Experimental validation of the discrete-time MCS adaptive strategy

Bernardo, Mario di; Gaeta, Alessandro di; Montanaro, Umberto; Olm, Josep M.; Santini, Stefania

doi:10.1016/j.conengprac.2012.12.004

Cited by 22 publications

(24 citation statements)

References 45 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This approach for solving the one-delay problem has also been validated experimentally in [8,27] for the discrete-time MCS control of an ETB. The numerical results reported in next subsection confirm again the effectiveness of this choice.…”

Section: Dtmcsi-pp Implementation Detailsmentioning

confidence: 99%

“…In more detail, an unbounded drift of adaptive gains may arise when the plant is affected by unmodeled dynamics or disturbances even though they are matched (see, for example, [2,41] and references therein). Such a behavior is occasionally encountered by MCS practitioners, as for example in two relevant plants in the context of the automotive engineering, i.e., the Electronic Throttle Body and the Common Rail [8,26]. However, neither the MCSI algorithm nor the method proposed in [34] solve the gain locking problem systematically, and only slower gain drifting can be achieved.…”

Section: Introductionmentioning

confidence: 98%

“…For all those reasons, the ETB is an excellent device to test the adaptive law presented in Section 2. The reader is referred to [8] for further details on the ETB.…”

Section: Dtmcsi-pp Control Of An Automotive Actuatormentioning

confidence: 99%

“…In accordance with [11,8,27], in order to not complicate the control architecture with the use of observer [42] or to introduce noise and delays in the control system with the implementation of a derivative filter, here we implement the DTMCSI-PP algorithm using only position measurements. In so doing, we provide an additional evidence of the robustness and ease of implementation of MCS-based control strategies.…”

Section: Dtmcsi-pp Implementation Detailsmentioning

confidence: 99%

“…A Discrete-Time counterpart of the MCS (DTMCS), including a formal stability proof, was proposed in [12], experimentally validated in [8], and extended to piecewise linear systems in [13]. Recently, discrete-time MCS algorithms including integral action (DTMCSI) and integral plus switching action have been proposed in [26,27], respectively.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Discrete-time integral MRAC with minimal controller synthesis and parameter projection

Montanaro

Olm

2015

Journal of the Franklin Institute

Self Cite

View full text Add to dashboard Cite

Model reference adaptive controllers with Minimal Control Synthesis are effective control algorithms to guarantee asymptotic convergence of the tracking error to zero not only for disturbance-free uncertain linear systems, but also for highly nonlinear plants with unknown parameters, unmodeled dynamics and subject to perturbations. However, an apparent drift in adaptive gains may occasionally arise, which can eventually lead to closed-loop instability. In this article, we address this key issue for discrete-time systems under L 2 disturbances using a parameter projection algorithm. A consistent proof of stability of all the closed-loop signals is provided, while tracking error is shown to asymptotically converge to zero. We also show the applicability of the adaptive algorithm for digitally controlled continuous-time plants. The proposed algorithm is numerically validated taking into account a discrete-time LTI system subject to parameter uncertainty, parameter variations and L 2 disturbances. Finally, as a possible engineering application of this novel adaptive strategy, the control of a highly nonlinear electromechanical actuator is considered.

show abstract

Section: Dtmcsi-pp Implementation Detailsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

“…For all those reasons, the ETB is an excellent device to test the adaptive law presented in Section 2. The reader is referred to [8] for further details on the ETB.…”

Section: Dtmcsi-pp Control Of An Automotive Actuatormentioning

confidence: 99%

Section: Dtmcsi-pp Implementation Detailsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Discrete-time integral MRAC with minimal controller synthesis and parameter projection

Montanaro

Olm

2015

Journal of the Franklin Institute

Self Cite

View full text Add to dashboard Cite

show abstract

Modelling and cycle‐by‐cycle control of an electromechanical engine valve actuator with impact speed limiter based on hydraulic damper

di Gaeta,

Montanaro,

Giglio

2024

Asian Journal of Control

View full text Add to dashboard Cite

Variable valve actuation (VVA) systems based on electromechanical valve actuators (EMVAs) enable a flexible operation and management of internal combustion engines that results in a reduction of pollutant emission and fuel consumption while improving engine power and efficiency. However, to achieve the benefits promised by EMVAs, it is of utmost importance to guarantee the catch of the valve by the system magnets with an acceptable impact speed of the valve at the end‐strokes. Valve catching failures can severely damage engine valves while high impact speeds induce mechanical wear and unacceptable noise. The soft landing control of the valve at the end‐strokes is challenging because of system nonlinearities, parameter uncertainties, and external disturbances. The complexity of the control design increases further when industrial cost‐effective solutions are sought. This paper presents a novel solution for reducing the impact speed of the valve without the use of costly position valve sensors. The proposed solution leverages on a small hydraulic damper (HD) embedded into the EMVA and a feedback cycle‐by‐cycle controller. The passive element reduces the impact speed while the controller adjusts at every engine cycle the EMVA coil currents based on pressure peaks detected in the HD to guarantee valve catching while steering the impact speed toward its minimum. The design of the controller exploits a physics‐based model that maps the pressure peak in the HD onto the impact speed and a gradient descent algorithm is adopted searching the minimum. The closed‐loop permanence is assessed in simulation by using an experimentally validated EMVA model and considering detailed dynamics of the HD. Numerical results show the ability of the proposed control architecture to reduce the impact speed and its robustness to parameters variations (oil temperature) and external disturbances (gas pressure forces).

show abstract

Cycle‐by‐Cycle Adaptive Force Compensation for the Soft‐Landing Control of an Electro‐Mechanical Engine Valve Actuator

Gaeta¹,

Velasco²,

Montanaro³

2014

Asian Journal of Control

Self Cite

View full text Add to dashboard Cite

Electro-mechanical valve actuators (EMVA) are a solution for implementing variable valve actuation in internal combustion engines. Their use can increase engine power, reduce fuel consumption and pollutant emissions, while significantly improving engine efficiency. The control of this actuator is a complex task since non-smooth nonlinearities, parameter variations and external forces strongly affect plant dynamics. In addition, the impact of the valve at its end-strokes translates into mechanical wear and unacceptable noise, and in the worst case the electromagnet may also fail to catch the valve, causing system failure. The design of effective control strategies to ensure valve capture with low impact velocities is therefore essential for the correct functioning of such a mechatronic device. In this paper, the control problem of reducing the impact velocity at "landing" known in the literature as soft landing control, is tackled via novel cycle-by-cycle adaptive force compensation control algorithms. Two schemes are presented: a discrete adaptive proportional integral controller to regulate landing velocity to a preassigned set-point, and a gradient descent method based controller to automatically achieve the minimum admissible impact velocity. The effectiveness of both methods in limiting landing velocities is shown numerically using a high predictive simulator of the EMVA system, when considering unknown varying environmental conditions, such as internal friction and external gas pressure forces.

show abstract

Experimental validation of the discrete-time MCS adaptive strategy

Cited by 22 publications

References 45 publications

Discrete-time integral MRAC with minimal controller synthesis and parameter projection

Discrete-time integral MRAC with minimal controller synthesis and parameter projection

Modelling and cycle‐by‐cycle control of an electromechanical engine valve actuator with impact speed limiter based on hydraulic damper

Cycle‐by‐Cycle Adaptive Force Compensation for the Soft‐Landing Control of an Electro‐Mechanical Engine Valve Actuator

Contact Info

Product

Resources

About