In this paper, a Kalman Filter type algorithm is introduced for the recursiveestimation of the states of linear systems whose outputs are measured through sensors with multilevel quantized outputs. It is also shown h...
详细信息
ISBN:
(纸本)9789874685933
In this paper, a Kalman Filter type algorithm is introduced for the recursiveestimation of the states of linear systems whose outputs are measured through sensors with multilevel quantized outputs. It is also shown how the algorithm can be adapted for the parameterestimation of linear systems in linear regression form, from quantized outputs. The influence of the quantization step on the performance of the proposed algorithms is analyzed through simulation examples.
A class of non-linear non-linear-in-the-parameters continuous-time systems are described. A least-squares parameterestimation scheme is derived and both non-recursive and recursive in time formulae are given. The cla...
详细信息
A class of non-linear non-linear-in-the-parameters continuous-time systems are described. A least-squares parameterestimation scheme is derived and both non-recursive and recursive in time formulae are given. The class of systems described includes a number of partially-known, or grey box, model structures and thus the method provides parameter identification techniques for such model structures. An example is given to illustrate the method and corresponding simulation results are presented.
The aim of the first part of our research, described in this paper, was to compare daily streamflow estimation techniques and models. A general application software named DebEst was developed for the purpose. The Sain...
详细信息
The aim of the first part of our research, described in this paper, was to compare daily streamflow estimation techniques and models. A general application software named DebEst was developed for the purpose. The Saint-Francois River basin was used as a physical test area because of the availability of several hydrometric stations in this region. All techniques and models used gave good results. However, principal component analysis and multiple regression applied to deterministic models gave better results than ARIMA models. The least square recursive algorithm was more flexible than the other techniques, although discrepancies sometimes appeared because of incorrect weighting of measuring and modelling noise. Results improved significantly when seasonal models were used and when the variation of parameters was taken into account as a function of flow. All techniques described yielded autocorrelated residuals, at least for the first three time lags. The amplitude of the residual autocorrelation function was reduced by seasonal models although it still remained high. In the second part of our research, the Kalman filter technique will be used in conjunction with the methods described above to extract residual information and generate truly independent residuals.
[1] The majority of canopy interception models, simulating the driving processes that control the energy and water exchange between the canopy and the atmosphere, contain parameters that can not be measured directly, ...
详细信息
[1] The majority of canopy interception models, simulating the driving processes that control the energy and water exchange between the canopy and the atmosphere, contain parameters that can not be measured directly, but can only be meaningfully inferred by calibration against a measured record of input-output data. The aim of the present paper is to explore the suitability of two different types of measurements for the identification of parameters in a single-layer forest canopy interception model. The first information source consists of measured throughfall dynamics, whereas the second consists of measured canopy water storage dynamics. The latter measurements were obtained using the attenuation of a microwave signal over a 12.5-m propagation line while scanning vertically through the forest canopy. Results demonstrate that measured throughfall dynamics contain only very limited information for the calibration of a canopy interception model and are particularly inadequate to identify the storage capacity and evaporation rate of the forest canopy. On the contrary, microwave-measured canopy water storage dynamics contain sufficient information to be able to identify the interception model parameters with a high degree of confidence.
The paper presents a computer program for recursive time series analysis of data which may come from MISO or SISO systems with discrete or continuous-time representations. The major novel features are the incorporatio...
详细信息
The paper presents a computer program for recursive time series analysis of data which may come from MISO or SISO systems with discrete or continuous-time representations. The major novel features are the incorporation of powerful model structure identification criteria and the provision to obtain asymptotically efficient estimates when the stochastic disturbances have rational spectral density. Some runs of the central algorithms are displayed to demonstrate the program's visual interactive mode of operation and flexibility.
In this paper the problem of robust real-time identification of linear discrete-tlme multivariable systems is considered. Three methodologically different approaches to the synthesis of such algorithms are presented. ...
详细信息
In this paper the problem of robust real-time identification of linear discrete-tlme multivariable systems is considered. Three methodologically different approaches to the synthesis of such algorithms are presented. They are based on the generalized least-squares criterion, the optimal one-step estimation and the optimization of the stochastic approximation algorithm with respect to its weighting matrix. Properties of the derived algorithms in the presence of approximately normal disturbances are analysed by Monte Carlo simulations. The obtained results indicate the most suitable algorithms for the application in the engineering practice.
A MIMO selftuning controller (MIMOSC) is proposed for multivariable, nonlinear chemical engineering processes which may have distributed character. MTMOSC consists of a Linear time series model combined with a linear ...
详细信息
A MIMO selftuning controller (MIMOSC) is proposed for multivariable, nonlinear chemical engineering processes which may have distributed character. MTMOSC consists of a Linear time series model combined with a linear quadratic optimal control strategy modified to permit on-line solution of the Riccati equation. The model structure is selected based upon a priori chemical engineering process dynamics knowledge and the model parameters are estimated usine a recursive extended least squares method. MIMOSC is investigated experimentally on a fixed-bed chemical reactor. The ability to hancdle load chanye and a tracking problem is investiganted.
First a short overview is given of different methods of adaptive and predictive control, which are promising for application in industry. Not only the methods but also the already commercially available control instru...
详细信息
First a short overview is given of different methods of adaptive and predictive control, which are promising for application in industry. Not only the methods but also the already commercially available control instrumentation is discussed. The main body of this survey consists of a review of applications of adaptive/predictive control in the pulp and paper industry.
A new recursive least squares estimation algorithm is proposed. The recursion is due to the fact that new estimates are computed from the estimates and internal signals computed in the previous sampling period plus in...
详细信息
A new recursive least squares estimation algorithm is proposed. The recursion is due to the fact that new estimates are computed from the estimates and internal signals computed in the previous sampling period plus incoming signals. Robustness is obtained by conditioning the estimates update to occur only in the presence of a minimal signal excitation level. The effects of noise and unmodelled dynamics are addressed by a variable forgetting factor.
暂无评论