Hybrid Loewner Data Driven Control

Vuillemin, Pierre; Kergus, Pauline; Poussot-Vassal, Charles

doi:10.1016/j.ifacol.2020.12.1574

Cited by 6 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, thanks to the advancements in solver technology and computation power, these methods have gained more attention [24]- [33]. The data used for synthesizing the controller may be in time-domain [24]- [26] or in frequency-domain [27]- [33]. One of the advantages of the latter is that the closed-loop sensitivity functions can be shaped directly in the frequency domain.…”

Section: Drts Hutmentioning

confidence: 99%

Stabilization and Performance Improvement of Power-Hardware-in-the-Loop Test Using a Robust Data-Driven Method

Madani

Karimi

2022

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Power-Hardware-in-the-Loop (PHIL) setups have gained high importance in validation of the performance of newly developed instruments and devices with low risk and implementation cost. However, the interconnection of a power hardware with a simulated model via a feedback loop using measured signals may make the closed-loop system unstable. In this paper, a systematic method is proposed to design a digital filter such that the closed-loop stability of the PHIL setup is guaranteed and its performance is optimized. The design of the filter is formulated as a specific controller synthesis problem that minimizes the deviation in measured feedback current and ensures a certain robustness margin. The design problem is written as a convex optimization and solved efficiently using available solvers. Due to the data-driven characteristic of the proposed method, there is no need for the model of the Hardware Under the Test (HUT) to guarantee stability and performance. Since the design is done in a systematic fashion using advanced control design techniques with a rigorous mathematical guarantee for stability, a high-performance filter can be designed with no need for laborious manual tuning that may lead to low-performance filters. This method is also extended to a multi-scenario case including different combinations of hardware and simulated systems. The PHIL experimental results validate the effectiveness of the proposed method while satisfying the required stability margin and improving the performance significantly compared to conventional filters.

show abstract

Section: Drts Hutmentioning

confidence: 99%

Stabilization and Performance Improvement of Power-Hardware-in-the-Loop Test Using a Robust Data-Driven Method

Madani

Karimi

2022

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

show abstract

“…Note that, in the end, a continuous controller K(s) is obtained and then discretized to obtain the control law K(z). An hybrid version of the L-DDC, allowing to obtain directly a discrete controller for a continuous system, also exists and can be found in [5], but has not been used here.…”

Section: Loewner Data-driven Controlmentioning

confidence: 99%

“…The real-time acquisition and control functions, including the control law K(z), the PWM and Average blocks as schematized on Figure 2, are implemented within a LabView interface. For the considered flow system, three different controllers denoted K f j (z) with the same structure as in (8), sampled at f s2 = 100Hz, corresponding to three different performance objectives M f j (s) as defined in (5), for j = {1, 3, 5}, are constructed. These may be read as,…”

Section: Considered Configurationsmentioning

confidence: 99%

Interpolatory-based data-driven pulsed fluidic actuator control design and experimental validation

Poussot-Vassal,

Kergus,

Kerhervé

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Pulsed fluidic actuators play a central role in the fluid flow experimental control strategy to achieve better performances of aeronautic devices. In this paper, we demonstrate, through an experimental test bench, how the interpolatory-based Loewner Data-Driven Control (L-DDC) framework is an appropriate tool for accurately controlling the outflow velocity of this family of actuators. The contributions of the paper are, first, to emphasize the simplicity and versatility of such a data-driven rationale in a constrained experimental setup, and second, to solve some practical fluid engineers concerns by detailing the complete workflow and key ingredients for successfully implementing a pulsed fluidic actuator controller from the data acquisition to the control implementation and validation stages.

show abstract

“…[14]). Recently, a hybrid version has also been proposed by [19]. The Loewner Framework (LF) is a data-driven model identification and reduction technique that was originally introduced in [16].…”

Section: Control Design Via Data-driven Approximationmentioning

confidence: 99%

On Loewner data-driven control for infinite-dimensional systems

Gosea¹,

Poussot-Vassal²,

Antoulas³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we address extensions of the Loewner Data-Driven Control (L-DDC) methodology. First, this approach is extended by incorporating two alternative approximation methods known as Adaptive-Antoulas-Anderson (AAA) and Vector Fitting (VF). These algorithms also include least squares fitting which provides additional flexibility and enables possible adjustments for control tuning. Secondly, the standard model reference data-driven setting is extended to handle noise affecting the data and uncertainty in the closed-loop objective function. These proposed adaptations yield a more robust data-driven control design.

show abstract

Hybrid Loewner Data Driven Control

Cited by 6 publications

References 7 publications

Stabilization and Performance Improvement of Power-Hardware-in-the-Loop Test Using a Robust Data-Driven Method

Stabilization and Performance Improvement of Power-Hardware-in-the-Loop Test Using a Robust Data-Driven Method

Interpolatory-based data-driven pulsed fluidic actuator control design and experimental validation

On Loewner data-driven control for infinite-dimensional systems

Contact Info

Product

Resources

About