FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2613222845> ?p ?o ?g. }

• W2613222845 abstract "Although Kalman filter (KF) was originally proposed for system control i.e. steering a system as desiredby monitoring the system states, its application for parameter estimation problems is widespread because of theexcellent similarity between these two apparently different problem types in state space description.In standard Kalman filter, system dynamics is described through the dynamics of certain internal variable,termed as states, evolving over time as defined by an assumed process model, while a measurement model mapsthese states to measurements. In some parameter estimation problems, the system is replaced by a state spaceformulation of the dynamic model with parameters appended in the unobserved states and collectively observedthrough the response measurements. Filtering based parameter estimation problems are thus inherently nonlineardue to the required nonlinear mapping of parameters to the corresponding observations.Being a linear estimator, Kalman Filter (KF) cannot be employed for such nonlinear system estimation andalternative filtering algorithms (eg. Particle filter) are therefore generally used. However, being model based,these filters optimally estimate the parameters of a quasi-static model of the real dynamic system. Consequently,any time variation in the system dynamics may completely diverge the estimation yielding a false or infeasiblesolution. By decoupling the estimation of system states and parameters, and applying concurrent filtering strategythat attempts conditional estimation of states based on parameters and vice versa, time varying systems can beestimated.This article attempts to combine KF with Particle filter (PF) and apply them for estimation of states and systemparameters respectively on a system with correlated noise in process and measurement. The idea is to nest abank of linear KFs for state estimation within a PF environment that estimates the parameters. This facilitatesemploying relatively less expensive linear KF for linear state estimation problem while costly PF is employedonly for parameter estimation. Additionally, the proposed algorithm also takes care of those systems for whichsystem and measurement noises are not uncorrelated as it is commonly idealized in standard filtering algorithms.As an example, for mechanical systems under ambient vibration it happens when acceleration response isconsidered as measurement. Thus the process and measurement noise in these system descriptions are obviouslycorrelated. For this, an improved description for the Kalman gain is developed. Further, to enhance the consistencyof particle filtering based parameter estimation involving high dimensional parameter space, a new temporalevolution strategy for the particles is defined. This strategy aims at restricting the solution from diverging (up tothe point of no return) because of an isolated event of infeasible estimation which is very much likely especiallywhen dealing with high dimensional parameter space." @default.
• W2613222845 created "2017-05-19" @default.
• W2613222845 creator A5023903141 @default.
• W2613222845 creator A5025041886 @default.
• W2613222845 creator A5076770415 @default.
• W2613222845 creator A5079817811 @default.
• W2613222845 creator A5083434766 @default.
• W2613222845 date "2017-04-23" @default.
• W2613222845 modified "2023-10-14" @default.
• W2613222845 title "Estimation of time varying system parameters from ambient response using improved Particle-Kalman filter with correlated noise" @default.
• W2613222845 hasPublicationYear "2017" @default.
• W2613222845 type Work @default.
• W2613222845 sameAs 2613222845 @default.
• W2613222845 citedByCount "0" @default.
• W2613222845 crossrefType "proceedings-article" @default.
• W2613222845 hasAuthorship W2613222845A5023903141 @default.
• W2613222845 hasAuthorship W2613222845A5025041886 @default.
• W2613222845 hasAuthorship W2613222845A5076770415 @default.
• W2613222845 hasAuthorship W2613222845A5079817811 @default.
• W2613222845 hasAuthorship W2613222845A5083434766 @default.
• W2613222845 hasConcept C115961682 @default.
• W2613222845 hasConcept C154945302 @default.
• W2613222845 hasConcept C157286648 @default.
• W2613222845 hasConcept C163294075 @default.
• W2613222845 hasConcept C206833254 @default.
• W2613222845 hasConcept C2775924081 @default.
• W2613222845 hasConcept C29265498 @default.
• W2613222845 hasConcept C41008148 @default.
• W2613222845 hasConcept C47446073 @default.
• W2613222845 hasConcept C52421305 @default.
• W2613222845 hasConcept C79334102 @default.
• W2613222845 hasConcept C99498987 @default.
• W2613222845 hasConceptScore W2613222845C115961682 @default.
• W2613222845 hasConceptScore W2613222845C154945302 @default.
• W2613222845 hasConceptScore W2613222845C157286648 @default.
• W2613222845 hasConceptScore W2613222845C163294075 @default.
• W2613222845 hasConceptScore W2613222845C206833254 @default.
• W2613222845 hasConceptScore W2613222845C2775924081 @default.
• W2613222845 hasConceptScore W2613222845C29265498 @default.
• W2613222845 hasConceptScore W2613222845C41008148 @default.
• W2613222845 hasConceptScore W2613222845C47446073 @default.
• W2613222845 hasConceptScore W2613222845C52421305 @default.
• W2613222845 hasConceptScore W2613222845C79334102 @default.
• W2613222845 hasConceptScore W2613222845C99498987 @default.
• W2613222845 hasLocation W26132228451 @default.
• W2613222845 hasOpenAccess W2613222845 @default.
• W2613222845 hasPrimaryLocation W26132228451 @default.
• W2613222845 hasRelatedWork W2005459091 @default.
• W2613222845 hasRelatedWork W2064665624 @default.
• W2613222845 hasRelatedWork W2111676726 @default.
• W2613222845 hasRelatedWork W2126889921 @default.
• W2613222845 hasRelatedWork W2127224101 @default.
• W2613222845 hasRelatedWork W2381377965 @default.
• W2613222845 hasRelatedWork W2505458652 @default.
• W2613222845 hasRelatedWork W2570122939 @default.
• W2613222845 hasRelatedWork W2906127373 @default.
• W2613222845 hasRelatedWork W4310470578 @default.
• W2613222845 isParatext "false" @default.
• W2613222845 isRetracted "false" @default.
• W2613222845 magId "2613222845" @default.
• W2613222845 workType "article" @default.