8,932 research outputs found
Identification of multichannel AR models with additive noise: A Frisch scheme approach
No full textA new approach for estimating multichannel AR (M-AR) models from noisy observations is proposed. It relies on the so-called Frisch scheme, whose rationale consists in finding the solution of the identification problem within a locus of solutions compatible with the second order statistics of the noisy data. Once that the locus of solutions has been defined, it is necessary to introduce a suitable selection criterion in order to identify a single solution. The criterion proposed in the paper is based on the comparison of the theoretical statistical properties of the residual of the noisy M-AR model with those computed from the data. The results obtained by means of Monte Carlo simulations show that the proposed algorithm outperforms some existing methods
A three-step identification procedure for ARARX models with additive measurement noise
No full textThis paper concerns the identification of extended ARARX models that consider also an additive white noise affecting the output. This model allows to take into account the presence of both a process disturbance and an additive measurement noise. A three-step identification procedure is described for identifying the extended ARARX model. The first step consists in an iterative bias-compensated least squares algorithm while the subsequent steps are based on simple (non-iterative) least squares equations. Simulation results are included to show the effectiveness of the proposed method
Frisch scheme-based identification of multivariable errors-in-variables models
No full textThis paper describes an identification procedure for minimally parametrized multivariable models in the Errors–in–Variables (EIV) context of the Frisch scheme that considers additive white observation noise on the process inputs and outputs. This procedure relies on the geometric approach described in (Guidorzi and Diversi, 2009) that associates EIV models to directions in the noise space. The proposed procedure has been tested by means of a Monte Carlo simulation that confirms its effectiveness
The Frisch scheme in multivariable errors-in-variables identification
No full textThis paper concerns the identification of multivariable errors-in-variables (EIV) models, i.e. models where all inputs and outputs are assumed as affected by additive errors. The identification of MIMO EIV models introduces challenges not present in SISO and MISO cases. The approach proposed in the paper is based on the extension of the dynamic Frisch scheme to the MIMO case. In particular, the described identification procedure relies on the association of EIV models with directions in the noise space and on the properties of a set of high order YuleâWalker equations. A method for estimating the system structure is also described
Structural health monitoring application of errors-in-variables identification
No full textStructural Health Monitoring denotes a set of
methodologies oriented to the description of the dynamical
behavior of a structure in view of damage detection. These
methodologies have taken advantage from the development
of sensor, modeling and network techniques and constitute,
today, a well established area. One of the most used methods consists in deducing dynamic models from the observations and in comparing these models with reference ones, concerning integrity conditions of the monitored structure. In many cases the excitations can be considered as White noise in the range of frequencies of interest and, in these cases, the structure can be described by means of autoregressive models. When this approximation is not realistic it is necessary to use input/output models that take into account also the characteristics of the excitation. This last case is considered in this paper making reference to the use of Errors–in–Variables (EIV) models and to data collected on a real structure during a small seismic event
Identification of errors-in-variables models with colored output noise
No full textThis paper deals with the problem of identifying linear errors-in-variables (EIV) models corrupted by white noise on the input and colored noise on the output. This allows to take into account the presence of both measurement errors and process disturbances. The proposed approach is based on a nonlinear system of equations whose unkowns are the system parameters and the input noise variance. The obtained set of equations allows mapping the EIV identification problem into a quadratic eigenvalue problem that, in turn, can be mapped into a linear generalized eigenvalue problem. The performance of the proposed approach is illustrated by means of Monte Carlo simulations and compared with those of existing techniques
Residual generation and disturbance de-coupling for a chemical process
No full textThe paper presents some results concerning fault diagnosis for
dynamic processes using dynamic system identification and
disturbance de--coupling techniques. The first step of the
considered approach consists of exploiting input--output
descriptions of the monitored system. In particular, the disturbance
term of that model can be used to take into account unknown inputs
affecting the system. The next step of the scheme leads to define a
set of relations that can be used as residual signals since they are
insensitive to the disturbance term. The proposed fault diagnosis
scheme has been tested on an real industrial chemical process in the
presence of sensor, actuator and component faults. The results and
concluding remarks have been finally reported
Identification of noisy input-output FIR models with colored output noise
No full textThis paper deals with the identification of FIR models corrupted by white input noise and
colored output noise. An identification algorithm that exploits the properties of both the dynamic Frisch
scheme and the high-order Yule-Walker (HOYW) equations is proposed. It is shown how the HOYW
equations allow to define a selection criterion for identifying the input noise variance (and then the FIR
coefficients) within the Frisch locus of solutions. The proposed approach does not require any a priori knowledge about the input and output noise variances. The algorithm performance is assessed by means of Monte Carlo simulations
Identification of residual generators for fault detection of linear dynamic models
No full textClassical model-based fault detection schemes for
linear multivariable systems require the definition of suitable
residual functions. This paper shows the possibility of identifying
residual generators even when the system model is
unknown, by following a black-box approach. The result is
obtained by using canonical input-output polynomial forms
which lead to characterise in a straightforward fashion the basis
of the subspace described by all possible residual generators.
The performance of the proposed identification method is tested
by means of Monte Carlo simulations
Recursive identification of errors-in-variables models with correlated output noise
No full textAbstract: The identification of Errors-in-variables (EIV) models refers to systems where the available
The identification of Errors-in-variables (EIV) models refers to systems where the available measurements of their inputs and outputs are corrupted by additive noise. A large variety of solutions are
available when dealing with this estimation problem, in particular when the corrupting noises are white
processes. However, the number of available solutions decreases when the output noise is assumed as
a colored process, which is a case of great practical interest. On the other hand, many applications
require estimation algorithms to work on-line, tracking a dynamical system behavior for control, signal
processing, or diagnosis. In many cases, they even have to take into account computational constraints.
In this paper, we propose an estimation method that is able to both lay out an algorithm to solve
the identification problem of EIV systems with arbitrarily correlated output nois
- …
