Deep Model Predictive Control with Online Learning for Complex Physical Systems

Abstract

The control of complex systems is of critical importance in many branches of science, engineering, and industry. Controlling an unsteady fluid flow is particularly important, as flow control is a key enabler for technologies in energy (e.g., wind, tidal, and combustion), transportation (e.g., planes, trains, and automobiles), security (e.g., tracking airborne contamination), and health (e.g., artificial hearts and artificial respiration). However, the high-dimensional, nonlinear, and multi-scale dynamics make real-time feedback control infeasible. Fortunately, these high-dimensional systems exhibit dominant, low-dimensional patterns of activity that can be exploited for effective control in the sense that knowledge of the entire state of a system is not required. Advances in machine learning have the potential to revolutionize flow control given its ability to extract principled, low-rank feature spaces characterizing such complex systems. We present a novel deep learning model predictive control (DeepMPC) framework that exploits low-rank features of the flow in order to achieve considerable improvements to control performance. Instead of predicting the entire fluid state, we use a recurrent neural network (RNN) to accurately predict the control relevant quantities of the system. The RNN is then embedded into a MPC framework to construct a feedback loop, and incoming sensor data is used to perform online updates to improve prediction accuracy. The results are validated using varying fluid flow examples of increasing complexity.

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• TensorFlow: A system for large-scale machine learning
• Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control
• Chaos as an Intermittently Forced Linear System
• Artificial Neural Networks trained through Deep Reinforcement Learning discover control strategies for active flow control
• Data-Driven Sparse Sensor Placement for Reconstruction
• Koopman operator-based model reduction for switched-system control of PDEs
• Deep Learning and Model Predictive Control for Self-Tuning Mode-Locked Lasers
• Deep Dynamical Modeling and Control of Unsteady Fluid Flows
• Low-order model for successive bifurcations of the fluidic pinball
• Cluster-based feedback control of turbulent post-stall separated flows
• Neural-inspired sensors enable sparse, efficient classification of spatiotemporal data
• Jet mixing optimization using machine learning control
• Multiobjective Optimal Control Methods for Fluid Flow Using Reduced Order Modeling
• Controlling nonlinear PDEs using low-dimensional bilinear approximations obtained from data

Cited by in corpus (16)

• Super-Resolution Analysis via Machine Learning: A Survey for Fluid Flows
• Operator learning for predicting multiscale bubble growth dynamics
• Data-Driven Model Predictive Control using Interpolated Koopman Generators
• Vehicular Applications of Koopman Operator Theory -- A Survey
• BubbleNet: Inferring micro-bubble dynamics with semi-physics-informed deep learning
• Bio-inspired variable-stiffness flaps for hybrid flow control, tuned via reinforcement learning
• Machine-learned control-oriented flow estimation for multiactuator multi-sensor systems exemplified for the fluidic pinball
• State space models for building control: how deep should you go?
• Model predictive control for robust quantum state preparation
• Deep Model Predictive Control with Stability Guarantees
• Real-time optimal control of high-dimensional parametrized systems by deep learning-based reduced order models
• Latent feedback control of distributed systems in multiple scenarios through deep learning-based reduced order models
• On the Universal Transformation of Data-Driven Models to Control Systems
• Self-tuning model predictive control for wake flows
• Godunov Loss Functions for Modelling of Hyperbolic Conservation Laws
• Pruning deep neural networks generates a sparse, bio-inspired nonlinear controller for insect flight