Neural Network Based Adaptive Robust Control of a Single-Axis Hydraulic Shaking Table

Wen, Jiabao; Zhao, Chengcheng; Shi, Zhan

doi:10.1109/iecon49645.2022.9968641

Cited by 2 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mathematical model of electro-hydraulic actuators has been thoroughly explored in numerous articles, but most focus on displacement trajectory control. The dynamic model adopted in this paper primarily builds upon [14,16], incorporating some modifications.…”

Section: Problem Formulation and Nonlinear Modelmentioning

confidence: 99%

“…The 3D model of the considered system is illustrated in Figure 1. As introduced in [14,16], a typical single-axis shaking table system comprises a table for supporting the test object, a guide rail for sliding support of the table, sensors for gathering information and providing feedback communication, and the hydraulic actuator, which is the main driving force for the shaking table's motion. Our control objective is to ensure that the signal measured by the acceleration sensor fixed on the table (without load) or the acceleration sensor attached to the test object on the table follows the target acceleration trajectory as closely as possible.…”

Section: Problem Formulation and Nonlinear Modelmentioning

confidence: 99%

“…This priority differs from other hydraulic applications like active suspensions [12] or robotic manipulators [13], where displacement precision is paramount. Although displacement and acceleration seem intrinsically connected by Newton's law, this relationship often fails to hold in practice, making much of the existing research on hydraulic actuator displacement control less applicable to shaking tables [14]. Although displacement and velocity are integrals of acceleration, the smoothing impact of double integration can yield similar displacement results from substantially divergent acceleration trajectories [15].…”

Section: Introductionmentioning

confidence: 99%

“…The robust integral of the sign of the error (RISE) controller mitigates discontinuous inputs in hydraulic control [26,27]. Neural network controllers offer a solution to inherent non-linearity and unmodelled dynamics but require extensive parameter tuning [14,16]. Disturbance observers offer a more practical alternative by estimating unmodelled dynamics and disturbances as a lumped variable [28], but may underperform with significant parameter uncertainties [29].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Extended‐state‐observer‐based adaptive robust control of a single‐axis hydraulic shaking table

Wen,

Zhao,

Wang

et al. 2023

IET Control Theory & Appl

Self Cite

View full text Add to dashboard Cite

The shaking table test has been widely recognized as one of the most reliable methods for assessing the dynamic response of structures and systems when exposed to a range of vibrations. Replicating these vibrations, recorded as acceleration signals, necessitates precise controller designs. However, traditional controller designs primarily rely on linear system assumptions, leading to complications in incorporating parametric uncertainty and uncertain non‐linearity. This paper presents a novel approach to the acceleration trajectory control problem of single‐axis hydraulic shaking table systems from a non‐linear perspective. This strategy employs a controller combined with an extended state observer (ESO) and adaptive control, offering a design that more closely reflects real‐world scenarios and requires less parameter adjustment. The ESO estimates the uncertain non‐linearity using the displacement signal, while the parametric uncertainty is tackled through adaptive control. The proposed controller's effectiveness is validated through theoretical proof using Lyapunov analysis. The analysis demonstrates that asymptotic tracking performance is guaranteed when the uncertain non‐linearity is time‐invariant. Moreover, with time‐variant uncertain non‐linearity, the controller can also ensure both prescribed transient tracking performance and final tracking accuracy. Comparative experiment results underscore the superior performance of the proposed controller.

show abstract

Section: Problem Formulation and Nonlinear Modelmentioning

confidence: 99%

Section: Problem Formulation and Nonlinear Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Extended‐state‐observer‐based adaptive robust control of a single‐axis hydraulic shaking table

Wen,

Zhao,

Wang

et al. 2023

IET Control Theory & Appl

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our conference paper [16] presents the analysis and design of an adaptive robust controller based on plain multilayer neural networks. Although the results in [16] achieve considerable enhancements compared to the PID controller, there is still room for improvement. For example, errors may accumulate over time for a highly time‐related application.…”

Section: Introductionmentioning

confidence: 99%

LSTM‐based adaptive robust nonlinear controller design of a single‐axis hydraulic shaking table

Wen

Zhao

Shi

2022

IET Control Theory & Appl

Self Cite

View full text Add to dashboard Cite

The shaking table has been used extensively in the structure test field to verify the structure's performance against various vibrations, for example earthquakes. In order to replicate the vibrations, which are measured by the acceleration signal, the model of the shaking table should be thoroughly constructed to design the controller. However, parametric uncertainty and strong nonlinearity, such as the nonlinear friction, make it an obstacle to obtaining an accurate model. A neural network-based controller, specifically a long short-term memory neural network-based neural network controller, is designed in this paper to address this issue. The nonlinear systems are estimated by the neural network's universal approximation characteristics, and a long short-term memory neural network is utilized to optimize the time-series-related errors. Furthermore, a robust sliding mode controller is utilized to compensate for the residual error of the neural network and other uncertainties. The semi-global asymptotic stability of the controller is proved by Lyapunov analysis. Comparative experimental results indicate the superiority of the proposed controller.

show abstract

Neural Network Based Adaptive Robust Control of a Single-Axis Hydraulic Shaking Table

Cited by 2 publications

References 24 publications

Extended‐state‐observer‐based adaptive robust control of a single‐axis hydraulic shaking table

Extended‐state‐observer‐based adaptive robust control of a single‐axis hydraulic shaking table

LSTM‐based adaptive robust nonlinear controller design of a single‐axis hydraulic shaking table

Contact Info

Product

Resources

About