Input Design For System Identification Research Articles

Application-Oriented Input Design in System Identification: Optimal Input Design for Control [Applications of Control

Model-based control design plays a key role in today's industrial practice, and industry demands cuttingedge methods for identifying the necessary models. However, additional tools are needed to handle the increasingly stringent conditions on cost and performance related to identifying the models.

Adaptive Input Design for Identification of Output Error Model with Constrained Output

Optimal input design for system identification is an area of intensive modern research. This paper considers the identification of output error (OE) model, for the case of constrained output variance. The constraint plays a very important role in the process industry, in the reduction of degradation of product quality. In this paper, it is shown, in the form of a theorem, that the optimal input signal, with constrained output, is achieved by a minimum variance controller together with a stochastic reference. The key problem is that the optimal input depends on the system parameters to be identified. In order to overcome this problem, a two-stage adaptive procedure is proposed: obtaining an initial model using PRBS as input signal; application of adaptive minimum variance controller together with the stochastic variable reference, in order to generate input signals for system identification. Theoretical results are illustrated by simulations.

Plant friendly input design for system identification in closed loop

Optimal experiment or input design is the scientific exercise of designing informative excitation signals for the identification of a real-life dynamic system. In the least costly input design framework, the input is designed such that the identification cost is minimized while meeting desired specifications on the quality of the identified model. Identification of real-life processes require that the identification be “plant-friendly”. These are typically imposed as constraints on experiment time, input and output amplitudes or input move sizes. This work focusses on an LMI based plant friendly input design in the least costly framework.

Optimal Plant Friendly Input Design for System Identification

The primary objective in solving optimal input design problems is to obtain maximally informative inputs to be used as perturbation signals in system identification experiments. In plant-friendly identification, the designer has to respect constraints on experiment time, input and output amplitudes or input move sizes. This work focuses on plant friendly input design with constraints on input move size and output power. We present a convex relaxation to the problem of designing an informative input subject to input move size and output power constraints. The problem is finitely parametrized using ideas from Tchebycheff systems and reformulated as a SemiDefinite Programme.

Input design for worst-case system identification with uniformly quantized measurements

Optimal input design for system identification with quantized sensors is tackled in a worst-case setting. Under the assumption of uniformly quantized measurements, the paper addresses the one-step input design problem for static gains. Results are provided for both noise-free and noisy data and are instrumental for designing suboptimal inputs for FIR systems of arbitrary order. The input design problem over a time horizon of length N is also investigated, in order to characterize the time complexity of the considered identification problem.

ROBUST INPUT DESIGN USING SUM OF SQUARES CONSTRAINTS

The objective of this paper is to illustrate how recent advances in convex programming can be used in input design in system identification. It is shown that sum of squares techniques and a recent generalization of the KYP lemma can be used to transform input design problems with frequency dependent bounds on the frequency function to convex optimization problems. Another important problem that can be handled with sum of squares is robust input design where initial uncertainty in the true parameters is accounted for.

MULTI-OBJECTIVE OPTIMIZATION APPROACH TO OPTIMAL INPUT DESIGN FOR AUTOREGRESSIVE MODEL IDENTIFICATION

Optimality criteria of the experimental design in dynamic system identification may sometimes be conflicting for structure determination and parameter estimation steps. This leads the necessity of the tradeoffs between the performances in these two steps. Here, the optimal input design in system identification is investigated as a multi-objective optimization problem. The Pareto-optimal set of inputs is derived and how to use it in system identification is discussed.

Optimal input design using generalized binary sequence

The generalized binary sequence (GBS) of Tulleken offers an attractive alternative in input design for system identification. In terms of time-domain responses, the GBS ranges from a square-wave sequence to a step input as the non-switching probability changes from zero to unity. In terms of frequency responses, the emphasis of input spectrum changes from low frequency to the Nyquist frequency as the non-switching probability varies. In this work, analytical expression for the optimal non-switching probability for the GBS is derived when the specification is given on a single-frequency value. Moreover, for a more realistic case, if the specification is given as a frequency interval, the optimal non-switching probability can also be obtained analytically. A linear example is used to illustrate this GBS-based input design procedure.