Machine learning-based input-augmented Koopman modeling and predictive control of nonlinear processes

Computers & Chemical Engineering, Год журнала: 2024, Номер 191, С. 108854 - 108854

Опубликована: Авг. 24, 2024

Язык: Английский

---

Accelerated MPC: A real-time model predictive control acceleration method based on TSMixer and 2D block stochastic configuration network imitative controller

Process Safety and Environmental Protection, Год журнала: 2024, Номер 208, С. 837 - 852

Опубликована: Июль 15, 2024

Язык: Английский

---

Efficient data-driven predictive control of nonlinear systems: A review and perspectives

Digital Chemical Engineering, Год журнала: 2025, Номер unknown, С. 100219 - 100219

Опубликована: Янв. 1, 2025

Язык: Английский

---

Optimizing vehicle handling through Koopman-based model predictive torque vectoring: An experimental investigation

Control Engineering Practice, Год журнала: 2025, Номер 158, С. 106272 - 106272

Опубликована: Фев. 15, 2025

Язык: Английский

---

Robust model predictive control framework for multi-motor driving cutterhead system with model errors in hard rock TBMs

Proceedings of the Institution of Mechanical Engineers Part I Journal of Systems and Control Engineering, Год журнала: 2025, Номер unknown

Опубликована: Фев. 28, 2025

Tunnel boring machine (TBM) is a safe and effective equipment for excavating tunnels. The advance control of the driving cutterhead system plays an important role in hard rock TBM excavation. This work presents robust model predictive (MPC) optimizing torques motors. First, established with state-space representation subject to constraints additional disturbances. Based on real operating data, parameters are identified by using prediction error method. To address robustness issue, output disturbance constructed. states estimated state feedback design Kalman filter. state, MPC designed presenting compensation strategy considered together. Extensive simulations based given test performance under tracking rejection system.

Язык: Английский

---

Data Enabled Predictive Control for Water Distribution Systems Optimization

Water Resources Research, Год журнала: 2025, Номер 61(4)

Опубликована: Апрель 1, 2025

Abstract Recent developments in control theory, coupled with the growing availability of real‐time data, have paved way for improved data‐driven methodologies. This study explores application Data‐Enabled Predictive Control (DeePC) algorithm to optimize operation water distribution systems (WDS). WDS are characterized by inherent uncertainties and complex nonlinear dynamics. Hence, classic strategies involving physical model‐based or state‐space methods often difficult implement scale. The DeePC method suggests a paradigm shift utilizing approach. technique employs finite set input‐output samples (control settings measured data) learn an unknown system's behavior derive optimal policies, effectively bypassing need explicit mathematical model system. In this study, is applied two applications pressure management chlorine disinfection scheduling, demonstrating superior performance compared standard strategies.

Язык: Английский

---

Min–max tracking model predictive control for linear parameter‐varying systems using polyhedral invariant sets

Asian Journal of Control, Год журнала: 2025, Номер unknown

Опубликована: Апрель 14, 2025

Abstract In this paper, a min–max tracking model predictive control (MPC) method for linear parameter‐varying (LPV) systems using polyhedral invariant sets is proposed. The aims to expand the error state stabilizable domain and improve dynamic performance while handling asymmetric system constraints guaranteeing robust stability under parameter uncertainty, with low computational burden. Firstly, augmented formulation constructed based on original state‐space reference trajectory obtain state. Secondly, optimization problem considering variation formulated Thirdly, sequence of optimal laws offline obtained by solving design nested corresponding sets. These have larger than ellipsoidal An interpolation applied during online control. Finally, simulation results comparative analysis substantiate effectiveness proposed MPC method.

Язык: Английский

---

Auto-tuning strategy for deep Koopman robust model predictive control design based on advanced Metaheuristics

Neurocomputing, Год журнала: 2025, Номер unknown, С. 130197 - 130197

Опубликована: Апрель 1, 2025

Язык: Английский

---

Research on prediction algorithm of effluent quality and development of integrated control system for waste-water treatment

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Июнь 2, 2025

Research is implemented to protect the environment from an epidemic of chemical materials that could render living conditions hazardous. In order efficiently use productivity while maintaining a constant and reliable level waste quality, severe regulations regarding Waste-Water Treatment Control Systems (WWTCS) must be adopted mitigate serious nature water pollution impure performance. Suboptimal treatment efficiency resources are results methods used for WWTCS, which not highly susceptible changing impact features complex biological systems. The present study presented prediction algorithm Integrated System (ICS) address problems conventional methods. This research proposes Deep Learning (DL) quality wastewater employs Quantile Regression-Random Forest (QR-RF) meta-learner when combined with Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU). proposed method has been into practice tested at Asia's Jiangsu Province Metropolitan Plant (WWTP). With Root Mean Absolute Error (RMSE) 4.76 mg/L 24-h horizons (MAE) 0.85 1-h predictions, model outperforms in terms accuracy. ICS superior standard WWTCS by vital error boundary, minimizing energy consumption 17% boosting chemical-based optimization 24%. average removal rate 94.23% Chemical Oxygen Demand (COD) compared 88.76% systems, findings experiments exhibited significant performance improvements.

Язык: Английский

---

Data-driven predictor of control-affine nonlinear dynamics: Finite discrete-time bilinear approximation of koopman operator

Applied Mathematics and Computation, Год журнала: 2024, Номер 487, С. 129068 - 129068

Опубликована: Окт. 12, 2024

Язык: Английский

---