Developing autonomous hydraulic excavators is constrained by limited access to physical machines and the high cost of real-world experimentation. This paper proposes a simulation-to-real framework for learning a system-level digital surrogate using Long Short-Term Memory (LSTM) networks. Instead of modeling internal dynamics, the excavator is treated as an input-output operator, and the surrogate is trained to reproduce its closed-loop behavior under identical control inputs. The approach is first validated in a MuJoCo simulation environment and then transferred to a real excavator. To address measurement inconsistencies in real-world data, a consistency-aware state estimation method based on adaptive Kalman filtering is introduced. Experimental results demonstrate that the learned surrogate achieves high fidelity in both angular velocity and long-horizon trajectory reproduction under closed-loop autoregressive evaluation. These results confirm that the proposed model can serve as a drop-in surrogate for both simulation and physical systems, enabling scalable and efficient development of excavation automation algorithms.
: The distribution of hydraulic fractures and production control strategies have significant influences on the fluid flow and production performance of low-permeability waterflooding reservoirs. Traditional approaches typically focus solely on fracture parameters while overlooking production control. To address this limitation, this work proposes a joint optimization framework that simultaneously integrates hydraulic fracturing design and production control. However, the joint optimization of hydraulic fracture distribution and production control requires a large amount of reservoir numerical simulation. To solve this problem, a novel adaptive multi-surrogate-assisted differential evolution (AMSADE) algorithm is developed. The AMSADE algorithm utilizes a surrogate model pool comprising radial basis functions, polynomial response surfaces, and deep neural networks. Additionally, an embedded discrete fracture model (EDFM) is adopted for simulation of fractured reservoir flow and optimization evaluation. The proposed method was applied to a two-dimensional waterflooding reservoir model. The results demonstrate that the algorithm converges rapidly, requiring only 200 numerical simulations to achieve optimal performance. Compared with the classical differential evolution algorithm, the net present value was improved by 17.5%. Overall, the proposed joint optimization framework based on the AMSADE algorithm successfully and simultaneously determines the optimal fracturing and production control parameters.
Xiaopeng Ma, Bin Zhang, Jinsheng Zhao et al.· Energy Engineering· 0 citations
This paper presents an applied study of data-driven Model Predictive Control (MPC) based on the Koopman operator framework for a three-tank hydraulic benchmark. The central contribution is a compact, physics-informed lifting strategy: observable functions are chosen directly from Torricelli’s law governing turbulent orifice flow, yielding a seven-dimensional Extended Dynamic Mode Decomposition (EDMD) model that captures the dominant nonlinearities with fewer basis functions than generic dictionaries. The resulting Koopman-MPC replaces the nonconvex optimization of nonlinear MPC (NMPC) with a convex quadratic program, achieving comparable tracking accuracy (0.65 cm vs. 0.64 cm mean error) while reducing the average per-step computation time by 36.7× (50 ms vs. 1807 ms in Python/SciPy). A Moving Horizon Estimator (MHE) operating on the full nonlinear model reconstructs the unmeasured tank level with accuracy comparable to the 2 mm measurement noise floor. These results provide a quantitative benchmark for Koopman-based predictive control on a nonlinear hydraulic system with bidirectional inter-tank coupling and regime-dependent flow transitions.
Wilder Hernandez Manosalva, H. Ramirez-Murillo, D. Tellez-Castro· International Conference on...· 0 citations
The release of stores from internal carriage configurations in modern aircraft requires precise modeling due to the highly unsteady flow dynamics involved. Traditional methods for predicting store trajectories are often computationally intensive, making real-time predictions challenging. In this study, we develop a data-driven reduced-order model (ROM) capable of accurately predicting the real-time trajectory of a store released from an internal cavity. Using data generated from two-dimensional inviscid computational fluid dynamics (CFD) simulations, we employ dynamic-time-warping-based clustering to group similar cases and proper orthogonal decomposition (POD) for mode reduction. A support-vector-machine-based regression model is then used to predict the POD components for new cases. The proposed ROM successfully predicts the [Formula: see text]-position, [Formula: see text]-position, and orientation of the store with moderate to satisfactory accuracy when compared to CFD simulations. However, limitations in predicting higher-order POD modes, particularly for the [Formula: see text]-position, indicate areas for future improvement. Despite these challenges, the proposed ROM shows significant promise for efficiently predicting store trajectories, making it a viable solution for real-time analysis in defense and aerospace applications.
Arpan Das, Errol Hale, Michael Candon et al.· Journal of Aircraft· 0 citations
DRL is highlighted as a promising data-driven strategy for robust and adaptive control of nonlinear structural systems under partial observability by addressing a critical limitation of passive systems and accelerates the decay of residual vibrations.
Takehiko Asai· Journal of Structural Engine...· 0 citations
Bioretention systems are widely implemented to mitigate urban runoff; however, conventional passive designs lack the ability to adapt discharge behavior to evolving rainfall conditions, often resulting in premature drainage and inefficient utilization of available storage. This study develops and evaluates a forecast-informed, rule-based control framework for optimizing the long-term hydraulic performance of bioretention systems. The proposed approach integrates continuous valve actuation with a kernel-based rainfall forecast indicator embedded directly within a process-based hydraulic model implemented in OpenHydroQual. Control parameters governing forecast interpretation and valve responsiveness are optimized offline using a genetic algorithm over multiyear simulations. System performance is evaluated using three complementary metrics: exceedance flow at a probability level of
p
=
0.001
, total discharged volume, and a residence-time objective, defined as the outlet load of a Laplace-transformed age variable. Simulation-based evaluation using a field-calibrated model of a dual bioretention system demonstrated that exceedance-only optimization reduces extreme discharge from
1.99
×
10
2
to
2.86
×
10
1
m
3
/
day
(approximately 86%) relative to passive operation. Multiobjective optimization revealed a well-defined Pareto trade-off between exceedance suppression and residence-time enhancement. Increasing the residence-time weight yielded a 52% reduction in the transformed-age load while increasing exceedance by less than 20% relative to the exceedance-only solution. Notably, moderate weighting achieved approximately 40% improvement in residence time with less than 3% increase in exceedance, indicating a region of hydraulically efficient operating conditions. Results demonstrate that forecast-informed rule-based control can substantially improve both extreme-flow mitigation and storage persistence without requiring online optimization or computationally intensive model predictive control. The proposed framework provides a transparent, computationally lightweight, and physically interpretable approach for enhancing the operational performance of existing bioretention infrastructure under long-term, variable hydrologic conditions.
Fadi Gabbani, S. Hamidi, A. Massoudieh· Journal of Sustainable Water...· 0 citations
To address cleaning operations in high-interference power cooling ponds, this paper proposes a physics-aware Markov decision process (PA-MDP) framework. This framework is designed to mitigate trajectory deviations caused by the decoupling of decision-making logic and physical dynamics in dynamically constrained underwater environments. The core innovation lies in the explicit coupling of computational fluid dynamics (CFD) prior fields with stochastic differential equation integration, enabling the quantitative physical characterization of robot motion uncertainty. On this basis, a multi-objective reward function—balancing cleaning coverage gains and nonlinear flow field energy costs—is reconstructed to induce physically consistent optimal operational paths. To handle non-stationary disturbances, an asynchronous replanning mechanism based on Mahalanobis distance monitoring is introduced. Furthermore, within the stochastic model predictive control framework, the mean-square ultimate boundedness of the closed-loop system is proved using Lyapunov theory. Simulation and prototype experiments demonstrate that, compared to NMPC and DOB-ALC algorithms, PA-MDP reduces steady-state error by approximately 33% under strong disturbances and shortens recovery time by 44% following sudden disruptions. The proposed method achieves a superior performance balance between dynamic robustness and energy efficiency optimization. These findings provide an integrated solution characterized by theoretical rigor and engineering resilience for autonomous robotic operation and maintenance in complex, constrained aquatic environments.
Jingwei Hu, Wei Jiang, Yu Yan et al.· Transactions of the Institut...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
