Optimal sensor placement for modal identification using system-realization methods

Kammer, Daniel C.

doi:10.2514/3.21688

Cited by 89 publications

(42 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After output selection, the outputs that are needed to accurately describe the system are known, but a model still has to be derived. As in Kammer (1996), the derived output sets may be well suited for modeling, but not for control where the outputs may have di!erent tasks. This is also recognized by Roh and Park (1997), who consider actuator selection for control and for identi"cation, using di!erent criteria depending on the task.…”

Section: Ezciency Of Estimationmentioning

confidence: 99%

“…This is also recognized by Roh and Park (1997), who consider actuator selection for control and for identi"cation, using di!erent criteria depending on the task. In Kammer (1996), the criterion for selecting sensors for modal identi"cation of #exible structures is maximization of the determinant of the observability Gramian, which is also among the criteria proposed by MuK ller and Weber (1972) for selecting control sensors.…”

Section: Ezciency Of Estimationmentioning

confidence: 99%

See 1 more Smart Citation

A review of methods for input/output selection

2001

View full text Add to dashboard Cite

Control system design involves input/output (IO) selection, that is, decisions on the number, the place, and the type of actuators and sensors. The choice of inputs and outputs a!ects the performance, complexity, and costs of the control system. Due to the combinatorial nature of the selection problem, systematic methods are needed to complement one's intuition, experience, and physical insight. This paper reviews the currently known IO selection methods, which aids the control engineer in picking a suitable method for the problem at hand. The methods are grouped according to the control system property that is addressed and applications are grouped according to the considered control systems. A set of criteria is proposed that a good IO selection method should possess. It is used to assess and compare the methods and it could be used as a guideline for new methods. The state of the art in IO selection is sketched and directions for further research are mentioned.

show abstract

Section: Ezciency Of Estimationmentioning

confidence: 99%

Section: Ezciency Of Estimationmentioning

confidence: 99%

A review of methods for input/output selection

2001

View full text Add to dashboard Cite

show abstract

“…In [14] it is shown that the EfI method produces a final set that is optimal from a system observability perspective. Thus, rejecting sensors that were in fact selected by EfI will result in an less optimal system observability, but this is again a consequence of the candidate set containing duplicate information.…”

Section: Gramian Based Rejectionmentioning

confidence: 99%

Redundant Information Rejection in Sensor Localisation Using System Gramians

Gibanica

Abrahamsson

Kammer

2016

Topics in Modal Analysis &Amp; Testing, Volume 10

Self Cite

View full text Add to dashboard Cite

Sensors placement is important in vibration testing. The method of effective independence, recently extended to account for triaxial sensors, is widely used for this purpose in case a finite element model of the structure is available. In this paper a criteria is added to reject redundant information that usually arises in symmetric structures or finite element models with high candidate sensor density. A sensor placement strategy is proposed in which, initially, the method of effective independence is used to select the best sensors from a candidate set by maximising the Fisher information matrix determinant. Next, the gramians of a balanced realisation is used to compare the information between systems consisting of only previously added sensors to the final set with systems of the previous and candidate sensors. Sensors with redundant information will have negligible effect on the gramian and can be rejected. The method is fast, as gramians of systems with only one or two outputs are evaluated. It is sub-optimal in the sense that all possible sensor placement combinations are not evaluated for optimality. A test case, consisting of a rectangular plate, is presented, but the method has been used on a large scale industrial model with good results.

show abstract

“…To this end, the blades are clamped at the root and are subjected to three point forces; two tip forces in the flapwise and edgewise directions, and a point force in the flapwise direction at the node aimed at introducing strong torsional motion. In the baseline model, the Effective Independence [13] (EI) indices are computed to select a set of 50 output directions in which the mode shapes with natural frequency below 8 Hz are well identifiable. In the beam model, a corresponding set of output nodes is created and connected to the blade pitch axis using mass-less beam elements, shown in Fig.…”

Section: Model Updating Of the Beam Modelmentioning

confidence: 99%

Development of Simplified Models for Wind Turbine Blades with Application to NREL 5 MW Offshore Research Wind Turbine

Vakilzadeh

Johansson

Lindholm³

et al. 2014

Dynamics of Coupled Structures, Volume 1

View full text Add to dashboard Cite

Integration of complex models of wind turbine blades in aeroelastic simulations places an untenable demand on computational resources and, hence, means of speed-up become necessary. This paper considers the process of producing simplified rotor blade models which accurately approximate the dynamics of interest. The novelty, besides applying an efficient model updating procedure to the wind turbine blade, is to challenge the conventional beam element formulation utilized in the majority of aeroelastic codes. First, a 61.5 m blade, previously reported by the National Renewable Energy Laboratory, is selected as a case study and a verified industry-standard three dimensional shell model is developed based on its actual geometry. Next, given the reported spanwise cross sectional properties of the blade, a calibrated beam model is developed, using an efficient model updating process, that shows an excellent agreement to the low frequency dynamics of the baseline model in terms of mode shapes, resonance frequency and frequency response function. The simulation study provides evidence that a beam model cannot capture all the important features found in a large-scale 3D blade. This motivates a departure from conventional beam element formulation and suggests addressing the problem of producing simplified models in the framework of model reduction techniques. A modified modal truncation algorithm is applied to the baseline model to produce a simpler model which accurately approximates its input-output behavior in a given frequency range. It is concluded that besides the computational efficiency of the reduction algorithm, the resulting approximation error is guaranteed to be bounded and the yielded low-order model can, in turn, be served in wind turbine design codes. Keywords Model calibration • Model reduction • Wind turbine blade • Frequency response calibration • Beam modeling IntroductionWind energy as one of the most promising sources of renewable energy is going to provide 7-8 % of the demanded world's electricity power by 2020. However, for wind energy to meet such an impressive rate of growth, future wind turbines must be designed for maximum energy conversion efficiency. In response to this demand, the size of wind turbine blades has dramatically increased within the last few years and it is predicted that the wind turbines installed within the next 10-15 years will have rotor diameter of 180-200 m [1]. To design for future wind turbine blades, development of models will become more and more crucial to performing entire wind turbine simulations by using aeroelastic modeling tools and with the purpose of aerodynamic load estimation, reliability analysis and so forth. While coupling high fidelity large scale three dimensional Finite Element (FE) blade models in such aeroelastic settings results in more accurate prediction of performance, it also poses overwhelming demands on computational resources making the development of simplified engineering models become more and more necessary. For this reason, the pr...

show abstract

Optimal sensor placement for modal identification using system-realization methods

Cited by 89 publications

References 3 publications

A review of methods for input/output selection

A review of methods for input/output selection

Redundant Information Rejection in Sensor Localisation Using System Gramians

Development of Simplified Models for Wind Turbine Blades with Application to NREL 5 MW Offshore Research Wind Turbine

Contact Info

Product

Resources

About