을 분석, 평가하는 방법으로 검토 시 누락의 가능성을 배제하고 비교적 객관화된 평가서를 작성할 수 있는 장점으로 인하여 널리 이용되고 있다. 그러나 대규모 화학공장에서는 HAZOP 기법을 이용한 위험성 평가의 실시 및 그 결과의 활용이 잘 이루어지고 있으나, 중 소 화학공장들의 경우 기법 적용이 용이하지 못하여 현실적으로 많은 어려움을 겪고 있다. 이에 본 연구에서는 기존에 많이 사용되고 있는 위험성 평가 기법들의 장 단점을 분석 검토하여 중 소 화학공장의 현장에 적용하기 쉬운 위험성 평가 기법을 제시함으로써 사업장 위험 관리 수준의 향상에 기여하고자 하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1999년도 제14차 학술회의논문집
• /
• pp.145-148
• /
• 1999

It is known that Volterra kernel models can represent a wide variety of nonlinear chemical processes. Also, it is necessary for Volterra model identification to excite the process to be identified with a signal having wide range of frequency spectrum and high enough amplitude of input signals. Kashiwagi[4 ∼ 7] has recently shown a method for measuring Volterra kernels up to third order using pseudorandom M-sequence signals. However, in practice, since it is not always possible to apply such input sequences to the actual chemical plants. Even when we can apply such a pseudorandom signal to the process, it takes much time to obtain higher order Volterra kernels. Considering these problems, the authors propose here a new method for practical identification of Volterra kernels by use of approximate open differential equation (ODE) model and simple plant test. Simulation results are shown for verifying the usefulness of our method of identification of nonlinear chemical processes.

• Korean Chemical Engineering Research
• /
• 제46권5호
• /
• pp.949-957
• /
• 2008

공정의 계단응답으로부터 일차시간지연모델을 구하고, 이로부터 PID(Proportional-Integral-Derivative) 제어기를 설계하는 방법은 공정제어 분야에서 널리 사용되는 오랜 것이다. 이를 위한 간단한 그래프에 근거한 방법에서 복잡한 수치적 최소자승법에 근거한 방법까지 다양한 방법들이 제안되어 있는데, 노이즈가 섞여 있는 계단응답에 대응하는 간단한 방법들은 의외로 잘 알려져 있지 못하다. 최근 계단응답 신호의 적분을 이용하는 방법들이 제안되어 있는데 이들을 비교 검토하여 실용적인 방법을 찾고자 한다. 계단응답의 적분 신호는 현 제어 계기에서 쉽게 얻어진다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.430-435
• /
• 1992

Hazard identification is one of the most important task in process design and operation. This work has focused on the development of a knowledge-based expert system for HAZOP (Hazard and Operability) studies which are regarded as one of the most systematic and logical qualitative hazard identification methodologies but which require a multidisciplinary team and demand much time-consuming, repetitious work. The developed system enables design engineers to implement existing checklists and past experiences for safe design. It will increase efficiency of hazard identification and be suitable for educational purposes. This system has a frame-based knowledge structure for equipment failures/process material properties and rule networks for consequence reasoning which uses both forward and backward chaining. To include wide process knowledge, it is open-ended and modular for future expansion. An application to LPG storage and fractionation system shows the efficiency and reliability of the developed system.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.26.2-26
• /
• 2001

It is known that Volterra kernel model can represent a wide variety of nonlinear chemical processes. But almost all Volterra kernel models which appeared in the literature are up to second order, because it was difficult to measure higher order Volterra kernels. Kashiwagi has recently shown a method for measuring Volterra kernels up to third order using pseudorandom M-sequence signals. In this paper, the authors verified the applicability of this method for chemical processes using polymerization reactor simulation. Also, the authors have recently proposed a practical Identification method for chemical processes, which is based on the combination of off-line nonlinear identification and on-line linear identification. This method is also applied to the identification of polymerization reactor, and we obtained ...

• 분석과학
• /
• 제34권3호
• /
• pp.99-114
• /
• 2021

물질확인은 화학물질 등록법에서 수행해야 하는 첫 단계로 올바른 물질확인은 이어지는 유해성 및 위해성 평가를 정확하게 효율적으로 하는데 필수적이다. 이에 톤수별 물질 등록을 모두 마친 EU REACH의 물질확인 절차를 유럽화학물질청 홈페이지 및 관련 지침서와 컨소시움 자료를 참고하여 살펴 보았고 물질확인과 물질의 동질성 확인 그리고 이를 위한 물질확인 프로파일 생성의 중요성을 고찰해보았다. 또한 기기분석에 의한 분석 결과 중 어떤 정보를 어떻게 정리하여 물질확인 프로파일을 생성하는지에 초점을 맞추어 몇 개 주요 유형별 물질을 사례로 제시해보았다. 분석과학은 화학물질 등록법이 정확하고 효율적으로 이행될 수 있도록 첫 단계인 물질확인에서 기여를 할 수 있는데 단, 법에 대한 이해를 기반으로 같은 물질의 등록자들의 분석 결과를 물질확인 프로파일로 최종 정리하여 공동 등록을 하는 등록자들과 관계 당국과 소통하는 것이 필수적이다.

• Transactions on Control, Automation and Systems Engineering
• /
• 제4권2호
• /
• pp.109-113
• /
• 2002

Van de Vusse reactor is known as a highly nonlinear chemical process and has been considered by a number of researchers as a benchmark problem for nonlinear chemical process. Various identification methods for nonlinear system are also verified by applying these methods to Van de Vusse reactor. From the point of view of identification, only the Volterra kernel of second order has been obtained until now. In this paper, the authors show that Volterra kernels of nonlinear Van de Vusse reactor of up to 3rd order are obtained by use of M-sequence correlation method. A pseudo-random M-sequence is applied to Van de Vusse reactor as an input and its output is measured. Taking the crosscorrelation function between the input and the output, we obtain up to 3rd order Volterra kernels, which is the highest order Volterra kernel obtained until now for Van de Vusse reactor. Computer simulations show that when Van de Vusse chemical process is identified by use of up to 3rd order Volterra kernels, a good agreement is observed between the calculated output and the actual output.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.1762-1765
• /
• 1997

Some ill-conditioned processes are very sensitive to small element-wise uncertainties arising in classical element-by-element model identifications. For such processes, accurate identification of simgular values and right singular vectors are more important than theose of the elements themselves. Singular values and right singular vectors can be found by iteraive identification methods which implement the input and output transformations iteratively. Methods based on SVD decomposition, QR decomposition and LU decomposition are proposed and compared with the Kuong and Mac Gregor's method. Convergence proofs are given. These SVD and QR mehtods use normal matrices for the transformations which cannot be calculated analytically in general and so they are hoard to apply to dynamic processes, whereas the LU method used simple analyitc transformations and can be directly applied to dynamic processes.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.1883-1887
• /
• 1991

Identification theory for linear discrete system has been presented by a great many reference, but research works for identification of continuous-time system are less than preceding identification. In fact, a great man), systems for engineering are continuous-time systems, hence, research for identification of continuous-time system has important meaning. This paper offers the following results: 1. Corresponding relations for the parameters of continuous-time model and discrete model may be shown, when single input-output system has general characteristic roots. 2. To do identification of single variable continuity linear system with stability constraints from samples of input-output data, it is necessary to use optimization with stability constraints. 3. Main results of this paper may be explained by a simple example.

• International Journal of Control, Automation, and Systems
• /
• 제5권5호
• /
• pp.508-514
• /
• 2007

Properties and potential applications of the block pulse response circulant matrix (PRCM) and its singular value decomposition (SVD) are investigated in relation to MIMO control and identification. The SVD of the PRCM is found to provide complete directional as well as frequency decomposition of a MIMO system in a real matrix form. Three examples were considered: design of MIMO FIR controller, design of robust reduced-order model predictive controller, and input design for MIMO identification. The examples manifested the effectiveness and usefulness of the PRCM in the design of MIMO control and identification. irculant matrix, SVD, MIMO control, identification.