Robust identification of continuous-time low-order models using moments of a single rectangular pulse response

Kim, Young Chol; Jin, Lihua

doi:10.1016/j.jprocont.2013.03.002

Cited by 8 publications

(11 citation statements)

References 16 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As selected by designer, one of the FOTD and SOTD models is identified by means of the closed form formula using only moments of the acquired output data. Only if we calculate a few moments from the pulse response data, the parameters of the model can be determined by using the recent result [14]. In particular, it is pointed up that this identification method is very suitable for autotuning algorithm because the testing input is very simple and can be easily applied to any actual systems, and moreover it takes very short period of time compared with a step input test.…”

Section: Architecture Of the Data Based Design Methodsmentioning

confidence: 99%

“…In particular, it is pointed up that this identification method is very suitable for autotuning algorithm because the testing input is very simple and can be easily applied to any actual systems, and moreover it takes very short period of time compared with a step input test. In spite of a simple testing input, it was shown in [14] that this identification yields high accuracy and is very robust against noise and order mismatching as well.…”

Section: Architecture Of the Data Based Design Methodsmentioning

confidence: 99%

“…Then Moments of the impulse response of a LTI process model can be represented in terms of moments of its input and output in a recursive manner [14].…”

Section: A Novel Methods For Identification Of Fotd and Sotd Models Frmentioning

confidence: 99%

“…The identification method in Section 2.3 has been originally developed for this DBLCD. After extending this approach to different models, all the details including these cases have been reported in [14]. The DBLCD algorithm begins with acquiring a pulse response data and provides the outcomes although we are not required to fulfil any processes for identification and synthesis directly.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Data Based Lower-Order Controller Design: Moment Matching Approach

Kim¹,

Jin²

2012

The Transactions of The Korean Institute of Electrical Engineer

Self Cite

View full text Add to dashboard Cite

-This paper presents a data based low-order controller design algorithm for a linear time-invariant process with a time delay. The algorithm is composed by combining an identification step based on open loop pulse test with a low-order controller design step to obtain the entire set of controllers achieving multiple performance specifications. The initial information necessary for this algorithm are merely the width and amplitude of a rectangular pulse, a controller of four types (PI, PD, PID, first-order), and design objectives. Various parametric approaches that have been developed are merged in the controller design algorithm. The resulting controller set satisfying the design objectives are displayed on the 2D and 3D graphics and thus it is very easy for us to pick a controller inside the admissible set because we can check the corresponding closed-loop performances visually.

show abstract

Section: Architecture Of the Data Based Design Methodsmentioning

confidence: 99%

Section: Architecture Of the Data Based Design Methodsmentioning

confidence: 99%

“…Then Moments of the impulse response of a LTI process model can be represented in terms of moments of its input and output in a recursive manner [14].…”

Section: A Novel Methods For Identification Of Fotd and Sotd Models Frmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Data Based Lower-Order Controller Design: Moment Matching Approach

Kim¹,

Jin²

2012

The Transactions of The Korean Institute of Electrical Engineer

Self Cite

View full text Add to dashboard Cite

show abstract

“…실험을 위해 가감속 페달에 구동장 치와 속도제어기를 갖는 실제 자율주행차량을 이용하며, 가감속 시 바퀴에서의 회전력을 측정하기 위해 휠하중측정장치(wheel force transducer: WFT)을 이용한다. 모델의 파라미터 식별 (identification)은 Kim과 Jin의 펄스응답 기반 모멘트 정합기법 [9] …”

unclassified

Experimental Modeling of Acceleration and Brake Systems for Autonomous Vehicle

Lee¹,

Kim²

2016

The Transactions of The Korean Institute of Electrical Engineer

Self Cite

View full text Add to dashboard Cite

-For the acceleration and brake systems of an autonomous vehicle, the dynamic models from acceleration (brake) pedal input to driving(braking) torque at the vehicle wheel are represented by a set of linear transfer functions in this paper. We present an experimental method that can identify these models using a single rectangular pulse response data. Various magnitude of inputs with different running speeds are applied to experimental tests. All the identified models are demonstrated by the measured data. Both acceleration and brake models have been also validated by comparing the velocity of a full vehicle model associated with the proposed models with the measured vehicle velocity.

show abstract

Identification from step response – The integral equation approach

Ahmed

2016

Can J Chem Eng

View full text Add to dashboard Cite

An overview of identification of continuous‐time models from step responses using the integral equation approach is presented. Both open loop and closed loop identification as well as identification of multiple‐input‐multiple‐output (MIMO) models are considered. Solutions to practical implementation problems are provided and methods for identification with transient initial conditions using raw data as well as estimation algorithms in the presence of disturbances are outlined. The methodologies are presented in a simplified way using the example of a first order model; however, the algorithms are applicable for models with higher orders. Solution techniques for the estimation equations are also discussed. Identification results under different experimental conditions and data quality are presented to demonstrate the performance of the algorithms. A number of experimental and simulation examples are presented to demonstrate the applicability of the approach.

show abstract

Robust identification of continuous-time low-order models using moments of a single rectangular pulse response

Cited by 8 publications

References 16 publications

Data Based Lower-Order Controller Design: Moment Matching Approach

Data Based Lower-Order Controller Design: Moment Matching Approach

Experimental Modeling of Acceleration and Brake Systems for Autonomous Vehicle

Identification from step response – The integral equation approach

Contact Info

Product

Resources

About