Abstract:This article is concerned with the parameter identification of output-error bilinear-parameter models with colored noises from measurement data. An auxiliary model least squares-based iterative method is developed through the overparameterization model. It examines the difficulty of estimating the overparameterized vector, which usually presents a heavy computational burden in the identification process. To overcome this drawback, a parameter separation technique is introduced and the nonlinear model is reform… Show more
“…The simulation results verify that the hierarchical estimation methods based on the separation have better performance than the modeling methods without separation. The proposed methods in this paper can combine other identification methods 61‐65 to investigate the parameter estimation problems of other linear and nonlinear models and control systems and can be applied to other fields 66‐73 such as signal processing, prediction and engineering application systems 74‐79 and so on.…”
This paper studies the problem of parameter estimation for the multifrequency sine signals, which have multiple characteristic parameters such as the amplitudes, phases, and frequencies. It is noted that the signal output is nonlinear with respect to the phase and frequency parameters while it is linear with respect to the amplitude parameters. This feature inspires us to separate all of the characteristic parameters into a linear parameter set and a nonlinear parameter set, where the linear set is composed of the amplitude parameters and the nonlinear set is composed of the phase parameters and the frequency parameters. After the parameter separation, two identification submodels are constructed for optimizing the linear parameter set and the nonlinear parameter set. Then the nonlinear identification model becomes a linear identification submodel and a nonlinear identification submodel. Therefore, the nonlinear optimization for minimizing the objective function is converted into the combination of the quadratic optimization and nonlinear optimization. Based on the separable identification submodels, a recursive least squares subalgorithm and a recursive gradient subalgorithm are proposed for identifying the linear parameters and nonlinear parameters, respectively. Moreover, an interactive estimation algorithm is designed to remove the related parameter sets between the subalgorithms and a hierarchical identification method is presented by combining the subalgorithms. For the purpose of tracking the time‐varying, a forgetting factor is introduced to improve the convergence speed. The numerical examples are provided to qualify the performance of the proposed method based on some performance measures.
“…The simulation results verify that the hierarchical estimation methods based on the separation have better performance than the modeling methods without separation. The proposed methods in this paper can combine other identification methods 61‐65 to investigate the parameter estimation problems of other linear and nonlinear models and control systems and can be applied to other fields 66‐73 such as signal processing, prediction and engineering application systems 74‐79 and so on.…”
This paper studies the problem of parameter estimation for the multifrequency sine signals, which have multiple characteristic parameters such as the amplitudes, phases, and frequencies. It is noted that the signal output is nonlinear with respect to the phase and frequency parameters while it is linear with respect to the amplitude parameters. This feature inspires us to separate all of the characteristic parameters into a linear parameter set and a nonlinear parameter set, where the linear set is composed of the amplitude parameters and the nonlinear set is composed of the phase parameters and the frequency parameters. After the parameter separation, two identification submodels are constructed for optimizing the linear parameter set and the nonlinear parameter set. Then the nonlinear identification model becomes a linear identification submodel and a nonlinear identification submodel. Therefore, the nonlinear optimization for minimizing the objective function is converted into the combination of the quadratic optimization and nonlinear optimization. Based on the separable identification submodels, a recursive least squares subalgorithm and a recursive gradient subalgorithm are proposed for identifying the linear parameters and nonlinear parameters, respectively. Moreover, an interactive estimation algorithm is designed to remove the related parameter sets between the subalgorithms and a hierarchical identification method is presented by combining the subalgorithms. For the purpose of tracking the time‐varying, a forgetting factor is introduced to improve the convergence speed. The numerical examples are provided to qualify the performance of the proposed method based on some performance measures.
“…In the future work, we will further investigate whether these algorithms can be applied to systems with missing data. The iterative algorithm in this paper is proposed for bilinear stochastic systems but the idea can be extended to other linear and nonlinear stochastic systems with colored noises [66][67][68][69][70][71][72][73][74][75] and can be applied to other literatures [76][77][78][79][80][81][82][83] such as signal modeling, pattern cognition, information processing, and engineering application systems.…”
Because of the product item of the control input and the state vector, the identification of bilinear systems is difficult. This paper considers the combined parameter and state estimation problems of bilinear state-space systems. On the basis of the observability canonical form and the model transformation, an identification model with a linear combination of the system parameters is obtained. Using the hierarchical principle, the identification model is decomposed into three submodels with fewer variables, and a three-stage least squares-based iterative (3S-LSI) algorithm is presented to estimate the system parameters. Furthermore, we derive a state estimator (SE) for estimating the unknown states, and present an SE-3S-LSI algorithm for estimating the unknown parameters and states simultaneously. After that, the least squares-based iterative algorithm is presented as a comparison. By analyzing the estimation results and the calculation amount, these two algorithms can identify the bilinear system effectively but the 3S-LSI algorithm can improve the computational efficiency. The simulation results indicate the effectiveness of the proposed algorithms.
“…3). The proposed approaches in this paper can combine other mathematical tools and strategies [43][44][45][46][47][48] to study the parameter estimation algorithms of other linear stochastic systems and non-linear stochastic systems with different structures and disturbance noises [49][50][51][52][53][54] and can be applied to literatures [55][56][57][58][59][60][61][62][63] such as paper-making systems. Remark 3: For the SO-F-PC-GSG algorithm, the initial values of the parameter vectors to be estimated can be arbitrary.…”
Section: State Observer Based Filtering Partially-coupled Generalisedmentioning
Parameter estimation is an important tool for modelling a real system. This study considers the parameter estimation problem of a multi-input multi-output state-space system with unmeasurable states. By employing the negative gradient search and cutting down redundant parameter estimates, the authors derive a partially-coupled generalised stochastic gradient (PC-GSG) algorithm to estimate the parameters. Considering the unmeasurable states, they present a new state observer which replaces the unknown parameters with their estimates to generate state estimates. By combining the PC-GSG algorithm and the new state observer, they obtain a state observer based partially-coupled generalised stochastic gradient (SO-PC-GSG) algorithm to estimate the parameters and states. In order to eliminate the interference of the coloured noise and strengthen the performance of the SO-PC-GSG algorithm, they propose a state observer based filtering PC-GSG algorithm by means of the data filtering technique. Finally, the effectiveness of the proposed algorithms is investigated in a simulation study.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.